The role of endocannabinoids in pregnancy.

pm2[Epub ahead of print]

The role of endocannabinoids in pregnancy.


H Chan, Centre for Clinical Research, University of Queensland Centre for Clinical Research, Herston, Australia.


Endocannabinoids are a family of lipid signalling molecules. As with prostaglandins (PGs), endocannabinoids are derived from poly-unsaturated fatty acids and affect cell function via receptor-mediated mechanisms. They also bind prostaglandin receptors although at a lower affinity. The endocannabinoid network is regulated in pregnancy from embryo development to labour onset. Even small changes in endocannabinoid exposure can retard embryo development and affect implantation success. There is now compelling evidence that aberrant expression of the endocannabinoid pathway in the placenta and circulating lymphocytes results in spontaneous miscarriage and poor pregnancy outcomes. It is likely that competition between endocannabinoids, prostaglandins and other similar lipids ultimately determines how phospolipid/fatty acid substrates are metabolised and, thus, the balance between the synthesis of uterotonic and tocolytic activities. We, therefore, hypothesise that endocannabinoid profiles may be used as a biomarker to predict and/or identify spontaneous labour onset.



[PubMed – as supplied by publisher]

Publication Types, MeSH Terms, Substances

LinkOut – more resources

Full Text Sources



The endocannabinoid system

Current Issue CoverEndocannabinoids are endogenous ligands derived from membrane phospholipids that bind to a family of G protein-coupled receptors, termed cannabinoid receptors. Five endocannabinoids have been identified to date in human tissues: N-arachidonoylethanolamide (anandamide; Devane et al. 1992), 2-arachidonoylglycerol (2-AG; Sugiura et al. 1995), homo-γ-linolenoylethanolamide (Hanus et al. 1993), docosatetraenoylethanolamide (Hanus et al. 1993), noladin ether (Hanus et al. 2001) and virodhamine (Porteret al. 2002). The two best-characterised endocannabinoids are anandamide and 2-AG, and of the five family members, only these molecules have been implicated in the processes of pregnancy and delivery. This review, therefore, focuses on anandamide and 2-AG. These endocannabinoids are not stored within cells, but are synthesised and released in response to increased substrate availability and synthase activity (Sugiura et al. 2002). Anandamide is thought to be released by a two-step enzymatic reaction in which arachidonic acid is transferred to a phospholipid precursor, phosphatidylethanolamine, by N-acyltransferase to produce N-arachidonoyl-phosphatidylethanolamine (NAPE), which is then cleaved by phospholipase D (PLD) (Fig. 1Di Marzo et al. 1997). 2-AG, however, is synthesised from diacylglycerol (produced from phosphoinositides by phospholipase C) through the action of diacylglycerol lipase (DAGL; Bisogno et al. 2005). Phospholipase C-independent routes of 2-AG production involving retrograde transmission in the brain have also been reported (Di Marzo et al. 1994). Endocannabinoid degradation is mediated by integral membrane proteins, monoacylglyceride lipase (MAGL) and fatty acid amide hydrolase (FAAH), and results in the generation of arachidonic acid and subsequently prostaglandins (PG) (Fig. 1Pacher et al. 2006). Other studies have demonstrated that endocannabinoids undergo oxidative metabolism by a number of fatty oxygenases: cyclo-oxygenases, lipoxygenases and cytochrome P450s, to produce PG glycerol esters (for 2-AG) and ethanolamides (for anandamide) (Bornheim et al. 1993Edgemond et al. 1998Kozak et al. 2000). The effects of endocannabinoids are dependent upon their half-life in the extracellular matrix, which is a product of the rate of synthesis and cellular uptake and the rate of degradation.

Figure 1

Metabolism of endocannabinoids. Anandamide and 2-AG are synthesised by a two-step enzymatic process from phospholipids, phosphoinositides and phosphatidylethanolamine. The endocannabinoids are converted to arachidonic acid by FAAH and MAGL respectively. Arachidonic acid is subsequently metabolised by PGHS-1 and PGHS-2 to produce prostaglandins (PGs). Alternatively, anandamide and 2-AG may be oxygenated by PGHS-2 and lipoxygenases to generate PG ethanolamides (prostamides (PMs)) and glycerol esters (-G). PLC, phospholipase C; DAGL, diacylglycerol lipase; NAT, N-acyltransferase; NAPE, N-arachidonoyl-phosphatidylethanolamine; PLD, phospholipase D; EA, ethanolamine.

Two cannabinoid receptors, CB1 and CB2, have been identified and cloned (Matsuda et al. 1990Munro et al. 1993). These receptors primarily couple to Gi/o proteins and activate multiple signalling pathways via both the α and βγ subunits of G proteins: G classically leads to adenylate cyclase inhibition, while Giβγ modulates calcium and potassium channels (Pertwee 2006). Other important actions include the activation of MAPKs and class IA and IB PI3K. CB receptors are desensitised (but not internalised) by the G protein-coupled receptor kinases and β-arrestin-2 (Jin et al. 1999). The CB1 receptor is abundantly expressed in the CNS and peripheral tissues, whereas the CB2 receptor is predominantly detected in immune cells (Matsuda et al. 1990,Munro et al. 1993).

Local endocannabinoid signalling network in the placenta

Enzymes and proteins involved in the production and signalling of endocannabinoids are expressed in the term placenta, particularly in the villous trophoblast tissues (Table 1). This suggests that endocannabinoids can be produced locally in the placenta; however, this remains to be unequivocally established. Immunohistochemical studies have characterised changes in the expression of endocannabinoid-producing and -metabolising enzymes as well as cannabinoid receptors in first-trimester gestational tissue. The expression of human FAAH and CB1 proteins increases at weeks 9–10 of gestation before returning to basal levels (Habayeb et al. 2008Taylor et al. 2011). The implication of these changes is that the expression of the CB1 receptor and FAAH is dramatically elevated (week 9) prior to the commencement of blood flow to the intervillous space (gestation weeks 10–12). This observation coupled with the low expression of NAPE-PLD suggests that local endocannabinoid production is limited and any available anandamide and 2-AG (presumably from the maternal circulation) are readily converted to other products (Fig. 1). The biological significance of the increased expression of the CB1 receptor remains to be elucidated.

Table 1

Presence of expression/activity of endocannabinoid network proteins in the placental tissue.

More members of the endocannabinoid signalling network are expressed in the term placenta than in the first-trimester placenta (Table 1). This may be an indication that the developed placenta requires endocannabinoid regulation for the maintenance of pregnancy and preparation for labour.

Endocannabinoids, embryo development and implantation

Cannabinoid receptor expression during embryo development

The expression of the CB1 receptor is first detected at the two-cell to the blastocyst stage in the mouse embryo, and in later stages it is largely confined to the trophectoderm (Paria et al. 1995Yang et al. 1996Wang et al. 2003). In contrast, the CB2 receptor is present from the 1-cell to the blastocyst stage (Yanget al. 1996). Available data support a critical role of endocannabinoids in embryo development; for example, treatment of embryos with anandamide blocks blastocyst development, predominantly at the 8-cell/morula stage with some embryos arrested at the 2-cell stage (Paria et al. 1998). In the same study, 2-AG treatment arrested embryos at the 2-cell stage. The effects of both anandamide and 2-AG are reversible by pre-treatment with a CB1 receptor antagonist, but not with a CB2 receptor antagonist. Furthermore, the effect of endocannabinoids is primarily mediated by the CB1 receptor. These data are consistent with the hypothesis that the regulation of endocannabinoid synthesis is requisite for successful embryo development.

Paradoxically, Cnr1−/− and Cnr2−/− mouse embryos demonstrate impeded embryo development; by day 3 of pregnancy when wild-type mouse embryos develop into the 8-cell stage, these knockout embryos can mostly be found at the 4-cell stage (Paria et al. 2001Xie et al. 2012). In agreement with earlier studies,Paria et al. (2001) also showed that treatment of 2-cell embryos with anandamide for 72 h significantly reduced the number of embryos that develop to the blastocyst stage in wild-type mice, and this has also been observed in theCnr2−/− mouse embryos but not in the Cnr1−/− mouse embryos. Overall, CB receptors are required for embryo development, and the inhibitory effect of endocannabinoids on this process is likely to be mediated in part by receptor-independent mechanisms.

Endocannabinoid regulation of blastocyst development and implantation

Endocannabinoids are also instrumental in the embryo implantation process. The binding of anandamide to the blastocyst decreases prior to implantation (Paria et al. 2001), and the expression of the CB1 receptor is higher in the dormant blastocyst than in the activated blastocyst (Wang et al. 2003). Endocannabinoid signalling is regulated in the mouse uterus, where anandamide concentration increases almost threefold as the endometrium changes from a blastocyst-receptive to a non-receptive state (Paria et al. 2001). Once the blastocyst develops in vitro, anandamide and 2-AG exposure for 48 h promotes trophoblast differentiation and outgrowth, which occur through the CB1 receptor (Wang et al. 1999Liu et al. 2002Sun et al. 2010). After 96 h, however, the vehicle-treated blastocysts displayed a phenotype similar to that of the anandamide-treated cells. Similarly, a low anandamide concentration (14 nM) accelerated the attachment of blastocysts to a uterine epithelial cell monolayer, whereas a high anandamide concentration (56 nM) impeded the attachment (Liu et al. 2002). High anandamide concentrations also inhibited trophoblast differentiation (Wang et al. 1999). Liu et al. showed that these effects were mediated by the CB1 receptor, but not by the CB2 receptor. Recently, it has been reported that Cnr1−/− and Faah−/− mouse trophoblast stem cells demonstrate retarded cell migration, attachment and spreading (Xieet al. 2012) as well as reduced trophoblast stem cell proliferation (Sun et al. 2010).

Anandamide concentrations increase from days 1 to 4 following implantation in the mouse uterus, whereas 2-AG concentrations remain unchanged at levels that are markedly higher than that of anandamide (Wang et al. 2007). Additionally, anandamide and 2-AG are more abundant in the inter-implantation regions than in the implantation sites, which would serve to inhibit blastocyst attachment and trophoblast development as described earlier. Concomitantly, DAGLα (DAGLA) – which promotes 2-AG production – is highly expressed in the luminal epithelium of the inter-implantation sites and down-regulated in the myometrium after implantation. NAPE-PLD is expressed in the luminal epithelium; however, its expression is reduced at the time of implantation. It is more strongly expressed at the inter-implantation site, thus corresponding to the high anandamide concentration at this location (Guo et al. 2005Wang et al. 2007). Both FAAH and MAGL are primarily localised in the luminal and glandular epithelia at the site of implantation before and after blastocyst attachment. This ensures low endocannabinoid signalling at the time of implantation for successful blastocyst attachment. In summary, implantation success requires low anandamide binding and signalling in the blastocyst, trophoblast and endometrium.

Endocannabinoids in ectopic pregnancies

Surprisingly, the opposite is true for ectopic pregnancies in the fallopian tube. FAAH, NAPE-PLD, and CB1 and CB2 receptors can be strongly detected in the luminal border of the fallopian tube epithelium (Gebeh et al. 2012). FAAH activity in peripheral blood cell membranes is significantly lower in women with ectopic pregnancies than in those with normal pregnancies (Gebeh et al. 2013). Similarly, the expression of FAAH mRNA is also significantly lower in the fallopian tubes of women with ectopic pregnancies than in those of women in the luteal or follicular phase of their menstrual cycle (Gebeh et al. 2012). Furthermore, plasma anandamide levels are significantly higher in women with ectopic pregnancies than in those with normal pregnancies (Gebeh et al. 2013). Also, oviductal transport of embryos is impaired in Faah−/− and Cnr1−/− mice (Wang et al. 20042006). The absence of expression of the CB1 receptor does not affect embryo development; however, embryos of the Faah−/− mice show retarded development.

Presently, the contradictory findings that high plasma anandamide concentrations are associated with implantation in the fallopian tube (ectopic pregnancy) and failed implantation in the uterine wall (no pregnancy) cannot be resolved. The results suggest that the effect of abnormal anandamide exposure depends on its timing and location.

Endocannabinoids and labour

Changes in cannabinoid receptor expression in early-pregnancy miscarriages

Theoretically, increased endocannabinoid availability may up-regulate the activation and function of CB receptors. There are, however, discrepancies in the literature regarding the expression of cannabinoid receptors in miscarriages during the first trimester. It has been shown that the expression of placental CB1 receptor protein is elevated 2.5-fold in spontaneous miscarriage in the first trimester compared with levels observed in voluntary termination (Trabucco et al. 2009). In addition, Trabucco et al. found that the immunostaining of the CB1 receptor is poorly detected in placental villi from elective surgical terminations, which supports earlier observations from our laboratory that the expression of CB1 receptor mRNA and protein is not readily detected in the first-trimester placentae (weeks 9–13 of gestation; Helliwell et al. 2004). However, for gestational age-matched placentae of weeks 7–8, the expression of CB1 receptor protein in the trophoblast, mesenchymal core and decidua was approximately twofold lower in spontaneous miscarriage than in elective surgical termination (Taylor et al. 2011). The difference in results could be due to the tissue source as Trabucco et al. did not indicate which part of the placenta was analysed.

The expression of the CB2 receptor is relatively unchanged throughout the first trimester (Habayeb et al. 2008), and its expression in the second and third trimesters remains to be determined. The expression of the CB2 receptor in trophoblast cells and decidua is greatly up-regulated in placentae from spontaneous abortion compared with those from surgical terminations (Tayloret al. 2011). The significance of this is unknown, given that the effects of anandamide appear to be primarily dependent on the function of the CB1 receptor.

Alterations in endocannabinoid signalling promote miscarriage in early pregnancy

The expression of the anandamide-generating enzyme NAPE-PLD may be an indicator of anandamide concentrations and pregnancy outcomes; however, similar to those for the CB receptor, opposing results have been reported. The expression of NAPE-PLD was not significantly altered in the trophoblast, mesenchymal core or decidua of placentae from spontaneous miscarriage compared with that in placentae from elective surgical termination (Taylor et al. 2011). Conversely, the expression of NAPE-PLD mRNA is at least twofold higher in the first-trimester (weeks 9–12) placentae from elective surgical termination than in those from spontaneous miscarriage (Trabucco et al. 2009). This suggests that anandamide concentrations in the placenta are lower prior to spontaneous miscarriage. In direct contrast with these observations, several studies have found that plasma anandamide concentrations measured in the first trimester of women at risk of miscarriage are up to threefold higher in those who subsequently miscarried than in those who progressed to term (Habayeb et al. 20042008Taylor et al. 2011). A recent study, however, has failed to demonstrate a difference between plasma anandamide concentrations in women with normal pregnancies and those in women who subsequently miscarried (Tong et al. 2012). Collectively, the expression of NAPE-PLD and the content of anandamide in the placenta and their association with plasma anandamide concentrations require further characterisation.

FAAH is responsible for the conversion of anandamide to arachidonic acid, which then provides a source for labour-promoting PG production. As expected, anandamide content is higher in the uterus and placenta of Faah−/− mice than in wild-type mice (Sun et al. 2010). The expression and activity of FAAH mRNA and protein in peripheral lymphocytes have been found to be significantly lower in women who subsequently miscarried spontaneously or failed to maintain IVF pregnancy in the first trimester than in gestational age-matched women undergoing voluntary pregnancy termination or those whose pregnancies continued to term (Maccarrone et al. 200020012002). In the murine model,Faah−/− mice deliver prematurely (Wang et al. 2006). Similarly, no immunostaining of FAAH could be detected in the trophoblast cells of first-trimester placentae obtained from women who miscarried, whereas the expression of FAAH could be observed in the trophoblast cells of placentae obtained from women who underwent voluntary termination (Trabucco et al. 2009). The absence of FAAH drives prostamide (PM), not PG, production (Fig. 1). The concentrations of PMs F, E2 and D2 are at least threefold higher in the tissues of anandamide-treated Faah−/− mice than in those of their wild-type controls (Weber et al. 2004). These findings provide strong evidence that i) interactions of FAAH and anandamide are essential for the normal progression of pregnancy and ii) a decreased activity of FAAH results in increased PM concentrations, which then have a negative impact on pregnancy. Other studies have shown that while there is no difference in the expression of FAAH protein in trophoblast cells of women with normal pregnancy and recurrent miscarriage, FAAH is more abundant in the decidual stromal cells of women with recurrent miscarriage than in the normal-pregnancy placentae (Chamley et al. 2008). These results suggest that the expression of FAAH is regulated in a tissue-specific manner and low expression and activity of FAAH are associated with early delivery.

Anandamide represses myometrial contractions

Anandamide represses oxytocin-induced contractions in human myometrial tissue (Dennedy et al. 2004). Inhibition is dose dependent and mediated by the CB1 receptor, but not by the CB2 receptor. In addition, daily i.p. injections of anandamide into rats during late pregnancy increased the length of gestation and the number of stillbirths per litter and decreased the levels of serum PGs Fand F (Wenger et al. 1997). In support of these findings, Cnr1−/− mice demonstrated earlier labour onset compared with wild-type and Cnr2−/− mice (Wang et al. 2008). The same study also showed that treatment of wild-type mice with CB1 receptor antagonists from days 14 to 18 of pregnancy also resulted in premature labour. The absence of CB1 receptor signalling altered serum progesterone, oestrogen, corticosterone and corticotrophin-releasing hormone concentrations prior to parturition (Wang et al. 2008). Interestingly, a marked increase in human plasma anandamide concentrations has been reported as pregnancy progressed from the third trimester to term (but not during labour Habayeb et al. 2004). Because this was a cross-sectional study, it is difficult to determine whether or not this elevation plays a significant role in the timing of labour onset.

Once labour onset has commenced, the role of anandamide is unclear. When plasma anandamide concentrations in women undergoing labour induction by medical intervention were measured, no significant associations were found for pre-induction plasma anandamide concentrations and the length of induction to delivery time (Nallendran et al. 2010). This may be due to the variety of methods of labour induction that were used. In contrast, Habayeb et al. (2004)reported a moderate but positive correlation between plasma anandamide concentrations and duration of contractions/cervical dilatation at the time of sampling. Interestingly, although anandamide impedes labour induction, its concentration in plasma is elevated during active labour. Plasma anandamide concentrations were significantly elevated after induction compared with before induction; again, this may be confounded by the various method(s) used for inducing labour (Nallendran et al. 2010). The results are supported by other studies that reported a significant 1.7- to 3-fold increase in plasma anandamide concentrations of women at term in active labour (without prior intervention) than in women at term but not in labour (Habayeb et al. 2004Lam et al. 2008). The expression of the CB1 receptor is markedly down-regulated in the villi of labouring placentae than in non-labouring placentae at term (Acone et al. 2009). This suggests that the increased pool of anandamide during labour is unlikely to be activating the CB1 receptor. Overall, anandamide has an inhibitory effect on myometrial contractions and extends gestation, but does not appear to have an effect once labour commences.

Endocannabinoids may promote labour via PG production

PGs are lipid mediators that play a critical role in the mechanisms of miscarriage (Sugino et al. 2000) and parturition at both term and preterm (Challis et al. 2000). Our laboratory demonstrated that endocannabinoids and a synthetic cannabinoid (CP55, 940) stimulate fetal membrane production of PGE2 in a CB1 receptor-dependent manner (Mitchell et al. 2008). Similar results were also obtained using rat uterus, although in this tissue anandamide increased the levels of PGE2 and F in a CB2 receptor-dependent manner (Sordelli et al. 2012). PGs E and F have been implicated in the regulation of uterine contractions, cervical ripening and membrane rupture. PGI and thromboxanes contribute not only to these processes, but also to the vasculopathy of preeclampsia (Walsh & Wang 1998Challis et al. 2000). Therefore, endocannabinoids have labour-promoting effects via PG production.

Paradoxical effects of anandamide on labour

The seemingly paradoxical effects of anandamide on labour onset may be cell- or species-specific; however, they may also be explained by the actions of PMs. Anandamide can be converted into two different groups of fatty acids: it is converted to arachidonic acid by FAAH and subsequently by PGHS-2 to an intermediate endoperoxide, PGH2, which can be further converted to PGs (Fig. 1). Anandamide can also be directly converted to PGH2 by PGHS-2 and subsequently metabolised to form PMs; only anandamide-derived PGH2-ethanolamine (PMH2) is metabolised to PMs. Synthetic PG-amide (Bimatoprost, which activates PM-sensitive receptors) weakly elicits contractions in both pregnant and non-pregnant human isolated myometrial tissues, whereas the tissues are responsive to 17-phenylprostaglandin F (Chenet al. 2005). The weak contractility induced by PMs could result from their low affinity for PG receptors (Ross et al. 2002). The affinity of PME2 for PGE2receptors is at least 500-fold lower than that of PGE2. If anandamide is being hydrolysed to PMs, then it may be that the increased PM concentrations may impede myometrial contractility and promote labour quiescence (Fig. 2). The paradox of the activity of anandamide may, therefore, be resolved by it being dependent upon its route of metabolism and the concurrent activity of FAAH; that is, the conditions that result in anandamide-inducing PG production are not the same as those that would have an overall relaxant effect on regular myometrial contractions (Fig. 2). If FAAH activity is high, increased amounts of anandamide would be metabolised via the central pathway (shown in Fig. 1), leading to elevated PG production. Alternatively, if FAAH activity is low, less amounts of anandamide would be converted to arachidonic acid, limiting the rate of PG production, but leaving anandamide available for PM production via the pathway shown on the right of Fig. 1.

Figure 2

A new model for the spontaneous onset of preterm labour that implicates the metabolism of anandamide and its effect on the prostaglandin:prostamide (PG:PM) ratio as a biological marker for the risk of initiation of uterine contractions. The presence and absence of fatty acid amide hydrolase (FAAH) and PGH synthase (PGHS) are variables in the new model, leading to three possible outcomes from the introduction or production of anandamide: FAAH absent or reduced – anandamide is unable to be converted to arachidonic acid, resulting in increased anandamide concentrations – i) anandamide may be metabolised by PGHS-2, which increases PM production and therefore delays labour induction and prolongs labour, ii) anandamide may bind to the CB1 receptor, which has been shown to stimulate PG production by our laboratory, and this may lead to an increased PG:PM ratio and drive labour induction and iii) lastly, normal or increased amounts of FAAH convert anandamide to arachidonic acid, increasing the availability of PG substrate and, in doing so, increasing the PG:PM ratio during labour induction.

A new model of preterm spontaneous onset of labour

Based on the studies described above and recent data from our laboratory, we propose a new model for the spontaneous onset of labour, where the endocannabinoid pathway is coupled to PG synthesis in the placenta (Fig. 2). According to the new model, increased plasma anandamide concentrations with normal FAAH activity predispose to labour onset. We have demonstrated that synthetic endocannabinoids stimulate the CB1 receptor and induce the production of labour-promoting PGs. In contrast, if FAAH activity is low or absent, anandamide is not readily converted to arachidonic acid and PGs. Instead, anandamide may be metabolised by PGHS-2 to produce PMs or may bind to the CB1 receptor, leading to PG production (Fig. 2). Each scenario results in distinctly opposing actions. Hence, the establishment of a normal range of PG:PM ratio for gestational age, and accurate measurement of the ratio in clinical samples, would be a useful tool in assessing the risk of premature labour.

Challenges in PG quantification

The measurement of PG concentrations in fluids and tissues is traditionally performed using RIAs or ELISAs. The sensitivity of these assays is dependent upon the sensitivity and specificity of antibodies raised against PG metabolites. PGs and endogenous cannabinoid metabolites share the same lipid backbone with differing polar head groups at exactly the same position (Fig. 3; reviewed inMcKirdy et al. (2013)). Small lipid molecules display low immunogenicity – to elicit an immune response, they must be conjugated to a carrier protein (commonly BSA). We report a major cross-reactivity of commercial antisera raised against PG PGE2 with its endocannabinoid counterpart, PM PME2 (Glasset al. 2005). Similarly, this has been observed for PGF and PMF. Additionally, these antibodies demonstrate a higher affinity for the endocannabinoid metabolites than their supposed PG target.

Figure 3

Structural similarities between arachidonic acid and the endocannabinoids, anandamide and 2-arachidonoylglycerol (2-AG), and prostaglandin (PG) E2and prostamide (PM) E2. Note that PME2 is closer in structure to the antigen than PGE2 itself due to its larger moiety at the carboxyl end. Adapted from McKirdy NC, Rice GE & Mitchell MD 2013 Eicosanoids as diagnostics for preterm labor. Reproductive Biology Insights 6 1–10.

This has important implications for the current literature and may change our view of the role that PGs play in human pregnancy and parturition. To illustrate, a hypothetical dataset representing the quantification of eicosanoids by immunoassay and mass spectrometry is presented in Fig. 4. The RIA data demonstrate a general increase in PG concentrations with labour (the two left columns) sufficient for statistical significance; however, the wide distribution of data points prevents their clinical utility in diagnostic testing. The immunoassay-based measurement of PGE2 is a conflation of antibody reactivity with both PGE2 (middle two columns) and PME2 (the two right columns). The contribution of PGE2 and PME2 to the observed immunoreactivity can be resolved using mass spectrometry approaches. Such deconvolution of the immunoassay data, thus, may provide a different insight into labour-associated changes in the concentrations of these lipid-derived mediators. The capacity to establish labour-associated changes in individual lipid metabolites also affords an opportunity to develop multivariate index assays that may improve diagnostic efficiency to identify labour onset at both term and preterm.

Figure 4

A hypothetical dataset demonstrating the putative confounding effects of antibody cross-reactivity on the quantification of immunoreactive PGE2. Immunoassay-based measurement of PGE2 (grey squares) is a conflation of antibody reactivity with both PGE2 (open squares) and PME2 (black squares). PGE2 and PME2 contributions can be resolved using mass spectrometry approaches. Such deconvolution of the immunoassay data, thus, may provide a different insight into labour-associated changes in the concentrations of these lipid-derived mediators.


Endocannabinoid signalling is regulated during embryo development, blastocyst implantation, placentation and labour. There is direct evidence that alterations in endocannabinoids and the enzymes that regulate their production and metabolism lead to retarded embryo development, poor blastocyst attachment and miscarriage. It is becoming increasingly evident that elevated anandamide concentrations promote spontaneous contractions, and this is likely to be mediated by PG production. We propose that differences in the profile of endocannabinoids present in biofluids may be of utility in the prediction and/or diagnosis of spontaneous labour onset at both term and preterm.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the review reported.


G E Rice was in receipt of a National Health & Medical Research Council of Australia Principle Research Fellowship.

  • Received 16 December 2012
  • Revision requested 29 January 2013
  • Revision received 28 May 2013
  • Accepted 5 June 2013
  • Made available online as an Advance Publication on 6 June 2013


    1. Acone G,
    2. Trabucco E,
    3. Colacurci N,
    4. Cobellis L,
    5. Mackie K,
    6. Meccariello R,
    7. Cacciola G,
    8. Chioccarelli T,
    9. Fasano S,
    10. Pierantoni R
    11. et al.

     2009 Low type I cannabinoid receptor levels characterize placental villous in labouring delivery. Placenta 30 203–205. (doi:10.1016/j.placenta.2008.11.018)

    1. Bisogno T,
    2. Ligresti A &
    3. Di Marzo V

     2005 The endocannabinoid signalling system: biochemical aspects. Pharmacology, Biochemistry, and Behavior 81 224–238. (doi:10.1016/j.pbb.2005.01.027)

    1. Bornheim LM,
    2. Everhart ET,
    3. Li J &
    4. Correia MA

     1993 Characterization of cannabidiol-mediated cytochrome P450 inactivation. Biochemical Pharmacology 451323–1331. (doi:10.1016/0006-2952(93)90286-6)

    1. Challis JRG,
    2. Matthews SG,
    3. Gibb W &
    4. Lye SJ

     2000 Endocrine and paracrine regulation of birth at term and preterm. Endocrine Reviews 21 514–550. (doi:10.1210/er.21.5.514)

    1. Chamley LW,
    2. Bhalla A,
    3. Stone PR,
    4. Liddell H,
    5. O’Carroll S,
    6. Kearn C &
    7. Glass M

     2008Nuclear localisation of the endocannabinoid metabolizing enzyme fatty acid amide hydrolase (FAAH) in invasive trophoblasts and an association with recurrent miscarriage. Placenta 29 970–975. (doi:10.1016/j.placenta.2008.08.003)

    1. Chen J,
    2. Senior J,
    3. Marshall K,
    4. Abbas F,
    5. Dinh H,
    6. Dinh T,
    7. Wheeler L &
    8. Woodward D

    2005 Studies using isolated uterine and other preparations show bimatoprost and prostanoid FP agonists have different activity profiles. British Journal of Pharmacology 144 493–501. (doi:10.1038/sj.bjp.0706044)

    1. Dennedy MC,
    2. Friel AM,
    3. Houlihan DD,
    4. Broderick VM,
    5. Smith T &
    6. Morrison JJ

     2004Cannabinoids and the human uterus during pregnancy. American Journal of Obstetrics and Gynecology 190 2–9.(discussion 3A)(doi:10.1016/j.ajog.2003.07.013)

    1. Devane WA,
    2. Hanus L,
    3. Breuer A,
    4. Pertwee RG,
    5. Stevenson LA,
    6. Griffin G,
    7. Gibson D,
    8. Mandelbaum A,
    9. Etinger A &
    10. Mechoulam R

     1992 Isolation and structure of a brain constituent that binds to the cannabinoid receptor. Science 258 1946–1949. (doi:10.1126/science.1470919)

    1. Di Marzo V,
    2. Fontana A,
    3. Cadas H,
    4. Schinelli S,
    5. Cimino G,
    6. Schwartz JC &
    7. Piomelli D

    1994 Formation and inactivation of endogenous cannabinoid anandamide in central neurons. Nature 372 686–691. (doi:10.1038/372686a0)

    1. Di Marzo V,
    2. De Petrocellis L,
    3. Bisogno T &
    4. Maurelli S

     1997 The endogenous cannabimimetic eicosanoid, anandamide, induces arachidonate release in J774 mouse macrophages. Advances in Experimental Medicine and Biology 407341–346.

    1. Edgemond WS,
    2. Hillard CJ,
    3. Falck JR,
    4. Kearn CS &
    5. Campbell WB

     1998 Human platelets and polymorphonuclear leukocytes synthesize oxygenated derivatives of arachidonylethanolamide (anandamide): their affinities for cannabinoid receptors and pathways of inactivation. Molecular Pharmacology 54 180–188.

    1. Gebeh AK,
    2. Willets JM,
    3. Marczylo EL,
    4. Taylor AH &
    5. Konje JC

     2012 Ectopic pregnancy is associated with high anandamide levels and aberrant expression of FAAH and CB1 in fallopian tubes. Journal of Clinical Endocrinology and Metabolism 97 2827–2835. (doi:10.1210/jc.2012-1780)

    1. Gebeh AK,
    2. Willets JM,
    3. Bari M,
    4. Hirst RA,
    5. Marczylo TH,
    6. Taylor AH,
    7. Maccarrone M&
    8. Konje JC

     2013 Elevated anandamide and related N-acylethanolamine levels occur in the peripheral blood of women with ectopic pregnancy and are mirrored by changes in peripheral fatty acid amide hydrolase activity. Journal of Clinical Endocrinology and Metabolism 98 1226–1234. (doi:10.1210/jc.2012-3390)

    1. Glass M,
    2. Hong J,
    3. Sato TA &
    4. Mitchell MD

     2005 Misidentification of prostamides as prostaglandins. Journal of Lipid Research 46 1364–1368. (doi:10.1194/jlr.C500006-JLR200)

    1. Guo Y,
    2. Wang H,
    3. Okamoto Y,
    4. Ueda N,
    5. Kingsley PJ,
    6. Marnett LJ,
    7. Schmid HH,
    8. Das SK &
    9. Dey SK

     2005 N-acylphosphatidylethanolamine-hydrolyzing phospholipase D is an important determinant of uterine anandamide levels during implantation.Journal of Biological Chemistry 280 23429–23432. (doi:10.1074/jbc.C500168200)

    1. Habayeb OM,
    2. Taylor AH,
    3. Evans MD,
    4. Cooke MS,
    5. Taylor DJ,
    6. Bell SC &
    7. Konje JC

    2004 Plasma levels of the endocannabinoid anandamide in women – a potential role in pregnancy maintenance and labor? Journal of Clinical Endocrinology and Metabolism 89 5482–5487. (doi:10.1210/jc.2004-0681)

    1. Habayeb OM,
    2. Taylor AH,
    3. Bell SC,
    4. Taylor DJ &
    5. Konje JC

     2008 Expression of the endocannabinoid system in human first trimester placenta and its role in trophoblast proliferation. Endocrinology 149 5052–5060. (doi:10.1210/en.2007-1799)

    1. Hanus L,
    2. Gopher A,
    3. Almog S &
    4. Mechoulam R

     1993 Two new unsaturated fatty acid ethanolamides in brain that bind to the cannabinoid receptor. Journal of Medicinal Chemistry 36 3032–3034. (doi:10.1021/jm00072a026)

    1. Hanus L,
    2. Abu-Lafi S,
    3. Fride E,
    4. Breuer A,
    5. Vogel Z,
    6. Shalev DE,
    7. Kustanovich I &
    8. Mechoulam R

     2001 2-Arachidonyl glyceryl ether, an endogenous agonist of the cannabinoid CB1 receptor. PNAS 98 3662–3665. (doi:10.1073/pnas.061029898)

    1. Helliwell RJ,
    2. Chamley LW,
    3. Blake-Palmer K,
    4. Mitchell MD,
    5. Wu J,
    6. Kearn CS &
    7. Glass M

     2004 Characterization of the endocannabinoid system in early human pregnancy. Journal of Clinical Endocrinology and Metabolism 89 5168–5174. (doi:10.1210/jc.2004-0388)

    1. Jin W,
    2. Brown S,
    3. Roche JP,
    4. Hsieh C,
    5. Celver JP,
    6. Kovoor A,
    7. Chavkin C &
    8. Mackie K

    1999 Distinct domains of the CB1 cannabinoid receptor mediate desensitization and internalization. Journal of Neuroscience 19 3773–3780.

    1. Kozak KR,
    2. Rowlinson SW &
    3. Marnett LJ

     2000 Oxygenation of the endocannabinoid, 2-arachidonylglycerol, to glyceryl prostaglandins by cyclooxygenase-2. Journal of Biological Chemistry 275 33744–33749. (doi:10.1074/jbc.M007088200)

    1. Lam PM,
    2. Marczylo TH,
    3. El-Talatini M,
    4. Finney M,
    5. Nallendran V,
    6. Taylor AH &
    7. Konje JC

     2008 Ultra performance liquid chromatography tandem mass spectrometry method for the measurement of anandamide in human plasma. Analytical Biochemistry 380 195–201. (doi:10.1016/j.ab.2008.05.033)

    1. Liu WM,
    2. Duan EK &
    3. Cao YJ

     2002 Effects of anandamide on embryo implantation in the mouse. Life Sciences 71 1623–1632. (doi:10.1016/S0024-3205(02)01928-8)

    1. Maccarrone M,
    2. Valensise H,
    3. Bari M,
    4. Lazzarin N,
    5. Romanini C &
    6. Finazzi-Agro A

    2000 Relation between decreased anandamide hydrolase concentrations in human lymphocytes and miscarriage. Lancet 355 1326–1329. (doi:10.1016/S0140-6736(00)02115-2)

    1. Maccarrone M,
    2. Valensise H,
    3. Bari M,
    4. Lazzarin N,
    5. Romanini C &
    6. Finazzi-Agro A

    2001 Progesterone up-regulates anandamide hydrolase in human lymphocytes: role of cytokines and implications for fertility. Journal of Immunology 166 7183–7189.

    1. Maccarrone M,
    2. Bisogno T,
    3. Valensise H,
    4. Lazzarin N,
    5. Fezza F,
    6. Manna C,
    7. Di Marzo V&
    8. Finazzi-Agro A

     2002 Low fatty acid amide hydrolase and high anandamide levels are associated with failure to achieve an ongoing pregnancy after IVF and embryo transfer. Molecular Human Reproduction 8 188–195. (doi:10.1093/molehr/8.2.188)

    1. Matsuda LA,
    2. Lolait SJ,
    3. Brownstein MJ,
    4. Young AC &
    5. Bonner TI

     1990 Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature 346561–564. (doi:10.1038/346561a0)

    1. McKirdy NC,
    2. Rice GE &
    3. Mitchell MD

     2013 Eicosanoids as diagnostics for preterm labor. Reproductive Biology Insights 6 1–10. (doi:10.4137/RBI.S11602)

    1. Mitchell MD,
    2. Sato TA,
    3. Wang A,
    4. Keelan JA,
    5. Ponnampalam AP &
    6. Glass M

     2008Cannabinoids stimulate prostaglandin production by human gestational tissues through a tissue- and CB1-receptor-specific mechanism. American Journal of Physiology. Endocrinology and Metabolism 294 E352–E356. (doi:10.1152/ajpendo.00495.2007)

    1. Munro S,
    2. Thomas KL &
    3. Abu-Shaar M

     1993 Molecular characterization of a peripheral receptor for cannabinoids. Nature 365 61–65. (doi:10.1038/365061a0)

    1. Nallendran V,
    2. Lam PM,
    3. Marczylo TH,
    4. Bankart MJ,
    5. Taylor AH,
    6. Taylor DJ &
    7. Konje JC

     2010 The plasma levels of the endocannabinoid, anandamide, increase with the induction of labour. British Journal of Obstetrics and Gynaecology 117863–869. (doi:10.1111/j.1471-0528.2010.02555.x)

    1. Okazaki T,
    2. Sagawa N,
    3. Bleasdale JE,
    4. Okita JR,
    5. MacDonald PC &
    6. Johnston JM

     1981aInitiation of human parturition: XIII. Phospholipase C, phospholipase A2, and diacylglycerol lipase activities in fetal membranes and decidua vera tissues from early and late gestation. Biology of Reproduction 25 103–109. (doi:10.1095/biolreprod25.1.103)

    1. Okazaki T,
    2. Sagawa N,
    3. Okita JR,
    4. Bleasdale JE,
    5. MacDonald PC &
    6. Johnston JM

     1981bDiacylglycerol metabolism and arachidonic acid release in human fetal membranes and decidua vera. Journal of Biological Chemistry 256 7316–7321.

    1. Pacher P,
    2. Batkai S &
    3. Kunos G

     2006 The endocannabinoid system as an emerging target of pharmacotherapy. Pharmacological Reviews 58 389–462. (doi:10.1124/pr.58.3.2)

    1. Paria BC,
    2. Das SK &
    3. Dey SK

     1995 The preimplantation mouse embryo is a target for cannabinoid ligand-receptor signaling. PNAS 92 9460–9464. (doi:10.1073/pnas.92.21.9460)

    1. Paria BC,
    2. Ma W,
    3. Andrenyak DM,
    4. Schmid PC,
    5. Schmid HH,
    6. Moody DE,
    7. Deng H,
    8. Makriyannis A &
    9. Dey SK

     1998 Effects of cannabinoids on preimplantation mouse embryo development and implantation are mediated by brain-type cannabinoid receptors. Biology of Reproduction 58 1490–1495. (doi:10.1095/biolreprod58.6.1490)

    1. Paria BC,
    2. Song H,
    3. Wang X,
    4. Schmid PC,
    5. Krebsbach RJ,
    6. Schmid HH,
    7. Bonner TI,
    8. Zimmer A &
    9. Dey SK

     2001 Dysregulated cannabinoid signaling disrupts uterine receptivity for embryo implantation. Journal of Biological Chemistry 27620523–20528. (doi:10.1074/jbc.M100679200)

    1. Park B,
    2. Gibbons HM,
    3. Mitchell MD &
    4. Glass M

     2003 Identification of the CB1 cannabinoid receptor and fatty acid amide hydrolase (FAAH) in the human placenta.Placenta 24 990–995. (doi:10.1016/S0143-4004(03)00165-6)

    1. Pertwee RG

     2006 The pharmacology of cannabinoid receptors and their ligands: an overview. International Journal of Obesity 30 (Suppl 1) S13–S18. (doi:10.1038/sj.ijo.0803272)

    1. Porter AC,
    2. Sauer JM,
    3. Knierman MD,
    4. Becker GW,
    5. Berna MJ,
    6. Bao J,
    7. Nomikos GG,
    8. Carter P,
    9. Bymaster FP,
    10. Leese AB
    11. et al.

     2002 Characterization of a novel endocannabinoid, virodhamine, with antagonist activity at the CB1 receptor. Journal of Pharmacology and Experimental Therapeutics 301 1020–1024. (doi:10.1124/jpet.301.3.1020)

    1. Ross RA,
    2. Craib SJ,
    3. Stevenson LA,
    4. Pertwee RG,
    5. Henderson A,
    6. Toole J &
    7. Ellington HC

     2002 Pharmacological characterization of the anandamide cyclooxygenase metabolite: prostaglandin E2 ethanolamide. Journal of Pharmacology and Experimental Therapeutics 301 900–907. (doi:10.1124/jpet.301.3.900)

    1. Sordelli MS,
    2. Beltrame JS,
    3. Cella M,
    4. Franchi AM &
    5. Ribeiro ML

     2012Cyclooxygenase-2 prostaglandins mediate anandamide-inhibitory action on nitric oxide synthase activity in the receptive rat uterus. European Journal of Pharmacology 685 174–179. (doi:10.1016/j.ejphar.2012.04.034)

    1. Sugino N,
    2. Nakata M,
    3. Kashida S,
    4. Karube A,
    5. Takiguchi S &
    6. Kato H

     2000 Decreased superoxide dismutase expression and increased concentrations of lipid peroxide and prostaglandin F(2α) in the decidua of failed pregnancy. Molecular Human Reproduction 6 642–647. (doi:10.1093/molehr/6.7.642)

    1. Sugiura T,
    2. Kondo S,
    3. Sukagawa A,
    4. Nakane S,
    5. Shinoda A,
    6. Itoh K,
    7. Yamashita A &
    8. Waku K

     1995 2-Arachidonoylglycerol: a possible endogenous cannabinoid receptor ligand in brain. Biochemical and Biophysical Research Communications 215 89–97. (doi:10.1006/bbrc.1995.2437)

    1. Sugiura T,
    2. Kobayashi Y,
    3. Oka S &
    4. Waku K

     2002 Biosynthesis and degradation of anandamide and 2-arachidonoylglycerol and their possible physiological significance.Prostaglandins, Leukotrienes, and Essential Fatty Acids 66 173–192. (doi:10.1054/plef.2001.0356)

    1. Sun X,
    2. Xie H,
    3. Yang J,
    4. Wang H,
    5. Bradshaw HB &
    6. Dey SK

     2010 Endocannabinoid signaling directs differentiation of trophoblast cell lineages and placentation. PNAS107 16887–16892. (doi:10.1073/pnas.1010892107)

    1. Taylor AH,
    2. Finney M,
    3. Lam PM &
    4. Konje JC

     2011 Modulation of the endocannabinoid system in viable and non-viable first trimester pregnancies by pregnancy-related hormones. Reproductive Biology and Endocrinology 9 152. (doi:10.1186/1477-7827-9-152)

    1. Tong S,
    2. Ngian GL,
    3. Onwude JL,
    4. Permezel M,
    5. Saglam B,
    6. Hay S,
    7. Konje JC,
    8. Marczylo TH,
    9. Fleming G,
    10. Walker SP
    11. et al.

     2012 Diagnostic accuracy of maternal serum macrophage inhibitory cytokine-1 and pregnancy-associated plasma protein-A at 6–10 weeks of gestation to predict miscarriage. Obstetrics and Gynecology 1191000–1008. (doi:10.1097/AOG.0b013e3182518fd3)

    1. Trabucco E,
    2. Acone G,
    3. Marenna A,
    4. Pierantoni R,
    5. Cacciola G,
    6. Chioccarelli T,
    7. Mackie K,
    8. Fasano S,
    9. Colacurci N,
    10. Meccariello R
    11. et al.

     2009 Endocannabinoid system in first trimester placenta: low FAAH and high CB1 expression characterize spontaneous miscarriage. Placenta 30 516–522. (doi:10.1016/j.placenta.2009.03.015)

    1. Walsh SW &
    2. Wang Y

     1998 Maternal perfusion with low-dose aspirin preferentially inhibits placental thromboxane while sparing prostacyclin. Hypertension in Pregnancy 17 203–215. (doi:10.3109/10641959809006076)

    1. Wang J,
    2. Paria BC,
    3. Dey SK &
    4. Armant DR

     1999 Stage-specific excitation of cannabinoid receptor exhibits differential effects on mouse embryonic development.Biology of Reproduction 60 839–844. (doi:10.1095/biolreprod60.4.839)

    1. Wang H,
    2. Matsumoto H,
    3. Guo Y,
    4. Paria BC,
    5. Roberts RL &
    6. Dey SK

     2003 Differential G protein-coupled cannabinoid receptor signaling by anandamide directs blastocyst activation for implantation. PNAS 100 14914–14919. (doi:10.1073/pnas.2436379100)

    1. Wang H,
    2. Guo Y,
    3. Wang D,
    4. Kingsley PJ,
    5. Marnett LJ,
    6. Das SK,
    7. DuBois RN &
    8. Dey SK

    2004 Aberrant cannabinoid signaling impairs oviductal transport of embryos. Nature Medicine 10 1074–1080. (doi:10.1038/nm1104)

    1. Wang H,
    2. Xie H,
    3. Guo Y,
    4. Zhang H,
    5. Takahashi T,
    6. Kingsley PJ,
    7. Marnett LJ,
    8. Das SK,
    9. Cravatt BF &
    10. Dey SK

     2006 Fatty acid amide hydrolase deficiency limits early pregnancy events. Journal of Clinical Investigation 116 2122–2131. (doi:10.1172/JCI28621)

    1. Wang H,
    2. Xie H,
    3. Sun X,
    4. Kingsley PJ,
    5. Marnett LJ,
    6. Cravatt BF &
    7. Dey SK

     2007Differential regulation of endocannabinoid synthesis and degradation in the uterus during embryo implantation. Prostaglandins & Other Lipid Mediators 83 62–74. (doi:10.1016/j.prostaglandins.2006.09.009)

    1. Wang H,
    2. Xie H &
    3. Dey SK

     2008 Loss of cannabinoid receptor CB1 induces preterm birth. PLoS ONE 3 e3320. (doi:10.1371/journal.pone.0003320)

    1. Weber A,
    2. Ni J,
    3. Ling KH,
    4. Acheampong A,
    5. Tang-Liu DD,
    6. Burk R,
    7. Cravatt BF &
    8. Woodward D

     2004 Formation of prostamides from anandamide in FAAH knockout mice analyzed by HPLC with tandem mass spectrometry. Journal of Lipid Research45 757–763. (doi:10.1194/jlr.M300475-JLR200)

    1. Wenger T,
    2. Fragkakis G,
    3. Giannikou P,
    4. Probonas K &
    5. Yiannikakis N

     1997 Effects of anandamide on gestation in pregnant rats. Life Sciences 60 2361–2371. (doi:10.1016/S0024-3205(97)00296-8)

    1. Xie H,
    2. Sun X,
    3. Piao Y,
    4. Jegga AG,
    5. Handwerger S,
    6. Ko MS &
    7. Dey SK

     2012 Silencing or amplification of endocannabinoid signaling in blastocysts via CB1 compromises trophoblast cell migration. Journal of Biological Chemistry 287 32288–32297. (doi:10.1074/jbc.M112.381145)

    1. Yang ZM,
    2. Paria BC &
    3. Dey SK

     1996 Activation of brain-type cannabinoid receptors interferes with preimplantation mouse embryo development. Biology of Reproduction 55 756–761. (doi:10.1095/biolreprod55.4.756)

    The role of endocannabinoids in pregnancy.

potp font 1

Tags: , , , , , , , , , , , , , , , , ,