Cytokine-Leukotriene Receptor Interactions

Biochemical and pharmacological studies have identified the structure of leukotrienes, the pathways that lead to their synthesis, and the signaling events they trigger when they interact with their cognate receptors. A privileged interaction exists between these lipid mediators and another group of molecules essential for inflammation and immune modulation, namely, cytokines. Whereas leukotrienes can trigger the synthesis and release of selected cytokines in distinct cell populations, many cytokines can affect cellular responsiveness to leukotrienes by modulating leukotriene receptor expression. As we progressively begin to unravel these complex interactions, new areas of cell-cell communication and eventual therapeutic interventions will emerge.


Leukotriene
Cytokine Effect* Target Cells Ref.
Human natural killer (NK) cells can be activated by IL-2. LTB 4 was found to up-regulate the expression of the β chain of the IL-2 receptor, thus rendering the cells more responsive to the stimulatory effects of IL-2 [14]. NK cells pre-exposed to LTB 4 became more cytotoxic to target cells when treated with IL-2 than unexposed NK cells.
Resting monocytes do not express the α chain of the IL-2 receptor complex. LTB 4 was shown to induce the expression of IL-2Rα and to augment the expression of IL-2Rβ in human monocytes [15], thus allowing the cells to express the heterotrimeric, high-affinity IL-2 receptor complex. LTB 4 -pretreated monocytes were thus rendered more sensitive to IL-2 and consequently produced greater amounts of TNFα in response to lower concentrations of IL-2.

Cysteinyl-Leukotrienes
Because cysteinyl-leukotrienes (CysLTs) are major protagonists in the pathophysiology of human asthma, and because neutrophils are involved in the more severe form of asthma, we studied the potential for LTD 4 to induce synthesis of the chemokine IL-8 through activation of the CysLT 1 receptor. We found LTD 4 to induce IL-8 gene expression in monocytic THP-1 cells and human dendritic cells with complete abrogation by selective CysLT 1 antagonists [20]. Human embryonic kidney (HEK)-293 cells stably transfected with CysLT 1 were used to further study the transcriptional regulation of the IL-8 promoter. Stimulation of the cells with graded concentrations of LTD 4 resulted in a time-and concentrationdependent induction of IL-8 transcription and protein synthesis. Use of IL-8 promoter mutants with substitutions in their NF-κB, activator protein (AP)-1, and NF-IL-6 binding elements revealed a requirement for NF-κB and AP-1, but not NF-IL-6, in LTD 4 -induced activation of the IL-8 promoter.
Overexpression of dominant-negative IκB inhibited the IL-8 transactivation induced by LTD 4 . NF-κB DNA binding activity was induced by LTD 4 , as determined by electrophoretic mobility shift assays, and could be supershifted by antibodies against p50 and p65. Supershift assays after LTD 4 stimulation also indicated the formation of a c-Jun/c-Fos complex. Moreover, our results demonstrate that LTD 4 upregulates the expression of c-fos and c-jun at the mRNA level. Our data showed for the first time that LTD 4 , via the CysLT1 receptor, could transcriptionally activate IL-8 production, with involvement of the transcription factors p50, p65, Fos, and Jun. These findings provide mechanistic and potentially therapeutic elements for modulation of the inflammatory component of asthma.
In this context, CysLTs have been reported to induce the production of other chemokines, such as monocyte chemoattractant protein (MCP)-1 in human monocytes/macrophages [21,22] and Regulated upon Activation, Normal T-cell Expressed, and Secreted (RANTES) in a murine model of asthma [23]. In a rat alveolar macrophage line, LTD 4 enhanced mRNA and protein expression of MIP-1α and TNFα induced by LPS [24].
CysLTs have also been shown to induce or augment the expression of TGFβ in eosinophils and airway epithelial cells [25,26]. We have recently confirmed these findings and extended them to show that LTD 4 -induced TGFβ can synergize with fibroblast-growth factor-2 to induce bronchial smooth muscle cell proliferation [27,28]. This further suggests a mechanism for the airway remodeling potential of CysLTs. The latter may also be involved in inflammatory phenomena within the vessel wall. Indeed, in vascular smooth muscle cells, LTD 4 was found to induce IL-1β production [29], suggesting that it may play a role in the pathogenesis of atherosclerosis.
IL-4-primed mast cells were shown to secrete IL-5, TNFα, and MIP-1β when stimulated with either LTC 4 or LTD 4 [30]. Moreover, the CysLT 1 antagonist MK571 significantly attenuated the generation of IL-5 and TNFα by mast cells activated by FcεRI cross-linkage [30], suggesting that endogenous CysLTs also participated in IgE-dependent cytokine production in these cells via the CysLT 1 receptor. The same group also indicated that IL-8 secretion was stimulated by LTC 4 and LTD 4 in IL-4-primed mast cells (although no baseline IL-8 production was shown) [31]. In contrast to the other cytokines, however, IL-8 secretion was reportedly resistant to blockade by MK571, suggesting dependence on the CysLT 2 receptor for LT-induced mast cell production of this cytokine. We have also recently shown that LTC 4 , more than LTD 4 , could induce IL-8 production in endothelial cells, expressing exclusively CysLT 2 [32]. Moreover, in cells stably transfected with CysLT 2 , LTC 4 induced IL-8 promoter transactivation, which was dependent on AP-1 and NF-κB transcription factors and which involved PKC δ and ε signaling, respectively. LTC 4 was shown to induce IL-4 production in cord blood eosinophils[33] and to induce IL-4 release from blood eosinophils through an intracrine mechanism [34]. CysLTs were also found to regulate dendritic cell functions in a murine model of asthma and to modulate the production of IL-10 and IL-5 in the lungs [35]. Conversely, the CysLT 1 antagonist pranlukast was shown to inhibit IL-5 production in rats[36] and guinea pigs [37] in models of antigen-dependent late asthmatic response. Pranlukast was also shown to attenuate allergen-specific TNFα production and NF-κB translocation in monocytes of asthmatic patients [38].
Of note, some effects of CysLT 1 antagonists have been shown to be independent of their receptor antagonist activity, especially at higher concentrations. Thus, they can modulate iNOS function[39], reduce IL-5 production[40], and inhibit NF-κB activation [41].

MODULATION OF LEUKOTRIENE RECEPTOR EXPRESSION BY CYTOKINES
Since the level of expression of receptors for inflammatory mediators may directly affect the degree of responsiveness of cells to these mediators, we and others have studied the potential for modulation of leukotriene receptor expression by biological response modifiers, including cytokines.

BLT 1
Although TNFα was reported to modulate the affinity state of the LTB 4 receptor in human PMN[42], the expression of the high-affinity receptor for LTB 4 (BLT 1 ) has been found to be relatively resistant to regulation. When they characterized the promoter region of the BLT 1 gene, Kato and colleagues [43] found that it contains cis-elements mainly associated with basal expression, such as SP1, which was found to be a major activator of basal transcription of BLT 1 . The promoter region of the BLT 1 gene includes the open reading frame (ORF) of the BLT 2 gene, which encodes a low-affinity receptor for LTB 4 [44]. BLT 1 is predominantly expressed on leukocytes and endothelial cells. Interestingly, the CpG sites of the BLT 1 promoter region are highly methylated in BLT 1 -nonexpressing cells, but not methylated in BLT 1 -expressing cells. Furthermore, methylation of this region in vitro inhibited the promoter activity of the BLT 1 gene. Thus, methylation at CpG sites in the promoter region is important for cell-specific transcription of BLT 1 .
We have shown BLT 1 expression in human neutrophils to be up-regulated by the corticosteroid dexamethasone (DEX)[45] ( Table 2). DEX up-regulated the steady-state levels of BLT 1 mRNA in human neutrophils in a time-and concentration-dependent manner. The effect was dependent on transcriptional activity, whereas BLT 1 mRNA stability was not affected. DEX-induced up-regulation of BLT 1 expression was prevented by pretreatment with the LTB 4 antagonist LY 255283. Moreover, LTB 4 itself up-regulated the expression of BLT 1 mRNA. BLT 1 protein expression on neutrophils exposed to DEX for 24 h was also up-regulated two-to threefold, and DEX-treated as well as LTB 4 -treated cells showed enhanced responsiveness to LTB 4 in terms of intracellular Ca ++ mobilization and chemotaxis. Whereas DEX and LTB 4 alone decreased neutrophil apoptosis by approximately 50%, neutrophils treated with both LTB 4 and DEX showed greater than 90% survival at 24 h. Moreover, BLT 1 antagonists prevented the increased neutrophil survival induced by DEX as well as by LTB 4 . Taken together, our results suggest that DEXinduced up-regulation of BLT 1 expression in neutrophils may be one mechanism through which glucocorticoids can prolong neutrophil survival, namely, by enhancing cell responses to the antiapoptotic effect of LTB 4 .
Pettersson and colleagues [46] reported that proinflammatory mediators, such as IFNγ, TNFα, and LPS, down-regulated BLT 1 expression in monocytes, whereas the anti-inflammatory cytokine, IL-10, and DEX up-regulated BLT 1 expression. The effect of IFNγ appeared to be exerted through a block in transcriptional activity. Receptor down-modulation following IFNγ stimulation resulted in a diminished chemotactic response to LTB 4 . In our hands, however, BLT 1 expression in monocytes, macrophages, or neutrophils was unaffected by any of a number of stimuli, including the bacterial products LPS and formyl-methionyl-leucyl-phenylalanine, or the cytokines IL-1β, TNFα, IL-4, IL-6, IL-10, or GM-CSF (unpublished observations). * The plus (+) sign indicates augmentation of receptor expression in the indicated cells by the corresponding cytokine, whereas the negative (-) or zero (0) sign indicates reduction of expression or no effect, respectively.
On the other hand, endothelial cells, which normally express very little BLT 1 , were recently shown to express BLT 1 in response to LPS, LTB 4 , and IL-1β [47]. Moreover, stem cell factor was shown to inhibit BLT 1 mRNA expression in murine bone marrow-derived mast cells and reduce their migration to LTB 4 [48].

BLT 2
BLT 2 , which binds LTB 4 with a lower affinity, is expressed ubiquitously, in contrast to BLT 1 . Since its open reading frame is located in the promoter region of the BLT 1 gene, it is suggested that there may be shared transcriptional regulation of these two receptors [44]. Very little information is available on the modulation of BLT 2 expression. Treatment of human umbilical vein endothelial cells with LPS leads to a greater than tenfold increase in the levels of BLT 1 mRNA without any significant effects on BLT 2 mRNA[47]. TNFα increases the expression of BLT 2 mRNA approximately six times above basal levels with only a modest increase in BLT 1 mRNA. IL-1β causes variable and parallel increases of both BLT 1 and BLT 2 mRNA. On the other hand, stem cell factor was found to reduce BLT 2 (and BLT 1 ) mRNA expression in murine bone marrow-derived mast cells [48].
In conclusion, although LTB 4 can modulate the production of a number of cytokines and thus appear to play an important role in inflammation and immune defenses, the expression of its receptors is relatively resistant to modulation by cytokines or other biological modifiers. This allows BLT 1 -expressing cells to maintain a consistent responsiveness to LTB 4 with relative independence from environmental influences.
Modulation of CysLT 1 expression was first demonstrated using the cytokine IL-5[52]. We had previously shown that HL-60 cells differentiated toward the eosinophilic lineage (HL-60/eos) developed specific functional LTD 4 receptors[53]. Work was then undertaken to study the potential modulation of CysLT 1 expression in HL-60/eos by IL-5, an important regulator of eosinophil function. IL-5 could rapidly up-regulate CysLT 1 mRNA expression, with consequently enhanced CysLT 1 protein expression and function in HL-60/eos. CysLT 1 mRNA expression was augmented two-to 15-fold following treatment with IL-5. The effect was seen after 2 h, was maximal by 8 h, and maintained at 24 h. Although CysLT 1 mRNA was constitutively expressed in undifferentiated HL-60 cells, its expression was not modulated by IL-5 in the absence of differentiation. Differentiated HL-60/eos cells pretreated with IL-5 for 24 h showed enhanced CysLT 1 expression on the cell surface, as assessed by flow cytometry using a polyclonal anti-CysLT 1 antibody. They also showed enhanced responsiveness to LTD4, but not to LTB 4 or PAF, in terms of Ca ++ mobilization, and augmented chemotactic response to LTD4. Recently, Early and colleagues [54] reported that IL-13 could also augment CysLT 1 expression in eosinophils.
In a subsequent study, we also reported that the Th2 cytokine IL-13 could upregulate CysLT 1 mRNA levels, with consequently enhanced CysLT 1 protein expression and function in human monocytes and monocyte-derived macrophages[55]. CysLT 1 mRNA expression was augmented two-to fivefold following treatment with IL-13 and was due to enhanced transcriptional activity. The effect was observed after 4 h, was maximal by 8 h, and maintained at 24 h. IL-4, but not IFNγ, induced a similar pattern of CysLT 1 up-regulation. Interestingly, IFNγ could prevent the upregulation of CysLT 1 mRNA induced by IL-4 or IL-13.
Monocytes pretreated with IL-13 or IL-4 for 24 h also showed enhanced CysLT 1 protein expression, as assessed by flow cytometry using a polyclonal anti-CysLT 1 antibody[55]. They also showed enhanced responsiveness to LTD4, but not to LTB 4 , in terms of Ca ++ mobilization, as well as augmented chemotactic activity. These studies suggested a possible mechanism by which IL-5, IL-13, and IL-4 could modulate CysLT 1 expression on eosinophils, monocytes, and macrophages, and consequently their responsiveness to LTD 4 , and thus contribute to the pathogenesis of asthma and allergic diseases.
Woszczek and colleagues [22] confirmed our findings with IL-4 and determined the activation of a STAT6 responsive element in the CysLT 1 promoter by IL-4. Interestingly, treating cord blood-derived mast cells with IL-4 did not alter CysLT 1 receptor mRNA or cell surface protein expression [56].
Airway remodeling is a feature of chronic asthma and involves a number of structural changes, including BSMC hyperplasia and hypertrophy. CysLTs have been suggested to play a role in airway remodeling in addition to their numerous other physiopathological effects. In a subsequent study [57], we aimed at characterizing the potential modulation by cytokines of CysLT 1 receptor expression in BSMC and the eventual functional relevance of this modulation. When human BSMC were exposed to TGFβ, IL-13, or IFNγ, their expression of CysLT 1 receptor was significantly augmented in a time-and concentration-dependent manner. The positive effect of IFNγ on BSMC proliferation was also reported by Amrani and colleagues [58]. Interestingly, IL-4 had no significant effect on CysLT 1 receptor expression in BSMC. Moreover, IL-13 and IFNγ, but not TGFβ, were able to increase CysLT 1 mRNA levels. Finally, when BSMC were pretreated with TGFβ or IL-13, but not IFNγ, their responsiveness to LTD 4 was markedly enhanced in terms of BSMC proliferation. Whereas TGFβ, IL-13, or LTD 4 alone had little effect on BSMC proliferation, pre-exposure of the cells to TGFβ or IL-13 for 24 h resulted in a significant increase in proliferation in response to LTD 4 . The enhanced proliferation was totally prevented by pretreating the cytokine-primed BSMC with the selective CysLT 1 receptor antagonist, montelukast. Taken together, these findings indicated a synergy between certain cytokines and CysLTs, mediated by the augmented expression of the CysLT 1 receptor and subsequent LTD 4 -triggered BSMC proliferation. IL-13 was also found to augment CysLT 1 expression in fibroblasts [59]. These findings support a role for CysLTs in the airway remodeling observed in asthmatic patients and may provide a rationale for preventive and therapeutic intervention.
We also found that B lymphocytes expressed CysLT 1 at both the mRNA and protein levels. Moreover, a two-to threefold enhancement of CysLT 1 expression was observed following B-cell exposure to a combination of activating anti-CD40 antibody and IL-4 [60]. The expression of CysLT 1 was also enhanced when B lymphocytes were cocultured with CD154-transfected fibroblasts in the presence of IL-4. Moreover, IL-4 and CD40-activated B lymphocytes showed an increased responsiveness to LTD 4 in terms of cytosolic calcium flux, which was totally prevented by the selective CysLT 1 antagonist montelukast. Stimulation of IL-4 and CD40-activated B lymphocytes with picomolar concentrations of LTD 4 induced mature epsilon transcripts and up-regulated immunoglobulin (Ig)E and IgG production two-to threefold. Expression of CysLT 1 could thus be up-regulated in B lymphocytes following stimulation with CD154 and IL-4, with consequent increased responsiveness of the cells to LTD 4 in terms of Ig production. Early and colleagues[54] also recently reported that IL-4 could enhance CysLT 1 expression in B cells as well as T cells.
Dendritic cells (DCs) acquire, during their maturation, the expression of the chemokine receptor CCR7 and the ability to migrate to lymph nodes in response to CC chemokine ligand 19 (CCL19). This migration is impaired in mice lacking the LTC 4 transporter and restored by addition of exogenous LTC 4 . To define the role of LT in human DC function, we studied the expression and function of the CysLT receptors during DC differentiation from monocytes and subsequent maturation [61]. Maturation of DC with LPS, a classic Toll-like receptor 4 agonist, reduced CysLT 1 expression by 50%, whereas CysLT 2 expression was increased. In contrast, the Toll-like receptor 3 agonist poly inosinic and cytidylic acid (polyI:C) had no effect on CysLT 1 or CysLT 2 receptor expression. Down-regulation of CysLT 1 expression by LPS could not be mimicked by TNFα alone or in combination with IL-1β or IL-6. It was, however, prevented by inhibitors of COX and could be reproduced by a combination of TNFα and prostaglandin E 2 . Immature DCs and DCs matured with polyI:C, but not with LPS, responded to LTD 4 with a robust cytosolic calcium flux, which was prevented by the CysLT 1 antagonist montelukast. LTD 4 induced DC chemotaxis and enhanced DC migration in response to CCL19 in DCs matured with polyI:C, but only weakly in DCs matured with LPS. Our data suggest that human DCs may differentially respond to leukotriene, depending on their maturational stimuli.
Interestingly, LTD 4 priming of immature DC enhanced IL-4 production more than IFNγ production by naïve CD4 + CD45RA + T cells, with a twofold increase at 1 nM LTD 4 . Similarly, LTD 4 priming of poly I:C-matured DC enhanced IL-4 production by naïve T cells fourfold. In contrast, LPS-matured DC were not affected by LTD 4 priming in inducing T-cell production of either IL-4 or IFNγ (unpublished observations).

CysLT 2
CysLT 2 is coexpressed with CysLT 1 on most leukocytes, including neutrophils, eosinophils, basophils, mast cells, monocytes, macrophages, DCs, and B lymphocytes. It is also expressed in the absence of CysLT 1 in a number of cell types, including vascular smooth muscle cells and endothelial cells, heart Purkinje-conducting fiber cells, cardiomyocytes, and adrenal chromaffin cells [62].
CysLT 2 expression was shown to be modestly up-regulated by IL-4 in mast cells [31]. It was also upregulated by IFNγ, but not by IL-1β, IL-4, IL-5, IL-13, or TNFα, in eosinophils from asthmatic patients [63]. However, we reported that CysLT 2 expression levels were not increased in human bronchial smooth muscle cells by IL-4, IL-13, TGFβ, or IFNγ [57]. Recently, Woszczek and colleagues [64] reported that IFNγ could up-regulate CysLT 2 expression in human endothelial cells with enhanced responsiveness of the cells to CysLTs. As with CysLT 1 , CysLT 2 expression appears to be controlled differently in each cell type.
In conclusion, a number of cytokines can modulate the expression of CysLT 1 and/or CysLT 2 in different cell populations. This is in contrast to the receptors for LTB 4 and suggests that responsiveness of various cells to LTC 4 and LTD 4 may be quite dependent on environmental factors and subject to a finer degree of control.
Globally, as reviewed in this paper, LTs and cytokines are involved in a complex network of interactions, which appear to be ligand-, receptor-, and cell-specific. At this time, there are still very limited data available on the mechanisms and signaling pathways involved in the bidirectional modulation of cytokines and LTs. For instance, not all signaling by the G-protein coupled LT receptors is G-protein dependent. Moreover, several orthosteric receptor antagonists can act as inverse agonists and new allosteric ligands may provide precious tools to help understand LT receptor signaling. As we learn more about the interactions between LTs and cytokines, our understanding of this complex network should help us to devise more selective approaches for intervention in disease states where enhancement or reduction of lipid mediator and cytokine effects are sought.