Healthy endometrium undergoes rapid and varied phenotypic changes to mediate the dynamic proliferation, secretion and regression events associated with the phases of the hormonal cycle. These extensive phenotypic changes are a unique characteristic of endometrium and their dysregulation likely plays a role in the myriad of disorders associated with this tissue. Here we have studied endometriosis, a condition where the endometrium, which normally provides the inner lining of the uterus, grows in other areas of the body, typically on the ovaries, bowel, rectum, bladder and pelvic lining. It is a leading cause of chronic recurring pelvic pain, dysmenorrhoea (painful menstruation) and infertility in millions of women of reproductive age (Bulletti et al., 2010). The precise pathogenesis of endometriosis is unknown, but the disorder does exhibit a positive correlation with estrogens and symptom severity is often synchronous with the menstrual cycle. Proposed explanations for the development of endometriosis favour mechanisms that allow endometrial cells to exit the uterine cavity, implant in other tissues and proliferate. Therefore, cell motility, adhesion and proliferation are phenotypes of endometrium that are especially relevant to endometriosis; however, the exact cellular signalling involved in driving these changes is poorly understood. Here, we show that the endogenous cannabinoid (endocannabinoid) system is part of the regulatory system that stimulates human endometrial migration and suggest that a dysregulation of this system plays a role in pathophysiologies of the uterus.
Anandamide (N-arachidonoyl ethanolamine; AEA) is an endogenous lipid produced throughout the body and was identified as the first endocannabinoid, activating the cannabinoid CB1 and CB2 receptors (Devane et al., 1992; receptor nomenclature follows Alexander et al., 2011). In the mouse, uterine AEA production varies according to state of pregnancy, uterine receptivity and embryo implantation (Schmid et al., 1997; Sun and Dey, 2008). Fatty acid amide hydrolase (FAAH), the principal inactivating enzyme for AEA, is also active in the mouse uterus (Maccarrone et al., 2000), localized in the endometrial epithelium. Down-regulation of this FAAH activity has been shown in early pregnancy (Maccarrone et al., 2002) and over the oestrous cycle in rats and mice, suggesting a role for hormonal regulation of this enzyme and thus on signalling in the endocannabinoid system (Xiao et al., 2002; Klinger et al., 2006). FAAH also controls one of the two distinct biosynthetic pathways for the formation of N-arachidonoyl glycine (NAGly) from AEA. (Burstein et al., 2000; Bradshaw et al., 2009). Like AEA, NAGly is produced throughout the body (Huang et al., 2001; Bradshaw et al., 2009) but, unlike AEA, it has no activity at either CB1 or CB2 receptors (Sheskin et al., 1997). We have demonstrated that NAGly production in the uterus varies with the hormonal cycle, suggesting a role for NAGly in uterine physiology (Bradshaw and Allard, unpublished experiments).
The CB1 and CB2 receptors, which are Gi/o-GPCRs, modulate cell migration (Derocq et al., 2000; Song and Zhong, 2000; Walter et al., 2003). Stimulation of both neurons and microglia selectively increased production of the endocannabinoid, 2-arachidonoyl glycerol (2-AG), which, together with other related lipids, triggered microglial cell migration by activating CB2 and the abnormal cannabidiol (Abn-CBD) receptors (Walter et al., 2003). Also, NAGly potently induced concentration-dependent migration of immortalized microglia, more effectively than any previously described cannabinoid ligands (McHugh et al., 2010). The agonist profile of the Gi/o-coupled Abn-CBD receptor characteristically includes activation by Abn-CBD and O-1602, compounds inactive at CB1 and CB2 receptors (Begg et al., 2003; Mackie and Stella, 2006). Other agonists include 2-AG and AEA (Járai et al., 1999; Wagner et al., 1999; Walter et al., 2003). Migration assays with wildtype HEK293 and cells stably transfected with the GPCR, GPR18 support our working hypothesis that NAGly is acting via Gi/o-coupled GPR18 receptors (Kohno et al., 2006; McHugh et al., 2010). Abn-CBD and O-1602 also induce migration of GPR18-transfected cells but not wildtype cells, suggesting that GPR18 is a receptor target for Abn-CBD (McHugh et al., 2010). Significantly, NAGly-induced migration in GPR18-transfected cells is blocked by O-1918, arachidonoyl serine and cannabidiol (CBD), which are weak partial agonists/antagonists of the Abn-CBD receptor (McHugh et al., 2010).
Up-regulation of genes associated with ‘migration-capable’ cell phenotypes occurs in late secretory phase endometrial tissue (Talbi et al., 2006). Human endometrial cells and the derived endometrial cell lines, HEC-1A and HEC-1B migrate in vitro towards oestrogen (Fujimoto et al. 1996; Acconcia et al., 2006). This capacity of human endometrial cells to migrate in response to oestrogen relates to the oestrogen-dependence of endometriosis (Kitawaki et al., 2002). If oestrogen is a chemoattractant for human endometrial cells and endometriosis symptoms are positively correlated with oestrogen availability, chemotaxis may be a mechanism by which endometrial cells exit the uterus.
Here, using the HEC-1B human endometrial cell line, we have shown that AEA, Δ9-tetrahydrocannabinol (Δ9-THC) and NAGly potently induced migration of human endometrial cells through a CB1 receptor-independent mechanism. We have also tested the hypothesis that NAGly acted through GPR18 in both HEC-1B and in an overexpressing GPR18-transfected HEK293 cell line. MAPK activation assays in HEK293-GPR18 cells disclosed a novel pharmacology for some established CB1 and CB2 receptor ligands at GPR18 receptors, including Δ9-THC, which activated MAPK at nanomolar concentrations. Furthermore, HEC-1B migration and MAPK activation by NAGly and Δ9-THC were antagonized by AM251, CBD and Pertussis toxin (PTX), adding to the novel cannabinoid pharmacology revealed at GPR18.