LPA3-mediated lysophosphatidic acid signalling in embryo implantation and spacing

Abstract

Every successful pregnancy requires proper embryo implantation. Low implantation rate is a major problem during infertility treatments using assisted reproductive technologies. Here we report a newly discovered molecular influence on implantation through the lysophosphatidic acid (LPA) receptor LPA3 (refs 2-4). Targeted deletion of LPA3 in mice resulted in significantly reduced litter size, which could be attributed to delayed implantation and altered embryo spacing. These two events led to delayed embryonic development, hypertrophic placentas shared by multiple embryos and embryonic death. An enzyme demonstrated to influence implantation, cyclooxygenase 2 (COX2) (ref. 5), was downregulated in LPA3-deficient uteri during pre-implantation. Downregulation of COX2 led to reduced levels of prostaglandins E2 and I2 (PGE2 and PGI2), which are critical for implantation. Exogenous administration of PGE2 or carbaprostacyclin (a stable analogue of PGI2) into LPA3-deficient female mice rescued delayed implantation but did not rescue defects in embryo spacing. These data identify LPA3 receptor-mediated signalling as having an influence on implantation, and further indicate linkage between LPA signalling and prostaglandin biosynthesis.

Figure 1

LPA₃ mRNA expression in WT uterus and effects of LPA₃-deficiency on implantation. a,b. Quantification of uterine LPA₃ mRNA during pregnancy and in E3.5 luminal endometrial epithelium (Epi), stroma (Str), and myometrium (Myo). c,d. In situ localization of LPA₃ in E3.5 WT uterus. *. Glandular endometrial epithelium. Scale bars=100 μm. e. Flushed blastocysts from E3.5 and E4.5 uteri. f,g. Number and location of implantation sites at E4.5 and E5.5 uteri. Blue bands (arrows): implantation sites; Brackets: clustered implantation sites. *P<0.001. In all figures, error bars are standard deviations, (+/+), (+/−), and (−/−) represent WT, Het, and LPA₃-deficient mice, respectively.

Figure 2

Multiple embryos at individual implantation sites and placental hypertrophy in LPA₃-deficient uteri. a. Samples of multiple embryos at individual implantation sites at E10.5. White arrowheads: embryos. Scale bars: 2 mm. b. Cross-sections of E10.5 uteri revealing two less developed embryos sharing one placenta in a LPA₃-deficient uterus. Scale bars: 1 mm. c. A placenta shared by three embryos at E18.5 in a LPA₃-deficient uterus. Scale bar: 8 mm. b and c, red arrowheads: embryos; yellow arrowheads: placentas. d. Weight of placentas at E18.5.

Figure 3

Delayed postimplantational development of embryos and increased embryonic death in LPA₃-deficient uteri. a. Representative embryos from E10.5 uteri. Scale bar: 2 mm. b. E10.5 embryo weight. c. Weights of E18.5 embryos and P0 pups. d. The average numbers of embryos implanted (Imp), at E10.5 and E18.5, and P0 pups. The numbers of embryos implanted were calculated as described in Material and Methods. *P<0.001, **P<0.05.

Figure 4

Reduced COX-2 mRNA and prostaglandin levels in LPA₃-deficient uteri, and exogenous prostaglandin rescue of delayed implantation. a. Expression of COX-2 during early pregnancy in WT and LPA₃-deficient uteri. *P<0.05. b. Reduced PGE₂ and PGI₂ levels in E3.5 LPA₃-deficient uteri. *P<0.05. c. Significantly increased percentage of LPA₃-deficient females showing on-time implantation upon PGE₂ and cPGI (a stable PGE₂ analog) treatment at E3.5. *P=0.003. d. Images of E4.5 LPA₃-deficient uteri with or without prostaglandin treatment. Red arrows: implantation sites. Supplementary Figure S1a, S1b, S1c