Coordinate expression of inducible nitric oxide synthase and cyclooxygenase-2 genes in uterine tissues of endotoxin-treated pregnant mice.

OBJECTIVES: Our purpose was to investigate the relationship between expression of cyclooxygenase-2 and inducible nitric oxide synthase genes after labor induction with bacterial lipopolysaccharide in a murine model of preterm parturition. STUDY DESIGN: Pregnant C57B1/6 mice were given Escherichia coli lipopolysaccharide (20 micrograms per mouse) by intraperitoneal injection on day 16 of gestation, and the animals were followed up for signs of labor. Control mice received an equivalent volume of 0.9% saline solution. The latency from lipopolysaccharide injections until appearance of the first pup was recorded. Two separate groups of mice were given either aminoguanidine or indomethacin (5 mg/kg intragastric) 24 hours before induction of preterm labor. In a separate set of experiments mice were treated with lipopolysaccharide as described and were killed at intervals from 0.5 to 72 hours and intrauterine tissues (uterus, placenta, and fetal membranes) were removed and snap frozen in liquid nitrogen. Total protein and ribonucleic acid were extracted for Western and Northern blot analysis of cyclooxygenase-2 and inducible nitric oxide synthase protein and messenger ribonucleic acid, respectively. RESULTS: Northern blots from uterine, placental, and fetal membrane tissues of lipopolysaccharide- and saline solution-treated mice revealed that cyclooxygenase-2 and inducible nitric oxide synthase messenger ribonucleic acid transcripts were rapidly (within 0.5 to 2 hours) up-regulated after lipopolysaccharide administration but were unchanged in mice injected with saline solution. Immunoblot analysis with isoform-specific antibodies revealed that both enzymes were expressed in uterus, placenta, and fetal membranes in a coordinated fashion with peak expression seen at 6 to 8 hours. Although the steady-state accumulation of messenger ribonucleic acid transcripts encoding cyclooxygenase-2 and inducible nitric oxide synthase peaked at 6 hours and declined to baseline by 16 hours after injection with lipopolysaccharide, expression of cyclooxygenase-2 and inducible nitric oxide synthase was sustained through the period when premature delivery was observed. Nitric oxide-dependent cyclooxygenase-2 and inducible nitric oxide synthase expression was demonstrated by the elimination of accumulation of both messenger ribonucleic acid transcripts in mice pretreated with aminoguanidine before injection with lipopolysaccharide. CONCLUSIONS: These data indicate that nitric oxide synthesis may be a prerequisite for subsequent stimulation of cyclooxygenase-2 and inducible nitric oxide synthase gene expression. Taken together, the data suggest that cyclooxygenase-2 and inducible nitric oxide synthase are expressed in a coordinated manner in the uterus of endotoxin-challenged pregnant mice and that their enzymatic products may contribute to the signaling of uterine activity or cervical changes culminating in expulsion of the fetus.

Interleukin-1 receptor antagonist blocks interleukin-1-induced expression of cyclooxygenase-2 in endometrium.

OBJECTIVE: Our purpose was to test the hypothesis that the interleukin-1 receptor antagonist can inhibit interleukin-1-induced prostaglandin production and de novo expression of the inducible cyclooxygenase-2 isoform in a human endometrial epithelial cell line. STUDY DESIGN: A continuous line of human endometrial epithelial cells was established from a hysterectomy specimen from a nonmalignant uterus. Cells were maintained as a monolayer culture in medium 199 supplemented with 10% fetal bovine serum and 50 micrograms/ml gentamicin. Cultures were treated with cytokines (interleukin-1 alpha or interleukin-1 beta, interleukin-1 receptor antagonist, or tumor necrosis factor-alpha), and media were collected for analysis of prostaglandin E2 and prostaglandin F2 alpha) by radioimmunoassay, whereas cells were harvested for ribonucleic acid and protein extractions and subsequent Northern blot or Western blot analyses, respectively. RESULTS: When endometrial cells were incubated with interleukin-1 alpha or interleukin-1 beta, each cytokine was shown to stimulate the production of prostaglandin E2 and prostaglandin F2 alpha in a time- and dose-dependent fashion, with interleukin-1 alpha being far more potent than interleukin-1 beta. Interleukin-1 receptor antagonist inhibited interleukin-1 alpha- and interleukin-1 beta-induced prostaglandin formation, with 50% inhibitory concentration values of 30 ng/ml for prostaglandin E2 and 90 ng/ml for prostaglandin F2 alpha. When Northern blots of interleukin-1 alpha-treated cells were probed with a complementary deoxyribonucleic acid fragment specific for either cyclooxygenase-1 or cyclooxygenase-2, rapid de novo induction of cyclooxygenase-2 messenger ribonucleic acid was observed; however, cyclooxygenase-1 expression was constant regardless of interleukin-1 alpha concentration or incubation time. Coincubation of cells with interleukin-1 alpha (10 ng/ml) and cycloheximide caused superinduction of cyclooxygenase-2 messenger ribonucleic acid but had no effect on the expression of cyclooxygenase-1 messenger ribonucleic acid. Actinomycin D completely abolished interleukin-1 alpha-induced cyclooxygenase-2 messenger ribonucleic acid expression, suggesting that the cytokine caused transcriptional activation of the cyclooxygenase-2 gene. Experiments were conducted to examine whether interleukin-1 receptor antagonist could suppress interleukin-1-induced cyclooxygenase-2 expression. Cells were preincubated for 30 minutes with interleukin-1 receptor antagonist and then challenged with interleukin-1 alpha. Northern and Western analyses revealed that interleukin-1 receptor antagonist blocked interleukin-1 alpha-induced expression of cyclooxygenase-2 messenger ribonucleic acid transcripts and the subsequent appearance of cyclooxygenase-2 protein. Interleukin-1 receptor antagonist had no effect on the constitutive expression of cyclooxygenase-1 messenger ribonucleic acid and protein. Interleukin-1 receptor antagonist failed to alter prostaglandin E2 formation in response to tumor necrosis factor-alpha, indicating that the antagonist is specific for interleukin-1 family cytokines. Finally, interleukin-1 receptor antagonist acted as a partial agonist in some experiments in that relatively high concentrations (> 100 ng/ml) caused a modest increase in prostaglandin E2 and F2 alpha production. CONCLUSIONS: These data indicate that interleukin-1 receptor antagonist is a potent inhibitor of interleukin-1-induced arachidonic acid metabolism and could possibly serve as an endogenous or exogenous modulator of interleukin-1 action in the endometrial epithelium.