Prostaglandin D2 as a Mediator of Allergic Asthma

See allHide authors and affiliations

Science  17 Mar 2000:
Vol. 287, Issue 5460, pp. 2013-2017
DOI: 10.1126/science.287.5460.2013


Allergic asthma is caused by the aberrant expansion in the lung of T helper cells that produce type 2 (TH2) cytokines and is characterized by infiltration of eosinophils and bronchial hyperreactivity. This disease is often triggered by mast cells activated by immunoglobulin E (IgE)–mediated allergic challenge. Activated mast cells release various chemical mediators, including prostaglandin D2 (PGD2), whose role in allergic asthma has now been investigated by the generation of mice deficient in the PGD receptor (DP). Sensitization and aerosol challenge of the homozygous mutant (DP−/−) mice with ovalbumin (OVA) induced increases in the serum concentration of IgE similar to those in wild-type mice subjected to this model of asthma. However, the concentrations of TH2 cytokines and the extent of lymphocyte accumulation in the lung of OVA-challenged DP−/− mice were greatly reduced compared with those in wild-type animals. Moreover, DP−/− mice showed only marginal infiltration of eosinophils and failed to develop airway hyperreactivity. Thus, PGD2 functions as a mast cell–derived mediator to trigger asthmatic responses.

The chronic airway inflammation associated with asthma is characterized by infiltration of both T lymphocytes that produce TH2 cytokines and eosinophilic leukocytes (1). Large numbers of eosinophils and high concentrations of TH2 cytokines, such as interleukin-4 (IL-4), IL-5, and IL-13, are thus present in both the airway and bronchial alveolar lavage (BAL) fluid of individuals with asthma. The importance of TH2 cytokines in asthma has been demonstrated in animal models, in which either disruption of the genes encoding these proteins or their antibody-mediated neutralization prevents eosinophilia and attenuates various pathological changes, such as airway hyperreactivity, associated with this condition (2). Symptoms of asthma are induced by exposure to specific antigens. Affected individuals produce IgE antibodies to these antigens, and antigen-antibody–mediated cross-linking of the IgE receptors on the surface of mast cells and the consequent activation of these cells are suggested to be important in initiation and development of bronchial asthma (3).

Activated mast cells produce a variety of chemical mediators, one of which, prostaglandin D2 (PGD2), is the major cyclooxygenase metabolite of arachidonic acid produced by these cells in response to antigen challenge (4). PGD2 is released in large amounts during asthmatic attacks in humans, and it has been proposed as a marker of mast cell activation in asthma (5). However, the role of PGD2 in allergic asthma remains unclear. PGD2 elicits its biological actions through interaction with the PGD receptor (DP), a heterotrimeric GTP-binding protein–coupled, rhodopsin-type receptor that is specific for this PG (6). To clarify the role of PGD2 in asthma, we generated and characterized mice deficient in DP.

The mouse DP gene was disrupted by insertion of a neomycin resistance gene into the first coding exon (exon 2) (Fig. 1A), and mice chimeric for the resulting mutant allele were generated and mated with C57BL/6 animals to produce mice heterozygous for this allele (7). Interbreeding of the heterozygotes produced homozygous mutant (DP−/−) mice (Fig. 1B) in a ratio expected from Mendelian inheritance, indicating that the lack of a functional DP gene does not result in fetal death. Reverse transcription and polymerase chain reaction (RT-PCR) analysis confirmed a higher molecular weight transcript corresponding to the Neo-inserted DP mRNA in the homozygous mutants (Fig. 1C). Loss of functional DP protein in these animals was confirmed by a bioassay with tracheal smooth muscle (8). Whereas PGD2 and the DP agonist BW245C [5(6-carboxyhexyl)-1-(3-cyclohexyl-3-hydroxypropyl)hydantoin] (9) each induced relaxation of tracheal smooth muscle from wild-type mice, no such effect was apparent in muscle derived from DP−/− animals (Fig. 1D). The DP−/− mice showed no apparent behavioral, anatomic, or histological abnormalities during l year of observation under specific pathogen-free conditions. To exclude possible effects of genetic background, we backcrossed DP−/− animals with C57BL/6 mice for five generations. The resulting heterozygous mice were intercrossed, and progenies of the resulting wild-type and DP−/− littermates (N5) were subjected to the analysis below (10).

Figure 1

Disruption of the mouse DP gene. (A) Strategy for targeted disruption. Organization of the DP gene, construction of the targeting vector (TK, thymidine kinase gene; Neo, neomycin resistance gene), and structure of the targeted genome are shown. Restriction sites are indicated: N, Nhe I; and X, Xba I. (B) Southern blot analysis. Genomic DNA from newborn littermates of heterozygote intercrosses was digested with Xba I, and the resulting fragments were subjected to analysis with a Nhe I–Xba I fragment of the genomic DNA as a probe. The positions of 3.6-kb (wild-type) and 4.7-kb (mutant) hybridizing fragments are shown for mice of the indicated genotypes. (C) RT-PCR analysis. Polyadenylated RNA from the ileum of wild-type mice (lanes 1 and 2) and DP−/− mice (lanes 3 and 4) was subjected to PCR amplification in the absence (lanes 1 and 3) or presence (lanes 2 and 4) of reverse transcriptase (RT). The estimated positions of nucleotides of 1972 and 889 base pairs (bp) are indicated on the right. (D) Relaxation of tracheal smooth muscle by PGD2 or BW245C. Tracheal rings from wild-type (top traces) and DP−/− (bottom traces) mice were suspended in an organ bath and induced to contract with carbachol. Relaxation in response to 3 μM PGD2 or 100 nM BW245C (or saline) applied at the times indicated by the arrows was assessed.

The role of DP in asthma was investigated with an ovalbumin (OVA)-induced asthma model in which PGD2 is generated in response to antigen challenge, as it is in humans with this condition (11). Wild-type and DP−/− mice were sensitized with intraperitoneal (ip) injections of OVA on day 0 and day 12 and were then exposed to aerosolized OVA on days 22, 26, and 30 (Fig. 2A) (12); control animals received saline instead of OVA. To determine the efficiency of this sensitization procedure, we analyzed serum concentrations of IgE. The concentrations of both total IgE and OVA-specific IgE in mice were markedly increased in response to ip injection of OVA and were boosted by subsequent inhalation of OVA; no substantial differences were apparent in this regard between wild-type and DP−/− mice (Fig. 2B). Repeated antigen inhalation in immunized wild-type mice resulted in a significant increase in total cell number in BAL fluid compared with that for the corresponding saline-treated control animals [22.2 (±10.4) × 105 versus 2.5 (±0.2) × 105, P < 0.05] (13). The infiltrated cells consisted predominantly of eosinophils, although the number of lymphocytes in the BAL fluid was also significantly increased (Fig. 2C). In contrast, only marginal increases in the numbers of eosinophils and lymphocytes in BAL fluid were apparent in DP−/− mice exposed to OVA challenge (Fig. 2C). Repeated OVA challenge in this animal model results in the development of airway hyperreactivity. We measured such hyperreactivity to acetylcholine 24 hours after the third inhalation of antigen (14). The OVA challenge significantly increased the sensitivity to acetylcholine in the wild-type mice, whereas little increase was detected in DP−/− animals (Fig. 2D).

Figure 2

Effects of DP deficiency on asthmatic responses in an OVA-induced asthma model. (A) Protocol for OVA immunization and challenge. (B) Time courses of the serum concentrations of total (left) and OVA-specific (right) IgE in wild-type and DP−/− mice. ○, Wild-type mice injected with saline; •, wild-type mice injected with OVA; □, DP−/− mice injected with saline; ▪, DP−/−mice injected with OVA. (C) Infiltration of inflammatory cells in BAL fluid. The numbers of total cells (upper left), eosinophils (upper right), macrophages (lower left), and lymphocytes (lower right) recovered in BAL fluid of OVA-challenged or saline-treated wild-type and DP−/− mice are shown. N.D., not detected. (D) Reactivity of the airway to acetylcholine in wild-type and DP-deficient mice. The dose-response curves of acetylcholine-induced bronchoconstriction in saline-treated (○) and OVA-challenged (•) wild-type mice and saline-treated (▵) and OVA-challenged (▴) DP−/− mice are shown. All data are means ± SEM of values from six mice per group. *P < 0.05 versus the respective saline control group; †P < 0.05 the OVA-challenged wild-type versus DP−/− mice (26).

Given the essential role of TH2 cytokines in evoking asthmatic responses (2), we measured the concentrations of IL-4, IL-5, and IL-13 in BAL fluid from wild-type and DP−/− mice (15). Challenge with OVA induced significant increases in the concentrations of all three of these TH2 cytokines in BAL fluid from wild-type mice (Fig. 3, A through C). Antigen challenge also increased the concentrations of these cytokines in DP−/− mice, but to a significantly lesser extent than in wild-type mice. In contrast, OVA challenge induced no difference in the concentration of the TH1 cytokine interferon-γ (IFN-γ) in BAL fluid between wild-type and homozygous DP-deficient mice (Fig. 3D).

Figure 3

Effects of OVA challenge on cytokine concentrations in BAL fluid of wild-type and DP−/− mice. The amounts of IL-4 (A), IL-13 (B), IL-5 (C), and IFN-γ (D) are expressed as means ± SEM of values from 10 mice per group. N.D., not detected. *P < 0.05 versus the respective saline control group; †P < 0.05 versus the OVA-challenged wild-type mice (26).

In humans with asthma, infiltration of numerous lymphocytes is apparent in the lung and is thought to be responsible for the increased abundance of TH2 cytokines. Lymphocytes and eosinophils accumulate in the bronchial submucosa and around blood vessels and, in some areas, lymphocytes form bronchus-associated lymphoid tissue (BALT). Histological analysis of OVA-challenged wild-type mice revealed extensive cell infiltration (Fig. 4A) and occasional BALT formation (Fig. 4B) in the lungs of all animals (16). In contrast, little cell infiltration was detected in the lungs of OVA-challenged DP−/− animals (Fig. 4C). We next examined mucus secretion by airway epithelial cells, given that hypersecretion of mucus is one of the characteristic features of asthmatic airways both in humans and animal models (17). Whereas many mucus-containing cells were apparent in wild-type mice challenged with OVA (Fig. 4D), few such cells were detected in antigen-challenged DP−/−mice (Fig. 4E).

Figure 4

Histological examination of lung tissue of OVA-challenged wild-type and DP−/− mice. (A to C) Hematoxylin and eosin staining. Accumulation of inflammatory cells (A) and formation of BALT (B) are apparent in the lungs of OVA-challenged wild-type mice, but not in those of antigen-challenged DP−/− mice (C). (Dand E) Periodic acid-Schiff staining. Mucus secretion by the airway epithelial cells is apparent in antigen-challenged wild-type mice (D) but not in OVA-challenged DP−/−mice (E). Original magnification, ×200

Finally, we examined the expression and localization of DP in the lung. Northern blot analysis (7) detected little expession of DP in the lung of nonimmunized as well as immunized mice before the antigen challenge (Fig. 5A). However, the OVA challenge to the airway markedly enhanced the DP expression in the lung. No induction was found in the spleen before and after the antigen challenge. We next performed immunofluorescence and immunoelectron microscopy using a specific antibody to mouse DP (18). Weak DP receptor immunoreactivity was detected in the cells surrounding bronchioles and alveoli of the lung of the immunized wild-type mouse before challenge, and the immunoreactivity was markedly enhanced by the airway exposure to OVA (Fig. 5B). In contrast, no DP receptor immunoreactivity was detected in the lung of DP−/− mice before and after the challenge. These results not only verified that our gene-targeting strategy successfully disrupted the DP gene, but also confirmed the induction of the DP receptor in the asthmatic lung suggested by the Northern blot analysis. Immunoelectron microscopy further identified the DP receptor–expressing cells as ciliated and nonciliated epithelial cells in the bronchioles and type II alveolar epithelial cells (Fig. 5C). Moderate DP receptor immunoreactivity was also detected in the type I alveolar epthelial cells and infammatory white blood cells (19).

Figure 5

Expression and localization of the DP receptor in the lung. (A) Northern blot analysis. Expression of the DP receptor mRNA was examined in the ileum (lanes 1 to 3), the spleen (lanes 4 to 6), and the lung (lanes 7 to 9) of unimmunized (lanes 1, 4, and 7) and immunized wild-type mice before (lanes 2, 5, and 8) and after (lanes 3, 6, and 9) the OVA challenge. Positions of DP mRNA are indicated. (B) Immunofluorescence microscopy for DP in the lung of immunized wild-type mice before (left) and after (middle) the OVA challenge and in immunized DP−/− mice after the challenge (right). Original magnification, ×20. (C) Immunoelectron microscopy for DP in the lung of immunized wild-type mice after the OVA challenge. Immunogold particles are seen on the plasma and intracellular membranes of bronchiolar (left) and alveolar (right) epithelial cells. Original magnification, ×8000.

We have shown that DP−/− mice do not develop asthmatic responses in an OVA-induced asthma model, indicating that PGD2 and its receptor (DP) are important for such responses. Our observation that the serum concentrations of IgE were similar in immunized wild-type and DP−/− mice suggests that the loss of DP does not affect the primary immune response. Indeed, OVA induced the production of similar amounts of TH2 cytokines by splenocytes prepared from immunized wild-type and DP−/− mice (19). In contrast, the concentrations of such cytokines in BAL fluid after OVA challenge were significantly lower in DP−/− mice than in wild-type mice, suggesting that the effect of DP deficiency is manifested locally at the site of challenge. The absence of lymphocyte accumulation in OVA-challenged DP−/− mice is suggestive of a defect in the recruitment of lymphocytes to the site of allergen challenge. Eosinophilic infiltration in allergic asthma is thought to be a consequence of the activation of TH2 lymphocytes. Consistent with this notion, infiltration of eosinophils did not occur to a significant extent in OVA-challenged DP−/−mice. Furthermore, the in vivo administration of PGD2 into the airway of dogs induced marked eosinophilic infiltration (20). Transgenic expression of PGD synthetase in the lung also increased both the concentrations of IL-4 and IL-5 and the extent of eosinophilic infiltration in BAL fluid in a mouse OVA-induced asthma model (21). These various observations suggest that PGD2 produced in response to allergic challenge acts at DP in the lung to recruit lymphocytes to the site of challenge. Indeed, we observed the marked expression of the DP receptor in bronchiolar and alveolar epithelial cells in the asthmatic airway. The airway epithelium is proposed as a source of proinflammatory cytokines and chemokines in asthma (3), raising the possibility that PGD2 acting at DP in the epithelium may stimulate the production and release of these mediators.

Is PGD2 the sole, obligatory mediator of asthma? Although individuals with asthma usually exhibit high concentrations of IgE in serum, the IgE concentration is often not correlated with the incidence of asthma attacks. This dissociation has led to the suggestion that the IgE- and mast cell–mediated pathway is important in triggering asthma, but plays a limited role in the chronic phase of an established asthmatic state. The mast cell–derived PGD2 therefore appears to play an important role that is restricted to the initiation process, and other redundant pathways that evoke asthmatic responses exist. Consistently, we have found that excessive challenge with OVA overcomes the effect of DP deficiency (22). It is also possible that PGD2 produced by cells other than mast cells during asthmatic attacks contributes to trigger and/or enhance allergic responses. It was reported that PGD2 is produced also by macrophages and dendritic cells (23).

In summary, we have shown that PGD2 functions as a mediator of allergic asthma. In addition to being produced in the lung, PGD2 is produced in various other tissues in response to allergic stimuli (24), suggesting that it may also play an important role in other allergic disorders, such as allergic rhinitis and atopic dermatitis. The DP receptor may thus represent a new therapeutic target for the treatment of such allergic reactions (25).

  • * To whom correspondence should be addressed. Department of Pharmacology, Kyoto University Faculty of Medicine, Yoshida, Sakyo-ku, Kyoto 606-8501, Japan. E-mail: snaru{at}


View Abstract

Stay Connected to Science

Editor's Blog

Navigate This Article