Interaction between lysophosphatidic acid, prostaglandins and the endocannabinoid system during the window of implantation in the rat uterus

Abstract

Bioactive lipid molecules as lysophosphatidic acid (LPA), prostaglandins (PG) and endocannabinoids are important mediators of embryo implantation. Based on previous published data we became interested in studying the interaction between these three groups of lipid derivatives in the rat uterus during the window of implantation. Thus, we adopted a pharmacological approach in vitro using LPA, DGPP (a selective antagonist of LPA3, an LPA receptor), endocannabinoids' receptor selective antagonists (AM251 and AM630) and non selective (indomethacin) and selective (NS-398) inhibitors of cyclooxygenase-1 and 2 enzymes. Cyclooxygenase isoforms participate in prostaglandins' synthesis. The incubation of the uterus from rats pregnant on day 5 of gestation (implantation window) with LPA augmented the activity and the expression of fatty acid amide hydrolase, the main enzyme involved in the degradation of endocannabinoids in the rodent uteri, suggesting that LPA decreased endocannabinoids' levels during embryo implantation. It has been reported that high endocannabinoids are deleterious for implantation. Also, LPA increased PGE2 production and cyclooxygenase-2 expression. The incubation of LPA with indomethacin or NS-398 reversed the increment in PGE2 production, suggesting that cyclooxygenase-2 was the isoform involved in LPA effect. PGs are important mediators of decidualization and vascularization at the implantation sites. All these effects were mediated by LPA3, as the incubation with DGPP completely reversed LPA stimulatory actions. Besides, we also observed that endocannabinoids mediated the stimulatory effect of LPA on cyclooxygenase-2 derived PGE2 production, as the incubation of LPA with AM251 or AM630 completely reversed LPA effect. Also, LPA augmented via LPA3 decidualization and vascularization markers. Overall, the results presented here demonstrate the participation of LPA3 in the process of implantation through the interaction with other groups of lipid molecules, prostaglandins and endocannabinoids, which prepare the uterine milieu for embryo invasion during the window of implantation.

Figure 1. Expression of LPA3 receptor and Lyso-PLD enzyme in the rat uterus during the window of implantation.

LPA3 messenger (A) and protein (B) and Lyso-PLD messenger (D) and protein (E) were detected by RT-PCR and Western Blot. Also, LPA3 localization (C) was determined by immunhistochemistry in day 5. Black arrows indicate positive staining (200x). In A *** p<0.001 day 4 vs day 5, ** p<0,01 day 6IM vs day 6II. Results are shown as means ± sem. N = 4–6 for each point. d5: uterus from rats pregnant on day 5 of gestation; IM: implantation sites; II: inter-implantation sites.

Figure 2. Effect of LPA on FAAH activity and expression in the rat uterus during the window of implantation.

Uterine strips from rats pregnant on day 5 of gestation (implantation window) were incubated with different concentrations of LPA (10, 20, 50 and 100 µM) for 6 h and the expression of FAAH mRNA (A) and protein (B) was studied. Afterwards, uterine strips from rats pregnant on day 5 of gestation (implantation window) were incubated with LPA 50 µM for 3, 6 and 12 h and the expression of FAAH mRNA (C) and protein (D) was determined. In A, B, C and D: ** p<0,01 vs the rest. Results are shown as means ± sem. N = 4–6 for each point. C: control; L: LPA.

Figure 3. LPA3 mediated LPA effect on FAAH activity and expression in the rat uterus during the window of implantation.

Uterine strips from rats pregnant on day 5 of gestation (implantation window) were incubated with LPA 50 µM for 6 h in the presence of DGPP 10 µM (a selective LPA3 antagonist) and the activity (A) and the expression of FAAH mRNA (B) and protein (C) were determined. In A: * p<0,05 vs the rest; In B and C: ** p<0,01 vs the rest. Results are shown as means ± sem. N = 4–6 for each point. C: control.

Figure 4. Effect of LPA on the production of prostagandins in the rat uterus during the window of implantation.

Uterine strips from rats pregnant on day 5 of gestation (implantation window) were incubated with LPA 50 µM for 6 h and the production of PGF2α (A) and PGE2 (B) was determined by radioimmunoassay. In B: *** p<0.001 vs C. Results are shown as means ± sem. N = 4–6 for each point. C: control.

Figure 5. Effect of LPA on COX-1 and COX-2 expression in the rat uterus during the window of implantation.

Uterine strips from rats pregnant on day 5 of gestation (implantation window) were incubated with LPA 50 µM for 6 h and the expression of COX-1 mRNA (A) and protein (B) and of COX-2 mRNA (C) and protein (D) was studied. In C: *** p<0,001 vs C; In D: ** p<0,01 vs C. Results are shown as means ± sem. N = 4–6 for each point. C: control.

Figure 6. LPA3 mediated LPA effect on the production of PGE2 and on COX-2 expression in the rat uterus during the window of implantation.

Uterine strips from rats pregnant on day 5 of gestation (implantation window) were incubated with LPA 50 µM for 6 h in the presence of DGPP 10 µM (a selective LPA3 antagonist) and the production of PGE2 (A) and the expression of COX-2 mRNA (B) and protein (C) were determined. In A and B: *** p<0,001 vs the rest; In C: ** p<0,01 vs the rest. Results are shown as means ± sem. N = 4–6 for each point. C: control.

Figure 7. Endocannabinoids mediated LPA effect on the production of PGE2 and on COX-2 expression in the rat uterus during the window of implantation.

Uterine strips from rats pregnant on day 5 of gestation (implantation window) were incubated with LPA 50 µM for 6 h in the presence of AM251 10^{− 8}M (a selective CB1 antagonist) or AM630 10^{− 8}M (a selective CB2 antagonist) or both. The production of PGE2 (A) and the expression of COX-2 mRNA (B) and protein (C) were determined. In A: *** p<0,001 vs C and 251+630; In B: *** p<0,001 vs the rest; In C: ** p<0,01 vs the rest. Results are shown as means ± sem. N = 4–6 for each point. C: control.

Figure 8. Effect of LPA on decidualization and vascularization markers in the rat uterus during the window of implantation.

Uterine strips from rats pregnant on day 5 of gestation (implantation window) were incubated with LPA 50 µM for 6 and 12 h and the expression of IGFBP-1 (A) and IL-10 (B and C) was studied. In A: *** p<0,001 vs the rest; In C: *** p<0,001 vs the rest, ** p<0.01 vs C. Results are shown as means ± sem. N = 4–6 for each point. C: control.

Figure 9. Model of interaction between LPA, prostaglandins and endocannabinoids at the implantation site.