Leukotriene B₄, an activation product of mast cells, is a chemoattractant for their progenitors

Mast cells cultured in IL-3 were analyzed in terms of purity and maturation state for up to 10 wk, with the aim of studying chemotactic responses in cells of minimal maturity at a time when numbers were such that experiments were practicable. Cells were analyzed weekly by flow cytometry; results are shown in some cases for 2-, 6-, and 10-wk-old BMMCs (Fig. 1, c and e). After 1 wk, ∼1% of cells exhibited high c-kit expression, but this rose to ∼50% by 2 wk in culture (Fig. 1 a). At 4 wk, >85% of cells were c-kit⁺, and this was maintained up to 10 wk (Fig. 1 a). The intensity of c-kit expression was high at 1 wk; this increased until 3 wk and was then maintained (Fig. 1 b). Identification of c-kit⁺ cells as mast cells was confirmed by double labeling with antibodies to other mast cell markers, including FcεR1, CD34, T1/ST2 (22), and CD13 (Fig. 1 c). Expression of FcεR1 and TI/ST2 did not change detectably from 2 to 10 wk in culture, whereas CD34 and CD13 did not reach maximum levels until 3–4 wk. Expression of the integrins α₄ and β₇ was observed on 1-wk-old c-kit⁺ BMMCs, which was consistent with the progenitor mast cells previously described as c-kit⁺α₄^hiβ₇⁺ (12). This expression decreased from 1 to 4 wk and remained at low levels up to 10 wk in culture (Fig. 1 d). At 2 wk in culture, the granularity of the cell population measured as side scatter by flow cytometry was low. By 6 wk, some cells showed an increase in side scatter; at 10 wk, most of the cell population exhibited substantial granularity, which was consistent with mature mast cells (Fig. 1 e).

Chemotactic responses of 2-wk-old BMMCs were measured using 96-well microchemotaxis plates. Migrated cells were recovered from the lower chambers after 3 h at 37°C and double stained with monoclonal antibodies to c-kit (mast cells) and Gr-1 (a marker for the major contaminating population). The c-kit⁺ cells migrated in response to SCF when present in the lower chambers. With SCF used as the positive control in further experiments, none of the following were found to be chemotactic for c-kit⁺ cells under the conditions of our experiments: (a) cytokines/growth factors (NGF, TGF-β_1, activin-A, and VEGF); (b) CC chemokines (JE [CCL2], MIP-1α [CCL3], RANTES [CCL5], Eotaxin-1 [CCL11], MIP-3β [CCL19], and MEC [CCL28]); (c) CXC chemokines (KC and MIP-2 [orthologues of the human GRO family], MIG [CXCL9], IP-10 [CXCL10], I-TAC [CXCL11], and SDF-1α [CXCL12]); and (d) peptides (fMLP, bradykinin, and substance P). In the same assays, Gr-1⁺ cells present in the upper chambers migrated in response to KC, MIP-2, MIP-1α, JE, RANTES, and fMLP (e.g., KC at 0.1, 1, and 10 nM induced the migration of 49 ± 29, 118 ± 35, and 219 ± 34 Gr-1⁺ cells, respectively, compared with 9 ± 4 cells for the control; n = 4 experiments).

We investigated the possibility that mature mast cells, on activation, release a chemoattractant for immature cells. 10-wk-old BMMCs were passively sensitized with IgE and activated by the specific antigen, DNP, conjugated to human serum albumin (HSA). Supernatants were then tested in chemotaxis assays on 2-wk-old immature and 10-wk-old mature BMMCs. Supernatants from nonactivated cells induced negligible chemotaxis; however, supernatants from activated cells induced pronounced migration of 2-wk-old cells with a bell-shaped concentration–response curve and a maximal response at a 1:9 dilution (Fig. 2 a). In comparison, only very weak responses were obtained using 10-wk-old cells in the chemotaxis assay (Fig. 2 b).

Supernatants from activated and nonactivated 10-wk-old BMMCs were applied to reverse phase HPLC and fractions were collected for chemotaxis assay on 2-wk-old BMMCs. Virtually all the chemotactic activity from activated mast cell supernatants was present in fraction 20 (41% acetonitrile [ACN]), with a small amount in fraction 21 (Fig. 3). No chemotactic activity was detectable in the other fractions or in any of the fractions from the nonactivated cell supernatants (Fig. 3). Known products of mast cell activation were subsequently run through the same column as authentic standards and their retention times were recorded by absorbance at 280 nm (for LTs) or 200 nm (for non-LT standards). LTB₄ eluted with a retention time indistinguishable from that of the activated mast cell–derived activity (fraction 20), in contrast to the peptido-LTs, LTC₄ (fraction 17/18), LTD₄ (fraction 21), and LTE₄ (fraction 21/22). As shown in Fig. 3, other mast cell activation products—PGD₂, histamine, and 5-hydroxytryptamine—eluted much earlier in the gradient than the mast cell–derived chemotactic activity, as did the 20-hydroxy metabolite of LTB₄. Thus, the chemotactic activity in the supernatants of activated mast cells coeluted with authentic LTB₄ and not with any other known product of acute activation. Furthermore, ELISA of the HPLC fractions of activated mast cell supernatants showed that the majority of LTB₄ immunoreactivity was in fraction 20 (10.6 nM) with a smaller amount (4.6 nM) in fraction 21.

The generation of chemotactic activity for 2-wk-old BMMCs was abolished by pretreating mature mast cells with MK-886, an inhibitor of 5-lipoxygenase (5-LO) activating protein before activation (Fig. 4 a). Supernatants were assayed on 2-wk-old BMMCs as before. MK-886 at 100 nM reduced chemotactic activity substantially, whereas 1 μM gave complete inhibition. Supernatants from untreated activated mast cells contained 46 nM LTB₄ (3.1 ng/10⁶ cells), measured by ELISA, which was reduced to 4 nM and <0.06 nM with 100 nm and 1 μM MK-886, respectively.

The chemotactic activity of lipid mediators tested on 2-wk-old BMMCs was compared with that of SCF (Fig. 4 b). LTB₄ had a high potency and efficacy. SCF had a much lower efficacy and the peptido-LTs C₄, D₄, and E₄ were inactive. In addition, PGD₂ (Fig. 4 b), 5-hydroxytryptamine, histamine, and platelet activating factor (unpublished data) were also inactive as chemotactic agents on these cells. 20-hydroxy-LTB₄, a metabolite of LTB₄, exhibited chemotactic activity, but was 100-fold less potent than its parent (unpublished data). The c-kit⁺ cells migrated in response to both LTB₄ and SCF when the agents were present in the lower chambers, but not when added with the cells above the filter, characteristic of a chemotactic response requiring a gradient, rather than a chemokinetic response, representing increased random movement (Fig. 4, c and d). As well as being active as a chemoattractant for c-kit⁺ cells cultured for 2 wk in IL-3, LTB₄ was also active on cells cultured for 2 wk in a combination of IL-3 and SCF (unpublished data).

The chemotactic effect of LTB₄ was examined on BMMCs at different stages of maturity. LTB₄ was highly potent and efficacious on immature mast cells cultured for 2 wk. By 6 wk, chemotactic responses were very weak and remained so at 10 wk (Fig. 5 a). In contrast, no notable decrease in responsiveness to SCF was observed as the cells matured (Fig. 5 b).

The loss of responsiveness to LTB₄ with mast cell maturation was associated with a reduction in mRNA expression of BLT1. Cells cultured for 2, 6, and 10 wk were purified by positive selection using CD117 (c-kit) immunomagnetic beads. Fig. 6 a shows, using RT-PCR, the decrease in mRNA as the mast cells matured. Using real-time PCR analysis, 2-wk-old BMMCs were shown to express fourfold more BLT1 mRNA than 6-wk-old cells, and 10-fold more than 10-wk-old cells (Fig. 6 b).

Fresh cell suspensions of mouse femoral bone marrow were tested in 96-well chemotaxis assays using LTB₄ and SCF at concentrations effective on cultured BMMCs (Fig. 4 b). After 3 h of incubation, migrated cells were recovered from the lower chambers and cultured for 2 wk in IL-3, IL-9, SCF, and TGF-β1, a medium favoring rapid maturation toward a mucosal mast cell phenotype (23, 24) to enable quantitation of mast cell numbers by their content of mast cell–specific proteases. Cells were then lysed by freeze-thawing and the released murine mast cell proteases (mMCP)-1 and -2 were quantitated using ELISA. Although the number of mast cell progenitors in the bone marrow was low (25) and, thus, the migrating cells were few in number, this system provided amplification to enable a measurement in proportion to the number of migrated mast cell progenitors. LTB₄ induced a concentration-related increase in both mMCP-1 and -2, with a maximal effect, seen at 10 nM (Fig. 7, a and b). Interestingly, SCF failed to induce a detectable response; no proteases were detected nor were any cells observed microscopically in the cultures before the lytic step (Fig. 7, a and b). Chemokines acting via CCR3, Eotaxin-1, and CXCR2, KC and MIP-2, also failed to demonstrate a chemotactic effect in this system (unpublished data).

2-wk-old 5-chloromethylfluorescein diacetate (CMFDA)–labeled BMMCs were either negatively selected by immunomagnetic bead separation to remove the contaminating Gr-1⁺ cells, or c-kit⁺ BMMCs were positively selected. The isolated cells were injected intravenously into mice, followed by i.d. injections of 150 pmol LTB₄ or vehicle control at separate sites in the dorsal skin of each mouse. Animals were killed 1 h after i.d. injections, and the skin was removed and processed for microscopy. Negatively selected CMFDA-labeled BMMCs accumulated at the LTB₄ injection sites, compared with the few cells present in the vehicle control injection sites (Fig. 8, a–c; P < 0.05). The same was seen for positively selected CMFDA-labeled BMMCs (Fig. 8 d), with 23.2 ± 4.7 cells/50 mm² accumulating at the LTB₄ site compared with 9.4 ± 1.9 cells/50 mm² accumulating at the vehicle control site (n = 5, P < 0.05). The majority of CMFDA-labeled BMMCs observed were extravascular. Fig. 8 (b–d) shows representative images of CMFDA-labeled BMMCs detected in the skin by confocal microscopy.

Mast cells were cultured from human umbilical cord blood using an established method (26). A uniform population of c-kit⁺ cells was evident at ∼4 wk that exhibited the characteristics of immature mast cells, such as detectable but low FcεR1 expression, low side scatter, and lack of specific markers for other cell types, including CD14 (monocytes), CD16 (neutrophils), and CD123 (basophils). In subsequent weeks of culture, a population of mast cells with a more mature phenotype was also observed, with higher expression of both c-kit and FcεR1, but no expression of CD14, CD16, or CD123, and increased side scatter, indicating increased granularity. By 8–9 wk, only the more mature c-kit^high FcεR1^high mast cell population was present, also in agreement with Ochi et al. (26). The two populations are illustrated in Fig. 9 (a–e) in a representative 7-wk-old culture. Chemotaxis of the cultured cells was investigated between 5 and 7 wk when both the c-kit⁺ and c-kit^high mast cell populations were present. The proportions of the two populations varied between cord blood preparations and over the 5–7-wk period, from 32 to 80% (56.2 ± 8.3%, n = 5) c-kit⁺ and 20 to 68% (43.8 ± 8.3%, n=5) c-kit^high cells. Chemotaxis experiments were performed in 96-well plates using the mixed cultured cells, and the numbers of migrated cells from each of the two populations were determined according to intensity of c-kit expression by flow cytometry. LTB₄ was a highly potent and efficacious chemoattractant for the immature c-kit⁺ mast cell population (Fig. 9 f), but the mature c-kit^high mast cells were unresponsive (Fig. 9 g). In contrast, the mature cells responded to SCF, albeit with low efficacy (Fig. 9 g), whereas the immature cells showed no response (Fig. 9 f).

Cell migration is of critical importance at several stages during the life history of the mast cell. Committed mast cell progenitors, derived from pluripotential stem cells in the bone marrow stroma, must first transit toward and through the sinus endothelium to gain access to the general circulation. The progenitors then attach to and migrate through the microvascular endothelium of peripheral tissues. There, under the influence of local factors, the progenitors begin to mature and can migrate away from blood vessels toward their final destination, where they exhibit a phenotype characteristic of their location. We are particularly interested in mucosal mast cells whose numbers increase markedly in apposition to epithelial surfaces in response to allergic stimuli, a process associated with the migration of cells from the lamina propria and thought to include recruitment from microvessels in this region.

To investigate potential chemoattractant molecules involved in progenitor recruitment from the blood, we cultured mouse bone marrow cells with IL-3 and measured their responses in chemotaxis chambers using cells as immature as possible at a time when c-kit⁺ cells were of sufficient numbers for study. Under the conditions of our experiments, these immature c-kit⁺ cells were remarkably unmoved by a range of known chemotactic agents. The cells responded chemotactically to SCF but not to other factors tested, such as chemokines acting through CXCR2 or CCR3, receptors previously described on mast cells (26–28). We were surprised at this lack of responsiveness, which contrasts with many reports of chemotactic responses of mast cells to a range of mediators, including chemokines (27, 29–33). We believe that this difference is largely attributable to the stage of maturity of the cells as we concentrated on very immature mast cells, but differences in methodology may also play a part.

In light of the mast cell hyperplasia observed in sensitized individuals exposed to allergen, we investigated whether activation of mature mast cells by cross-linking their FcεR1 results in the production of a chemoattractant for immature mast cells. We demonstrate that immature mast cells cultured for 2 wk from mouse bone marrow are highly responsive in vitro to the chemotactic effect of LTB₄ released on activation of mature mast cells. In addition, immature cells were able to accumulate in response to intradermally injected LTB₄ in vivo. The results shown in Fig. 7 indicate that mast cell progenitors in fresh bone marrow also respond highly to LTB₄ in vitro, although it is not possible from this study to determine precisely when the receptor is first expressed. The responsiveness to LTB₄ is transient and is lost as the cells mature, correlating with the loss of mRNA for BLT1. These results are consistent with those in a recent study showing a weak chemotactic effect of LTB₄ on BMMCs (chemotactic index of ∼2) as cells of 4–6 wk in culture were used (34). Interestingly, results obtained with human CBMCs were very similar to those obtained with mouse BMMCs. Human immature mast cells were highly responsive to LTB₄, an effect lost on maturation.

The lack of responses of mature mast cells to LTB₄ may explain why this mediator has previously been identified only as a product of this cell rather than as its chemoattractant. Human and mouse mast cells have been reported to produce between 1 and 4 ng LTB₄/10⁶ cells after FcεR1 cross-linking (35–37). Accordingly, we found that our activated BMMC supernatants contained 3.1 ng LTB₄/10⁶ cells, and nonactivated mast cells produced no detectable LTB₄. BLT1 expression has been investigated previously on mast cells in tissues during allergic reactions, but was reported to be very low (38). In contrast, cysteinyl LT receptors are expressed on mast cells in tissues during allergic disease and by human CBMCs (38–40). The observation that immature mast cells respond strongly to a major activation product of mature cells may be more than a coincidence. The evanescent responsiveness of mast cells to LTB₄, correlating with transient BLT1 mRNA expression, strongly implies an important function at a critical stage in their life history. We suggest that LTB₄ produced by activated mast cells in tissues is able to recruit circulating mast cell progenitors and contribute to mast cell hyperplasia at sites of allergic reactions (41, 42). Furthermore, LTB₄ may contribute to the movement of immature mast cells from the lamina propria to the epithelium of affected tissues. Once at their final location, the mature cells then become insensitive to LTB₄ because of the loss of BLT1. The reason for this may be to ensure that mature mast cells do not congregate exclusively at areas of focal allergen stimulation, leaving adjacent areas of epithelium depleted.

Recruitment of mast cell progenitors would also be expected to result from LTB₄ production by cells other than mast cells, e.g., neutrophils. Irrespective of the cellular source of LTB₄, progenitor recruitment would only result in population with mature cells if the appropriate maturation factors were present. The general mechanisms we describe may also operate in response to foreign organisms; e.g., helminths and microorganisms. It will be of interest to investigate the importance of these mechanisms in disease models in which mast cells have a critical role. One example is a mouse autoimmune arthritis model in which mast cells are obligatory (43). LTB₄ may drive both the recruitment of mast cell progenitors and neutrophils in this model. Similarly, a mouse model of experimental allergic encephalomyelitis is described as mast cell dependent (44–46), and an LTB₄ receptor antagonist has been shown to suppress disease pathology in this model (47).

Chemoattractants generated at sites of inflammation serve an important function in recruiting inflammatory cells. These same mediators, having gained access to the general circulation, can act remotely and release leukocytes from the bone marrow by establishing a chemotactic gradient across the sinus endothelium (15), as exemplified by eotaxin-1, that can release both mature eosinophils and their progenitors (14, 16). The experiments described here using freshly isolated bone marrow cell suspensions demonstrate that mast cell progenitors in the marrow can respond to low concentrations of LTB₄. Thus, circulating LTB₄ may induce progenitor release in vivo. Although c-kit, the receptor for SCF, is a classical marker for mast cells, it may be important that fresh mouse mast cell progenitors and the immature human CBMCs were shown to be unresponsive to SCF as a chemoattractant. SCF clearly has a vital role in mast cell development, but we suggest that this factor may not have a role in releasing mast cell progenitors from the bone marrow or their recruitment into tissues. Indeed, SCF gradients established in the bone marrow as the result of local production in the stroma would be in the wrong direction to mediate movement of cells into sinuses by chemotaxis, although theoretically such gradients could be reversed by elevation of circulating SCF as a consequence of production in other tissues. The observation that the mouse bone marrow cells cultured for 2 wk with IL-3 did respond chemotactically to SCF (albeit with relatively low efficacy) suggests that these cells are a little more advanced in maturity than the bone marrow progenitor mast cells; these 2-wk-old cells retain responsiveness to LTB₄, but now have the c-kit receptor coupled to the locomotor machinery. Both mouse and human mature mast cell did respond chemotactically to SCF, so it is possible that this mediator is involved with relocating mature or maturing mast cells within tissues once the progenitors have been recruited. This would be an additional function to the critical role of SCF in mast cell proliferation and maturation.

Thus, LTB₄ may serve the function of an autocrine mediator produced by activated mast cells, mediating the supply of their progenitors to the tissue. This may be important in the mast cell hyperplasia that is overt in allergic reactions and responses to helminth parasite infections. If this is the case, it would follow that interventions resulting in an increase in mast cell activation would increase tissue mast cell numbers. Such an increase in numbers has been observed in a recent study after deletion of RabGEF1, which is a component of a negative feedback pathway that makes mast cells more readily activated (48). Rabgef1^−/− mice showed spontaneous inflammation in the skin associated with mast cell degranulation and increased mast cell numbers (48). Alternatively, interventions aimed at suppressing the production or actions of LTB₄ e.g., using 5-LO inhibitors or BLT1 antagonists, may suppress mast cell hyperplasia and, thus, be of benefit in allergic diseases, such as allergic rhinitis and asthma.

BSA, PBS, 2-ME, Histopaque 1077, human TGF-β1, FMLP, bradykinin, Substance P, anti–DNP IgE (clone SPE-7), HSA-DNP, and 0.22-μm spin filters were obtained from Sigma-Aldrich. TFA (peptide synthesis grade) was obtained from Rathburn, and ACN (far UV grade for HPLC) was obtained from Merck. RPMI 1640, DMEM, Hepes, penicillin, streptomycin, l-glutamine, FBS, fungizone, sodium pyruvate, and nonessential amino acids were purchased from Invitrogen. PE–rat anti–mouse CD117 (2B8), FITC–rat anti–mouse IgE (R35-72), FITC–rat anti–mouse CD34 (RAM 34), FITC–rat anti–mouse Gr-1 (RB6-8C5), FITC–rat anti–mouse CD13 (R3-242), rat anti–mouse CD16/32 (2.4G2), PE–mouse anti–human CD117 (YB5.B8), FITC–mouse anti–human CD14 (M5E2), FITC–mouse anti–human CD16 (3G8), Cy5–mouse anti–human CD123 (9F5), biotin–rat anti–mouse α4 (R1-2), FITC–rat anti–mouse β7 (M293), the appropriate isotype controls, and FACSFlow were obtained from BD Biosciences. FITC–rat anti–mouse T1/ST2 (2.4G2) was purchased from Morwell Diagnostics GmbH, and polyclonal rabbit anti–human FcεR1 was obtained from Upstate Ltd. Alexa 568–conjugated GSL-1 isolectin B4, AlexaFluor 488 goat anti–rabbit IgG (A-11008), TOPRO-3, and Cell Tracker green CMFDA were purchased from Molecular Probes. All eicosanoids were obtained from Cayman Chemical. Murine IL-3, SCF, VEGF, human SCF, IL-6 and IL-10, and all chemokines were purchased from PeproTech. Murine IL-9 and human Activin A were obtained from R&D Systems. Murine NGF was purchased from Promega. All microbeads were obtained from Miltenyi Biotec. RNeasy mini kit, HotStar Taq DNA Polymerase, and Omniscript Reverse Transcriptase were obtained from QIAGEN. TaqMan universal PCR mastermix, BLT1, GAPDH, and 18S-specific primers (Assays-on-Demand), and the sequence detection system were (model ABI Prism 7700) purchased from Applied Biosystems.

Murine BMMCs were obtained by culture of BALB/c femoral BM cells at 5 × 10⁵ cells/ml in RPMI 1640 with 10% FBS, 50 μM 2-ME, 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 5 ng/ml murine IL-3. BMMC cultures were maintained for up to 10 wk at 37°C in 5% CO₂. The flask was replaced weekly for up to 3 wk to remove adherent cells, with weekly replacements of half the medium throughout. An aliquot of cells was taken at weekly intervals for immunofluorescence staining. In some experiments, BM cells were cultured with 15 ng/ml SCF in addition to IL-3 (28).

Human mast cells were derived from heparin-treated umbilical cord blood obtained after routine deliveries, according to the method of Ochi et al. (26), with minor modifications. Whole blood was diluted 1:3 with RPMI 1640 and centrifuged through Histopaque 1077 to isolate the mononuclear cells. Residual erythrocytes were removed by hypotonic lysis, and the mononuclear cells were resuspended in RPMI 1640 with 10% FBS, 2 mM l-glutamine, 0.1 mM nonessential amino acids, 100 U/ml penicillin, and 100 μg/ml streptomycin. The cells were seeded at 10⁶ cells/ml and cultured in the presence of 100 ng/ml SCF, 50 ng/ml IL-6, and 10 ng/ml IL-10. Cells were cultured for up to 8 wk, and the culture medium and flask were replaced weekly. An aliquot of cells was taken at weekly intervals for immunofluorescence staining.

BMMCs or CBMCs were resuspended at 2–5 × 10⁶ cells/ml in ice-cold staining buffer (PBS, 0.1% sodium azide, and 0.5% BSA) containing either anti–mouse CD16/32 or goat IgG, respectively, to block Fc receptors. After a 10-min incubation on ice, specific mAbs were added at saturating concentrations and incubated for a further 20–30 min in the dark. Cells were then washed, resuspended in FACSFlow containing TOPRO-3 (1:10,000, to identify dead cells), and analyzed on a flow cytometer (model FACSCalibur; BD Biosciences). For FcεR1 staining, after Fc receptor blocking, BMMCs were incubated with anti-DNP IgE and then FITC-anti–mouse IgE. CBMCs were incubated with polyclonal rabbit anti-FcεR1, followed by AlexaFluor 488 goat anti–rabbit IgG (1:1,000).

Chemotactic responses of BMMCs were examined using 96-well chemotaxis plates (ChemoTx) with 5-μm pore size polycarbonate filters. 20 μl of BMMCs were added to the top wells at 2 × 10⁶ cells/ml in RPMI 1640 with 25 mM Hepes and 0.1% BSA (assay buffer), and 30 μl of agonist or buffer was added to the lower wells. After a 3-h incubation at 37°C, migrated cells were removed, the wells washed with 25 μl of assay buffer, and stained with PE-anti–mouse c-kit (1:130) and FITC-anti–mouse Gr-1 (1:160) for 10 min at 22°C to distinguish the mast cells from nonmast cells. 100 μl FACSFlow containing 1:10,000 TOPRO-3 was added to the cells before counting by flow cytometry. In some assays, the activation products from mature BMMCs were added to the lower wells. To investigate chemokinesis, BMMCs were mixed with agonist before addition to the upper assay chamber and migration to the buffer was determined. Assays were performed in duplicate, and results were expressed as the absolute number of cells migrating.

Chemotaxis of freshly isolated BM cells was measured in a similar way to the cultured BMMCs, except that 20 μl BM cells at 10⁷ cells/ml were added to the upper wells of chemotaxis plates immediately after isolation from femurs and aspiration through a 19-gauge needle. Migrated cells were removed and cultured for up to 14 d in DMEM with 10% FBS, 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, 2.5 μg/ml fungizone, 1 mM sodium pyruvate, 1 ng/ml TGF-β1, 1 ng/ml murine IL-3, 5 ng/ml mIL-9, and 50 ng/ml mSCF (24). Assays were performed in quadruplicate and, after cell lysis, cells were enumerated by specific ELISAs for mMCP-1 and -2 (49, 50).

2-wk-old BMMCs (2 × 10⁷ cells/ml) were labeled with 25 μM Cell Tracker Green CMFDA for 40 min at 37°C. Cells were purified by immunomagnetic separation by either negative selection using Gr-1 FITC followed by anti-FITC microbeads, to remove Gr-1⁺ contaminating cells, or by positive selection using anti-CD117 (c-kit) microbeads. The isolated cells (10⁶ cells/BALB/c mouse) were administered by tail vein injection 5 min before i.d. injections into dorsal skin of 50 μl LTB₄ in Tyrode's solution (150 pmol/site) and vehicle control in each mouse. Mice were killed 1 h after i.d. injection. The skin was removed and fixed in 4% paraformaldehyde for 24 h. A quarter segment from an 8-mm-diameter biopsy (Schuco International) was whole mounted in a hard+set mounting medium (VECTASHIELD; Vector Laboratories) for examination by standard epifluorescence microscopy. The number of CMFDA⁺ cells in each one fourth skin biopsy was counted. A second quarter segment of each skin biopsy was counterstained with 10 μg/ml Alexa 568–conjugated GSL-1 isolectin B4, which is a marker for mouse endothelial cells and neutrophils (unpublished data). This tissue was then whole mounted and examined by confocal microscopy using a TCS NT system (Leica). Home Office guidelines for animal welfare based on the Animals (Scientific Procedures) Act 1986 were strictly observed.

CBMCs cultured for 5–7 wk were used to measure chemotaxis of human mast cells in the same way as BMMCs, except that 20 μl of cells at 10⁶ cells/ml were used. The mast cells were identified by staining with PE-anti–human c-kit (1:100) and FITC-anti–human CD14 (1:100) for 10 min at 22°C before diluting with FACSFlow containing TOPRO-3 and counted. Assays were performed in duplicate, and the migration of c-kit⁺ or c-kit^high mast cells was expressed as a percentage of the total c-kit⁺ or c-kit^high cells added to the wells, respectively.

Mature (10 wk) BMMCs (5 × 10⁶ cells/ml) were incubated with 10 μg/ml anti-DNP IgE for 1 h at 37°C. Cells were washed in RPMI 1640 with 0.1% BSA, resuspended in HSA-DNP at 30 ng/ml, and incubated for 2 h at 37°C. Some cells were left untreated, or treated with either anti-DNP IgE or HSA-DNP alone to control for basal secretion of chemoattractant mediators. The supernatants were collected by centrifugation through a 0.22-μm filters. Supernatants were either used undiluted (NEAT) or diluted at 1:3, 1:9, or 1:27 into RPMI 1640 with 0.1% BSA. In some assays, cells were preincubated with MK-886 (5-LO inhibitor; Alexis Biochemicals) for 10 min at 37°C before activation, along with DMSO-only controls (vehicle) and cells that were treated with neither MK-886 nor DMSO.

400 μl of activated or nonactivated 10-wk-old mast cell supernatants, adjusted to pH 2.0 with 20% TFA and further diluted in 0.08% TFA, pH 2.0, or 500 pmol of eicosanoid standards, 400 nmol histamine, and 40 nmol 5-hydroxytryptamine in 0.08% TFA were applied to a μRPC C2/C18 PC 3.2/3 column (Pharmacia Smart System; GE Healthcare). The flow rate was one column volume (240 μl)/min in 0.08% TFA, and the gradient was started 15 min after sample injection: 0–30% ACN at 5% ACN/min, 30–50% ACN at 1% ACN/min, and 50–100% ACN at 5% ACN/min. Fractions were collected from the start of the gradient (2-min fractions for 0–10 min, 0.5-min fractions for 10–22 min, and 2-min fractions for the remainder of the gradient into carrier BSA), freeze-dried, and dissolved in 400 μl RPMI 1640 with a final concentration of 0.1% BSA for bioassay and ELISA.

LTB₄ was quantified using a specific ELISA kit (R&D Systems) with a detection limit of 60 pM.

2-, 6-, and 10-wk-old BMMCs were isolated with anti–mouse CD117 (c-kit) immunomagnetic beads and total RNA was extracted using an RNeasy mini kit. First strand cDNA was generated from 0.5 μg of total RNA using Omniscript Reverse Transcriptase with oligodeoxythymidine_12-18 primers. PCR was performed using primers for mouse BLT1 (available from GenBank/EMBL/DDBJ under accession no. AF044030) 5′-TAAAGTCTTCCATCTGCTCTTCGAA-3′ and 5′-ACTTCGAAGACTCAGGAATGGTGGA-3′ or a using GAPDH control primer set. PCR products were generated using HotStarTaq DNA Polymerase and resolved on a 1% agarose gel containing 0.025 μg/ml ethidium bromide.

Total RNA was extracted from purified BMMCs as described in the RT-PCR section and cDNA was synthesized using the SuperScript First-Strand Synthesis System (Invitrogen) with both oligodeoxythymidine and random hexamer primers. 25-μl PCR procedures were performed in triplicate using the sequence detection system (model ABI Prism 7700), with TaqMan universal PCR mastermix and BLT1, GAPDH, and 18S specific primers. Samples were standardized to 18S, and BLT1 mRNA was quantified using the comparative threshold for detection method (http://appliedbiosystems.com).

The data are expressed as mean ± SEM. Statistical differences between groups of three or more was estimated using a Kruskal Wallis test with pairs of groups compared with Dunn's multiple comparison test. For groups of two, the Mann-Whitney test was used. P < 0.05 was considered significant.

We thank Miss Gill Martin for culturing BMMCs, Miss Lorraine Lawrence for histopathology, and Dr. Tricia Denning-Kendall and the midwives at Southmead Hospital, Bristol for cord blood collection.

This work was supported by grants from the Wellcome Trust (038775 and 060312), Asthma UK (Grant 02/040 and Chair support), and the Norman Salvesen Emphysema Research Trust (R36177). Ph.D. studentships were funded by Novartis and Union Chemique Belge.

The authors have no conflicting financial interests.