S1P3 Receptor Mediates the Proinflammatory Effect of the Endocannabinoid 2-Arachidonoylglycerol in Endometriotic Epithelial Cells

Abstract

Endometriosis is a chronic inflammatory disease characterized by the ectopic implantation of endometrium outside the uterus associated with pelvic pain and infertility. The molecular mechanisms involved in the pathogenesis of endometriosis are complex and far from being fully elucidated. We recently showed that the signaling of the bioactive sphingolipid sphingosine 1-phosphate (S1P) is deeply dysregulated in endometriosis. The endocannabinoids anandamide (AEA) and 2-arachidonoylglycerol (2-AG), via ligation to G-protein coupled receptors, CB1, CB2, and GPR18 as well as the cation channel TRPV1, play a crucial role in the modulation of pain and inflammation. Here, the role of endocannabinoid signaling in endometriosis and its possible cross talk with the S1P signaling axis has been investigated. It has been found that CB1, CB2, GPR18, TRPV1 as well as the enzymes involved in endocannabinoid metabolism are expressed in endometriotic lesions. Furthermore, the effect of 2-AG and AEA in the modulation of inflammation has been established in human endometriotic epithelial cells. 2-AG, but not methanandamide (MAEA), the nonhydrolyzable AEA analogue, induced a marked increase in the expression of cyclooxygenase 2 and various pro-inflammatory interleukins (IL-1β, IL-6 and IL-8). Interestingly, S1P₃, whose expression is augmented by 2-AG, is crucial for transducing the biological action of the endocannabinoid. Indeed, S1P₃ pharmacological blockade or its specific silencing impaired the pro-inflammatory action of 2-AG. In conclusion, these findings demonstrate, for the first time, the occurrence of a functional interplay between endocannabinoids and S1P signaling in endometriosis, paving the way for novel pharmacological strategies to treat the disease.

Keywords: 2‐arachidonoylglycerol; S1P receptors; endocannabinoid system; endometriosis; inflammation; sphingosine 1‐phosphate.

FIGURE 1

Cannabinoid receptors and enzymes are expressed in human endometriotic lesions. (A) qPCR analysis was performed using TaqMan Gene Expression Assay probes specific for cannabinoid receptors CB1, CB2, GPR18, and TRPV1 in the endometrium of healthy women (HE) (n = 15) or endometriotic lesions (EL) of different localization (n = 20: 7 ovarian endometriosis and 13 deep infiltrating endometriosis). Results were analyzed employing the 2^−ΔCt method. (B) Representative immunohistochemical images of CB1, CB2, GPR18 and TRPV1 expression in endometriotic lesions (n = 15: 11 ovarian endometriosis and 4 deep infiltrating endometriosis). The staining with DAB produced a brown precipitate at the site of antibody binding, localized both in epithelial (black arrows) and stromal (red arrows) cells of the endometriotic lesions. The negative control was obtained by processing tissue sections in parallel with the same IHC protocol, omitting the primary antibody. Magnification: ×40, scale bar: 30 μm. (C) qPCR analysis was performed using TaqMan Gene Expression Assay probes specific for cannabinoid enzymes NAPE‐PLD, DAGL, FAAH, and MAGL in the endometrium of healthy women (HE) (n = 15) or endometriotic lesions (EL) of different localization (n = 20: 7 ovarian endometriosis and 13 deep infiltrating endometriosis). Results were analyzed employing the 2^−ΔCt method. NAPE‐PLD and MAGL mRNA levels are significantly increased in EL compared to HE (Student's t‐test *p < 0.05).

FIGURE 2

2‐AG, but not MAEA, increases the expression and release of different pro‐inflammatory factors in endometriotic epithelial cells. (A) Endometriotic epithelial cells were serum‐starved for 18 h and treated with increasing concentrations of 2‐AG and MAEA for 24 h. mRNA quantitative analysis of COX2, IL‐1β, IL‐6 and IL‐8 was performed by qPCR. Results, analyzed with the 2^−ΔΔCt method, were obtained using β‐Actin as a housekeeping gene and individual inflammatory factors of the unchallenged specimen as a reference gene. 2‐AG increases COX2, IL‐1β, IL‐6 and IL‐8 expression in a statistically significant manner (One‐way ANOVA, *p < 0.05; **p < 0.01; ****p < 0.0001). (B) Endometriotic epithelial cells were serum‐starved for 18 h and treated with 10 μM 2‐AG or 2.5 μM MAEA for 24 h. The obtained conditioned media were screened for the content of proinflammatory markers using the Human Inflammation Array as described in Section 2. Results were expressed as fold increase in respect to control. 2‐AG induces the extracellular release of IL‐1β, IL‐2, IL‐6, IL‐7 and IFNγ in a statistically significant manner (Student's t‐test *p < 0.05; **p < 0.01; ***p < 0.001).

FIGURE 3

2‐AG modulates S1P signaling axis in endometriotic epithelial cells. Endometriotic epithelial cells were serum‐starved for 18 h and treated with 10 μM 2‐AG for 24 h. (A) mRNA quantitative analysis of S1P metabolic enzymes (SK1, SK2, SPP1, SPP2 and SPL) as well as molecules implicated in S1P signaling (S1P_1–5, the specific transporter Spns2 and the SK1‐activating protein CIB1) was performed by qPCR. Results, analyzed with the 2^−ΔΔCt method, were obtained using β‐Actin as a housekeeping gene and individual targets of the unchallenged specimen as a reference gene. 2‐AG increases SK1, S1P₃, and Spns2 mRNA levels in a statistically significant manner (Student's t‐test *p < 0.05; **p < 0.01). (B) Protein lysates were analyzed using SDS‐PAGE electrophoresis and WB, using specific anti‐SK1, anti‐SK2, anti‐S1P₃, anti‐Spns2 and anti‐GAPDH antibodies. The histograms represent the densitometric analysis of 4 independent experiments. Data are the mean ± SD and are reported as band intensity normalized to the expression of GAPDH, fold change over control (set as 1). 2‐AG increases S1P₃ protein content in a statistically significant manner (Student's t‐test *p < 0.05).

FIGURE 4

2‐AG pro‐inflammatory action relies on S1P₃. (A) Serum‐starved endometriotic epithelial cells were pretreated or not with the S1P₃ antagonist TY‐52156 (10 μM) for 45 min before being challenged with 10 μM 2‐AG for 24 h. mRNA quantitative analysis of COX2, IL‐1β, IL‐6, and IL‐8 was performed by qPCR. Results, analyzed with the 2^−ΔΔCt method, were obtained using β‐Actin as a housekeeping gene and individual inflammatory factors of the unchallenged specimen as a reference gene. The blockade of S1P₃ on 2‐AG‐induced inflammatory effect (**p < 0.01; ***p < 0.001, ****p < 0.0001) was statistically significant by two‐way ANOVA followed by Bonferroni's post hoc test (^# p < 0.05; ^## p < 0.01; ^### p < 0.001). (B) Endometriotic epithelial cells transfected with SCR‐, S1P₁‐, S1P₂‐ and S1P₃‐siRNA were serum‐starved prior to being challenged with 10 μM 2‐AG for 24 h. mRNA quantitative analysis of COX2, IL‐1β, IL‐6 and IL‐8 was performed by qPCR. Results, analyzed with the 2^−ΔΔCt method, were obtained using β‐Actin as a housekeeping gene and individual inflammatory factors of the unchallenged specimen as a reference gene. The effect of S1P₃ downregulation in the reduction of 2‐AG‐induced inflammatory effect (*p < 0.05; ***p < 0.001; ****p < 0.0001) was statistically significant by two‐way ANOVA followed by Bonferroni's post hoc test (^# p < 0.05; ^## p < 0.01; ^### p < 0.001; ^#### p < 0.0001).

FIGURE 5

2‐AG pro‐inflammatory action is SK‐independent in endometriotic epithelial cells. (A) Serum‐starved endometriotic epithelial cells were treated with 10 μM 2‐AG for different time intervals (5, 15 and 30 min). Protein lysates were analyzed using SDS‐PAGE electrophoresis and WB, using specific anti‐phospho‐SK1 (P‐SK1), anti‐phospho‐SK2 (P‐SK2) and anti‐GAPDH antibodies. The histograms represent the densitometric analysis of three independent experiments. Data are the mean ± SD and are reported as band intensity normalized to the expression of GAPDH, fold change over control (set as 1). (B) Serum‐starved endometriotic epithelial cells were treated for 30 min with 10 μM 2‐AG before being harvested and then subjected to C17‐S1P quantification by LC–MS/MS as described in the Section 2. Results are the mean ± SEM of three independent experiments and are reported as pmol of S1P normalized on cell number. (C) Endometriotic epithelial cells transfected with SCR‐, SK1‐ and SK2‐siRNA were serum‐starved prior to being challenged with 10 μM 2‐AG for 24 h. mRNA quantitative analysis of COX2, IL‐1β, IL‐6 and IL‐8 was performed by qPCR. Results, analyzed with the 2^−ΔΔCt method, were obtained using β‐Actin as a housekeeping gene and individual inflammatory factors of the unchallenged specimen as a reference gene. The effect of SK1 or SK2 downregulation did not significantly affect the 2‐AG inflammatory effect (*p < 0.05; **p < 0.01; ***p < 0.001).