Isoform-Specific Compensation of Cyclooxygenase (Ptgs) Genes during Implantation and Late-Stage Pregnancy

Abstract

The participation of cyclooxygenase (COX) in embryo implantation and parturition has been studied extensively. However, the distinct role of the two COX isoforms in these processes still remains unclear. Using three characterized mouse lines where the Ptgs1 and Ptgs2 genes substitute for one another, this study focused on the reproductive significance of their distinct roles and potential biological substitution. In both non-gravid and gravid uteri, the knock-in COX-2 is expressed constitutively, whereas the knock-in COX-1 is slightly induced in early implantation. The delayed onset of parturition previously found in COX-1 null mice was corrected by COX-2 exchange in COX-2>COX-1 mice, with normal term pregnancy, gestation length and litter size. In contrast, loss of native COX-2 in COX-1>COX-2 mice resulted in severely impaired reproductive functions. Knock-in COX-1 failed to substitute for the loss of COX-2 in COX-1>COX-2 mice during implantation, indicating that COX-1 may be replaced by COX-2, but not vice versa. A panel of prostaglandins detected in uterus and ovary demonstrates that prostaglandin biosynthesis preferentially depends on native COX-1, but not COX-2. More interestingly, preferential compensations by the COX isoforms were sustained despite weak dependency on their role in prostaglandin biosynthesis in the uterus and ovary.

Figure 1

Ptgs gene exchange and estrous cyclicity. (a) Western blot analysis of COX-1 and COX-2 expression in non-pregnant female uterus, showing knockout of COX-1 and knock-in of COX-2. Images are representative of 3 separate experiments. (b) Average estrous cycle length of four groups of mice (10–12 weeks old). At least two cycles were generally captured for each mouse and then averaged to get a cycle length of each mouse. Data are presented as mean ± SEM, n = 5–6. (c) Average weight of both ovaries from non-gravid mice (10–12 weeks old). Data are presented as mean ± SEM, n = 7.

Figure 2

Implantation and plasma sex hormones on gestation day 4.5 (d4.5). (a) The number of implantation sites and positive pregnancy rates. A positive pregnancy in implantation experiments is defined as a mouse that displayed at least one visible implantation site on d4.5 of pregnancy. As such, the numbers within the bars indicate the pregnancy (%) that denotes the proportion of mice with implantation sites out of the total mice plugged. Data are presented as mean ± SEM, ^*P < 0.05 vs WT, one-way ANOVA analysis; ^#P < 0.05 vs WT, Chi-square test. (b) Representative photographs of uteri with implantation sites stained with 1% Evan’s blue on d4.5. (c) Plasma estradiol and (d) progesterone concentration on gestation d0.5 and d4.5, respectively. Blood was collected via submandibular vein from mice of each genotype (n = 4–7) and concentration was determined by sequential competitive immunoassay. Data are presented as mean ± SEM, ^*P < 0.05 vs WT, one-way ANOVA analysis.

Figure 3

Uterine capacity to generate prostaglandins on gestation day 4.5 (d4.5). (a) Western blot analysis of COX-1 and COX-2 expression in uteri on gestation d4.5. Images are representative of 3 separate experiments. (b), (c), and (d) Uterine prostaglandin F_2α (PGF_2α), 6-keto-prostaglandin F_1α (6-keto-PGF_1α), and prostaglandin E₂ (PGE₂), respectively. Prostaglandins were extracted, determined by competitive immunoassay, and further normalized with the protein level. Data are presented as mean ± SEM, n = 5–6, ^*P < 0.05 vs WT, one-way ANOVA analysis. BLD, below the limit of detection.

Figure 4

Ovarian histology and prostaglandin profile on gestation day 4.5. (a) Representative light photomicrographs (original magnification, × 50) of ovarian sections on gestation day 4.5. Numbers under each micrograph are means ± SEM of corpora lutea counted on sections. Data are presented as mean ± SEM, n = 5–7, ^*P < 0.05 vs WT, one-way ANOVA analysis. (b), (c), and (d) Ovarian prostaglandin F_2α (PGF_2α), 6-keto-prostaglandin F_1α (6-keto-PGF_1α), and prostaglandin E₂ (PGE₂), respectively. Frozen ovaries (2–3 pooled) from the same group were subjected to prostaglandin extraction and measurement. Data are presented as mean ± SEM, n = 3, ^*P < 0.05 vs WT, nonparametric tests.

Figure 5

Fetal development and plasma sex hormones during late-stage pregnancy. (a) Representative uteri with fetuses from WT, COX-2>COX-1, and Reversa mice on gestation day 19 before the onset of parturition. These data were collected from female WT, COX-2>COX-1, and Reversa mice mating with males of the same genotype. No COX-1>COX-2 mice were pregnant in these experiments. A representative resorbed fetus (red arrow) was observed in the uterus of a Reversa mouse on d19 of pregnancy. Numbers under each gross specimen are means ± SEM of viable fetuses. Data are presented as mean ± SEM, n = 6–7, ^*P < 0.05 vs WT, one-way ANOVA analysis. (b) Plasma estradiol and (c) progesterone concentration on gestation day 18–19. Blood was collected via submandibular vein from mice of each genotype (n = 5–6) and concentration was determined by sequential competitive immunoassay. Data are presented as mean ± SEM. ^*P < 0.05 vs WT, one-way ANOVA analysis.

Figure 6

Uterine COX expression and prostanoid profiles in mice during late-stage pregnancy. (a) Western blot analysis of COX-1 and COX-2 expression in uteri on gestation day 19. (b), (c), and (d) Uterine prostaglandin F_2α (PGF_2α), 6-keto-prstaglandin F_1α (6-keto-PGF_1α), and prostaglandin E₂ (PGE₂), respectively. Prostaglandins were extracted from uteri isolated from each group and determined by competitive immunoassay, and further normalized with the protein level. Data are presented as mean ± SEM, n = 5–6, ^*P < 0.05 vs WT, one-way ANOVA analysis.

Figure 7

Corpus luteum histology and ovarian prostaglandin profiles on gestation day 19. (a) Representative H&E stained ovarian sections (original magnification, ×50) on gestation day 19. Numbers under each micrograph are means ± SEM of corpora lutea counted. Data are presented as mean ± SEM, n = 5–7. (b) Ovarian weights from pregnant mice on gestation day 19. Weights from both ovaries of each mouse were measured. Data are presented as mean ± SEM, n = 6–7. (c), (d) and (e) Ovarian prostaglandin F_2α (PGF_2α), 6-keto-prostaglandin F_1α (6-keto-PGF_1α), and prostaglandin E₂ (PGE₂), respectively. Frozen ovaries (2–3 pooled) were subjected to prostaglandin extraction prior to competitive immunoassay. n = 3, ^*P < 0.05 vs WT, nonparametric tests.