Differential impact of prostaglandin H synthase 1 knockdown on platelets and parturition

Abstract

Platelet activation is a hallmark of severe preeclampsia, and platelet PGH synthase 1-derived (PGHS1-derived) thromboxane A(2) (TxA(2)) has been implicated in its pathogenesis. However, genetic disruption of PGHS1 delays parturition. We created hypomorphic PGHS1 (PGHS1(Neo/Neo)) mice, in which the substantial but tissue-dependent variability in the inhibition of PGHS1-derived eicosanoids achieved by low-dose aspirin treatment is mimicked, to assess the relative impact of this strategy on hemostatic and reproductive function. Depression of platelet TxA(2) by 98% in PGHS1(Neo/Neo) mice decreased platelet aggregation and prevented thrombosis. Similarly, depression of macrophage PGE(2) by 75% was associated with selectively impaired inflammatory responses. PGF(2alpha) at 8% WT levels was sufficient to induce coordinated temporal oxytocin receptor (OTR) expression in uterus and normal ovarian luteolysis in PGHS1(Neo/Neo) mice at late gestation, while absence of PGHS1 expression in null mice delayed OTR induction and the programmed decrease of serum progesterone during parturition. Thus, extensive but tissue-dependent variability in PG suppression, as occurs with low-dose aspirin treatment, prevents thrombosis and impairs the inflammatory response but sustains parturition. PGHS1(Neo/Neo) mice provide a model of low-dose aspirin therapy that elucidates how prevention or delay of preeclampsia might be achieved without compromising reproductive function.

Figures

Figure 1

Generation of PGHS1Neo mice. (A) Targeting strategy. Numbers indicate known coding exons. Dashed lines indicate regions for homologous recombination; dotted lines represent fragments generated by PCR genotyping. TK, thymidine kinase. Restriction sites: A, ApaLI; B, BamHI; H, HindIII; X, XbaI. (B) Identification of PGHS1Neo allele by Southern blot analysis. The 4.1- and 6.0-kb bands represent WT and targeted alleles, respectively, when ApaLI was used for DNA digestion, while the 5.3- and 3.7-kb bands represent the corresponding alleles with HindIII digestion. (C) PCR genotyping of tail biopsies from WT (317 bp) and PGHS1Neo (752 bp) mice.

Figure 2

PGHS1 is downregulated in PGHS1Neo/Neo mice. (A) Western blot analysis of PGHS1 in nonstimulated (top panel) and PGHS2 in LPS-stimulated (bottom panel) peritoneal macrophages. PGHS1 and PGHS2 protein expression in nonstimulated (B) and LPS-stimulated (C) macrophages, respectively. (D) PGE2 production by nonstimulated peritoneal macrophages from PGHS1Neo/Neo mice (4 mice/group; repeated 3 times). (E) Real-time RT-PCR for determination of PGHS1 mRNA levels in stomach and kidney tissue of PGHS1Neo/Neo, WT, and PGHS1-KO mice (n = 8).

Figure 3

Induction and evaluation of inflammatory response of PGHS1Neo/Neo mice. Data are expressed as ear weight increase of an 8-mm diameter biopsy after AA (A), TPA (B), or capsaicin (C) treatment (left ear) compared with vehicle treatment (right ear). *P < 0.05 versus WT (n = 6–12). (D) Carrageenan-induced paw edema. Results are displayed as increase in paw thickness. For the celecoxib treatment group, 8 week old mice were fed diet containing 800 parts per million celecoxib for 4 weeks. *P < 0.05 versus WT (n = 6–9).

Figure 4

AA-induced aggregation of platelets and bleeding time of PGHS1Neo/Neo, PGHS1-KO, and WT mice. (A) Representative aggregation tracings for each genotype tested at different AA concentrations are depicted. ASA/WT, WT mice received low-dose aspirin in their drinking water (30 mg/l) for 1 week. (B) Tail bleeding times. Bars represent the mean value (n = 11–15). Each symbol represents the bleeding time of an individual mouse.

Figure 5

Protective effect against AA- and photochemical-induced thrombosis in hypomorphic PGHS1 mice. Mice were injected with AA via the tail vein (A) or with Rose Bengal via the jugular vein, followed by laser excitation at the right carotid artery (B and C). Percent survival of each group in A is indicated (*P < 0.001, as determined by Fisher’s exact test). (B) Representative Doppler flow tracings after photochemical injury in PGHS1-KO (top panel), PGHS1Neo/Neo (middle panel), and WT (bottom panel) mice. Rose Bengal injection and clot occlusion times are marked with arrows. (C) Combined data for photochemical-induced arterial injury model. The number of mice examined is indicated in parentheses. #P < 0.05 vs. PGHS1-KO and PGHS1Neo/Neo mice.

Figure 6

PGHS1 expression and PG profiles in uterus during late-stage pregnancy. (A) Immunohistochemical analysis of PGHS1 in uterus at gestation day 18.5 (magnification, ×200). Red arrows represent PGHS1 staining within decidua (endometrium). (B) PGHS1 gene expression in uterus at gestation day 17.5 by real-time RT-PCR detection. (C) PG profiles in PGHS1Neo/Neo, PGHS1-KO, and WT mice at gestation day 19. PGs were extracted from uteri isolated from each group and measured by LC/MS/MS. PGF2α, PGE2, and PGD2 levels in PGHS1Neo/Neo mice (n = 6) were 8%, 3.3%, and 3.2%, respectively, of those in WT mice (n = 10).

Figure 7

Serum sex hormones and uterine OTR expression in PGHS1Neo/Neo, PGHS1-KO, and WT mice at late gestation. (A and B) Serum progesterone and estradiol concentrations during late gestation, respectively. Blood was collected via the saphenous vein from mice of each genotype group (n = 4–10) and concentration was determined by sequential competitive immunoassay. *P < 0.001 versus PGHS1Neo/Neo and WT mice at day 19. (C) mRNA expression of OTR in uterus at gestation day 19. Semiquantitative RT-PCR was performed with primers spanning one 11-kb intron to avoid genomic DNA contamination. (D) Relative level of OTR to β-actin expression.

Figure 8

Corpus luteum histology and ovarian PGF2α levels at day 19 of gestation in WT, PGHS1Neo/Neo, and PGHS1-KO mice. (A) Representative light photomicrographs (magnification, ×40) of ovarian sections from WT, PGHS1Neo/Neo, and PGHS1-KO females at gestation day 19. (B) Ovarian weights from nongravid females and pregnant mice at gestation day 19. Weights from both ovaries of each mouse were measured. n = 3–6; *P < 0.05 versus PGHS1Neo/Neo and WT mice at day 19 of gestation. (C) Ovarian PGF2α levels from nongravid females and pregnant mice at gestation day 19. Frozen ovaries from different groups were pooled after weighing and subjected to PG extraction prior to LC/MS/MS analysis.