Low fatty acid amide hydrolase and high anandamide levels are ...

3 downloads 0 Views 168KB Size Report
Jul 27, 2001 - Lambert, D.M. and Di Marzo, V. (1999) The palmitoylethanolamide and oleamide enigmas: are these two fatty acid amides cannabimimetic?
Molecular Human Reproduction Vol.8, No.2 pp. 188–195, 2002

Low fatty acid amide hydrolase and high anandamide levels are associated with failure to achieve an ongoing pregnancy after IVF and embryo transfer* Mauro Maccarrone1,6, Tiziana Bisogno3, Herbert Valensise2, Natalia Lazzarin4, Filomena Fezza3, Claudio Manna5, Vincenzo Di Marzo3 and Alessandro Finazzi-Agro`1 1Department

of Experimental Medicine and Biochemical Sciences and 2Division of Obstetrics and Gynaecology, University of Rome Tor Vergata, I-00133 Rome, 3Endocannabinoid Research Group, Istituto per la Chimica di Molecole di Interesse Biologico, Pozzuoli, 4Fatebenefratelli Association for Research, San Giovanni Calibita Hospital, Rome and 5Genesis, Center for Human Reproduction and Therapy of Infertility, Rome, Italy 6To

whom correspondence should be addressed at: Department of Experimental Medicine and Biochemical Sciences, University of Rome Tor Vergata, Via Montpellier 1, I-00133 Rome, Italy. E-mail: [email protected] *This paper is dedicated to Prof. Carlo Romanini, who passed away on May 14, 2001 and is sadly missed.

Human reproduction is a rather inefficient process, yet the molecular reasons for this inefficiency remain unknown. IVF and embryo transfer (IVF–embryo transfer) also results in a high frequency of implantation failures and early spontaneous abortions. Here we show that the anandamide (AEA)-degrading enzyme, fatty acid amide hydrolase (FAAH), had significantly lower activity (46 ⍨ 17 versus 161 ⍨ 74 pmol/min per mg protein) and protein content (0.10 ⍨ 0.03 versus 0.23 ⍨ 0.06 units) in lymphocytes of IVF–embryo transfer patients who failed to achieve an ongoing pregnancy than in those who become pregnant, and this was paralleled by a significant increase in blood AEA (4.0 ⍨ 2.2 pmol/ml and 0.9 ⍨ 1.0 pmol/ml respectively). The blood levels of the other endocannabinoid, 2arachidonoylglycerol, or of the AEA congener, N-palmitoylethanolamine, which are metabolized by enzymes different from FAAH, was not different between the pregnant and nonpregnant women, nor was there any difference in the activity of the AEA membrane transporter or the amounts of cannabinoid receptors in lymphocytes. Taken together with the reported negative effects of AEA on embryo implantation, this study indicates that low FAAH activity and subsequent increased AEA levels in blood might be one of the causes of implantation failure or pregnancy loss, thereby leading to a better understanding of the pathophysiological and therapeutic implications of endocannabinoids in human fertility. Key words: endocannabinoids/miscarriage/pregnancy/receptor/transporter

Introduction About half of all conceptuses are lost before the expected term, thus making human reproduction a rather inefficient process (Lockwood, 2000). IVF and embryo transfer (IVF– embryo transfer) also results in a high frequency of implantation failure and spontaneous abortions (Lockwood, 2000), resulting in an average birth rate per IVF–embryo transfer cycle of ~25% per oocyte retrieval (Nygren and Andersen, 2001). This percentage has scarcely improved during the past few years, despite the introduction of enhanced protocols for ovarian stimulation and improved embryo culturing techniques (Plachot, 2000), suggesting that defects at the fetal–maternal interface might be critical for the outcome of IVF. Remarkably, implantation failure at the preclinical stage is often misdiagnosed and leads to pain, suffering and medical costs. 188

Several unrelated factors have been implicated in the reproduction of animal models, e.g. biotransformation enzymes (Zusterzeel et al., 2000), superoxide dismutase (Sugino et al., 2000) and a novel prothrombinase, fg12 (Clark et al., 2001). However, the molecular mechanisms responsible for infertility remain elusive. In women, the association between chronic marijuana smoking and early pregnancy loss has long been known, and indeed the marijuana active component ∆9-tetrahydrocannabinol has been shown to adversely affect the course and outcome of pregnancy in several animal studies, through activation of cannabinoid (CB) receptors (Paria and Dey, 2000). Additionally, the ‘endocannabinoid’ anandamide (N-arachidonoylethanolamine, AEA) has adverse effects on reproductive functions in mice, including retarded embryo development, fetal loss and pregnancy failure (Paria and Dey, © European Society of Human Reproduction and Embryology

Anandamide degradation and IVF

2000). In particular, the levels of AEA in mouse uterus have been shown to inversely correlate with the activity of the AEA-degrading enzyme fatty acid amide hydrolase (FAAH), which is higher at the implantation site than at areas away from the implantation site (Paria et al., 1996). This suggests that FAAH might modulate the local concentration of AEA conducive to embryo development and implantation (Paria and Dey, 2000). In this line of evidence, dysregulated cannabinoid signalling disrupts uterine receptivity for embryo implantation in mice (Paria et al., 2001), and in rats sexual receptivity depends on the interplay between cannabinoids, progesterone and dopamine (Mani et al., 2001). In keeping with these data, we have recently reported the association between decreased levels of FAAH in maternal lymphocytes and early pregnancy loss in humans (Maccarrone et al., 2000a). We have also shown that lymphocyte FAAH, but not the AEA membrane transporter (AMT) nor the CB receptors, is stimulated by progesterone and Th2-type cytokines (Maccarrone et al., 2001), which favour human fertility (Piccinni et al., 1995; Piccinni and Romagnani, 1996). Moreover, we have found that addition of AEA to human lymphocytes in vitro inhibits the release of leukemia inhibitory factor (Maccarrone et al., 2001), which is critical for implantation and maintenance of the fetus in humans (Piccinni et al., 1998; Sharkey, 1998). Taken together, these findings indicate that an active FAAH in maternal lymphocytes is needed for successful pregnancy, hence suggesting that the high AEA levels that might follow the defective expression of FAAH could adversely affect gestation in humans. However, such an inverse correlation between FAAH activity and AEA concentration has not yet been demonstrated in any human tissue, and remains a matter of speculation. In this study, we sought to investigate whether the concentration of FAAH in maternal lymphocytes and the levels of AEA in the blood of the same subjects were associated with, and predictive of, the outcome of embryo transfer after IVF. We chose patients undergoing IVF–embryo transfer because they offer the advantage of an exact timing of when the embryo is transferred to the uterus. Moreover, since IVF is widely used to correct infertility (Plachot, 2000; Nygren and Andersen, 2001), we reasoned that a marker which predicts the outcome of embryo transfer might be of value for both improving the protocols and assisting the patients. Since AEA and congeners bind to CB receptors (Pertwee, 2000) and their degradation by intracellular FAAH requires cellular uptake through AMT (Jarrahian et al., 2000), we also measured CB binding and AMT activity in maternal lymphocytes of IVF–embryo transfer patients. In parallel, the levels of β-HCG were measured, because of its role in pregnancy (Goldstein, 1994) and its use as a marker to monitor human gestation (Kovalevskaya et al., 1999).

Materials and methods Patients and treatments Twenty-one patients and four healthy, non-pregnant controls were enrolled in this study. This sample size was chosen on the basis of our experience on the levels and SD values of FAAH activity in the

lymphocytes of aborting and non-aborting human subjects (Maccarrone et al., 2000a). Thus, we could calculate through power analysis, performed by means of the InStat program (GraphPAD Software for Science, San Diego, CA, USA), that for SD ⫽ 30%, eight subjects were needed in each group to have 90% power to detect a difference between means of 50%, with two-tailed P values ⬍ 0.05. In our hands, this number of patients failing to achieve an ongoing pregnancy was usually found in every 20 subjects who underwent IVF–embryo transfer. Women were similar in age (33.9 ⫾ 1.4 years), number of years of sterility (2.4 ⫾ 0.8) and reason for entering the IVF–embryo transfer program, i.e. a severe andrological factor. We excluded from enrolment patients with chronic diseases (diabetes, hypertension or systemic lupus erythematosus), those with uterine anomalies or under long-term treatments (corticosteroids or low-dose aspirin) and those who smoked ⬎20 cigarettes per day. All patients underwent an ovarian stimulation protocol after suppression with the GnRH analogue, Buserelin (Suprefact, Hoechst, Milan, Italy), 0.3 ml s.c. from day 21 of the previous cycle for at least 14 days and until 17β-estradiol was ⬍30 pg/ml. This treatment was followed by ovarian stimulation with recombinant FSH (Gonal F; Serono, Rome, Italy), in a dose which ranged from 150–450 IU depending on the ovarian response of patients. When leading follicles had a diameter ⬎17 mm, HCG (Profasi; Serono) was injected (5000 IU i.m.). Oocyte retrieval was performed 37 h later by transvaginal ultrasound-guided probe aspiration of the ovarian follicles. Because of the severe andrological factor, ICSI was the procedure of choice, performed as previously described (Palermo et al., 1992). Three embryos were transferred to each patient 44– 46 h after ICSI. Women had a single blood test at 6.0 ⫾ 0.8 weeks of gestation, which is the earliest age where low FAAH correlates with spontaneous abortion in vivo (Maccarrone et al., 2000a). At this time point, βHCG was also determined by the AXSJM enzyme immunoassay kit, according to the manufacturer’s instructions (Abbott Labs, Wisbaden, Germany), and the outcome of IVF–embryo transfer was assessed by ultrasonography performed with a 6.5 MHz transvaginal probe and 3.5 MHz transabdominal probe (AU4 IDEA Esaote, Ansaldo, Milan, Italy), and results were kept blind to the biochemistry laboratories. All women were enrolled after giving informed consent, and the experimental procedures respected the principles of the Declaration of Helsinki and were approved by the Ethics Committee of the University of Rome Tor Vergata. Isolation of lymphocytes Blood samples (30 ml per donor) were drawn from the antecubital vein of patients and healthy controls, and were collected into heparinized sterile tubes. Peripheral lymphocytes were purified by gradient centrifugation using the density separation medium Lymphoprep (Nycomed Pharma, Oslo, Norway), as previously reported (Maccarrone et al., 2000a). Anandamide hydrolase activity and expression FAAH (N-arachidonoylethanolamine amidohydrolase, E.C. 3.5.1.4) activity was assayed by reversed phase high performance liquid chromatography (RP–HPLC), as previously described (Maccarrone et al., 1998). Cell homogenates (20 µg/lane) were prepared as described (Maccarrone et al., 1998) and were subjected to SDS– polyacrylamide gel electrophoresis (12%), under reducing conditions. Rainbow molecular weight markers (Amersham Pharmacia Biotech, Buckinghamshire, UK) were phosphorylase b (97.4 kDa), bovine serum albumin (66.0 kDa) and ovalbumin (46.0 kDa). For immunochemical analysis, gels were electroblotted onto 0.45 µm nitrocellulose filters (Bio-Rad, Richmond, CA, USA), and

189

M.Maccarrone et al. FAAH was detected with anti-FAAH polyclonal antibodies (Primm S.r.l., Milan, Italy), diluted 1:200 (Maccarrone et al., 1998). Goatanti rabbit antibodies conjugated with alkaline phosphatase (GAR-AP; Bio-Rad), diluted 1:2000, were used as the second antibody (Maccarrone et al., 1998). The same anti-FAAH antibodies were used to quantify FAAH protein content by ELISA. Wells were coated with human lymphocyte homogenates (20 µg/well), which were then reacted with anti-FAAH polyclonal antibodies (diluted 1:300), as the first antibody, and with GAR-AP, diluted 1:2000, as the second antibody (Maccarrone et al., 1998). Colour development of the alkaline phosphatase reaction was measured at 405 nm, using p-nitrophenyl phosphate as substrate. The A405 values could not be converted into FAAH concentrations, because the purified enzyme is not available to make calibration curves. However, the ELISA test was linear in the range 0–50 µg/well of cell homogenate and its specificity for FAAH was validated by antigen competition experiments (Maccarrone et al., 1998). RT–PCR was performed using total RNA isolated from human lymphocytes (10⫻106 cells), as previously described (Maccarrone et al., 1998). The amplification parameters were 2 min at 95°C, 45 s at 95°C, 30 s at 55°C and 30 s at 60°C, and linear amplification was observed after 20 cycles (Maccarrone et al., 1998). The primers were as follows: (⫹) 5⬘-TGGAAGTCCTCCAAAAGCCCAG, (–) 5⬘-TGTCCATAGACACAGCCCTTCAG, for FAAH; and (⫹) 5⬘-AGTTGCTGCAGTTAAAAAGC, (–) 5⬘-CCTCAGTTCCGAAAACCAAC, for 18S rRNA. A total of 5 µl of the reaction mixture was electrophoresed on a 6% polyacrylamide gel, which was then dried and subjected to autoradiography (Maccarrone et al., 1998). RT–PCR products were validated by size determination and sequencing, as previously reported (Maccarrone et al., 1998). Analysis of AEA uptake and CB receptors The activity of AMT in intact lymphocytes (2⫻106/test) was studied essentially as described (Maccarrone et al., 2000b). To discriminate non-carrier-mediated from carrier-mediated transport of 200 nmol/l [3H]AEA (223 Ci/mmol; NEN DuPont de Nemours, Ko¨ ln, Germany) through cell membranes, controlled experiments were carried out at 4°C (Hillard et al., 1997). For CB receptor studies, membrane fractions were prepared from lymphocytes (10⫻106/test) as reported (Maccarrone et al., 2000c), were quickly frozen in liquid nitrogen and stored at –80°C for no longer than 1 week. These membrane fractions were used in rapid filtration assays with the CB1/CB2 receptor high affinity ligand [3H]CP55.940 (126 Ci/mmol; NEN DuPont de Nemours) (Pertwee, 2000), used at 400 pmol/l as described previously (Deutsch et al., 1997; Maccarrone et al., 2000c). Nonspecific binding was determined in the presence of 10 µmol/l AEA (Deutsch et al., 1997). Quantitation of AEA and congeners in blood Serum and nucleated cells from blood (~5 ml per sample) were prepared and separated by centrifugation. Proteins from serum were eliminated by the addition of 5 ml of cold acetone, followed by centrifugation and evaporation of the excess acetone from the supernatant. The de-proteinized serum and nucleated cells were then reconstituted and lipids were extracted as reported (Di Marzo et al., 2000). A total of 100 pmol each of d8-AEA, d8-2-arachidonoylglycerol (d8-2-AG) and d4-N-palmitoylethanolamine (d4-PEA) were added as internal standards. The organic phases were dried under nitrogen and purified by normal phase–HPLC carried out as described previously (Bisogno et al., 1997; Di Marzo et al., 2000). Mono-arachidonoylglycerols and AEA/PEA standards were eluted after 18–23 and 27–32

190

min respectively. The mono-arachidonoylglycerol fraction contained both 1(3)- and 2-stereoisomers. HPLC fractions were dried under a flow of nitrogen and derived with 15 µl N-methyl-N-trimethylsilyl-trifluoroacetamide containing 1% trimethylchloro-silane for 2 h at room temperature, thus yielding the trimethylsilyl derivatives of AEA, PEA and 2-AG. The two derived fractions were analysed separately by gas chromatography– mass spectrometry (GC/MS) carried out as described previously (Bisogno et al., 1997; Di Marzo et al., 2000). The derivatives of both deuterated and non-deuterated AEA, PEA, 2-AG and 1(3)-AG standards were eluted after 17.8, 12.9, 18.2 and 18.7 min respectively. MS-detection was run in the selected ion monitoring mode to improve sensitivity. Selected ions for AEA were at m/z ⫽ 427 and 419, corresponding to the molecular ions for d8-AEA and nondeuterated AEA, and at m/z ⫽ 412 and 404, corresponding to the loss of a methyl group in both compounds. Selected ions for PEA were at m/z ⫽ 375 and 371, corresponding to the molecular ions for d4-PEA and non-deuterated PEA, and at m/z ⫽ 360 and 356, corresponding to the loss of a methyl group in both compounds. Selected ions for 2-AG were at m/z ⫽ 530 and 522, corresponding to the molecular ions of d8-2-AG and non-deuterated 2-AG, and at m/z ⫽ 515 and 507, corresponding to the loss of a methyl group in both compounds. The compounds were identified on the basis of the presence, at the same retention time as the deuterated internal standards, of the corresponding MS signals with the appropriate relative abundance. The amounts of AEA, PEA and 2-AG were calculated from the peak area ratios between the signals at m/z ⫽ 404 and 412, m/z ⫽ 360 and 356, and m/z ⫽ 507 and 515 respectively. A linear correlation between these area ratios and the amounts of standards was observed in separate studies (Bisogno et al., 1997; Di Marzo et al., 2000). Statistical analysis Data reported in this paper are the mean (⫾ SD) of at least three independent determinations, each in duplicate. Statistical analysis was performed by the non-parametric Mann–Whitney test, elaborating experimental data by means of the InStat program.

Results Failure to achieve an ongoing pregnancy occurred in nine out of 21 patients (Table I). Between those nine and the 12 patients with a physiological single pregnancy there were no significant differences in age (33.7 ⫾ 1.4 versus 34.2 ⫾ 1.3 years), gestational age (6.0 ⫾ 0.9 versus 6.1 ⫾ 0.8 weeks), body mass index (21.6 ⫾ 1.6 versus 21.4 ⫾ 1.5 kg/m2) or βHCG content (3411 ⫾ 2535 versus 4800 ⫾ 1963 IU/ml; median ⫽ 4000 IU/ml). However, lymphocyte FAAH activity, as assessed by RP–HPLC, was significantly lower in patients who failed to become pregnant than in those who became pregnant (46 ⫾ 17 versus 161 ⫾ 74 pmol/min per mg protein) (Figure 1a). From the observational data, a threshold value of ~75 pmol/ min per mg protein was calculated: women with FAAH activity below this value were not pregnant, whereas those with FAAH activity of this value or above were pregnant. The enzyme activities in lymphocytes of women who achieved pregnancy and of those who did not were related to the FAAH protein and mRNA concentrations, as assessed qualitatively by Western blot and RT–PCR (Figure 2). Western blot analysis of lymphocyte extracts showed that specific anti-FAAH antibodies recognized a single immuno-

Anandamide degradation and IVF

Table I. Subject characteristics, endocannabinoid level and βHCG content in the blood of women undergoing IVF and embryo transfer Subjects (age, years)

1 (34) 2 (32) 3 (33) 4 (34) 5 (35) 6 (35) 7 (36) 8 (33) 9 (36) 10 (35) 11 (36) 12 (34) 13 (32) 14 (33) 15 (32) 16 (32) 17 (34) 18 (33) 19 (35) 20 (34) 21 (35)

Gestational age (weeks)

Body mass index (kg/m2)

5 6 5 6 7 5 5 6 7 6 7 7 6 5 6 7 7 7 6 6 5

Endocannabinoid content (pmol/ml)

21.0 21.2 22.2 19.3 20.9 24.5 23.1 20.9 23.0 22.5 18.9 18.7 21.3 20.7 19.5 21.0 23.4 23.0 22.0 21.9 22.5

AEA

PEA

nd 2.1 1.8 nd nd 2.0 nd 3.6 nd 0.9 7.6 1.5 nd 7.1 1.4 5.4 3.9 1.0 2.2 3.2 2.2

14.2 32.7 21.0 10.3 12.4 21.3 9.4 20.5 16.4 5.1 5.1 2.9 17.1 7.0 4.0 3.1 4.9 3.1 8.4 4.8 2.6

βHCG (IU/ml)

Outcome

6800 3000 1300 7000 3300 1000 4000 1500 7000 4500 6400 7000 6000 3000 3000 7000 1000 5000 6500 1500 2500

⫹ – – ⫹ ⫹ – ⫹ – ⫹ ⫹ – ⫹ ⫹ – ⫹ – – ⫹ – ⫹ ⫹

AEA ⫽ anandamide; PEA ⫽ N-palmitoylethanolamine; nd ⫽ not detectable (below the detection limit of the assay, i.e. ⬍0.2 pmol/ml); ⫹ ⫽ ongoing pregnancy; – ⫽ not pregnant. Data are shown as means of triplicate experiments (SD 艋10%).

Table II. Subject characteristics, fatty acid amide hydrolase (FAAH) activity and content, AEA membrane transporter (AMT) activity, cannabinoid (CB) receptor binding and blood endocannabinoid level in nonpregnant women Subjects (age, years)

1 2 3 4

(32) (36) (34) (33)

Body mass index (kg/m2)

21.2 23.0 20.1 19.7

FAAH1 protein

140 165 180 170

Protein

0.22 0.24 0.25 0.24

AMT activity2

52 50 55 49

Binding to CB receptors3

22000 18500 19500 21000

Endocannabinoid content (pmol/ml) AEA

PEA

nd nd 1.4 nd

15.7 16.4 6.3 16.0

1FAAH

activity was expressed as pmol/min per mg protein, FAAH protein content as absorbance units (after ELISA) at 405 nm. activity was expressed as pmol/min per mg protein. 3Binding to CB receptors was expressed as c.p.m. per mg protein. nd ⫽ not detectable (below the detection limit of the assay, i.e. ⬍0.2 pmol/ml). Data are shown as means of triplicate experiments (SD 艋10%). 2AMT

reactive band of the molecular size expected for FAAH (Giang and Cravatt, 1997; Ueda et al., 2000). These antibodies were used to quantify FAAH content in lymphocytes by ELISA, showing that FAAH protein (0.10 ⫾ 0.03 versus 0.23 ⫾ 0.06 absorbance units at 405 nm) was significantly (P ⬍ 0.0001) lower in patients who were not pregnant than in those who were (Figure 1b). Again, from these observational data, a threshold value of ~0.15 absorbance units was calculated, and this was in keeping with a previous study (Maccarrone et al., 2000a). Unlike FAAH, the activity of the AEA membrane transporter (53 ⫾ 7 versus 56 ⫾ 6 pmol/min per mg protein) and the binding of [3H]CP55.940 to CB receptors (20556 ⫾ 1758 versus 21333 ⫾ 1614 c.p.m. per mg protein) were not significantly different in lymphocytes of women who failed to

achieve pregnancy compared with those who became pregnant (Figure 3a,b). AEA and its congener PEA (Lambert and Di Marzo, 1999) were measured for the first time in human blood, yielding chromatograms such as those shown in Figure 4 for AEA. From the analysis of these chromatograms, the concentrations reported in Table I were calculated. The decrease in FAAH activity and content observed in lymphocytes of women who were not pregnant compared with those who were was paralleled by a significant (P ⫽ 0.0003) increase in blood AEA, from 0.9 ⫾ 1.0 pmol/ml in blood of pregnant women to 4.0 ⫾ 2.2 pmol/ml in blood of the non-pregnant women (Table I). Blood PEA was not significantly different in nonpregnant and pregnant patients (13.8 ⫾ 10.3 versus 8.5 ⫾ 191

M.Maccarrone et al.

Figure 2. Expression of FAAH in lymphocytes isolated from women who became pregant (patient 7) or who failed to become pregnant (patient 8), as assessed by Western blotting (a) and RT–PCR (b). Results are representative of three independent experiments.

On the other hand, FAAH activity (164 ⫾ 17 pmol/min per mg protein) and content (0.24 ⫾ 0.01 absorbance units), and AEA level (0.4 ⫾ 0.7 pmol/ml) in control subjects were not significantly different when compared with those of pregnant women, but differed significantly when compared with those of patients who failed to achieve pregnancy (P ⫽ 0.0028 in all cases).

Figure 1. Activity (a) and expression (b) of anandamide hydrolase (FAAH) in lymphocytes isolated from women who became pregnant (white circles) or who failed to become pregnant (black circles). Biochemical data were associated a posteriori with the outcome of pregnancy. In both panels, the horizontal line is the threshold value between the group of pregnant women and the group of those who failed to achieve an ongoing pregnancy.

5.5 pmol/ml), which is in keeping with the notion that PEA is only a poor substrate of lymphocyte FAAH (Maccarrone et al., 2001) and does not affect the release of fertilitypromoting cytokines from lymphocytes (Maccarrone et al., 2001). The other endocannabinoid, 2-AG (Sugiura and Waku, 2000), which is also cleaved by hydrolases other than FAAH (Sugiura and Waku, 2000; Ueda et al., 2000), was undetectable (below the detection limit of 4.5 pmol/ml) in blood. Finally, the data reported in Table II showed that there were no significant differences between non-pregnant controls and women experiencing either successful pregnancy or failure to achieve ongoing pregnancy, with respect to subject age (33.7 ⫾ 1.7 years versus non-aborting or aborting subjects respectively), body mass index (21.0 ⫾ 1.5 kg/m2), AMT activity (52 ⫾ 3 pmol/min per mg protein), [3H]CP55.940 binding to CB receptors (20250 ⫾ 1555 c.p.m. per mg protein), and blood PEA (13.6 ⫾ 4.9 pmol/ml). 192

Discussion In this study, we report the complete analysis in each patient of the level of blood endocannabinoids, of their degradation by lymphocyte AMT or FAAH, and of their binding to lymphocyte CB receptors. We show that high FAAH activity in lymphocytes is paralleled by AEA concentration in blood, an unprecedented finding for a human tissue, and that low FAAH and high AEA activity and content are associated with failure to achieve an ongoing pregnancy after embryo transfer. The role of FAAH in reproduction seems to be critical, because the lower enzyme activity in women who were not pregnant compared with those who were was due to down-regulation of expression at the mRNA and protein levels. These data extend to IVF, previous findings on the association between low FAAH and in-vivo miscarriage in humans (Maccarrone et al., 2000a). These data, showing that low FAAH is a means to maintain high AEA concentrations, also provide a basis for multiple studies in animal models demonstrating adverse effects of exogenously added AEA on the reproductive function (Paria and Dey, 2000). Unlike FAAH, AMT activity was the same in patients who were pregnant and in those who were not, suggesting that transport is not a check-point for AEA degradation during pregnancy. Keeping in mind that AEA is taken up through a facilitated diffusion mechanism (Hillard et al.,

Anandamide degradation and IVF

Figure 3. (a) Activity of anandamide membrane transporter (AMT), and (b) binding of [3H]CP55.940 to cannabinoid (CB) receptors in lymphocytes isolated from women who became pregnant (white circles) or who failed to become pregnant (black circles). Biochemical data were associated a posteriori with the outcome of pregnancy.

1997; Maccarrone et al., 2000b), and that FAAH activity can contribute to the maintenance of the AEA gradient between the outer and inner side of the plasma membrane (Day et al., 2001; Deutsch et al., 2001), these data suggest that despite residual FAAH activity in lymphocytes of non-pregnant women, a steady-state uptake of AEA is still maintained. Lymphocytes also express CB receptors, which are, however, are not modulated by pregnancy. These findings support the concept that degradation by FAAH is the only critical event controlling AEA level/action during gestation. These data also corroborate the previous findings that FAAH, but not AMT or CB receptors, are modulated by progesterone (Maccarrone et al., 2001). The FAAH decrease was not due to impaired production of βHCG, suggesting that low FAAH activity and subsequently increased AEA levels might be one of the causes of failure to achieve pregnancy. This finding might open a potential avenue

Figure 4. Gas chromatographic-mass spectrometric (GC/MS) analysis (select ion chromatogram at m/z ⫽ 404, arbitrary units on the y axis) of lipids purified from the blood of women who became pregnant [(a) middle trace, patient 18] or who failed to become pregnant [(a) lower trace, patient 17]. Only the data for anandamide are shown. In (a) the select ion chromatogram for d8-anandamide (m/z ⫽ 412, upper trace, 5-fold reduced scale) used as the internal standard is also shown. In (b) the select ion mass spectrum of the high performance liquid chromatography fraction from the aborting patient is shown. Results are representative of three independent experiments.

for prevention, and seems to identify FAAH and AEA as early markers for implantation failure and early miscarriage, both in vitro and in vivo (Maccarrone et al., 2000a). It is noteworthy that non-pregnant controls had the same FAAH activity and content as the subjects with normal gestation, suggesting that a down-regulation of FAAH occurred in lymphocytes of patients who failed to achieve pregnancy. This observation might be explained by at least two hypotheses: either the network of maternal signals which keep high FAAH levels during physiological pregnancy is impaired, or ‘defective’ blastocysts release an inhibitor of FAAH expression. This putative FAAH suppressor might act as a ‘suicide factor‘, which favours the elimination of blastocysts committed to die. The latter hypothesis is in keeping with the proapoptotic potential of AEA demonstrated in mouse blastocysts (Maccarrone et al., 2000d). The present data also provide the grounds for the previous observation that AEA exogenously added to human lymphocytes inhibits the release of leukemia inhibitory factor in vitro (Maccarrone et al., 2001), a cytokine critical for fertility (Piccinni et al., 1998; Sharkey, 1998). Increased blood AEA 193

M.Maccarrone et al.

in patients who failed to achieve pregnancy might have the same effect in vivo, thus contributing to the failure. An interesting link might also exist between AEA and leptin, an adipocyte-derived hormone that is involved in the regulation of food intake and energy expenditure, and this also may act as an important regulator of reproductive function (Yura et al., 2000). Defective leptin signalling, which causes sterility in leptin-deficient ob/ob mice (Yura et al., 2000), has been recently associated with elevated levels of hypothalamic endocannabinoids in the same animals (Di Marzo et al., 2001), whereas leptin treatment, which restores fertility (Yura et al., 2000), reduces hypothalamic endocannabinoids (Di Marzo et al., 2001). Therefore, it is tempting to speculate that an inverse relationship between leptin and AEA also exists in the reproductive system, as it does in the brain, and contributes to the regulation of fertility. That the balance between AEA and leptin might control reproduction has been recently proposed (Mechoulam and Fride, 2001), and the data presented here seem to favour this hypothesis. Future investigations hold the answer to the molecular signals which might bridge the leptincontrolled networks and the endocannabinoid system. Altogether, this study has demonstrated that high FAAH activity is paralleled by low AEA levels in blood. It has shown that low FAAH activity and increased AEA levels might be one of the causes of implantation failure or early pregnancy loss, thereby leading to a better understanding of the pathophysiological and therapeutic implication of endocannabinoids in human fertility. The findings reported here open a potential avenue for prevention of implantation failure and miscarriage, both in vivo and in vitro, and suggest that FAAH and AEA assays might be used for monitoring human gestation.

Acknowledgements We wish to thank Monica Bari, Katia Falciglia and Natalia Battista for expert assistance and helpful discussions. This investigation was partly supported by Istituto Superiore di Sanita` (III AIDS Program), Ministero dell’Universita` e della Ricerca Scientifica e Tecnologica (PRIN Program to AFA, 3933 to VDM) and INTAS (97/1297 to VDM).

References Bisogno, T., Maurelli, S., Melck, D., De Petrocellis, L. and Di Marzo, V. (1997) Biosynthesis, uptake and degradation of anandamide and palmitoylethanolamide in leukocytes. J. Biol. Chem., 272, 3315–3323. Clark, D.A., Ding, J.W., Yu, G., Levy, G.A. and Gorczynski, R.M. (2001) Fgl2 prothrombinase expression in mouse trophoblast and decidua triggers abortion but may be countered by OX-2. Mol. Hum. Reprod., 7, 185–194. Day, T.A., Rakhshan, F., Deutsch, D.G. and Barker, E.L. (2001) Role of fatty acid amide hydrolase in the transport of the endogenous cannabinoid anandamide. Mol. Pharmacol., 59, 1369–1375. Deutsch, D.G., Glaser, S.T., Howell, J.M., Kunz, J.S., Puffenbarger, R.A., Hillard, C.J. and Abumrad, N. (2001) The cellular uptake of anandamide is coupled to its breakdown by fatty acid amide hydrolase (FAAH). J. Biol. Chem., 276, 6967–6973. Deutsch, D.G., Goligorsky, M.S., Schmid, P.C., Krebsbach, R.J., Schmid, H.H.O., Das, S.K., Dey, S.K., Arreaza, G., Thorup, C., Stefano, G. et al. (1997) Production and physiological actions of anandamide in the vasculature of the rat kidney. J. Clin. Invest., 100, 1538–1546. Di Marzo, V., Hill, M.P., Bisogno, T., Crossman, A.R. and Brotchie, J.M. (2000) Enhanced levels of endogenous cannabinoids in the globus pallidus are associated with a reduction in movement in an animal model of Parkinson’s disease. FASEB J., 14, 1432–1438.

194

Di Marzo, V., Goparaju, S.K., Wang, L., Liu, J., Batkai, S., Jarai, Z., Fezza, F., Miura, G.I., Palmiter, R.D., Sugiura, T. et al. (2001) Leptin-regulated endocannabinoids are involved in maintaining food intake. Nature, 410, 822–825. Giang, D.K. and Cravatt, B.F. (1997) Molecular characterization of human and mouse fatty acid amide hydrolase. Proc. Natl Acad. Sci. USA, 94, 2238–2242. Goldstein, S.R. (1994) Embryonic death in early pregnancy: a new look at the first trimester. Obstet. Gynecol., 84, 294–297. Hillard, C.J., Edgemond, W.S., Jarrahian, A. and Campbell, W.B. (1997) Accumulation of N-arachidonoylethanolamide (anandamide) into cerebellar granule cells occurs via facilitated diffusion. J. Neurochem., 69, 631–638. Jarrahian, A., Manna, S., Edgemond, W.S., Campbell, W.B. and Hillard, C.J. (2000) Structure-activity relationships among N-arachidonylethanolamine (anandamide) head group analogues for the anandamide transporter. J. Neurochem., 74, 2597–2606. Kovalevskaya, G., Birken, S., Kakuma, T., Schlatterer, J. and O’Connor, J.F. (1999) Evaluation of nicked human chorionic gonadotropin content in clinical specimens by a specific immunometric assay. Clin. Chem., 45, 68–77. Lambert, D.M. and Di Marzo, V. (1999) The palmitoylethanolamide and oleamide enigmas: are these two fatty acid amides cannabimimetic? Curr. Med. Chem., 6, 757–773. Lockwood, C.J. (2000) Prediction of pregnancy loss. Lancet, 355, 1292–1293. Maccarrone, M., Valensise, H., Bari, M., Lazzarin, N., Romanini, C. and Finazzi-Agro`, A. (2000a) Relation between decreased anandamide hydrolase concentrations in human lymphocytes and miscarriage. Lancet, 355, 1326–1329. Maccarrone, M., Bari, M., Lorenzon, T., Bisogno, T., Di Marzo, V. and Finazzi-Agro`, A. (2000b) Anandamide uptake by human endothelial cells and its regulation by nitric oxide. J. Biol. Chem., 275, 13484–13492. Maccarrone, M., Fiorucci, L., Erba, F., Bari, M., Finazzi-Agro`, A. and Ascoli, F. (2000c) Human mast cells take up and hydrolyze anandamide under the control of 5-lipoxygenase, and do not express cannabinoid receptors. FEBS Lett., 468, 176–180. Maccarrone, M., De Felici, M., Bari, M., Klinger, F., Siracusa, G. and FinazziAgro`, A. (2000d) Down-regulation of anandamide hydrolase in mouse uterus by sex hormones. Eur. J. Biochem., 267, 2991–2997. Maccarrone, M., Valensise, H., Bari, M., Lazzarin, N., Romanini, C., and Finazzi-Agro`, A. (2001) Progesterone up-regulates anandamide hydrolase in human lymphocytes. Role of cytokines and implications for fertility. J. Immunol., 166, 7183–7189. Maccarrone, M., van der Stelt, M., Rossi, A., Veldink, G.A., Vliegenthart, J.F.G. and Finazzi-Agro`, A. (1998) Anandamide hydrolysis by human cells in culture and brain. J. Biol. Chem., 273, 32332–32339. Mani, S.K., Mitchell, A. and O’Malley, B.W. (2001) Progesterone receptor and dopamine receptors are required in ∆9-tetrahydrocannabinol modulation of sexual receptivity in female rats. Proc. Natl Acad. Sci. USA, 98, 1249–1254. Mechoulam, R. and Fride, E. (2001) A hunger for cannabinoids. Nature, 410, 763–764. Nygren, K.G. and Andersen, A.N. (2001) Assisted reproductive technology in Europe, 1997. Results generated from European registers by ESHRE. European IVF-Monitoring Programme (EIM), for the European Society of Human Reproduction and Embryology (ESHRE). Hum. Reprod., 16, 384–391. Palermo, G., Joris, H., Devroey, P. and Van Steirteghem, A.C. (1992) Pregnancy after intracytoplasmic sperm injection. Lancet, 340, 17–18. Paria, B.C. and Dey, S.K. (2000) Ligand-receptor signaling with endocannabinoids in preimplantation embryo development and implantation. Chem. Phys. Lipids, 108, 211–220. Paria, B.C., Deutsch, D.D. and Dey, S.K. (1996) The uterus is a potential site for anandamide synthesis and hydrolysis: differential profiles of anandamide synthase and hydrolase activities in the mouse uterus during the periimplantation period. Mol. Rep. Dev., 45, 183–192. Paria, B.C., Song, H., Wang, X., Schmid P.C., Krebsbach, R.J., Schmid, H.H., Bonner, T.I., Zimmer, A. and Dey, S.K. (2001) Dysregulated cannabinoid signaling disrupts uterine receptivity for embryo implantation. J. Biol. Chem., 276, 20523–20528. Pertwee, R.G. (2000) Cannabinoid receptor ligands: clinical and neuropharmacological considerations, relevant to future drug discovery and development. Exp. Opin. Invest. Drugs, 9, 1553–1571. Piccinni, M.P. and Romagnani, S. (1996) Regulation of fetal allograft survival by hormone-controlled Th1- and Th2-type cytokines. Immunol. Res., 15, 141–150.

Anandamide degradation and IVF Piccinni, M.P., Beloni, L., Livi, C., Maggi, E., Scarselli, G. and Romagnani, S. (1998) Defective production of both leukemia inhibitory factor and type 2 T-helper cytokines by decidual T cells in unexplained recurrent abortions. Nature Med., 4, 1020–1024. Piccinni, M.P., Giudizi, M.G., Biagiotti, R., Beloni, L., Giannarini, L., Sampognaro, S., Parronchi, P., Manetti, R., Annunziato, F., Livi, C. et al. (1995) Progesterone favors the development of human T helper cells producing Th2-type cytokines and promotes both IL-4 production and membrane CD30 expression in established Th1 cell clones. J. Immunol., 155, 128–133. Plachot, M. (2000) Fertilization. Hum. Reprod., 15, 19–30. Sharkey, A. (1998) Cytokines and implantation. Rev. Reprod., 3, 52–61. Sugino, N., Nakata, M., Kashida, S., Karube, A., Takiguchi, S. and Kato, H. (2000) Decreased superoxide dismutase expression and increased

concentrations of lipid peroxide and prostaglandin F(2alpha) in the decidua of failed pregnancy. Mol. Hum. Reprod., 6, 642–647. Sugiura, T. and Waku, K. (2000) 2-Arachidonoylglycerol and the cannabinoid receptors. Chem. Phys. Lipids, 108, 89–106. Ueda, N., Puffenbarger, R.A., Yamamoto, S. and Deutsch, D.G. (2000) The fatty acid amide hydrolase (FAAH). Chem. Phys. Lipids, 108, 107–121. Yura, S., Ogawa, Y., Sagawa, N., Masuzaki, H., Itoh, H., Ebihara, K., AizawaAbe, M., Fujii, S. and Nakao, K. (2000) Accelerated puberty and late-onset hypothalamic hypogonadism in female transgenic skinny mice overexpressing leptin. J. Clin. Invest., 105, 749–755. Zusterzeel, P.L., Nelen, W.L., Roelofs, H.M., Peters, W.H., Blom, H.J. and Steegers, E.A. (2000) Polymorphisms in biotransformation enzymes and the risk for recurrent early pregnancy loss. Mol. Hum. Reprod., 6, 474–478. Submitted on July 27, 2001; accepted on November 19, 2001

195