Leukotriene B₄ Receptors BLT1 and BLT2: Expression and Function in Human and Murine Mast Cells

Leukotriene B₄ (LTB₄)² is a potent proinflammatory lipid mediator derived from arachidonic acid via the 5-lipoxygenase pathway (1, 2, 3, 4). LTB₄ is known to be a chemotactic factor and activator of leukocytes, particularly granulocytes and T cells (5, 6, 7, 8), and has been implicated in several allergic and inflammatory diseases. The action of LTB₄ is mediated by two G protein-coupled receptors, BLT1 and BLT2, high- and low-affinity receptors, respectively (9, 10, 11, 12, 13, 14). BLT1 is only activated by LTB₄, whereas BLT2 can be activated by LTB₄ and several hydroxyeicosotetraenoic acids (HETEs) (12). The tissue distribution of the two receptors is quite different. Although BLT1 expression in both mice and humans has been reported to be predominantly restricted to peripheral leukocytes, BLT2 expression in humans seems to be fairly ubiquitous; with the highest level in the spleen, liver, and lymphocytes (10, 11, 13, 14).

Mast cells (MCs) have an established role in a wide range of both acute and chronic inflammatory conditions. In allergy, cross-linking of their surface receptors for IgE results in degranulation and release of inflammatory mediators. However, MCs also can be recruited and activated without degranulation in response to chemokines in a number of inflammatory conditions, including parasitic infections, allergic responses, and wound healing (15). Upon activation, MCs have been shown to secrete a wide range of cytokines and chemokines that activate and recruit neutrophils, eosinophils, and effector T cells (7, 8, 16). However, little is known about the mechanism of MC recruitment and migration to the site of inflammation. Several chemotactic factors for mouse bone marrow-derived MCs (BMMCs) in vitro have been described, including stem cell factor (SCF), IL-3, MCP-1, MIP-1α, and RANTES (17, 18, 19, 20). For both human (h)HMC-1 and cord blood-derived MCs (CBMCs), SCF (18), IL-8 (20), C3a, and C5a (21) have been reported as chemotactic factors, and RANTES for CBMCs only (18). Kitaura et al. (22) reported mBMMC migration to LTB₄ in a chemotaxis assay but did not identify the receptors involved. Until recently, there have been no published data on LTB₄ receptor expression or function in MCs. Recently, Weller and colleagues (23) reported BLT1 expression and function in mouse and hMC progenitors, although the authors did not investigate potential expression or a functional role for BLT2. In the present study, we have established the expression of both LTB₄ receptors, BLT1 and BLT2, in hMCs and murine (m)MCs and have evidence to suggest that both receptors contribute to the migration of MCs to LTB₄. We show dose-dependent migration and ERK phosphorylation in response to 12-(S)-HETE, a BLT2-selective agonist, thus further demonstrating functional activity of BLT2 in MCs. We also demonstrate that SCF down-regulates expression of both receptors in mBMMCs, resulting in the loss of their chemotactic response to LTB₄.

The hMC line, HMC-1, was cultured in IMDM supplemented with 10% heat-inactivated FCS, 2 mM l-glutamine, 100 IU/ml penicillin, 50 μg/ml streptomycin, and 1.2 mM α-thioglycerol (Sigma-Aldrich).

mBMMCs were obtained by culturing primary femoral BM cells from BALB/c mice in RPMI 1640 medium supplemented with 10% FCS, 0.1 mM nonessential amino acids, 100 IU/ml penicillin, 50 μg/ml streptomycin, 10 ng/ml mIL-3, and 20% WEHI-3 cell (line TIB-68; American Type Culture Collection) conditioned medium. Nonadherent cells were selected and medium replaced weekly, maintaining cell number at ∼5 × 10⁵ cells/ml. After 4 wk of culture, the cells typically consisted of >98% MCs as determined by c-kit receptor immunostaining and flow cytometry analysis. Cells were routinely tested for migration to 50 ng/ml mSCF and 100 nM LTB₄ as described below, before use in experiments.

For mBMMC migration assays, 24-well, 8-μm microchemotaxis chambers (Costar) were coated with 100 μl of 100 μg/ml bovine fibronectin (Sigma-Aldrich) in PBS and incubated for 30 min at room temperature. The fibronectin solution was aspirated from the wells, and a total of 2 × 10⁵ MCs in 100 μl of chemotaxis medium (RPMI 1640 medium, 25 mM HEPES, 0.25% BSA, 100 IU/ml penicillin, 50 μg/ml streptomycin) was added to the upper chamber. The lower chamber contained 600 μl of chemotaxis medium with and without 50 ng/ml mSCF (Biosource) or different concentrations of LTB₄ (Sigma-Aldrich) or 12-(S)-hydroxyeicosatetraenoic acid (HETE) (Cayman Chemical). Either BLT1 antagonist U-75302 (Cayman Chemical) or BLT2 antagonist LY255283 (Cayman Chemical) was added at the appropriate concentrations to both the upper chamber and the lower compartment. Chemokinetic analysis was performed with LTB₄ in both the upper and lower chambers. After incubation for 3 h at 37°C, in 5% CO₂, the Transwells were removed and discarded, and the cells in the lower chamber were quantitated in a 60-s aquisition via FACS.

The hHMC-1 cell migration was measured using a 96-well, 8-μm ChemoTx plate (Neuroprobe). The filter screen was coated with 100 μg/ml human fibronectin (Sigma-Aldrich) in PBS for 1 h at room temperature, after which it was allowed to air dry for 30 min. Wells were slightly overfilled (305 μl) with chemotaxis medium with or without hSCF (100 ng/ml; R&D Systems), LTB₄, or BLT antagonist. The microporous filter screen was attached, and 29 μl of cell suspension containing 1 × 10⁵ cells in chemotaxis medium with and without antagonists was applied in droplets to the top (as instructed by the manufacturer). Plates were incubated for 4 h at 37°C, 5% CO_2. Without removing the filter screen, the plate was tipped at a 45° angle, and the nonmigrating cells on the surface of the screen were removed by rinsing with RPMI 1640 medium and then gently wiping with a damp paper towel. The screen was then removed, and the cells that had migrated into the filter screen were fixed by immersion in methanol for 1 min, stained in Wright-Giemsa modified stain (WG-32; Sigma-Aldrich) on a rotary shaker for ∼20 min, destained with water, and air dried. The number of cells migrating into the filter was counted microscopically and reported as the total number of cells observed in four randomly selected high-power fields.

RNA was extracted from cells using RNeasy (Qiagen). Genomic DNA was removed using DNA-free, a DNase treatment and removal kit (Ambion). Briefly, ∼10 μg of RNA, 10 μl of 10× DNase buffer, and 2 μl of DNase I were adjusted to a total volume of 100 μl in diethyl pyrocarbonate-treated H₂O. The reaction was incubated for 30 min at 37°C, then 10 μl of DNase-removal slurry was added and incubated for an additional 2 min at room temperature, centrifuged for 1 min, and RNA was removed and frozen at −80°C.

RNA was reverse transcribed into cDNA and amplified using a OneStep RT-PCR kit (Qiagen) and a gene-specific antisense primer. Primers for mBLT1 were 5′-CCTGTGCCACTATGTCTGTGGAATAAGC-3′ (sense) and 5′-CACCAGGTGGTAAGGCAGCCAGAA-3′(antisense). The primers for mBLT2 were 5′-ATCGTGCTGGCCTTTGGCTTGCTC-3′ (sense) and 5′-GACTGTCTTTCTCCGTCTTGCCCC-3′ (antisense). The resulting PCR products for mBLT1 and mBLT2 were 444 and 387 bp, respectively.

The primers for hBLT1 were 5′-TATGTCTGCGGAGTCAGCATGTACGC-3′ (sense) and 5′-CCTGTAGCCGACGCCCTATGTCCG-3′ (antisense); and for hBLT2 were 5′-AGCCTGGAGACTCTGACCGCTTTCG-3′ (sense) and 5′-GACGTAGAGCACCGGGTTGACGCTA-3′ (antisense). The resulting PCR products for hBLT1 and hBLT2 were 345 and 320 bp, respectively. The hBLT1 and hBLT2 primer sequences were those designed by Yokomizo et al. (24). The reaction mixture (50 μl) contained 100–500 ng of total RNA, 1× Qiagen OneStep RT-PCR buffer mix, 1× Q-solution, 400 μM each of dNTP, 0.2 μM each of gene-specific primer, 5 U RNase inhibitor, and 2 μl of OneStep Enzyme mix (contains Omniscript and Sensiscript reverse transcriptase (RT) and HotStarTaq DNA polymerase). No-RT control contained HotStarTaq only. RT-PCR was performed on a Stratagene Robocycler, and the amplification profile was 50°C for 30 min (RT); 95°C 15 min (HotStartaq activation); and 35–40 cycles of 94°C for 1 min, 55°C for 1 min, and 72°C for 1 min. Cycling was completed with a 10-min extension at 72°C. The final PCR products were resolved on a 1.5% agarose gel containing ethidium bromide and visualized under UV light.

The mBMMCs were stimulated for 16 h with 50 ng/ml mSCF. Cells were then washed four times and used for chemotaxis as described above, or RNA was extracted and used for TaqMan real-time PCR. Supernatant was collected and assayed by ELISA for LTB₄ (Assay Designs Correlate EIA) and mIL-6 (R&D Systems).

DNase-treated RNA was reverse transcribed using random hexamers with the Applied Biosystems High-capacity cDNA archive kit as per the manufacturer’s recommendations. The cDNA was quantitated using PicoGreen (Molecular Probes). Briefly, 5 μl of cDNA diluted in 100 μl of Tris-EDTA was added to 100 μl of picogreen working solution. After 5 min at room temperature, the assay was read in a fluorescent plate reader at 485-nm excitation, 580-nm emission. A standard curve was made using the cDNA standard in the kit, and the concentration of the unknown cDNA was determined by linear regression.

The experimental cDNA was diluted to a final concentration of 20 ng/μl. Five microliters of cDNA (100 ng) was amplified in the presence of 12.5 μl of universal master mix, 1.25 μl of gene-specific TaqMan probe, and 6.25 μl of H₂O. Endogenous controls used gene-specific probes for β-actin. Probes used FAM as a reporter (hBLT1 Hs00609525_m1; hBLT2 Hs0025197342m1; mBLT1 Mm00521839_m1; mBLT2 Mm00498491_s1; Applied Biosystems).

Samples underwent the following amplification stages: stage 1, 50°C for 2 min; stage 2, 95°C for 10 min; and stage 3, 95°C for 15 s followed by 60°C for 1 min. Stage 3 was repeated 40 times. Gene-specific PCR products were measured by means of an Applied Biosystems 7500 Real-Time PCR instrument continuously during 40 cycles. Target gene expression was normalized between different samples based on the values of the endogenous control expression for each cDNA sample. The ΔΔC_T and the comparative expression levels were calculated as described previously (25).

For surface staining of BLT1, a total of 5 × 10⁵ cells was blocked with 10% human serum in FACS buffer (PBS with 25 mM HEPES, 1 mM EDTA, 1% FBS) for 30 min at 4°C. Ten micrograms per milliliter of either FITC-conjugated anti-hLTB₄ receptor (BLT1) mAb (Abcam 7B1) or isotype control was added and incubated for 30 min at 4°C and washed three times with FACS buffer, and BLT1 receptor expression was analyzed by flow cytometry. For intracellular staining, a total of 5 × 10⁵ cells, after blocking, was resuspended in 250 μl of BD Cytoperm/Cytofix reagent (BD Pharmingen) for 20 min at 4°C, washed twice with BD Perm/Wash buffer, resuspended in 100 μl of Perm/Wash buffer with 10 μg/ml mAb for 30 min at 4°C, washed twice, and analyzed by flow cytometry. An appropriate irrelevant isotype control Ab was used to measure the extent of nonspecific binding.

Activation of ERK and Akt by LTB₄ in mBMMCs was determined by immunoblotting using the phosphospecific anti-ERK p42/44 (Thr²⁰²/Tyr²⁰⁴) and anti-Akt (Ser⁴⁷³), respectively (Cell Signaling Technology). Briefly, a total of 3 × 10⁷ cells was washed with serum-free RPMI 1640 medium and starved with medium containing 0.5% serum for 4 h. Cells were then resuspended at 10⁷/ml with RPMI 1640 medium and stimulated with 10 and 100 nM LTB₄, respectively. Aliquots of cells (200 μl) were taken at 0, 2, 5, 10, 30, and 60 min and chilled in 1 ml of ice-cold PBS. For dose-dependent ERK activation, cells were treated with various concentrations of LTB₄ (1–100 nM) or 12-(S)-HETE (0.1 to 10 μM) for 2 min only. Cells were then lysed in buffer containing 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 50 mM β-glycerol phosphate, 5 mM EDTA, 1 mM Na₃VO₄, 10 mM sodium pyrophosphate, 1% Triton X-100, and complete protease inhibitor mixture (Roche). Approximately 20 μg of total proteins was separated on 10% SDS-PAGE and subjected to Western blot analysis using standard protocols.

We investigated the possibility of MC expression of the LTB₄ receptors, BLT1 and BLT2, by RT-PCR analysis with specific primers in mBMMCs and in an hMC line, HMC-1. As shown in Fig. 1,A, RT-PCR with primers specific for mBLT1 and mBLT2 yielded PCR products of the expected sizes (444 and 387 bp, respectively). Controls without RT confirmed that the source of the products was indeed from mRNA, and not from contaminating genomic DNA. The same primers also were used to amplify BLT1 and BLT2 from mouse T cell RNA (data not shown), and the products were sequenced to confirm that they corresponded to the expected segments of mBLT1 and mBLT2 (NM_008519 and NM_020490). In Fig. 1 B, the RT-PCR results for the corresponding human products from HMC-1 RNA are shown at 345 and 320 bp, for hBLT1 and hBLT2, respectively. These PCR products were confirmed by sequencing to be identical with the corresponding segments of the published sequences (NM_181657 and NM_019839). No-RT controls to test for genomic DNA contamination yielded no PCR products. Plasmids containing cDNA for each receptor were used as positive controls and yielded PCR products of the expected sizes in all cases (data not shown). Thus, mBMMC and hHMC-1 cells express mRNA for both BLT1 and BLT2.

To demonstrate BLT1 receptor protein expression, flow cytometry was performed after staining the Jurkat T cell leukemia line and the HMC-1 MC line with a specific anti-hBLT1 mAb. Staining of Jurkat cells showed a rightward shifted peak relative to the isotype control staining, indicating that BLT1 receptor was expressed on the cell surface (Fig. 2,A). In contrast in HMC-1 cells, there was a slight shoulder of positive surface staining indicating, at most, a very small population of BLT1 receptors on the surface (Fig. 2,B). However, staining of fixed and permeabilized HMC-1 cells (Fig. 2 C) showed a BLT1-positive cell population, indicating that the primary site of BLT1 protein expression in HMC-1 cells was intracellular. Incubations of HMC-1 cells with LTB₄ concentrations of 100 nM and 1 μM for different times varying from 15 min to overnight did not stimulate translocation of BLT1 to the surface (data not shown). There are currently no suitable flow cytometry Abs available for staining either hBLT2 or mBLT1/BLT2 proteins, thus precluding studies of their expression.

Because LTB₄ has been shown to be a chemotactic factor for leukocytes expressing LTB₄ receptors, we investigated whether LTB₄ could mediate chemotaxis of mBMMCs. Chemotaxis to SCF was used as a positive control for comparison. Fig. 3,A shows the relative levels of migration of mBMMCs to LTB₄, compared with SCF-induced migration mediated through the c-kit receptor. Although the chemotactic response was not as pronounced as that to SCF, mBMMCs clearly demonstrated LTB₄-dependent migration (p = 0.0001). The migration of mBMMCs to LTB₄ also was tested over a range of concentrations. The mean results of three separate experiments in Fig. 3,B indicate that mBMMCs migrated to LTB₄ in a dose-dependent manner, with an EC₅₀ of 5 nM, and reached maximum chemotaxis at 100 nM LTB₄, which was the highest concentration tested. To establish whether the LTB₄-dependent migration was due to chemotaxis and/or chemokinesis, migration was compared with only the lower well containing 100 nM LTB₄ vs 100 nM LTB₄ in both upper and lower wells. The results in Fig. 3 C indicate that the level of chemotaxis (LTB₄ in lower wells only) was clearly higher than chemokinesis (LTB₄ in both upper and lower wells). At 100 nM LTB₄, there was no statistical difference between the control well and the chemokinesis well, whereas there was a significant difference between the chemotaxis and chemokinesis wells (p = 0.01).

To investigate the existence of functional BLT2 chemotactic activity in mBMMC, chemotaxis assays were performed using a range of concentrations (0.1–10 μM) of 12-(S)-HETE, a BLT2-selective agonist (12). In three separate experiments, significant, dose-dependent chemotaxis to 12-(S)-HETE was observed, reaching similar or greater magnitude, compared with that observed for 100 nM LTB₄, as shown in Fig. 3 D. Although the most significant chemotactic activity was observed between 5 and 10 μM, statistically significant chemotaxis vs background was observed at 12-(S)-HETE concentrations as low as 1 μM (p < 0.05; Experiment 1); and 2.5 μM (p < 0.05; Experiment 3). In each experiment, a steep dose-response to 12-(S)-HETE was observed, reaching maximal levels at 10 μM, the highest concentration tested.

Chemotaxis assays also were performed in the presence of selective BLT1 and BLT2 receptor antagonists to investigate the extent of involvement of BLT1 and/or BLT2 in the LTB₄-dependent migration. Dose-reponses were performed with both antagonists in mBMMCs. The results in Fig. 4,A show that there was statistically significant inhibition of chemotaxis by U-75302 (a BLT1-specific antagonist), as low as 10 μM, and by LY255283 (a BLT2-specific antagonist), as low as 2.5 μM compound. Chemotaxis to 25 nM LTB₄ was significantly inhibited by 10 μM U-75302 or by 10 μM LY255283 (Fig. 4,B). There was no additive effect when these concentrations of the two compounds were combined. To further investigate the specificity of the BLT antagonists, we tested their effect on chemotaxis to SCF. Fig. 4 C shows that, at concentrations of 20 μM compound, there was no inhibition by either antagonist of chemotaxis to SCF, and there was, in fact, a significant enhancement effect. Furthermore, no evidence for cytotoxicity was observed after overnight incubation of mBMMCs with either antagonist at 20 mM (data not shown).

To further support the functional expression of BLT1 and BLT2 in mBMMCs, we examined LTB₄-mediated signaling events. It has been shown that ligation of LTB₄ to BLT1 and/or BLT2 activates PI-3 and MAPK in a variety of experimental systems, including cell lines, transfected cell lines, and primary cells (26, 27, 28, 29, 30, 31). We investigated ERK1/2, Akt, and p38 phosphorylation in mBMMCs in response to 10 nM and 100 nM LTB₄ or LPS. Background ERK1/2, Akt, and p38 phosphorylation was observed in the absence of stimulants, and even starvation of mBMMC in 0.5% serum for up to 24 h failed to eliminate the background (data not shown). After 10-min incubation with LPS, a small, transient increase over background in ERK1/2 and Akt phosphorylation was observed, which significantly decreased by 45 min (Fig. 5, last two lanes). LTB₄, at 10 or 100 nM, induced transient activation of ERK1/2, peaking at around 2 min and progressively declining thereafter to a much lower level by 30–60 min. Although the kinetics for ERK activation by the two different concentrations of LTB₄ were similar, 100 nM LTB4 seemed to result in higher and more sustained activation than 10 nM. In response to LTB₄, Akt phosphorylation initially decreased from the relatively high background, followed by transient phosphorylation, peaking at around 10 min and declining thereafter. In contrast with ERK1/2 and Akt, p38 MAPK phosphorylation in mBMMC was relatively unchanged in the presence of LTB₄ or LPS. Similar results were obtained in three separate experiments. Thus, transient LTB₄-mediated ERK1/2 phosphorylation and changes in Akt phosphorylation were observed in mBMMC at both 10 and 100 nM, and the relative magnitude of the responses was similar to the relative chemotactic responses observed at 10 vs 100 nM LTB₄.

In a separate experiment, ERK activation in mBMMCs was examined in response to a 2-min stimulation with LTB₄ (1–100 nM) or 12-(S)-HETE (0.1–10 μM). Dose-dependent ERK phosphorylation in response to LTB₄ was clearly observed, reaching maximal levels between 10 and 100 nM. Treatment of mBMMCs with 12-(S)-HETE also induced ERK phosphorylation. Marginal phosphorylation was observed at 2 μM, but a clear increase in phospho-ERK could be observed at 5 μM and even more so at 10 μM (see Fig. 5,B), similar to the concentration range where the most significant chemotaxis to 12-(S)-HETE was observed (Fig. 3 D).

Chemotaxis assays were performed using HMC-1 cells and, as for mBMMCs, LTB_4-dependent migration was observed (Fig. 6,A). Migration due to chemokinesis was measured as described above and was higher than spontaneous migration of untreated control cells (p = 0.01) but significantly lower than the LTB₄-dependent chemotactic response (p = 0.0008). Migration of HMC-1 cells was tested over a range of concentrations of LTB₄. The mean results of three separate experiments (Fig. 6,B) showed dose-dependent HMC-1 chemotaxis to LTB₄, with an EC₅₀ of ∼25 nM LTB₄. Chemotaxis of HMC-1 cells to 100 nM LTB₄ was completely inhibited by both the BLT1-specific receptor antagonist, U-75302, and the BLT2-specific antagonist, LY255283, at a concentration of 500 nM (Fig. 6 C). No difference was observed when combining the two compounds. Neither antagonist showed any evidence of cytotoxicity at 20 μM, after overnight incubation with HMC-1 cells (data not shown).

SCF is an important growth and differentiation factor for MCs and stimulates both chemotaxis and the production of chemokines and cytokines, including IL-6, in mBMMCs. We found that incubation of mBMMCs with SCF for 16 h resulted in dose-dependent LTB₄ production, with an EC₅₀ of 60 ng/ml SCF (Fig. 7,A). Similarly, SCF induced a dose-dependent production of IL-6 with an EC₅₀ of 75 ng/ml (Fig. 7,B), and also induced a small amount of MCP-1 (data not shown). However, similar incubations with LTB₄ did not induce IL-6 or MCP-1 (data not shown). Because SCF stimulated LTB₄ production in mBMMCs, we also explored the effects of SCF on BLT1 and BLT2 mRNA expression. Cells were treated for 16 h with 50 ng/ml SCF, after which RNA was extracted for RT and real-time PCR. The ΔΔCt method for relative quantitation of gene expression in the TaqMan system was used. The calculated fold differences in Tables I and II show that SCF treatment of mBMMCs down-regulated BLT1 and BLT2 receptor mRNA expression, respectively. This result is shown graphically in Fig. 8 A.

Because we observed down-regulation of BLT1 and BLT2 expression in SCF-treated mBMMC cells, we tested chemotaxis to LTB₄ and SCF in this same cell population. After a 16-h incubation with 50 ng/ml SCF, mBMMCs were washed four times to remove any residual SCF and tested in chemotaxis assays to 100 nM LTB₄ or 50 ng/ml SCF. Pretreatment with SCF completely eliminated the LTB₄-dependent migration of mBMMCs (Fig. 8,B), which correlated with the reduction observed in mRNA expression for both BLT1 and BLT2 receptors (Fig. 8,A). Conversely, the ability to migrate to SCF was enhanced >2-fold (Fig. 8 C).

LTB₄ is a potent leukocyte activator and chemoattractant, which mediates its biological action via the LTB₄ receptors, BLT1 and BLT2. Recently, BLT1 expression and LTB₄-dependent chemotaxis have been reported in immature MCs (23). In the present study, we have shown the first evidence for BLT2 expression in mBMMCs and in HMC-1 cells. Both HMC-1 cells and mBMMCs migrated to LTB₄ in a dose-dependent manner, and this migration was inhibited by either a BLT1- or BLT2-selective antagonist, suggesting that both receptors contributed to the chemotactic activity. Dose-dependent migration and ERK phosphorylation in mBMMCs in response to 12-(S)-HETE, a BLT2-selective agonist (12). further supported the existence of functional BLT2 receptors in mBMMCs. Stimulation of mBMMCs with LTB₄ induced transient, dose-dependent ERK signaling and changes in Akt phosphorylation. Pretreatment of mBMMCs with SCF significantly down-regulated expression of both BLT1 and BLT2 mRNA and inhibited their ability to migrate to LTB₄. Our study thus demonstrates the expression of functional LTB₄ receptors, both BLT1 and BLT2, in mMCs and hMCs, and a regulatory role for SCF in their expression.

Mature MCs are normally found interspersed throughout the tissue. Mast cell hyperplasia is a hallmark of many inflammatory conditions such as allergic rhinitis (32), asthma (33), scleroderma (34), psoriasis (35), tissue transplant rejection (36), rheumatoid arthritis (37), and mastocytosis (38). The factors regulating MC recruitment during inflammation are still not fully understood. This relocalization of tissue MCs has been hypothesized to be regulated by chemotaxins such as complement proteins, C3a and C5a (21), growth factors, and chemokines such as CXCL2, platelet-activating factor, CCL5/RANTES, TGF-β, SCF, IL-3, and/or MCP-1 (39, 40, 41, 42, 43). Our study and that of Weller et al. (23) show that LTB₄ also may play a significant role in regulating MC migration in vivo through both BLT1 and BLT2 receptors.

We also have shown LTB₄-dependent signaling events in mBMMCs. Although LTB₄ clearly induced transient phosphorylation of ERK, the predominant effect of LTB₄ on the Akt pathway (Fig. 5) appeared to be an initial dephosphorylation of Akt, followed by a transient rephosphorylation, which may represent LTB₄-stimulated phosphorylation, or a decline in LTB₄-induced inhibition of phosphorylation. Chemoattractant-induced inhibition of Akt phosphorylation has previously been observed: Heit et al. (44) demonstrated that fMLP inhibited the Akt phosphorylation induced in neutrophils by IL-8, suggesting that fMLP had a direct inhibitory effect on the PI3K/Akt pathway. The initial dephosphorylation of Akt that we observed at both 10 and 100 nM LTB₄ suggests that LTB₄ may have similar effects on this pathway.

Until very recently, there had been no reports of BLT2 expression in any murine cells or tissues. Tager and Luster (45) indicated by Northern blot analysis that mouse BLT2 expression was not detected in neutrophils, macrophages, T cells, lymph node, spleen, or lung. However, Iizuka et al. (30) recently detected mouse BLT2 expression predominantly in small intestine, followed by skin, and a small amount in spleen and colon. In addition to our data in this study showing expression of both BLT1 and BLT2 in mMCs, we also have evidence from RT-PCR analysis that murine T cells and T cell lines express mRNA for both receptors (K. Lundeen and A. Fourie, unpublished observations).

In the Jurkat human T cell leukemia line, we found that BLT1 receptor protein was expressed on the cell surface, whereas our staining of HMC-1 cells showed predominantly intracellular expression of BLT1. We observed chemotaxis to LTB₄ in HMC-1 cells, despite the apparent intracellular location of BLT1. It is difficult to attribute chemotaxis to a completely intracellular pool of receptors. It is possible that BLT1 would externalize under certain conditions, but we were unable to detect BLT1 translocation to the surface in response to LTB₄ at 100 nM or 1 μM in a time course ranging from 15 min to overnight. There is precedence for the intracellular localization of chemotactic factor receptors. fMLP is a neutrophil chemotactic factor for which the receptor is located in the secretory vesicles of neutrophils. Upon stimulation with fMLP, they are rapidly translocated to the membrane creating a dramatic up-regulation of fMLP receptors on the neutrophil surface (46). Lippert et al. (20) also have reported intracellular expression of the IL-8 receptor CXCR2 in HMC-1 cells, while CXCR1 was expressed on the cell surface. These authors also showed that IL-8 stimulation did not induce translocation of CXCR2 to the cell surface, similar to the apparent lack of translocation of BLT1 receptor to the surface we observed upon stimulation with exogenous LTB₄ of HMC-1 hMCs (data not shown). It is possible that a very small population of BLT1 on the surface may mediate chemotaxis or that there are additional surface BLT1 receptors not recognized by the particular mAb used for BLT1 staining.

Another possibility is that BLT2 plays a predominant role in mediating LTB₄-dependent chemotaxis of MCs. BLT2 is clearly capable of mediating chemotaxis when transfected into cells lacking LTB₄ receptors (24). We have evidence to suggest that, in mouse MCs and hMCs, both BLT1 and BLT2 receptors are functional. Chemotactic activity could be blocked by either a BLT1- or BLT2-selective LTB₄ receptor antagonist. U-75302 is a BLT1 antagonist with IC₅₀ values of 1 μM and >10 mM for BLT1- and BLT2-transfected Chinese hamster ovary cells, respectively. The IC₅₀ values for LY255283 are >10 μM for BLT1 and ∼1 μM for BLT2 (12, 47). In our studies, both antagonists resulted in inhibition of migration at concentrations where they would be selective for BLT1 or BLT2, respectively. We were unable to demonstrate an additive effect of the two antagonists. A similar observation was made by Kitaura et al. (22) where there was very little, if any, additive inhibitory effect when these two antagonists were combined, at 1 μM each, on BMMC migration to IgE-sensitized cells and Ag. One possible explanation for our data is that the two receptors do not function independently. For example, it is known that BLT1 forms homodimers, and there is a growing body of evidence showing that many G protein-coupled receptors form heterodimers, which can alter their signaling and their intracellular transport. Thus, it is possible that BLT1 and BLT2 may function as heterodimers. This could explain why both selective antagonists could maximally inhibit chemotaxis, even at concentrations where they should inhibit LTB₄ binding to only BLT1 or BLT2, respectively.

Further support for functional BLT2 receptors in mBMMCs was the evidence for dose-dependent chemotaxis to concentrations between 1 and 10 μM 12-(S)-HETE, previously shown by Yokomizo et al. (12) to be a selective agonist for BLT2. Although a minimal amount of chemotaxis for BLT1-transfected cells was observed in their study at 10 μM 12-(S)-HETE, the relative amount, compared with that observed for LTB₄ at 100 nM, was essentially negligible (12). In contrast, we have shown significant chemotaxis to 12-(S)-HETE at concentrations as low as 1 μM, reaching similar magnitude at 10 μM 12-(S)-HETE to that observed at 100 nM LTB₄. These results are very similar to those reported for BLT2-transfected cells (12) and support the existence of functional BLT2 receptors in mBMMCs.

SCF is known to be an important growth and differentiation factor that stimulates the production of numerous cytokines and chemokines in mBMMCs (48). Our observation that SCF-stimulated mBMMCs secreted LTB₄ in a dose-dependent manner, similarly to IL-6 and MCP-1 production, may thus have interesting implications in MC biology. Another consequence of SCF stimulation of mBMMCs was a reduction in BLT1 and BLT2 mRNA expression. The SCF-treated cells also exhibited a dramatic decrease in their ability to migrate to LTB₄, while their response to SCF was conversely enhanced. A similar observation in MCs was made by Sawada et al. (49). They found that preincubation of BMMCs and hCBMCs with SCF for even 1 h completely suppressed IgE-mediated chemotaxis to Ag, while their ability to migrate to nerve growth factor remained essentially unchanged. It is of interest that IgE-mediated chemotaxis to Ag has been shown to be mediated, in part, by LTB₄ (22). Sawada et al. (49) concluded that locally produced SCF may have an inhibitory effect on chemotaxis of MCs, contributing to their accumulation and enhancement of functions at the peripheral site of inflammation. Our observations of decreased LTB₄-mediated chemotaxis and the effects of SCF on the LTB₄ receptors seem to support this hypothesis and suggest that the progressive decrease observed by Weller et al. (23) in BLT1 expression and MC chemotaxis during maturation may have been partly induced by SCF in the culture medium.

In conclusion, our study confirms the expression of BLT1 in hMCs and mMCs, and in addition, shows intracellular expression of BLT1 protein in HMC-1 cells. We show, for the first time, the expression of BLT2 in MCs and dose-dependent chemotaxis and ERK phosphorylation in response to the BLT2-selective agonist, 12-(S)-HETE, as evidence of BLT2 function in these cells. Our data with selective antagonists suggest that both receptors play a role in mediating LTB₄-dependent chemotaxis of MCs and raise the possibility of a heterodimeric BLT1/BLT2 receptor form on MCs. We also present evidence for a role of SCF in regulation of LTB₄ receptor expression and thus, LTB₄-dependent MC chemotaxis. Although the mechanisms of MC migration and accumulation at sites of inflammation are still unclear, our demonstration of two different functional chemotactic receptors for LTB₄ in MCs clearly suggests that LTB₄-dependent migration contributes to this process.

We thank Pragnya J. Desai for providing us with mouse BMMCs.

The authors have no financial conflict of interest.