The CRTH2 agonist Pyl A prevents lipopolysaccharide-induced fetal death but induces preterm labour

Abstract

We have previously demonstrated that the anti-inflammatory prostaglandin 15-deoxy-Δ 12,14-prostaglandin J(2) (15dPGJ(2)) delays inflammation-induced preterm labour in the mouse and improves pup survival through the inhibition of nuclear factor-κB (NF-κB) by a mechanism yet to be elucidated. 15dPGJ(2) is an agonist of the second prostaglandin D(2) receptor, chemoattractant receptor homologous to the T helper 2 cell (CRTH2). In human T helper cells CRTH2 agonists induce the production of the anti-inflammatory interleukins IL-10 and IL-4. We hypothesized that CRTH2 is involved in the protective effect of 15dPGJ(2) in inflammation-induced preterm labour in the murine model. We therefore studied the effects of a specific small molecule CRTH2 agonist on preterm labour and pup survival. An intrauterine injection of lipopolysaccharide (LPS) was administered to CD1 mice at embryonic day 16, ± CRTH2 agonist/vehicle controls. Mice were killed at 4.5 hr to assess fetal wellbeing and to harvest myometrium and pup brain for analysis of NF-κB, and T helper type 1/2 interleukins. To examine the effects of the CRTH2 agonist on LPS-induced preterm labour, mice were allowed to labour spontaneously. Direct effects of the CRTH2 agonist on uterine contractility were examined ex vivo on contracting myometrial strips. The CRTH2 agonist increased fetal survival from 20 to 100% in LPS-treated mice, and inhibited circular muscle contractility ex vivo. However, it augmented LPS-induced labour and significantly increased myometrial NF-κB, IL-1β, KC-GRO, interferon-γ and tumour necrosis factor-α. This suggests that the action of 15dPGJ(2) is not via CRTH2 and therefore small molecule CRTH2 agonists are not likely to be beneficial for the prevention of inflammation-induced preterm labour.

Figure 1

Murine myometrial CRTH2 mRNA. The mRNA was isolated from murine uterus and converted to cDNA (n = 3 per treatment group). RT-PCR was used to amplify CRTH2 showing a product size of 344 bp. No difference in CRTH2 expression was seen between treatment groups. CRTH2 mRNA expression was comparable between mice. C = Non-template control, V = vehicle, PA = Pyl A, LPS = lipopolysaccharide.

Figure 2

The effect of Pyl A and GSKCRTH2X on CR3 (CD11b) expression on eosinophils. Pyl A (32 μm) was used to increase CR3 (CD11b) expression on eosinophils. Pre-treatment with the CRTH2 antagonist GSKCRTH2X (100 μm) was used to confirm this effect was via CRTH2. Eosinophils were identified by labelling with anti-CD49d and on the basis of forward and side scatter. A representative histogram reveals a clear shift to the right with Pyl A treatment indicative of an increase in CR3 (CD11b) expression. This effect was attenuated with CRTH2 antagonist pre-treatment (a). A summary of CR3 (CD11b) expressed in eosinophils with each treatment is shown in the graph (n = 3) (b). The effect of Pyl A on CR3 expression is identical to that of 15dPGJ₂. A representative histogram is shown for the effect of both 15dPGJ₂ and Pyl A on CR3 expression (c). V = vehicle, PA = Pyl A, GSK X = GSKCRTH2X. For statistical analysis, analysis of variance of repeated measures with Dunnett's post hoc test was used; **P < 0.01.

Figure 3

Dose–response of lipopolysaccharide (LPS) induced preterm labour. CD1 mice received an intrauterine injection of vehicle or LPS at E16 and were allowed to deliver, (n = 3–10 per treatment group). No vehicle control mice delivered preterm and LPS induced preterm labour in a dose–response effect, (a). No surviving pups were seen at an LPS dose of >10 μg (b). For statistical analysis, one-way analysis of variance with Dunnett's post hoc test comparing all groups to the vehicle control was used; **P < 0.01, ***P < 0.001.

Figure 4

The effect of Pyl A on lipopolysaccharide (LPS) -induced preterm labour. CD1 mice received an intrauterine injection of vehicle, LPS or Pyl A at E16 and were allowed to deliver, (n = 4 per treatment group). No vehicle control mice delivered preterm, and at 250 μg Pyl A alone did not induce preterm labour (a). LPS induced preterm labour in a dose–response manner, and this effect was augmented with the addition of Pyl A (b). V = vehicle, PA = Pyl A. For statistical analysis, one-way analysis of variance with Bonferroni's multiple comparison's test was used; **P < 0.01, ***P < 0.001.

Figure 5

Effect of Pyl A on intrauterine fetal viability at 4.5 hr post intrauterine injection. Dams were killed at 4.5 hr post intrauterine injection of 20 μg of lipopolysaccharide (LPS), 250 μg of Pyl A and pup viability was assessed (n = 3 dams). An average of 11–14 pups per dam was seen in each treatment condition. Pyl A significantly increased viability at 4.5 hr post injection from 20% survival to 100% (a). In a subsequent experiment mice were allowed to deliver spontaneously. No pups in the LPS-treated or LPS/Pyl A-treated groups survived premature delivery (b). V = vehicle, PA = Pyl A. For statistical analysis, one-way analysis of variance with Bonferroni's multiple comparison test was used; ****P < 0.0001.

Figure 6

The effect of lipopolysaccharide (LPS) and Pyl A on myometrial and pup brain nuclear factor-κB (NF-κB) and cyclo-oxygenase 2 (COX 2). Dams were killed at 4.5 hr post intrauterine injection and tissue was harvested for protein analysis of phospho-p65 and COX-2 (n = 3). Co-injection of LPS and Pyl A increased myometrial phospho p65 (a), with no effect on pup brain phospho-p65. Phospho-p65 was decreased in brains of pups from dams treated with LPS alone (b). Representative phospho-p65 immunoblots are shown for each treatment group with B actin used as a loading control. No significant changes in COX-2 protein expression were seen in any treatment group in both myometrium and pup brain (c,d). However, an increase in messenger RNA of COX-2 was seen in LPS-treated and LPS/Pyl A-treated mice (e). For statistical analysis, one-way analysis of variance with Bonferroni's multiple comparison test was used; *P < 0.05, ***P < 0.001.

Figure 7

Effect of lipopolysaccharide (LPS) and Pyl A on the pro-inflammatory cytokines in the myometrium. Dams were killed at 4.5 hr post intrauterine injection of 20 µg LPS/250 µg Pyl A and mRNA and protein was extracted from the myometrium and analysed by PCR or ELISA (n = 3). With co-injection of LPS and Pyl A, interferon-γ (IFN-γ) and tumour necrosis factor-α (TNF-α) mRNA were significantly increased (a) as well as protein levels of interleukin-12 (IL-12), IL-1β and KC-GRO (b). V = vehicle, PA = Pyl A. For statistical analysis, one-way analysis of variance with Bonferroni's multiple comparison's test was used; *P < 0.05, **P < 0.01.

Figure 8

Effect of lipopolysaccharide (LPS) and Pyl A on the anti-inflammatory cytokines in the myometrium. Dams were killed at 4.5 hr post intrauterine injection of 20 μg LPS/250 μg Pyl A and mRNA and protein was extracted from the myometrium and analysed by PCR or ELISA. No change in interleukin-4 (IL-4) mRNA expression was seen, and a non-significant increase in IL-10 was seen with co-administration of Pyl A and LPS (a). Although there was an increase in protein levels of IL-10 and IL-5 with co-injection of LPS and Pyl A, this did not reach statistical significance (b). V = Vehicle, PA = Pyl A.

Figure 9

The effect of CRTH2 agonists on circular muscle contractility. CD1 mice were killed on E15–16 and myometrial tissue was harvested to mount on the myograph in the circular orientation (n = 3). A cumulative dose–response of CRTH2 agonists or vehicle was performed and the average area under the curve was determined. An inhibition in myometrial contractility was seen with Pyl A, but no effect with 15dPGJ₂ or DK-PGD₂. Graphs showing the effects of the CRTH2 agonists are shown (a,c,e), with representative traces of myometrial contractility (b,d,f). V = vehicle, PA = Pyl A. For statistical analysis, analysis of variance of repeated measures was used with Dunnett's multiple comparison test; **P < 0.01, ***P < 0.001.