Cysteinyl Leukotrienes and Their Receptors: Cellular Distribution and Function in Immune and Inflammatory Responses¹

Leukotriene (LT)³ C₄, LTD₄, and LTE₄, collectively termed the cysteinyl LTs (cys-LTs), are peptide-conjugated lipids that are prominent products of activated eosinophils, basophils, mast cells (MCs), and macrophages. Originally identified on the basis of their contractile properties for intestinal and bronchial smooth muscle (1, 2, 3, 4, 5), they are now recognized as potent inflammatory mediators that initiate and propagate a diverse array of biologic responses. In the last decade, two receptors for the cys-LTs, termed the type 1 and type 2 cys-LT receptors (CysLT₁ and CysLT₂ receptors, respectively) have been identified, cloned, and characterized; null mouse strains respectively lacking the CysLT₁ receptor and the biosynthetic enzymes involved in cys-LT synthesis have been derived; and clinically efficacious receptor antagonists and inhibitors of cys-LT synthesis have been introduced to treat humans with asthma. These advances confirm the earlier observations regarding the actions of cys-LTs at the level of smooth muscle function in vivo. They have also revealed unanticipated functions for the cys-LTs, acting nonredundantly through the CysLT₁ and CysLT₂ receptors, in both innate and adaptive immune responses, as well as in the effector phase of inflammation, tissue repair, and fibrosis. We will review these findings and discuss their implications for the role of cys-LTs and their receptors in host defense and disease.

The cys-LTs are rapidly generated de novo from cell membrane phospholipid-associated arachidonic acid, which is liberated by cytosolic phospholipase A₂ (cPLA₂) in response to cell activation (6). Arachidonic acid is converted to 5-hydroperoxy-eicosatetraenoic acid (5-HPETE) and subsequently to an unstable intermediate, LTA₄, by the enzyme 5-lipoxygenase (5-LO) (7). 5-LO translocates from its cytosolic or nucleoplasmic location to the perinuclear envelope, where it acts in concert with 5-LO-activating protein (FLAP), which is required for 5-LO to function enzymatically in intact cells (8, 9). In neutrophils, LTA₄ is preferentially hydrolyzed to the dihydroxy leukotriene, LTB₄, by LTA₄ hydrolase (10), whereas eosinophils, basophils, MCs, and macrophages preferentially form LTC₄ through conjugation of LTA₄ to reduced glutathione by LTC₄ synthase (LTC₄S), the terminal enzyme involved in cys-LT synthesis (11, 12, 13) (Fig. 1). LTC₄ is exported to the cell surface via a specific, energy-dependent step (14) that requires multidrug resistance-associated protein 1 (MRP1) (15). It is converted extracellularly to LTD₄ by a γ-glutamyl transpeptidase (γ-GT) (16) or by a more functionally specific enzyme, γ-glutamyl leukotrienase (γ-GL) (17), and then to LTE₄ by a dipeptidase (18). LTE₄ is excreted in the urine without further chemical modification.

LTC₄S is structurally homologous to FLAP (12, 13). These proteins both belong to the microsomal-associated proteins involved in eicosanoid and glutathione metabolism family, which also includes microsomal GST-2 and PGE₂ synthase. Microsomal GST-2 conjugates LTA₄ to reduced glutathione and forms LTC₄ in vitro, but this activity is weak compared with that of authentic LTC₄S. Moreover, homogenates of most tissues from mice with a targeted deletion of LTC₄S are unable to perform this conjugation, and MCs derived from LTC₄S knockout mice generate no cys-LTs when activated ex vivo (19). Both FLAP and LTC₄S constitutively localize to the perinuclear envelope, the subcellular location where cys-LT synthesis is thought to occur. Interestingly, FLAP and LTC₄S also colocalize with 5-LO to inducible electron-dense cytoplasmic lipid bodies in activated eosinophils (20). Whether lipid bodies are accessory sites of LT synthesis or are transport organelles for eicosanoid-synthesizing enzymes is not known.

Although the enzymes of the 5-LO/LTC₄S pathway are expressed constitutively, their activity and expression in human and mouse MCs is subject to modulation by exogenous cytokines. Mouse MCs derived in vitro from bone marrow cells cultured in the presence of recombinant mouse stem cell factor and IL-10 respond to exogenous IL-3 with marked proliferation and progressive increments in their expression of 5-LO, FLAP, and LTC₄S (21). These IL-3-treated MCs respond to cross-linkage of their high-affinity FcR for IgE, FcεRI, by generating LTC₄ in markedly increased quantities as compared with replicates not treated with IL-3. Cord blood-derived human MCs respond to short-term (1–5 days) priming with recombinant human IL-4 with a striking up-regulation in LTC₄S protein and mRNA expression, accompanied by a parallel increase in enzymatic function (22). Inasmuch as IL-3 and IL-4 are expressed prominently by lymphocytes in inflamed mucosal surfaces, these studies suggest a mechanism by which endogenously produced cytokines may coordinately regulate cys-LT generation by MCs in the effector phase of the mucosal inflammation that accompanies helminth infection and allergic disease.

Genetic variants of 5-LO and LTC₄S have been described in humans. Allelic variants in the 5-LO core promoter result in the addition or deletion of consensus binding sites for the zinc-finger DNA binding proteins SP1 and early growth response-1, and diminished 5-LO promoter reporter activity compared with the wild-type allele (23). The lung function of individuals with asthma carrying these variant alleles improved substantially less in response to an orally administered 5-LO inhibitor than did the lung function of similarly treated individuals with asthma who were homozygous for the wild-type allele (24). A common polymorphism in the human LTC₄S gene results from an A-C substitution 444 bases upstream of the translational start site. The presence of this polymorphism correlates with increased levels of LTC₄ generation by calcium ionophore-stimulated peripheral blood eosinophils (25), diminished lung function in asthmatic children (26), increased clinical responses to CysLT₁ receptor antagonists (25, 27), and an increased incidence of aspirin sensitivity in some (28), but not all, human ethnic populations (29). Thus polymorphisms in cys-LT-generating enzymes may modify drug responses and certain phenotypic features in patients with asthma.

Early pharmacologic profiling studies of mammalian tissues predicted the existence of at least two types of receptors for the cys-LTs (30, 31). Subsequently, the CysLT₁ and CysLT₂ receptors were cloned in the mouse (32, 33, 34) and human (35, 36, 37). Both receptors are members of the G protein-coupled receptor family, which use both pertussis toxin (PTX)-sensitive and PTX-insensitive G proteins for their signaling. The different pharmacologic profiles and incompletely overlapping tissue distributions of the CysLT₁ and CysLT₂ receptors (Table I) correlate well with their pharmacology predicted from tissue profiling studies (38).

The sequences of the human CysLT₁ and CysLT₂ receptors are highly divergent (38% amino acid identity). Human and mouse CysLT₁ receptors are 87% identical, whereas the CysLT₂ receptors are 74% identical. The mouse gene encoding the CysLT₁ receptor consists of four exons and three introns, giving rise to two transcripts that result in long and short respective isoforms of the receptor (33). Exons I and IV encode the short isoform, whereas the long isoform also incorporates the small exons II and IV. Exon III of the mouse CysLT₁ receptor gene contains an in-frame start codon (ATG) that is 39-bp upstream of the ATG translation start site reported for the human CysLT₁ receptor. This sequence permits a 13-aa extension at the N terminus, resulting in a long receptor isoform with 352 deduced amino acid residues and a calculated mass of 40,715. The N-terminal extension is missing in the human CysLT₁ receptor because of a stop codon six nucleotides upstream of the ATG start codon within the coding exon; thus, only one isoform of the CysLT₁ receptor exists in the human (35). Both isoforms of the CysLT₁ receptor are expressed at the mRNA level in a variety of mouse tissues, including lung, skin, and small intestine. The gene for the mouse CysLT₁ receptor maps to band XD, and the human gene maps to a syntenic region of chromosome X. Both isoforms of the mouse CysLT₁ receptor and its single human counterpart bind LTD₄ with high affinity (K_d ∼ 1 nM), and bind LTC₄ and LTE₄ with progressively lower affinities (Table I).

The CysLT₂ receptor arises from a single exon gene on human chromosome 13q14. The mouse orthologue arises from a gene containing 6 exons on chromosome 14, with the entire coding sequence in exon 6 (34). Two splice variants of the receptor exist in mice, whereas a single variant is described in humans (37). The mouse and human CysLT₂ receptors both bind LTC₄ and LTD₄ with equal affinity (K_d ∼10 nM). The human CysLT₂ receptor mRNA is expressed in peripheral blood leukocytes, lymph nodes, spleen, heart, and several CNS regions (37, 39), a tissue distribution similar to that reported for the mouse CysLT₂ receptor (34). In situ hybridization localized CysLT₂ receptor mRNA to the adrenal medulla, as well as the cardiac Purkinje cells. Especially strong hybridization was noted in interstitial macrophages in the lung (37). Both HUVEC (40) and coronary smooth muscle cells (41) both express high levels of the CysLT₂ receptor protein and mRNA, suggesting a prominent function for this receptor in vascular responses. A single nucleotide polymorphism in the human CysLT₂ receptor was reported that results in the substitution of methionine for valine at amino acid residue 202. The resultant polymorphic receptor, when cloned and expressed heterologously, binds to LTD₄ and the receptor-selective partial agonist, BAY-u9773, at 4-fold lower affinity than the wild-type receptor (42). The presence of this polymorphic CysLT₂ variant conferred a substantially increased risk of atopy in inhabitants of Tristan da Cunha, a population characterized by both a founder effect and a high prevalence of atopy. The incompletely overlapping distributions and distinct ligand-binding properties of CysLT₁ and CysLT₂ receptors strongly suggest that they serve different functions in vivo. The existence of a third receptor for cys-LTs has been proposed on the basis of pharmacologic profiling studies of human pulmonary artery (43) and guinea trachea (30), although such a receptor has yet to be isolated and cloned.

When transfected into Xenopus oocytes, both CysLT₁ and CysLT₂ receptors use PTX-resistant Gq family proteins to initiate calcium flux (35, 37). HUVEC express both CysLT₁ (44) and CysLT₂ receptor transcripts (40), but CysLT₂ receptors dominate and are exclusively responsible for LTC₄-mediated calcium flux in this cell type (40). In contrast, CysLT₂ receptors fail to contribute to the robust cys-LT-mediated calcium fluxes in human MCs, even though these cells also express both receptors (45, 46). Blockade of CysLT₁ receptors also prevents cys-LT-mediated phosphorylation of ERK in human MCs, but does not affect p38 kinase phosphorylation or IL-8 secretion; the latter is inhibited by PTX (46). Stimulation of CysLT₁ receptors in human bronchial smooth muscle cells initiates calcium flux and translocation of both calcium-dependent protein kinase Cα, as well as calcium-independent protein kinase Cγ. Interestingly, sustained contraction of these smooth muscle cells in response to LTD₄ does not require extracellular calcium, even though it is CysLT₁ receptor-dependent (47). Thus, both CysLT₁ and CysLT₂ receptors can initiate signaling through more than one class of G proteins, and can induce both calcium-dependent and -independent events. These properties may vary depending on their relative levels of expression and the cell type involved.

Historically, cys-LTs have been recognized for their powerful bronchoconstricting effects and their role in asthma exacerbations. When administered by inhalation to human subjects, LTE₄ is substantially more potent than histamine in eliciting a decrease in airflow (48). Cys-LTs are present in bronchoalveolar lavage (BAL) fluid collected from atopic human subjects after endobronchial challenge with allergen (49), and levels of LTE₄ were elevated in urine samples collected from patients presenting to the emergency room for treatment of asthma exacerbations (50). Both an orally administered inhibitor of 5-LO (51) and CysLT₁ receptor-selective antagonists (52) were clinically efficacious in the treatment of humans with asthma. Pranlukast, one of the CysLT₁ receptor antagonists, substantially attenuated both the early- and late-phase decrements in airflow in allergic asthmatic individuals challenged with allergen by inhalation (53). Orally administered CysLT₁ receptor antagonists also attenuate bronchoconstrictive responses to challenges with exercise (54), cold air (55), and inhalation of adenosine (56). Intravenous administration of montelukast, a CysLT₁ receptor-selective antagonist, substantially increased measures of airflow compared with placebo in a group of patients presenting to the emergency room with acute asthma who also received standard treatment with bronchodilators and glucocorticoids (57). Taken together, these observations confirm the involvement of cys-LTs in the development of airflow obstruction in both experimentally induced and naturally occurring asthma in humans, and with a prominent role for the CysLT₁ receptor in exacerbations.

In addition to bronchconstriction, endobronchial instillation of LTE₄ into human subjects resulted in the subsequent influx of eosinophils, and to a lesser extent, neutrophils, into the BAL fluid (58), suggesting the capacity for cys-LTs to either directly or indirectly attract leukocytes to initiate inflammatory responses in vivo. A subsequent functional study revealed the expression of CysLT₁ receptors, but not CysLT₂ receptors, by human CD34⁺ peripheral blood-derived progenitor cells (59). cys-LTs induced transendothelial migration of these cells in vitro, suggesting a possible role for CysLT₁ receptors in the regulation of hemopoietic progenitor cell trafficking. Importantly, while cys-LT-mediated calcium fluxes were completely blocked by pretreatment of the CD34⁺ progenitor cells with montelukast, this pretreatment with antagonist failed to block calcium fluxes in mature peripheral blood leukocytes, indicating that the CysLT₂ receptor is acquired by leukocytes during hemopoietic development in vivo. Human peripheral blood monocytes (37, 60), eosinophils (61), and lung macrophages (35, 37) all express both CysLT₁ and CysLT₂ mRNA or protein. Priming of either human peripheral blood monocytes or monocyte-derived macrophages with recombinant human IL-4 or IL-13 increased their levels of CysLT₁ receptor mRNA and protein expression, and enhanced their chemotactic responses to LTD₄ in vitro (62). Similarly, CysLT₁ receptor expression and chemotaxis to LTD₄ in an eosinophilic subline of the human granulocytic leukemia cell line, HL60, were both enhanced by priming with rIL-5 (63). Both CysLT₁ and CysLT₂ receptors localized to eosinophils, mononuclear cells, and resident MCs in nasal biopsy tissue from humans with seasonal allergic rhinitis (64). In this study, the CysLT₁ receptor, but not the CysLT₂ receptor, also localized to a subset of lesional neutrophils. In another study, expression of the CysLT₁ receptor was up-regulated on CD45⁺ leukocytes in nasal biopsy specimens obtained from subjects with aspirin-sensitive chronic rhinitis and nasal polyposis, compared with the same leukocyte subsets in biopsy specimens from non-aspirin-sensitive subjects with rhinitis and nasal polyposis (65). Taken together, these observations support the hypothesis that cys-LTs may serve as chemotactic mediators and/or activating ligands for human effector leukocytes, and that the cys-LT receptor expression profiles of certain leukocytes are modified by available cytokines and other microenvironmentally derived factors in inflammation. In vitro studies support the capacity for cys-LTs to serve as eosinophil chemoattractants through a CysLT₁ receptor-dependent mechanism (66).

Although human eosinophils derived in vitro from umbilical cord blood mononuclear cells secrete IL-4 in response to cys-LTs in a CysLT₁ receptor-dependent manner (67), this response was not observed in freshly isolated peripheral blood eosinophils (68). Instead, peripheral blood eosinophils secreted IL-4 in response to cys-LTs only after permeabilization. The secretion of IL-4 by permeabilized eosinophils in response to cys-LTs was sensitive to PTX, and occurred at log-fold lower doses of LTC₄ than LTD₄. This response profile is not compatible with the function of either the CysLT₁ or CysLT₂ receptor. Moreover, the pretreatment of peripheral blood eosinophils with MK886, an inhibitor of both FLAP and LTC₄S, or with the 5-LO inhibitor AA861, inhibited their release of IL-4 in response to other transmembrane stimuli, such as IL-16, RANTES, or eotaxin. This study indicates an intracrine role for the cys-LTs, acting through an unidentified intracellular receptor, in activation-dependent signaling in eosinophils.

Like eosinophils, resident tissue MCs in nasal biopsies express both CysLT₁ and CysLT₂ receptors (64, 65). Human cord blood-derived MCs also express both receptors (45, 46), and respond to ex vivo stimulation by low nanomolar concentrations of LTC₄ and LTD₄ with strong, PTX-resistant calcium fluxes. As noted above, LTC₄ and LTD₄-mediated calcium fluxes were completely blocked by pretreatment of MCs with MK571, indicating an absolute requirement for CysLT₁ receptors in this response. Priming these MCs with recombinant human IL-4 markedly shifted their dose-response curve to LTC₄ (∼3 log-fold enhancement) to a greater extent than their response to LTD₄ (1.3 log-fold enhancement), but did not alter CysLT₁ receptor mRNA or cell surface protein expression (45), and only modestly up-regulated cell surface CysLT₂ protein (46). Unexpectedly, IL-4 also markedly enhanced calcium fluxes in response to UDP, a nucleotide ligand previously thought to be specific for a purinergic (P2Y) receptor termed P2Y6 (69). The calcium flux in response to UDP was MK571-sensitive and was cross-desensitized with LTC₄, but only incompletely with LTD₄. Chinese hamster ovary (CHO) cells stably expressing the long isoform of the mouse CysLT₁ receptor also respond to UDP with an MK571-sensitive calcium flux (45). IL-4-primed MCs, but not unprimed replicates, secreted several cytokines (IL-5, TNF-α, MIP-1β, and IL-8) when stimulated with either LTC₄, LTD₄, or UDP (46, 70). Of these cytokines, IL-8 was unique for the resistance of its secretion to blockade by MK571, indicating dependence on the CysLT₂ receptor (70). Blockade of CysLT₁ receptors by MK571, or inhibition of endogenous cys-LT production by MK886, significantly attenuated the generation of IL-5 and TNF-α by MCs activated by FcεRI cross-linkage (70). These studies indicate that the CysLT₁ and CysLT₂ receptors have distinct yet complementary functions for MCs, permitting cytokine generation through both autocrine and paracrine mechanisms. IL-4 could act by inducing the expression of an MK571-sensitive receptor for both LTC₄ and UDP, or could alter the sensitivity and ligand specificity of constitutively expressed CysLT₁ receptors by inducing posttranslational modification or heterodimerization with another receptor. Moreover, the induction of IL-5 generation by MCs is an additional mechanism by which cys-LTs could promote eosinophilia.

The role of cys-LTs and their receptors in immune responses, inflammation, and tissue repair in vivo has been assessed using both pharmacologic and genetic approaches in mice. Initial recognition that cys-LTs were involved in the maturation of DCs emerged from studies of DC migration in mice lacking MRP1, which is required for the export of LTC₄ to the extracellular space after it is synthesized (15, 71). In a model of contact hypersensitivity induced by topical application of FITC, DC migration was substantially attenuated in MRP1-deficient mice as compared with that observed in wild-type controls (72). The migration defect was corrected by the injection of LTC₄ or LTD₄ after FITC application. Migration in wild-type controls was blocked by the local injection of MK571, which acts as an antagonist of MRP1 at 25-fold higher doses than those required to block the CysLT₁ receptor (72). DCs from MRP1-deficient mice showed deficient ex vivo chemotactic responses to CCL19, a chemokine ligand specific for CCR7. Again, this defect was corrected by exogenous LTC₄ and LTD₄. In a subsequent study, myeloid DCs cultured ex vivo from mouse bone marrow were found to express CysLT₁ receptor mRNA, as well as mRNAs encoding 5-LO, FLAP, and LTC₄S (73). These DCs generated both IL-10 and IL-12 when challenged ex vivo with the dust mite Ag Der f. Treatment of these DCs with the CysLT₁ receptor-selective antagonists montelukast, zafirlukast, and pranlukast significantly blunted their Der f-mediated production of IL-10, but further enhanced their production of Th1 cytokine IL-12. Moreover, whereas exogenous LTC₄, LTD₄, and LTE₄ failed to directly elicit IL-10 or IL-12 generation by otherwise unstimulated DCs, each cys-LT amplified the Der f-mediated production of IL-10, and inhibited the generation of IL-12. Mice that received intranasal adoptive transfer of Der f-pulsed DCs that had been costimulated with LTD₄ ex vivo developed enhanced eosinophilia and increased BAL fluid levels of IL-5 after inhalation challenge with Der f, compared with eosinophil counts and IL-5 levels in mice that received transfer of DCs that had been stimulated with Der f alone. Finally, mice that received Der f-pulsed DCs treated with pranlukast ex vivo and then challenged with Der f developed strikingly diminished BAL fluid levels of eosinophilia and IL-5, compared with the other groups. Taken together, these studies indicate that LTC₄ produced endogenously by DCs (or provided exogenously by MCs or macrophages) during their initial exposure to Ag is a critical determinant of their homing to regional lymph nodes, and their repertoire of cytokines required to induce T cell responses. Either or both of these events, which depend at least in part on the CysLT₁ receptor, may play an important role in the afferent limb of acquired immunity. These findings may help to explain the fact that DC migration in response to FITC is defective in 5-LO-deficient mice (72), which also develop significantly lower IgG and IgE responses to systemic immunization with OVA than do wild-type controls (74).

BALB/c mice sensitized with an i.p. injection of OVA precipitated with alum were challenged with intranasal administration of OVA. Twenty-four hours later, levels of LTC₄ and LTB₄ had increased 5- and 3.4-fold, respectively, in the BAL fluid of the challenged mice compared with the levels in the saline controls (75). These increases were associated with widespread mucus occlusion of the airways, an influx of predominantly eosinophils in airway tissues and BAL fluid, and airway hyperresponsiveness to i.v. methacholine. The i.p. administration of leukotriene synthetic inhibitors (zileuton or MK-886) 30 min before nasal Ag challenge reduced eosinophil infiltration in tissues and BAL fluid by ∼85% and significantly blocked airway mucus release and plugging. In another study, the period of intranasal administration of OVA was extended to 75 days, so as to elicit smooth muscle hyperplasia and subepithelial fibrosis, characteristic features of airway remodeling in chronic asthma (76). In the chronic protocol, treatment with montelukast starting on day 26 significantly reduced the airway eosinophil infiltration, mucus plugging, collagen deposition, and smooth muscle hyperplasia, as determined on day 76. Levels of IL-13, IL-4, IL-10, and IL-5 mRNA were dramatically reduced in the lungs of the challenged montelukast-treated mice compared with placebo-treated challenged controls. Collectively, these studies indicate that cys-LTs, acting through the CysLT₁ receptor, initiate features of chronic inflammation and matrix remodeling in allergen-induced pulmonary disease, possibly by stimulating cytokine generation by resident inflammatory cells. The smooth muscle hyperplasia in the chronic model could also reflect direct proliferative effects, because LTD₄ is a mitogen for human bronchial smooth muscle cells when provided in vitro in combination with TGF β, IL-13 (77), or epidermal growth factor (78). It is noteworthy, however, that airway hyperresponsiveness to methacholine was not affected by inhibition of cys-LT synthesis or the CysLT₁ receptor in either model. Indeed, mouse bronchial smooth muscle does not constrict in response to cys-LTs (79, 80), and the airways of transgenic mice overexpressing the human CysLT₁ receptor specifically in smooth muscle were hyperresponsive to LTD₄, but not to methacholine (81).

Mice deficient in LTC₄S (19) and the CysLT₁ receptor (82) were generated by targeted gene disruption. Conjugation of LTA₄ to reduced glutathione to form LTC₄ was abolished in the brain, tongue, lung, spleen, stomach, and colon of LTC₄S-deficient mice. The activity was spared in the testis, suggesting the contribution of another enzyme with a highly restricted tissue distribution. IgE-dependent LTC₄ generation by bone marrow-derived MCs from the LTC₄S-deficient mice was abolished, while LTB₄ and PGD₂ generation were unaffected. Plasma protein leakage after i.p. challenge with zymosan or IgE-dependent passive cutaneous anaphylaxis was reduced by ∼50% in LTC₄S-deficient mice compared with the wild-type controls. Decrements in plasma leakage in the CysLT₁-deficient mice were comparable in magnitude to those in the LTC₄S knockout mice. Collectively, these studies confirm the absolute requirement for LTC₄S in cys-LT biosynthesis by immune effector cells in most organs and tissues, and indicate the prominent role of cys-LTs, acting through CysLT₁ receptors, in mediating increased vascular permeability in models of both innate and adaptive immunity.

Because LTC₄ is rapidly converted to LTD₄ in extracellular fluids by either γ-GT or the more functionally specific γ-GL, mouse strains lacking either or both enzymes were studied in the zymosan-induced model of peritonitis. Shi et al. showed that i.p. injection of zymosan into γ-GL-deficient mice and γ-GT/γ-GL double-deficient mice led to the accumulation of LTC₄ in the peritoneal cavity, indicating that γ-GL is the enzyme responsible for the conversion of LTC₄ to LTD₄ in this model (83). Plasma protein extravasation was not affected 1–24 h after zymosan injection, but neutrophil infiltration was significantly reduced 2–4 h after the injection in γ-GL-deficient mice. These findings suggest that LTC₄ does not require conversion to LTD₄ to increase vascular permeability through the CysLT₁ receptor, but that LTD₄ may play a role in neutrophil recruitment. Surprisingly, however, neutrophil recruitment was not affected in either LTC₄S- or CysLT₁ receptor-deficient mice subjected to the zymosan-induced peritonitis model (19, 82).

Intratracheal or systemic administration of the chemotherapeutic agent bleomycin induces chronic pulmonary inflammation and fibrosis in mice. These features were significantly blunted with respective disruptions of cPLA₂ (84) and 5-LO (85), suggesting a role for leukotrienes in this model. Several features of bleomycin-induced injury, including pulmonary macrophage and neutrophil recruitment, alveolar septal thickening, fibroblast accumulation, and collagen deposition were significantly reduced in LTC₄S-deficient mice compared with their wild-type littermates (86). Unexpectedly, CysLT₁ receptor-deficient mice were not protected from this injury; instead, these mice showed exaggerated alveolar septal thickening with reticular fiber deposition when compared with wild-type or LTC₄S-deficient mice. Additionally, cys-LT levels in the BAL fluids of the CysLT₁ receptor-deficient mice were roughly 2-fold greater than the levels recovered from the wild-type mice, with no change in the levels of LTB₄ or PGE₂. These results suggest that the cys-LTs are crucial to this macrophage-mediated chronic inflammatory and fibrotic insult, likely working through the CysLT₂ receptor. Because CysLT₂ receptor-deficient mice have not been reported, the direct role for the CysLT₂ receptor in bleomycin-induced pulmonary inflammation and fibrosis remains to be elucidated.

cys-LTs, initially thought to be restricted in their functions to acute, smooth muscle-mediated responses, clearly serve a diverse array of biologic functions (summarized in Fig. 2). The cloning of their two receptors, development of receptor-specific Abs, CysLT₁ receptor-selective antagonists, and null mice lacking biosynthetic enzymes and the CysLT₁ receptor have all expanded the scope of functions served by this mediator class. The involvement of cys-LTs in the afferent limb of adaptive immunity (particularly the induction of Th₂ responses in the lung via effects on DCs and cytokine generation), the recruitment and/or activation of effector cells (especially eosinophils and MCs), inflammation, and fibrosis have all been supported by animal models and await validation in humans. Moreover, the prominent expression of the CysLT₂ receptor in the brain, adrenals, heart, and vascular endothelium suggests additional unrecognized functions of cys-LTs. Studies in this area await the development of CysLT₂ receptor-deficient mice. Based on these observations, the scope of therapeutic applications for agents that alter cys-LT generation or reception are likely to expand.