Chemotactic action of prostaglandin E2 on mouse mast cells acting via the PGE2 receptor 3

Abstract

Mast cells are long-lived cells that are principally recognized for their effector function in helminth infections and allergic reactions. These cells are derived from pluripotential hematopoietic stem cells in the bone marrow that give rise to committed mast cell progenitors in the blood and are recruited to tissues, where they mature. Little is known about the chemotactic signals responsible for recruitment of progenitors and localization of mature mast cells. A mouse model was set up to identify possible mast cell progenitor chemoattractants produced during repeated allergen challenge in vivo. After the final challenge, the nasal mucosa was removed to produce conditioned medium, which was tested in chemotaxis assays against 2-wk murine bone marrow-derived c-kit+ mast cells (BMMC). A single peak of chemotactic activity was seen on reverse-phase HPLC with a retention time and electrospray mass spectrum consistent with prostaglandin E2 (PGE2). This lipid was found to be a highly potent chemoattractant for immature (2-wk) and also mature (10-wk) BMMC in vitro. Fluorescently labeled 2-wk c-kit+ BMMC, when injected intravenously, accumulated in response to intradermally injected PGE2. Analysis using TaqMan showed mRNA expression of the PGE2 receptors 3 (EP3) and 4 (EP4) on 2- and 10-wk BMMC. Chemotaxis induced by PGE2 was mimicked by EP3 agonists, blocked by an EP3 receptor antagonist, and partially inhibited by a MAPKK inhibitor. These results show an unexpected function for PGE2 in the chemotaxis of mast cells.

Fig. 1.

Purification of mast cell chemoattractant activity in MCM from OVA-sensitized/challenged mice. (a) MCM from sensitized/OVA-challenged (■) or sensitized/saline-challenged (▴) mice was tested for chemoattractant activity against 2-wk BMMC either undiluted or at 0.33 or 0.1 dilution. After 3 h, migrated cells were removed and c-kit⁺ BMMC counted. Data are mean ± SEM for n = 3 (∗, P < 0.05). No cells migrated to assay buffer alone. The total number of cells migrating varied between cell batches, but all experiments were internally controlled. (b) Each fraction eluted from the final reversed-phase HPLC column was assayed for chemoattractant activity against 2-wk BMMC. Arrows indicate the elution of subsequently applied LTB₄, 20-hydroxy LTB₄, PGE₂, and PGD₂ standards. (c) Concentration of PGE₂ and PGD₂ in fractions containing mast cell chemotactic activity.

Fig. 2.

PGE₂ is chemotactic for 2-wk BMMC in vitro and in vivo. (a) PGE₂ (0.1–1,000 nM) was assayed for chemoattractant activity against 2-wk (■) and 10-wk (▴) BMMC. Migrated cells were removed and c-kit⁺ BMMC counted. Data are mean ± SEM for n = 4 (P > 0.05). (b) PGE₂ was added to either the top (□) or bottom (■) of a chemotaxis plate, and 2-wk BMMC were added to the top. Migrated cells were removed and c-kit⁺ BMMC counted. Data are mean ± SEM for n = 5. No cells migrated to assay buffer alone. (c) Increased numbers of 2-wk CMFDA-labeled BMMC in skin injected with 0.15 or 1.5 nmol of PGE₂ and 0.15 nmol of LTB₄ compared with vehicle control (Tyrodes). Data are mean ± SEM for n = 7 mice, ∗, P < 0.05. (d) Representative confocal micrograph of whole-mount mouse skin from the 1.5-nmol PGE₂ injection site showing extravascular CMFDA⁺ BMMC (arrows). CMFDA-labeled c-kit⁺ BMMC are green in skin samples counterstained with GSL-1 Isolectin B4, staining endothelium, and neutrophils red. Inset, ×4.

Fig. 3.

The chemoattractant activity of PGE₂ acts via the EP3 receptor. (a) Chemotaxis of 2-wk BMMC to PGE₂ in the absence (■) or presence (□) of 1 μg/ml pertussis toxin. Data are mean ± SEM for n = 4. (b) Selective agonists; PGE₁ alcohol (■, EP2, and EP4), 17pt PGE₂ (▴, EP1), Sulprostone (▾, EP1 and EP3), Misoprostol (♦, EP2, EP3, and EP4) were added to the bottom wells of a chemotaxis plate, and 2-wk BMMC were added to the top. Data are mean ± SEM for n = 3 cell cultures. No cells migrated to assay buffer alone. (c) Real-time PCR analysis of EP1, EP2, EP3, and EP4 receptor mRNA expression in 2- and 10-wk BMMC. Data are mean ± SEM for n = 3 using RNA isolated from separate cell cultures.

Fig. 4.

Confirmation that the EP3 receptor mediates mast cell migration to PGE₂. Chemotaxis of 2-wk BMMC to either PGE₂ (100 nM, ■) or SCF (1 nM, ▴) in the presence of either an EP3 antagonist (L826,266) (a) or an EP4 antagonist (L161,982) (b). Data are mean ± SEM for n = 4. (c) mRNA expression of the EP3 receptor α, β, and γ isoforms in 2-wk BMMC were analyzed by RT-PCR and compared with the housekeeping gene GAPDH. The 400-bp PCR product for the β isoform was sequenced and confirmed as EP3 receptor β, and the additional 500-bp PCR product was sequenced and determined to be EP3 receptor α product. The 500-bp PCR product was probably synthesized because of the close proximity of the EP3 receptor β reverse primer to the splicing site between the α and β isoforms, producing the extra PCR product of 500 bp. Data are mean ± SEM for n = 3 using RNA isolated from Applied Biosystems separate cell cultures. (d) Chemotaxis of 2-wk BMMC to 100 nM PGE₂ in the presence of 10 μM LY294002, a phosphatidylinositol 3-kinase inhibitor (LY); SB202190, a p38MAPK inhibitor (SB); or U0126, a MAPKK inhibitor (U). Data are mean ± SEM for n = 6, ∗, P < 0.05.