Regulation of Gonadotropin-Releasing Hormone Secretion by Cannabinoids
http://www.100md.com
《内分泌学杂志》
Departments of Psychiatry and Behavioral Sciences, Cell Biology, and Medicine (Endocrinology) (C.M.G., G.M.F., W.X., W.C.W.), Duke University Medical Center, Durham, North Carolina 27710
Department of Biology (S.L.P.), Neuroscience and Behavior Program, and Center for Neuroendocrine Studies, University of Massachusetts, Amherst, Massachusetts 01003
Abstract
Cannabinoids (CBs) exert untoward effects on reproduction by reducing LH secretion and suppressing gonadal function. Recent evidence suggests these effects are due primarily to hypothalamic dysfunction; however, the mechanism is obscure. Using immortalized hypothalamic GnRH neurons, we find these cells produce and secrete at least two different endocannabinoids. After release, 2-arachidonyl monoacylglycerol and anandamide are rapidly transported into GnRH neurons and are degraded to other lipids by fatty-acid amide hydrolase. The immortalized GnRH neurons also possess CB1 and CB2 receptors that are coupled to Gi/Go proteins whose activation leads to inhibition of GnRH secretion. In perifusion experiments, CBs block pulsatile release of GnRH. When a CB receptor agonist is delivered into the third ventricle of adult female mice, estrous cycles are prolonged by at least 2 d. Although in situ hybridization experiments suggest either that GnRH neurons in vivo do not possess CB1 receptors or that they are very low, transcripts are localized in close proximity to these neurons. Inasmuch as GnRH neurons in vivo possess G protein receptors that are coupled to phospholipase C and increased intracellular Ca2+, these same neurons should also be able to synthesize endocannabinoids. These lipids, in turn, could bind to CB receptors on neighboring cells, and perhaps GnRH neurons, to exert feedback control over GnRH function. This network could serve as a novel mechanism for regulating GnRH secretion where reproductive functions as diverse as the onset of puberty, timing of ovulation, duration of lactational infertility, and initiation/persistence of menopause may be affected.
Introduction
CANNABINOIDS (CBs) have long been known to exert potent negative effects on rodent, primate, and human reproduction (1). Chronic administration of CBs to male rodents and humans reduces sperm counts, depresses serum testosterone concentrations, and may suppress levels of LH in serum (1, 2, 3). In females, chronic CB exposure delays sexual maturation, disrupts menstrual or estrous cycles, depresses follicular maturation of the ovary, alters uterine and vaginal cytology, and may reduce contents of LH and sex steroids in blood (1, 4, 5, 6). Injections of 9-tetrahydrocannabinol (THC) over 10 d can decrease hypothalamic GnRH concentrations in a dose-dependent manner (7). Despite these marked changes, some of the symptoms are alleviated over time due to tolerance (5), and there is rapid reestablishment of function after cessation of CB use or exogenous gonadotropin administration (1, 2). By contrast, acute THC exposure inhibits pulsatile LH secretion (8), and it also blocks estrogen or estrogen/progesterone-mediated positive feedback on the LH surge in ovariectomized rats (9). Although very low doses of CBs suppress LH secretion in rodents and humans (8, 10), dose-dependent effects are evident as the duration of blockade is prolonged (8). In contrast to LH, effects of CBs on FSH and prolactin secretion are more variable (1, 11, 12, 13, 14, 15). If THC is given on the afternoon of proestrus, the LH surge and ovulation are delayed by 24 h (13, 16). This blockade can be relieved through administration of exogenous gonadotropin or synthetic GnRH (8, 13, 14, 17). In male rats, administration of THC into the third ventricle depresses serum LH contents and increases concentrations of GnRH within the medial basal hypothalamus (15). By comparison, addition of the CB to cultured anterior pituitary cells exerts no effects on basal or GnRH-stimulated LH release (7, 15). Hence, CB effects on LH release appear to be upstream of the pituitary. Because GnRH is a major regulator of reproduction in mammals, these collective findings suggest that CBs may perturb reproduction through actions on hypothalamic GnRH neurons.
Although marijuana and THC have been known to exert biological actions for some time (1), CB receptors were only cloned within the last decade (18, 19). The CB1 receptor is expressed primarily within the central nervous system, whereas the CB2 receptor is found in the periphery and in immune cells. After identification of CB receptors, investigators began to search for their endogenous ligands. These compounds are lipids and include anandamide (ANA), 2-arachidonyl monoacylglycerol (2-AMG), noladin ether, virodhamine, and N-arachidonyldopamine (20). Interestingly, there is some suggestion that CBs may play a role in reproduction because ANA levels in hypothalamus are low before puberty, increase just before puberty, and return to intermediate values immediately after vaginal opening (21). In addition, whereas autoradiography and in situ hybridization have shown that binding and expression levels of the CB1 receptor are low in hypothalamus (22), the highest levels of expression in this brain region are in the medial preoptic area and arcuate nucleus (23). Because GnRH perikarya are located in the preoptic area and these neurons send their axons through the arcuate nucleus to the median eminence to regulate gonadotropin release (24), some of the CB effects on reproduction may be mediated through the hypophysiotrophic GnRH neurons. To examine this possibility, we have used the immortalized hypothalamic GnRH neurons to determine whether these neurons can produce and respond to CBs, and we have extended these studies to GnRH neurons in vivo.
Materials and Methods
Cell culture
The immortalized hypothalamic GnRH neurons (GT1 cells) and RAW 264.7 macrophage cells were cultured and maintained in 95% O2-5% CO2 at 37 C in DMEM/F-12 medium (Invitrogen, Grand Island, NY) as described (25). For lipid studies, GT1–7 neurons were grown to 70–75% confluence in six-well or 100-mm2 plates and labeled for 18 h with either 1.5 μCi/ml [3H]ethanolamine (PerkinElmer Life and Analytical Sciences, Inc., Boston, MA) or [3H]arachidonic acid (AA; PerkinElmer) in serum-free medium with 0.01% fatty acid-free BSA (Sigma-Aldrich Corp., St. Louis, MO) and 1% N-2 supplement (Invitrogen). After radiolabeling, neurons were washed two times with serum-free DMEM/F-12 medium (Invitrogen) or Krebs-Ringer bicarbonate glucose buffer (KB). Neurons were subsequently stimulated for various times with ionomycin (Calbiochem, La Jolla, CA) or with norepinephrine (NE; Sigma-Aldrich) in 50 μM ascorbic acid (Sigma-Aldrich). To ascertain whether the immortalized GnRH neurons contained a CB transporter, cells were radiolabeled with [3H]ethanolamine overnight and incubated in KB for 30 min in the absence or presence of the transporter inhibitor, AM 404 (Tocris Cookson Inc., Ellisville, MO). In another transporter experiment, GT1–7 neurons were incubated with 0.5 nM [3H]ANA (PerkinElmer) and 99 nM nonradioactive ANA (BIOMOL Research Laboratories, Inc., Plymouth Meeting, PA) in the absence or presence of AM 404. To determine whether the cells could degrade endocannabinoids, GT1–7 neurons were prelabeled with [3H]ethanolamine overnight and preincubated for 30 min in KB in the presence or absence of the fatty-acid amide hydrolase (FAAH) inhibitors oleylethanolamide or arachidonoyl trifluoromethyl ketone (Biomol). Alternatively, cells were labeled with [3H]AA overnight and stimulated with ionomycin in the absence or presence of the inhibitors. All experiments were terminated by removal of medium.
For GnRH secretion studies, approximately 6 x 105 GT1–7 cells were grown for 48 h in 24-well culture dishes as described (25). At the end of this time, neurons were washed with PBS and preincubated in KB for 1 h, followed by an additional incubation for 30 min, and then exposed to various agents for 30 min. To test effects of a CB receptor agonist on secretion, GT1 neurons were preincubated in KB in the absence or presence of different concentrations of WIN 55,212-2 (Sigma-Aldrich) or CP 55,940 (Tocris Cookson) for 30 min. Neurons were then depolarized with 56 mM [K+] in the absence or presence of the agonist. To determine specificity of the agonist, cells were preincubated with 50 μM WIN 55,212-2 in the presence or absence of 100 μM AM 251 (Tocris Cookson), a selective CB1 receptor antagonist, and stimulated with 56 mM [K+] in the presence and absence of the agonist and/or antagonist. To evaluate whether the CB receptor was coupled to Gi or Go proteins, GT1–7 neurons were incubated overnight with 0.1 μg/ml pertussus toxin (Calbiochem). The next day, cells were preincubated with 50 μM WIN 55,212-2 and stimulated with 56 mM [K+] in the presence and absence of the agonist. Finally, to ascertain whether a CB receptor agonist could influence pulsatile GnRH secretion, GT1–7 neurons were grown on Cytodex beads (Amersham Biosciences Corp., Piscataway, NJ) for 4 d (26). At the end of this time, cells/beads were loaded into 1-ml plastic syringes and perifused at 0.1 ml/min with KB continually gassed with 95% O2-5% CO2 at 37 C for 5 h, and 5-min fractions were collected over the last 4 h. After the first 60 min of perifusion in KB, baseline secretion was assessed in the same buffer over the next 2 h. At the end of this time, one half of the samples were perifused with KB alone or were exposed to 50 μM WIN 55,212-2 for 30 min. Thereafter, all samples were returned to KB for 1 h. Neurons were exposed to 56 mM [K+] for the last 30 min to evaluate their response to depolarization. Medium was collected and analyzed for GnRH contents by RIA (25, 26).
Animals
Adult virgin female C57BL/6J mice (Jackson Laboratory, Bar Harbor, ME) were housed in a temperature- and humidity-controlled room with a 14-h light, 10-h dark cycle (lights on at 0700 h). Rodent chow and water were provided ad libitum. Animals were euthanized by decapitation for tissue samples for RNA analyses. All studies were conducted with an approved protocol from the Duke University Institutional Animal Care and Use Committee.
To evaluate effects of CBs on reproduction, estrous cycles were followed for 3–8 cycles, and only females with consistent 4- to 6-d cycles were used. Mice were anesthetized on metestrus with 100 mg/kg ketamine-20 mg/kg xylazine (ip), and implanted with a 5-mm cannula (Alzet Corp., Palo Alto, CA) into the third ventricle (anterior posterior, –0.08 mm; lateral, 0.0 mm; ventral, –4.7 mm; see Ref.27). The cannula was attached to an Alzet minipump implanted (sc) along the back of the mouse. Pumps were activated with saline for 12 h before implantation and they delivered 0.5 μl/h vehicle (2% Tween 80 in artificial cerebrospinal fluid) or approximately 120 nmol/h of the CB receptor agonist, CP 55, 940 (Tocris Cookson). Animals were followed until proestrus, and then methylene blue was injected into the cannula to confirm its location.
Lipid analyses
After stimulation, cells and medium were collected and analyzed separately or, in certain cases, pooled. Lipids were extracted from samples by sequential addition of 2.8 ml chloroform-methanol (1:2, vol/vol), 0.9 ml chloroform, and 0.9 ml distilled water (28). Samples were vortexed and centrifuged for 5 min at 400 rpm at 23 C to separate the organic and aqueous phases. The aqueous phase was aspirated, and the organic phase was dried to completion under nitrogen gas. Samples were reconstituted in chloroform and loaded onto LK6D high-performance thin-layer chromatography (TLC) plates (Whatman, Inc., Clifton, NJ). Plates were run in a solvent system of ethyl acetate-isooctane-acetic acid-water (93:47:21:100, vol/vol/vol/vol). Lipids were visualized with iodine, identified by comigration with standards (Biomol), scraped from the plates, and quantitated by liquid scintillation counting. Alternatively, plates were sprayed with En3hance (PerkinElmer) and visualized by autoradiography with Biomax MS film (Eastman Kodak Co., Rochester, NY). The identities of 2-AMG and ANA were further confirmed by TLC analyses in an alternative solvent system consisting of petroleum ether-diethyl ether-acetic acid (20:80:1, vol/vol/vol) (29) or by HPLC (30). Briefly, the HPLC apparatus consisted of HP 1090 syringe pumps (Hewlett-Packard, Palo Alto, CA), a Rheodyne 7125 injector (Rheodyne, Cotati, CA), a Kratos 773 UV detector (Kratos Analytical Instruments, Ramsey, NJ), a HP 3395 integrator (Hewlett-Packard), and a Frac-200 fraction collector (Amersham). The separation was achieved on a 250 x 2.1 mm 5-μm C18 Supelcosil column (Supelco, Bellefonte, PA) with an isocratic gradient of acetonitrile-water (50:50, vol/vol) run at 1 ml/min (30). Lipids were identified according to coelution with standards (Biomol) at 214 nm and liquid scintillation counting.
Analyses of gene expression
Total RNA was isolated from GT1–1 and GT1–7 neurons, RAW 264.7 cells, as well as various mouse tissues, using TriReagent (Molecular Research Center, Inc., Cincinnati, OH). First-strand cDNA was synthesized from 5 μg deoxyribonuclease I-treated total RNA, using Superscript II (Invitrogen, Carlsbad, CA). After first-strand synthesis, PCR was performed. For the CB1 receptor, the 5'-GTGTGCTGTTGCTGTTCATTGTGTA-3' and 5'-GCTGTGTTATTGGCGTGCTTGT-3' primers yielded a product of approximately 449 bp that spanned nucleotides 1036–1483 of the murine transcript (accession no. U22948). The CB2 receptor reaction was run with primers 5'-GGTCCTCTCAGCATTGATTTCTTAC-3' and 5'-TTCACATCAGCCTCTGTTTCTGTA-3' that produced a product of approximately 541 bp that included nucleotides 657-1198 of the mouse sequence (accession no. 009924). The FAAH reaction was run with 5'-CCTTCTTACCAAACAACATACC-3' and 5'-CCTTTTTCATGCCCTTCTTC-3' primers that produced a product of approximately 518 bp spanning nucleotides 1091–1609 of the mouse sequence (accession no. MMU82536). The phospholipase D (PLD) primers included 5'-GCTGTAAGGATTCACATTGAC-3' and 5'-AAAAGCTCTATCATCGGTATTC-3' that yielded a product of approximately 201 bp that spanned nucleotides 1136–1315 of the mouse sequence (accession no. NM178728). Mouse hypothalamus, pituitary, and spleen served as positive controls; negative controls included samples run without primers or with RNA from GT1–7 cells that had not undergone first-strand synthesis. The conditions for the PCR consisted of an initial denaturation step at 94 C for 90 sec, followed by 40 cycles at 94 C for 40 sec, 52–58 C for 40 sec, and 72 C for 45 sec. The PCR products were separated on 2% agarose gel, verified by sequencing at the Duke University Medical Center facility, and subcloned into the pCR 2.1 plasmid using the TA cloning kit (Invitrogen).
For Northern blot analyses, total RNA from GT1–1, GT1–3, and GT1–7 cells and various mouse tissues was purified using the mRNA Purification kit (Amersham) as described (31). Ten micrograms of mRNA were fractionated on 1.2% agarose gel and transferred to nylon membrane (Biotrans, Aurora, OH). The cDNA probes for the CB1 and CB2 receptors were radiolabeled with [-32P]-deoxycytidine triphosphate (PerkinElmer) by random primer and diluted to 1.2 x 107 dpm/μg DNA. Blots were hybridized, washed, and exposed to Storage Phophor Plates for visualization using a Typhoon 9200 Variable Mode Imager (Molecular Dynamics, Inc., Sunnyvale, CA).
For in situ hybridization histochemistry (ISHH), mouse brains were collected and frozen in powdered dry ice, then wrapped in Parafilm and stored at –80 C in sealed tubes. Twelve-micrometer coronal cryosections were obtained from the rostral preoptic area encompassing the organum vasculosum of the lamina terminalis region through the medial preoptic area (27). Sections were affixed to gelatin-coated slides and stored at –80 C until use. The cDNA probe for the CB1 receptor was made by PCR as described above. The cDNA template for mouse GnRH (mGnRH) was a 246-bp fragment corresponding to bases 39–285 of the mGnRH cDNA (accession no. AY39940) prepared using a forward primer of 5'-CGGCATTCTACTGCTGACTG-3' and a reverse primer sequence of 5'-CTTCTTCTGCCTGGCTTCCT-3'. Fragments were cloned into a TOPO-TA vector (Invitrogen) and sequenced to verify identity. Standard in vitro transcription was used to prepare 35S-labeled cRNA probe to the CB1 receptor and digoxigenin-labeled probes to mGnRH mRNA. Dual-label ISHH was performed as described previously (32). Sections were thawed, fixed, and prehybridized, before applying a mixture of 35S-labeled cRNA probes for the CB1 receptor (1 x 106 cpm) and 0.5 μl of the digoxigenin-labeled mGnRH probes in 25 μl hybridization buffer. To verify specificity, we hybridized representative sections to 35S-labeled sense strand probes in buffer with or without digoxigenin-labeled cRNA probes for mGnRH mRNA. Sections were hybridized at 52 C overnight under glass coverslips, washed, and processed for immunocytochemical detection of digoxigenin-labeled probes for mGnRH mRNAs as described (32). After detection of the dioxigenin-labeled probes, we used standard emulsion autoradiographical procedures (NTB3 emulsion; Kodak) to visualize radiolabeled probes for the CB1 receptor mRNA (7-d exposure). To analyze dual-label ISHH studies, we used BioQuant Windows (R and M Biometrics, Nashville, TN) interfaced to a Leitz Laborlux microscope through a 3CCD color video camera (Hitachi Denshi America, Ltd., Woodbury, NY). Sections from the preoptic region containing GnRH neurons (four to six sections/animal) were examined in five animals, and the number of GnRH neurons with and without CB1 receptor transcripts was determined.
Statistics
The data are presented as means and SEM. Student’s t tests were used to compare two groups; other comparisons were performed by ANOVA with subsequent Dunnett t or Bonferroni post hoc tests. P < 0.05 was considered significant.
Pulsatile GnRH release was analyzed by several different procedures (26). First, the threshold for pulse detection was set at three times the coefficient of variation for assay for the concentration of that particular sample (33). Patterns of secretion were also examined using cluster analyses (34) and the DETECT program (35). Because these procedures gave similar findings, the present results are depicted from the DETECT program.
Results
Immortalized GnRH neurons release 2-AMG
Because endocannabinoids are synthesized by enzymes that are activated by Ca2+-dependent processes, many investigators use ionomycin to stimulate their production (see Ref.20). The immortalized neurons were radiolabeled overnight with [3H]AA and stimulated the next day with 5 μM ionomycin. This agent produced a time-dependent increase in 2-AMG accumulation with an enhancement observed within the first 10 min of stimulation (Fig. 1A). AA and diacylglycerol (DAG) release was also augmented over the same time (Fig. 1B). To ascertain whether endocannabinoid production was responsive to other agents, such as neurotransmitters, GT1–7 neurons were stimulated with 100 μM NE. The NE also released 2-AMG, DAG, and AA (Fig. 1, C and D). Collectively, these data demonstrate that the immortalized GnRH neurons synthesize an endocannabinoid and that NE and other agents that increase entry of Ca2+ into cells can stimulate release and accumulation of 2-AMG in the medium.
Immortalized GnRH neurons release and transport ANA
Because the neurons produce 2-AMG, we wanted to determine whether they could make other endocannabinoids. Recently, a PLD species has been identified that generates the endocannabinoid, ANA (36). RT-PCR analyses and subsequent sequencing reveal that the mouse hypothalamus and GT1 cells contain transcripts for this form of PLD (Fig. 1E). No products were found when the PCR was run with RNA alone or without primers.
To ascertain whether the immortalized neurons synthesized ANA, GT1–7 neurons were radiolabeled overnight with [3H]ethanolamine and incubated the next day for 30 min with 100 μM AM 404, an ANA transporter inhibitor. The presence of the inhibitor alone increased ANA accumulation in medium by more than 500% (Fig. 1F). ANA was also identified by TLC and autoradiography after stimulation with 5 μM ionomycin (Fig. 1G). To determine whether ANA could be transported into immortalized GnRH neurons, [3H]ANA was added to the medium, and uptake was examined in the absence or presence of AM 404. In the absence of the inhibitor, neurons rapidly removed the radioendocannabinoid from the medium, with half-maximal uptake occurring within approximately 4 min after its addition (Fig. 1H). By comparison, AM 404 suppressed intracellular uptake of [3H]ANA from the medium. Together, these findings show that the immortalized GnRH neurons synthesize ANA and can rapidly remove it from the medium.
Immortalized GnRH neurons degrade endocannabinoids
FAAH degrades endocannabinoids in many different cells (37). To investigate whether the immortalized GnRH neurons contained FAAH transcripts, RT-PCR was performed (Fig. 2A). PCR products of appropriate size were detected not only in mouse pituitary but also in hypothalamus (a brain region containing GnRH neurons in vivo) (24). Products of the same size were also present in the GT1 neurons. No products were found when the PCR was run with RNA alone or without primers.
To study whether the FAAH was functional, GT1–7 cells were radiolabeled overnight with [3H]ethanolamine and incubated the next day for 30 min with KB in the presence or absence of the enzyme inhibitors oleylethanolamide or arachidonyl trifluoromethyl ketone. Basal release of ANA was enhanced more than 180-fold by the inhibitors (Fig. 2B). To evaluate whether the enzyme could degrade 2-AMG, GT1–7 neurons were labeled with [3H]AA and incubated with KB or with 5 μM ionomycin in the absence or presence of each inhibitor. Ionomycin significantly stimulated release of 2-AMG, and the accumulation of the endocannabinoid was further enhanced by addition of FAAH inhibitors (Fig. 2C). Because FAAH is an enzyme found within cells, these findings show that the GT1 neurons can remove ANA and 2-AMG from the medium and degrade these endocannabinoids.
Immortalized GnRH neurons contain CB1 and CB2 receptor transcripts
Because the immortalized GnRH neurons produced, transported, and degraded endocannabinoids, we next examined whether they contained receptors for these bioactive lipids. Northern blots were run for the CB1 receptor with lysates from mouse cerebral cortex and hypothalamus, as well as from the three immortalized neuronal cell lines. The transcript for the CB1 receptor was approximately 6 kb in all samples (Fig. 3A), a size consistent with the CB1 receptor in rodents (18). By contrast, whereas transcripts for the CB2 receptor could be visualized in mouse spleen by Northern blot, they were not observable in mouse hypothalamus or in the three immortalized GnRH neuronal cell lines (data not shown).
Because levels of the CB1 receptor were low and concentrations of the CB2 receptor were undetectable by Northern blot, we used RT-PCR to further examine expression of these transcripts in the immortalized GnRH neurons. This more sensitive method more readily identified CB1 and CB2 receptor RT-PCR products in the GT1–1 and GT1–7 cells, as well as in mouse hypothalamus (Fig. 3, B and C). Subsequent sequencing confirmed the identity of these products. Hence, the immortalized GnRH neurons contain transcripts for both CB receptors, albeit at low levels.
CBs inhibit GnRH secretion from immortalized GnRH neurons
To determine whether the CB receptors were functional in the immortalized GnRH neurons, GT1–7 neurons were incubated in KB alone or were depolarized with 56 mM [K+] in the absence or presence of a CB receptor agonist, WIN 55,212-2. The [K+] stimulus augmented GnRH secretion more than 3-fold over baseline, so that any inhibitory responses to CBs could be clearly visualized (Fig. 4A). WIN 55,212-2 reduced this response in a dose-dependent manner. An identical effect was observed with another CB receptor agonist, CP 55,940 (data not shown). Parenthetically, neither agonist influenced basal GnRH release (data not shown). To evaluate response specificity, the experiment was repeated in the presence of the selective CB1 receptor antagonist, AM 251. Again, 56 mM [K+] stimulated GnRH secretion, and this was reduced by more than 2-fold with 50 μM WIN 55,212-2 (Fig. 4B). The reduction in [K+]-stimulated GnRH secretion by the agonist was blocked with 100 μM AM 251. By comparison, the CB1 receptor antagonist alone exerted no effects on [K+]-stimulated GnRH release (Fig. 4B) or on basal GnRH release (data not shown). To ascertain whether the WIN 55,212-2 response was mediated by Gi/Go proteins, GT1–7 neurons were incubated in medium alone or were treated overnight with 0.1 μg/ml pertussus toxin. Although the toxin did not significantly influence GnRH secretion under basal conditions or in response to [K+] stimulation (data not shown), inhibition of GnRH release by the CB receptor agonist was abrogated by the toxin (Fig. 4C). Collectively, these data demonstrate that agonist activation of CB receptors reduces [K+]-stimulated GnRH secretion through a Gi/Go-mediated mechanism.
An important characteristic of immortalized GT1 neurons and GnRH neurons in vivo is their pulsatile release of GnRH (26, 38, 39, 40). To determine whether CBs alter pulsatility, GT1–7 cells were perifused for 5 h. After 1 h of stabilization, medium was collected over the next 2 h to establish baseline. During this period, GnRH was secreted in a pulsatile manner, with a frequency of 1.41 ± 0.21 pulses/h, duration of 17.4 ± 2.2 min, and interpulse interval of 25.3 ± 2.8 min (Fig. 4, D and E). After exposure to 50 μM WIN 55,212-2 for 30 min, pulsatile activity ceased over the next 1 h (Fig. 4E); responses to vehicle did not affect the pattern of GnRH release (Fig. 4D). GnRH release remained very low, and pulsatility was suppressed in the CB agonist-treated cultures even when they were returned to perifusion with KB (Fig. 4E). All samples responded to 56 mM [K+] depolarization at the end of the experiment. Hence, the previously CB-treated cells can respond to depolarization; albeit not as robustly as cells treated with KB alone. These data show that a CB receptor agonist can perturb pulsatile GnRH secretion from the immortalized GnRH neurons.
CB effects in vivo
To determine whether CBs influence GnRH function in vivo, a CB receptor agonist was administered to estrous cycling female mice. Animals were surgically implanted on metestrus with osmotic minipumps that delivered vehicle or low levels of the CB receptor agonist (CP 55,940) into the third ventricle of the hypothalamus, a region adjacent to the perikarya of GnRH neurons. Before cannulation, estrous cycles were monitored over at least 2 wk, and they were found to be approximately the same duration for both groups (vehicle controls, 5.0 ± 0.41 d; agonist-treated mice, 4.9 ± 0.32 d; n = 7–9 females). After cannulation, vehicle controls maintained a 5-d estrous cycle, whereas proestrus was delayed by at least 2 d in the agonist-treated mice (Fig. 4F). These findings show that CBs can exert effects on reproduction at the level of the hypothalamus.
The effects of CBs on estrous cycle duration could be exerted on GnRH neurons themselves and/or on neurons that control GnRH function. To distinguish between these possibilities, dual-label ISHH experiments were conducted with female C57BL/6J mice. Animals were euthanized without regard to the stage of estrous cycle. An examination of a total of 117 GnRH neurons revealed that most neurons were devoid of autoradiographic grains, and few grains were seen in the remaining neurons (Fig. 5, A–C). Despite this fact, autoradiographic grains were prominent is areas surrounding GnRH neurons. An absence of either a GnRH or CB1 receptor signal in the sense-strand controls verified the specificity of the probes (data not shown). Hence, these findings suggest that if the CB1 receptor is expressed in GnRH neurons, it is at a very low level and most CB1 receptors are localized in neighboring cells.
Discussion
CBs have long been known to exert potent effects on reproduction (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17). Although some of these influences may be mediated at the levels of the pituitary and gonads (41, 42, 43), the primary effects have been ascribed to hypothalamic action because synthetic GnRH can rescue ovulation in CB-treated animals (1, 8, 13, 14, 17). Despite this fact, the mechanism of action of CBs on GnRH function is obscure. Although ANA has been identified in hypothalamus (21), the present studies demonstrate that the immortalized hypothalamic GnRH neurons contain a complete and functional CB system. The presence of this system in the immortalized neurons, coupled with the reported in vivo effects of CB agonists on LH secretion and/or ovulation in female rodents and humans (1, 5, 6, 10, 13, 16, 17), suggests that endocannabinoids may play an important role in GnRH physiology and reproduction.
In the present report, we have found that the immortalized GnRH neurons synthesize at least two different endocannabinoids, 2-AMG and ANA. Besides endocannabinoids, Ca2+-ionophore and NE produce corresponding increases in DAG and free AA. These latter findings are consistent with a report showing calcium-dependent hydrolysis of phosphoinositides to be associated with liberation of inositol phosphates, as well as DAG and AA in GT1–7 cells (25). In the present experiment the delayed accumulation of 2-AMG, with respect to DAG, suggests that this endocannabinoid is derived from phospholipid hydrolysis and is modified subsequently by DAG lipase to produce 2-AMG (44). By contrast, ANA production may proceed through the action of PLD (38, 45). In this regard, it is noteworthy that the immortalized GnRH neurons possess transcripts for a PLD form that has been reported to generate ANA (36). Recently, PLD has been shown to be activated by GnRH (46). Because GnRH can inhibit its own secretion from the immortalized neurons (47) and because an ultrashort feedback system resides on GnRH neurons in vitro and in vivo (48, 49, 50), our results suggest that endocannabinoids may play a role in this process by autocrine; paracrine; or, more likely, juxtacrine interactions with cells in close apposition to GnRH neurons.
Besides synthesizing and releasing endocannabinoids, the immortalized GnRH neurons can also remove them from the medium. Under normal conditions, 2-AMG and ANA are transiently active and reenter the cell by passive diffusion and/or active transport (51, 52, 53). Although the existence of an endocannabinoid transporter is controversial (51, 52, 53), the present experiments show that GT1–7 cells can rapidly remove ANA from the medium and this action is blocked by a putative transporter inhibitor. Together, these data suggest that the immortalized GnRH neurons contain a CB transporter that can remove 2-AMG and/or ANA from the medium to rapidly curtail their biological actions.
Aside from removing endocannabinoids from the medium, the immortalized GnRH neurons also degrade these lipids. Two different inhibitors of FAAH, oleylethanolamide and arachidonyl trifluoromethyl ketone, enhance 2-AMG and ANA accumulation. Interestingly, in some systems, ANA is taken up and rapidly converted to other eicosanoids or is degraded by FAAH to AA and ethanolamine (54). The 2-AMG can be hydrolyzed to AA, by either a monoacylglycerol lipase or FAAH, and reesterified into phospholipid (55, 56). Furthermore, some in situ studies reveal that endocannabinoids recycled into AA can be converted into eicosanoids such as thromboxanes and prostaglandins (57). Because prostaglandin E2 can stimulate GnRH secretion from median eminence tissue fragments in vitro (58) and from the immortalized GnRH neurons (59), their potential synthesis from endocannabinoids suggests that CBs may be intimately linked with various eicosanoid signaling pathways to modulate GnRH release.
In addition to synthesizing endocannabinoids, the immortalized GnRH neurons also contain transcripts for the CB1 and CB2 receptors, and activation of these receptors inhibits GnRH secretion. In perifusion experiments, a CB receptor agonist completely disrupts pulsatile secretion of GnRH. It is noteworthy that, whereas GnRH secretion can be stimulated with [K+] depolarization subsequent to treatment with a CB agonist, the response is less robust than in cells previously exposed to KB alone. A similar relationship is seen in vivo, where THC is more efficient than atropine in maintaining blockade of LH responses to electrical stimulation of the preoptic area (60). Although a CB receptor agonist suppresses [K+]-stimulated GnRH secretion and this response is blocked by a CB receptor antagonist in the immortalized neurons, it should be emphasized that the levels of the CB1 and CB2 receptor transcripts in the immortalized cells are low. Moreover, [K+]-stimulated GnRH secretion is only partially inhibited by the highest concentration of the receptor agonist. Thus, CBs appear to exert modest effects on GnRH secretion from the immortalized neurons.
Effects of CBs were next evaluated in adult female C57BL/6J mice. A CB receptor agonist was found to prolong the estrus cycle in the mice by at least 2 d. These data are in agreement with other investigations where peripheral administration of CBs is reported to delay ovulation in a number of species, including the human (1, 5, 6, 10, 13, 16, 17). There are a number of different reasons to suspect that the reproductive effects of CBs are exerted primarily at the level of the hypothalamus. First, in the present study, delivery of the agonist into the third ventricle of the hypothalamus is sufficient to delay the estrous cycle in mice by several days. Second, peripherally administered GnRH restores LH secretion in CB-treated animals (8, 13, 14). Third, electrical stimulation of the medial preoptic area alleviates the THC-induced blockade of the preovulatory LH surge (61). Finally, ANA is reported to depress GnRH release from the medial basal hypothalamus in vitro (62). Although CBs influence reproduction through hypothalamic mechanisms, it has not been clear whether these lipids exert their effects directly on GnRH neurons or on neighboring cells that control GnRH release. Importantly, CB1 receptors are localized in the preoptic area and hypothalamus (63, 64), brain areas where the GnRH neurons reside. Results from our dual-label ISHH experiments suggest that few hypothalamic GnRH neurons contain CB1 receptor transcripts. By comparison, the distribution of autoradiographic grains indicates that many cells in the vicinity of GnRH neurons possess substantial levels of CB1 receptor transcripts. Importantly, a recent report by Herbison’s group (65) using single-cell PCR from transgenic mice has shown that CB2 receptors are expressed in approximately 25% of native hypothalamic GnRH neurons. Thus, these and the present findings suggest that CBs may exert their effects on reproduction not only through cells that control GnRH function but also through GnRH neurons themselves.
GnRH neurons in vivo have long been known to respond to ligands that stimulate phosphoinositide hydrolysis and increased intracellular calcium and DAG production (see Ref.25). This cascade may also lead to the synthesis of endocannabinoids, such as 2-AMG and ANA. These lipids, in turn, can bind to CB receptors on neighboring cells and GnRH neurons, thereby exerting some feedback control on GnRH function. In this regard, it is instructive that recent electrophysiological evidence from hippocampal and cerebellar neurons indicates that CBs can send retrograde signals back to presynaptic terminals to control neurotransmitter release (66, 67, 68). If the same processes occur in GnRH neurons in vivo, then the generation of endocannabinoids and their degradation to various eicosanoid products could serve as a novel mechanism for regulating GnRH secretion and, ultimately, reproduction in mammals.
Acknowledgments
We thank Dr. P. Mellon (University of California-San Diego, La Jolla, CA) for providing the GT1 cells, Dr. A. Arimura (Tulane University, New Orleans, LA) for the A772 antiserum, Dr. Nancy E. Buckley (National Institute of Neurological Disorders and Stroke, Bethesda, MD) for the mouse CB2 receptor cDNA, Dr. Clifford D. Carpenter (University of Massachusetts, Amherst, MA) for the dual labeling of cells for ISHH, and Ms. Jiechun Zhou (Duke University Medical Center, Durham, NC) for providing the RT-PCR results for PLD.
Footnotes
This work was partially supported by the Elizabeth and Ruth Gammon Research Fund (to C.M.G.); a Summer Research Fellowship from The Endocrine Society, a Howard Hughes Forum in Neuroscience Research Fellowship, and a Sigma Xi Grants-in-Aid Award (to G.M.F.); National Institutes of Health Grant HD27305 (to S.L.P.); and Unrestricted Research Funds (to W.C.W.).
Abbreviations: AA, [3H]Arachidonic acid; 2-AMG, 2-arachidonyl monoacylglycerol; ANA, anandamide; CB, cannabinoid; DAG, diacylglycerol; FAAH, fatty-acid amide hydrolase; ISHH, in situ hybridization histochemistry; KB, Krebs-Ringer bicarbonate glucose buffer; mGnRH, mouse GnRH; NE, norepinephrine; PLD, phospholipase D; THC, 9-tetrahydrocannabinol; TLC, thin-layer chromatography.
References
Bloch E, Thysen B, Morrill GA, Gardner E, Fugimoto G 1978 Effects of cannabinoids on reproduction and development. Vitam Horm. 36:203–259
Kolodny RC, Masters WH, Kolodner RM, Toro G 1974 Depression of plasma testosterone levels after chronic intensive marihuana use. N Engl J Med. 290:872–874
Collu R, Letarte J, Leboeuf G, Ducharme JR 1975 Endocrine effects of chronic administration of psychoactive drugs to prepuberal male rats. I: 9-tetrahydrocannabinol. Life Sci. 16:533–542
Solomon J, Cocchia AM, DiMartino R 1977 Effect of -9-tetracannabinol on uterine and vaginal cytology of ovariectomized rats. Science 195:875–877
Smith CG, Almirez RG, Berenberg J, Asch RH 1983 Tolerance develops to the disruptive effects of 9-tetrahydrocannabinol on primate menstrual cycle. Science 219:1453–1455
Wenger T, Croix D, Tramu G 1988 The effects of chronic prepubertal administration of marihuana (-9-tetrahydrocannabinol) on the onset of puberty and the postpubertal reproductive functions in female rats. Biol Reprod. 39:540–545
Chakravarty I, Shah PG, Sheth AR, Ghosh JJ 1979 Mode of action of -9-tetrahydrocannabinol on hypothalamo-pituitary function in adult female rats. J Reprod Fertil. 57:113–115
Tyrey L 1978 -9-Tetrahydrocannabinol suppression of episodic luteinizing hormone secretion in ovariectomized rats. Endocrinology 102:1808–1814
Steger RW, Silverman AY, Siler-Khodr TM, Asch RH 1980 The effects of 9-tetrahydrocannabinol on the positive and negative feedback control of luteinizing hormone release. Life Sci. 27:1911–1916
Mendelson JH, Mello NK, Ellingboe J, Skupny AST, Lex BW, Griffin M 1986 Marihuana smoking suppresses luteinizing hormone in women. J Pharmacol Exp Ther. 237:862–866
Kramer J, Ben-David M 1974 Suppression of prolactin secretion by acute administration of 9-tetrahydrocannabinol in rats. Proc Soc Exp Biol Med. 147:482–484
Daley JD, Branda LA, Rosenfeld J, Younglai EV 1974 Increase of serum prolactin in male rats by (-)-trans-9-tetrahydrocannabinol. J Endocr. 63:415–416
Ayalon D, Nir I, Cordova T, Bauminger S, Puder M, Naor Z, Kashi B, Zor U, Harrell A, Lindner HR 1977 Acute effect of 1-tetrahydrocannabinol on the hypothalamo-pituitary-ovarian axis in the rat. Neuroendocrinology 23:31–42
Smith CG, Besch NF, Smith RG, Besch PK 1979 Effect of tetrahydrocannabinol on the hypothalamic-pituitary axis in the ovariectomized rhesus monkey. Fertil Steril. 31:335–339
Wenger T, Rettori V, Snyder GD, Dalterio S, McCann SM 1987 Effects of -9-tetrahydrocannabinol on the hypothalamic-pituitary control of luteinizing hormone and follicle-stimulating hormone secretion in adult male rats. Neuroendocrinology 46:488–493
Nir I, Ayalon D, Tsafriri A, Cordova T, Lindner HR 1973 Suppression of the cyclic surge of luteinizing hormone secretion and of ovulation in the rat by 1-tetrahydrocannabinol. Nature 243:470–471
Asch RH, Fernandez EO, Smith CG, Pauerstein CJ 1979 Precoital single doses of 9-tetrahydrocannabinol block ovulation in the rabbit. Fertil Steril. 31:331–334
Matsuda LA, Lolait SJ, Brownstein MJ, Young AC, Bonner TI 1990 Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature 346:561–564
Munro S, Thomas KL, Abu-Shaar M 1993 Molecular characterization of a peripheral receptor for cannabinoids. Nature 365:61–65
Piomelli S 2003 The molecular logic of endocannabinoid signaling. Nat Rev Neurosci. 4:873–884
Wenger T, Gerendai I, Fezza F, González S, Bisogno T, Fernandez-Ruiz J, Di Marzo V 2002 The hypothalamic levels of the endocannabinoid, anandamide, peak immediately before the onset of puberty in female rats. Life Sci. 70:1407–1414
Mailleux P, Vanderhaeghen J-J 1992 Distribution of neuronal cannabinoid receptor in the adult rat brain: a comparative receptor binding radiography and in situ hybridization histochemistry. Neuroscience 48:655–668
Moldrich G, Wenger T 2000 Localization of the CB1 cannabinoid receptor in the rat brain. An immunohistochemical study. Peptides 21:1735–1742
Merchenthaler I, Grcs T, Sétàlò G, Petrusz P, Flerkò B 1984 Gonadotropin-releasing hormone (GnRH) neurons and pathways in the rat brain. Cell Tissue Res. 237:15–29
Kreda SM, Sumner M, Fillo S, Ribeiro CM, Luo GX, Xie W, Daniel KW, Shears S, Collins S, Wetsel WC 2001 1-Adrenergic receptors mediate LH-releasing hormone secretion through phospholipases C and A2 in immortalized hypothalamic neurons. Endocrinology 142:4839–4851
Wetsel WC, Valenca MM, Merchenthaler I, Liposits Z, Lopez F, Weiner RI, Mellon P, Negro-Vilar A 1992 Intrinsic pulsatile secretory activity of immortalized luteinizing hormone-releasing hormone-secreting neurons. Proc Natl Acad Sci USA. 87:4149–4153
Franklin KBJ, Paxinos G 1997 The mouse brain in stereotaxic coordinates. San Diego: Academic Press; 1–186
Gammon CM, Allen AC, Morell P 1989 Bradykinin stimulates phosphoinositide hydrolysis and mobilization of arachidonic acid in dorsal root ganglion neurons. J Neurochem. 53:95–101
Sugiura T, Kondo S, Sukagawa A, Nakane S, Shinoda A, Itoh K, Yamashita A, Waku K 1995 2-Arachidonoylgylcerol: a possible endogenous cannabinoid receptor ligand in brain. Biochem Biophys Res Commun. 215:89–97
Fontana A, Di Marzo V, Cadas H, Piomelli D 1995 Analysis of anandamide, an endogenous cannabinoid substance, and of other natural N-acylethanolamines. Prostaglandins Leukot Essent Fatty Acids 53:301–308
Wetsel WC, Liposits Z, Seidah NG, Collins S 1995 Expression of candidate pro-LHRH processing enzymes in rat hypothalamus and an immortalized hypothalamic neuronal cell line. Neuroendocrinology 62:166–177
Ottem EN, Godwin JG, Krishnan S, Petersen SL 2004 Dual-phenotype GABA/glutamate neurons in adult preoptic area: sexual dimorphism and function. J Neurosci. 24:8097–8105
Lumpkin MD, DePaolo LV, Negro-Vilar A 1984 Pulsatile release of follicle-stimulating hormone in ovariectomized rats is inhibited by porcine follicular fluid (inhibin). Endocrinology 114:201–206
Veldhuis J, Johnson M 1986 Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol. 250:E486–E493
Oerter KE., Guardabasso V, Rodbard D 1986 Detection and characterization of peaks and estimation of instantaneous secretory rate for episodic pulsatile hormone secretion. Comput Biomed Res. 19:170–190
Okamoto Y, Morishita J, Tsuboi K, Tonai T, Ueda N 2004 Molecular characterization of a phospholipase D generating anandamide and its congers. J Biol Chem. 279:5298–5305
Katayama K, Ueda N, Kurahashi Y, Suzuki H, Yamamoto S, Kato I 1997 Distribution of anandamide aminohydrolase in rat tissues with special reference to small intestine. Biochim Biophys Acta. 1347:212–218
Martinez de la Escalera G, Choi ALH, Weiner RI 1992 Generation and synchronization of gonadotropin-releasing hormone (GnRH) pulses: intrinsic properties of the GT1–1 GnRH neuronal cell line. Proc Natl Acad Sci USA. 89:1852–1855
Krsmanovic LZ, Stojilkovic S, Merelli F, Dufour S, Virmani M, Catt KJ 1992 Calcium signaling and episodic secretion of gonadotropin-releasing hormone in hypothalamic neurons. Proc Natl Acad Sci USA. 89:8462–8466
Knobil E 1980 The neuroendocrine control of the menstrual cycle. Recent Prog Horm Res. 36:52–88
Burstein S, Hunter SA, Shoupe TS 1979 Cannabinoid inhibition of rat luteal cell progesterone synthesis. Res Commun Chem Pathol Pharmacol. 24:413–416
Gerard CM, Mollereau C, Vassart G, Parmentier M 1991 Molecular cloning of a human cannabinoid receptor which is also expressed in testis. Biochem J. 279:129–134
Pagotto U, Marsicano G, Fezza F, Theodoropoulou M, Gruber Y, Stalla J, Arzberger T, Milone A, Losa M, Di Marzo V, Lutz B, Stalla GK 2001 Normal human pituitary gland and pituitary adenomas express cannabinoid receptor type 1 and synthesize endogenous cannabinoids: first evidence for a direct role of cannabinoids on hormone modulation at the human pituitary level. J Clin Endocrinol Metab. 86:2687–2696
Allen AC, Gammon CM, Ousley AH, McCarthy KD, Morell P 1992 Bradykinin stimulates arachidonic acid release through the sequential actions of an sn-1 diacylglycerol lipase and a monoacylglycerol lipase. J Neurochem. 58:1130–1139
Di Marzo V, Fontana A, Cadas H, Schinelli S, Cimino G, Schwartz JC, Piomelli D 1994 Formation and inactivation of endogenous cannabinoid anandamide in central neurons. Nature 372:686–691
Zheng L, Krsmanovic LZ, Vergara LA, Catt KJ, Stojilkovic SS 1997 Dependence of intracellular signaling and neurosecretion on phospholipase D activation in immortalized gonadotropin-releasing hormone neurons. Proc Natl Acad Sci USA. 94:1573–1578
Krazmanovic LZ, Stojilkovic SS, Mertz LM, Tomic M, Catt KJ 1993 Expression of gonadotropin-releasing hormone receptors and autocrine regulation of neuropeptide release in immortalized hypothalamic neurons. Proc Natl Acad Sci USA. 90:3908–3912
Hyyppa M, Motta M, Martini L 1971 ‘Ultrashort’ feedback control of follicle-stimulating hormone-releasing factor secretion. Neuroendocrinology 7:227–235
Castro JCB, Khorram O, McCann SM 1985 Possible negative ultra-short loop feedback of luteinizing hormone releasing hormone (LHRH) in the ovariectomized rat. Proc Soc Exp Biol Med. 179:132–135
Valenca MM, Johnston CA, Ching M, Negro-Vilar A 1987 Evidence for a negative ultrashort loop feedback mechanism operating on the luteinizing hormone-releasing hormone neuronal system. Endocrinology 121:2256–2259
Beltramo M, Stella N, Calignano A, Lin SY, Makriyannis A, Piomelli D 1997 Functional role of high-affinity anandamide transport, as revealed by selective inhibition. Science 277:1094–1097
Glaser ST, Abumrad NA, Fatade F, Kaczocha M, Studholme KM, Deutsch DG 2003 Evidence against the presence of an anandamide transporter. Proc Natl Acad Sci USA. 100:4269–4274
Fegley D, Kathuria S, Mercier R, Li C, Goutopoulos A, Makriyannis A, Piomelli D 2004 Anandamide transport is independent of fatty-acid amide hydrolase activity and is blocked by the hydrolysis-resistant inhibitor AM1172. Proc Natl Acad Sci USA. 101:8756–8761
Fowler CJ, Jonsson K, Tiger G 2001 Fatty acid amide hydrolase: biochemistry, pharmacology, and therapeutic possibilities for an enzyme hydrolyzing anandamide, 2-arachidonylglycerol, palmitoylethanolamine, and oleamide. Biochem Pharmacol. 62:517–526
Cravatt B, Giang D, Mayfield S, Bolger D, Lerner R, Guila N 1996 Molecular characterization of an enzyme that degrades neuromodulatory fatty acid amides. Nature 384:84–87
Di Marzo V 1999 Biosynthesis and inactivation of endocannabinoids: relevance to their proposed role as neuromodulators. Life Sci. 65:645–655
Kozak KR, Crews BC, Morrow JD, Wang L, Ma YH, Weinander R, Jakobsson P-J, Marnett LJ 2002 Metabolism of the endocannabinoids, 2-arachidonylglycerol and anandamide, into prostaglandin, thromboxane, and prostacyclin glycerol esters and ethanolamides. J Biol Chem. 277:44877–44885
Ojeda SR, Negro-Vilar A, McCann SM 1979 Release of prostaglandin E2 by hypothalamic tissue: evidence for their involvement in catecholamine-induced luteinizing hormone-releasing hormone release. Endocrinology 104:617–624
Ma YJ, Berg-von der Emde K, Rage F, Wetsel WC, Ojeda S 1997 Hypothalamic astrocytes respond to transforming growth-factor with the secretion of neuroactive substances that stimulate the release of luteinizing hormone-releasing hormone. Endocrinology 138:19–25
Tyrey L 1992 -9-Tetrahydrocannabinol attenuates luteinizing hormone release induced by electrochemical stimulation of the medial preoptic area. Biol Reprod. 47:262–267
Murphy LL, Tyrey L 1986 Induction of luteinizing hormone release by electrical stimulation of the medial preoptic area in 9-tetrahydrocannabinol-blocked proestrus rats. Neuroendocrinology 43:471–475
Fernández-Solari J, Scorticati C, Mohn C, De Laurentiis A, Billi S, Franchi A, McCann SM, Rettori V 2004 Alcohol inhibits luteinizing hormone-releasing hormone release by activating the endocannabinoid system. Proc Natl Acad Sci USA. 101:3264–3268
Herkenham M, Lynn AB, Johnson MR, Melvin LS, de Costa BR, Rice KC 1991 Characterization and localization of cannabinoid receptors in rat brain: a quantitative in vitro autoradiographic study. J Neurosci. 11:563–583
Matsuda LA, Bonner TI, Lolait SJ 1993 Localization of cannabinoid receptor mRNA in rat brain. J Comp Neurol. 327:535–550
Todman MG, Han S-K, Herbison AE 2005 Profiling neurotransmitter receptor expression in mouse gonadotropin-releasing hormone neurons using green fluorescent protein-promoter transgenics and microarrays. Neuroscience 132:705–712
Kreitzer AC, Regehr WG 2001 Retrograde inhibition of presynaptic calcium influx by endogenous cannabinoids at excitatory synapses onto purkinje cells. Neuron 29:717–727
Ohno-Shosaku T, Maejima T, Kano M 2001 Endogenous cannabinoids mediate retrograde signals from depolarized postsynaptic neurons to presynaptic terminals. Neuron 29:729–738
Wilson RI, Nicoll RA 2001 Endogenous cannabinoids mediate retrograde signaling at hippocampal synapses. Nature 410:588–592(C. Michael Gammon, G. Mar)
Department of Biology (S.L.P.), Neuroscience and Behavior Program, and Center for Neuroendocrine Studies, University of Massachusetts, Amherst, Massachusetts 01003
Abstract
Cannabinoids (CBs) exert untoward effects on reproduction by reducing LH secretion and suppressing gonadal function. Recent evidence suggests these effects are due primarily to hypothalamic dysfunction; however, the mechanism is obscure. Using immortalized hypothalamic GnRH neurons, we find these cells produce and secrete at least two different endocannabinoids. After release, 2-arachidonyl monoacylglycerol and anandamide are rapidly transported into GnRH neurons and are degraded to other lipids by fatty-acid amide hydrolase. The immortalized GnRH neurons also possess CB1 and CB2 receptors that are coupled to Gi/Go proteins whose activation leads to inhibition of GnRH secretion. In perifusion experiments, CBs block pulsatile release of GnRH. When a CB receptor agonist is delivered into the third ventricle of adult female mice, estrous cycles are prolonged by at least 2 d. Although in situ hybridization experiments suggest either that GnRH neurons in vivo do not possess CB1 receptors or that they are very low, transcripts are localized in close proximity to these neurons. Inasmuch as GnRH neurons in vivo possess G protein receptors that are coupled to phospholipase C and increased intracellular Ca2+, these same neurons should also be able to synthesize endocannabinoids. These lipids, in turn, could bind to CB receptors on neighboring cells, and perhaps GnRH neurons, to exert feedback control over GnRH function. This network could serve as a novel mechanism for regulating GnRH secretion where reproductive functions as diverse as the onset of puberty, timing of ovulation, duration of lactational infertility, and initiation/persistence of menopause may be affected.
Introduction
CANNABINOIDS (CBs) have long been known to exert potent negative effects on rodent, primate, and human reproduction (1). Chronic administration of CBs to male rodents and humans reduces sperm counts, depresses serum testosterone concentrations, and may suppress levels of LH in serum (1, 2, 3). In females, chronic CB exposure delays sexual maturation, disrupts menstrual or estrous cycles, depresses follicular maturation of the ovary, alters uterine and vaginal cytology, and may reduce contents of LH and sex steroids in blood (1, 4, 5, 6). Injections of 9-tetrahydrocannabinol (THC) over 10 d can decrease hypothalamic GnRH concentrations in a dose-dependent manner (7). Despite these marked changes, some of the symptoms are alleviated over time due to tolerance (5), and there is rapid reestablishment of function after cessation of CB use or exogenous gonadotropin administration (1, 2). By contrast, acute THC exposure inhibits pulsatile LH secretion (8), and it also blocks estrogen or estrogen/progesterone-mediated positive feedback on the LH surge in ovariectomized rats (9). Although very low doses of CBs suppress LH secretion in rodents and humans (8, 10), dose-dependent effects are evident as the duration of blockade is prolonged (8). In contrast to LH, effects of CBs on FSH and prolactin secretion are more variable (1, 11, 12, 13, 14, 15). If THC is given on the afternoon of proestrus, the LH surge and ovulation are delayed by 24 h (13, 16). This blockade can be relieved through administration of exogenous gonadotropin or synthetic GnRH (8, 13, 14, 17). In male rats, administration of THC into the third ventricle depresses serum LH contents and increases concentrations of GnRH within the medial basal hypothalamus (15). By comparison, addition of the CB to cultured anterior pituitary cells exerts no effects on basal or GnRH-stimulated LH release (7, 15). Hence, CB effects on LH release appear to be upstream of the pituitary. Because GnRH is a major regulator of reproduction in mammals, these collective findings suggest that CBs may perturb reproduction through actions on hypothalamic GnRH neurons.
Although marijuana and THC have been known to exert biological actions for some time (1), CB receptors were only cloned within the last decade (18, 19). The CB1 receptor is expressed primarily within the central nervous system, whereas the CB2 receptor is found in the periphery and in immune cells. After identification of CB receptors, investigators began to search for their endogenous ligands. These compounds are lipids and include anandamide (ANA), 2-arachidonyl monoacylglycerol (2-AMG), noladin ether, virodhamine, and N-arachidonyldopamine (20). Interestingly, there is some suggestion that CBs may play a role in reproduction because ANA levels in hypothalamus are low before puberty, increase just before puberty, and return to intermediate values immediately after vaginal opening (21). In addition, whereas autoradiography and in situ hybridization have shown that binding and expression levels of the CB1 receptor are low in hypothalamus (22), the highest levels of expression in this brain region are in the medial preoptic area and arcuate nucleus (23). Because GnRH perikarya are located in the preoptic area and these neurons send their axons through the arcuate nucleus to the median eminence to regulate gonadotropin release (24), some of the CB effects on reproduction may be mediated through the hypophysiotrophic GnRH neurons. To examine this possibility, we have used the immortalized hypothalamic GnRH neurons to determine whether these neurons can produce and respond to CBs, and we have extended these studies to GnRH neurons in vivo.
Materials and Methods
Cell culture
The immortalized hypothalamic GnRH neurons (GT1 cells) and RAW 264.7 macrophage cells were cultured and maintained in 95% O2-5% CO2 at 37 C in DMEM/F-12 medium (Invitrogen, Grand Island, NY) as described (25). For lipid studies, GT1–7 neurons were grown to 70–75% confluence in six-well or 100-mm2 plates and labeled for 18 h with either 1.5 μCi/ml [3H]ethanolamine (PerkinElmer Life and Analytical Sciences, Inc., Boston, MA) or [3H]arachidonic acid (AA; PerkinElmer) in serum-free medium with 0.01% fatty acid-free BSA (Sigma-Aldrich Corp., St. Louis, MO) and 1% N-2 supplement (Invitrogen). After radiolabeling, neurons were washed two times with serum-free DMEM/F-12 medium (Invitrogen) or Krebs-Ringer bicarbonate glucose buffer (KB). Neurons were subsequently stimulated for various times with ionomycin (Calbiochem, La Jolla, CA) or with norepinephrine (NE; Sigma-Aldrich) in 50 μM ascorbic acid (Sigma-Aldrich). To ascertain whether the immortalized GnRH neurons contained a CB transporter, cells were radiolabeled with [3H]ethanolamine overnight and incubated in KB for 30 min in the absence or presence of the transporter inhibitor, AM 404 (Tocris Cookson Inc., Ellisville, MO). In another transporter experiment, GT1–7 neurons were incubated with 0.5 nM [3H]ANA (PerkinElmer) and 99 nM nonradioactive ANA (BIOMOL Research Laboratories, Inc., Plymouth Meeting, PA) in the absence or presence of AM 404. To determine whether the cells could degrade endocannabinoids, GT1–7 neurons were prelabeled with [3H]ethanolamine overnight and preincubated for 30 min in KB in the presence or absence of the fatty-acid amide hydrolase (FAAH) inhibitors oleylethanolamide or arachidonoyl trifluoromethyl ketone (Biomol). Alternatively, cells were labeled with [3H]AA overnight and stimulated with ionomycin in the absence or presence of the inhibitors. All experiments were terminated by removal of medium.
For GnRH secretion studies, approximately 6 x 105 GT1–7 cells were grown for 48 h in 24-well culture dishes as described (25). At the end of this time, neurons were washed with PBS and preincubated in KB for 1 h, followed by an additional incubation for 30 min, and then exposed to various agents for 30 min. To test effects of a CB receptor agonist on secretion, GT1 neurons were preincubated in KB in the absence or presence of different concentrations of WIN 55,212-2 (Sigma-Aldrich) or CP 55,940 (Tocris Cookson) for 30 min. Neurons were then depolarized with 56 mM [K+] in the absence or presence of the agonist. To determine specificity of the agonist, cells were preincubated with 50 μM WIN 55,212-2 in the presence or absence of 100 μM AM 251 (Tocris Cookson), a selective CB1 receptor antagonist, and stimulated with 56 mM [K+] in the presence and absence of the agonist and/or antagonist. To evaluate whether the CB receptor was coupled to Gi or Go proteins, GT1–7 neurons were incubated overnight with 0.1 μg/ml pertussus toxin (Calbiochem). The next day, cells were preincubated with 50 μM WIN 55,212-2 and stimulated with 56 mM [K+] in the presence and absence of the agonist. Finally, to ascertain whether a CB receptor agonist could influence pulsatile GnRH secretion, GT1–7 neurons were grown on Cytodex beads (Amersham Biosciences Corp., Piscataway, NJ) for 4 d (26). At the end of this time, cells/beads were loaded into 1-ml plastic syringes and perifused at 0.1 ml/min with KB continually gassed with 95% O2-5% CO2 at 37 C for 5 h, and 5-min fractions were collected over the last 4 h. After the first 60 min of perifusion in KB, baseline secretion was assessed in the same buffer over the next 2 h. At the end of this time, one half of the samples were perifused with KB alone or were exposed to 50 μM WIN 55,212-2 for 30 min. Thereafter, all samples were returned to KB for 1 h. Neurons were exposed to 56 mM [K+] for the last 30 min to evaluate their response to depolarization. Medium was collected and analyzed for GnRH contents by RIA (25, 26).
Animals
Adult virgin female C57BL/6J mice (Jackson Laboratory, Bar Harbor, ME) were housed in a temperature- and humidity-controlled room with a 14-h light, 10-h dark cycle (lights on at 0700 h). Rodent chow and water were provided ad libitum. Animals were euthanized by decapitation for tissue samples for RNA analyses. All studies were conducted with an approved protocol from the Duke University Institutional Animal Care and Use Committee.
To evaluate effects of CBs on reproduction, estrous cycles were followed for 3–8 cycles, and only females with consistent 4- to 6-d cycles were used. Mice were anesthetized on metestrus with 100 mg/kg ketamine-20 mg/kg xylazine (ip), and implanted with a 5-mm cannula (Alzet Corp., Palo Alto, CA) into the third ventricle (anterior posterior, –0.08 mm; lateral, 0.0 mm; ventral, –4.7 mm; see Ref.27). The cannula was attached to an Alzet minipump implanted (sc) along the back of the mouse. Pumps were activated with saline for 12 h before implantation and they delivered 0.5 μl/h vehicle (2% Tween 80 in artificial cerebrospinal fluid) or approximately 120 nmol/h of the CB receptor agonist, CP 55, 940 (Tocris Cookson). Animals were followed until proestrus, and then methylene blue was injected into the cannula to confirm its location.
Lipid analyses
After stimulation, cells and medium were collected and analyzed separately or, in certain cases, pooled. Lipids were extracted from samples by sequential addition of 2.8 ml chloroform-methanol (1:2, vol/vol), 0.9 ml chloroform, and 0.9 ml distilled water (28). Samples were vortexed and centrifuged for 5 min at 400 rpm at 23 C to separate the organic and aqueous phases. The aqueous phase was aspirated, and the organic phase was dried to completion under nitrogen gas. Samples were reconstituted in chloroform and loaded onto LK6D high-performance thin-layer chromatography (TLC) plates (Whatman, Inc., Clifton, NJ). Plates were run in a solvent system of ethyl acetate-isooctane-acetic acid-water (93:47:21:100, vol/vol/vol/vol). Lipids were visualized with iodine, identified by comigration with standards (Biomol), scraped from the plates, and quantitated by liquid scintillation counting. Alternatively, plates were sprayed with En3hance (PerkinElmer) and visualized by autoradiography with Biomax MS film (Eastman Kodak Co., Rochester, NY). The identities of 2-AMG and ANA were further confirmed by TLC analyses in an alternative solvent system consisting of petroleum ether-diethyl ether-acetic acid (20:80:1, vol/vol/vol) (29) or by HPLC (30). Briefly, the HPLC apparatus consisted of HP 1090 syringe pumps (Hewlett-Packard, Palo Alto, CA), a Rheodyne 7125 injector (Rheodyne, Cotati, CA), a Kratos 773 UV detector (Kratos Analytical Instruments, Ramsey, NJ), a HP 3395 integrator (Hewlett-Packard), and a Frac-200 fraction collector (Amersham). The separation was achieved on a 250 x 2.1 mm 5-μm C18 Supelcosil column (Supelco, Bellefonte, PA) with an isocratic gradient of acetonitrile-water (50:50, vol/vol) run at 1 ml/min (30). Lipids were identified according to coelution with standards (Biomol) at 214 nm and liquid scintillation counting.
Analyses of gene expression
Total RNA was isolated from GT1–1 and GT1–7 neurons, RAW 264.7 cells, as well as various mouse tissues, using TriReagent (Molecular Research Center, Inc., Cincinnati, OH). First-strand cDNA was synthesized from 5 μg deoxyribonuclease I-treated total RNA, using Superscript II (Invitrogen, Carlsbad, CA). After first-strand synthesis, PCR was performed. For the CB1 receptor, the 5'-GTGTGCTGTTGCTGTTCATTGTGTA-3' and 5'-GCTGTGTTATTGGCGTGCTTGT-3' primers yielded a product of approximately 449 bp that spanned nucleotides 1036–1483 of the murine transcript (accession no. U22948). The CB2 receptor reaction was run with primers 5'-GGTCCTCTCAGCATTGATTTCTTAC-3' and 5'-TTCACATCAGCCTCTGTTTCTGTA-3' that produced a product of approximately 541 bp that included nucleotides 657-1198 of the mouse sequence (accession no. 009924). The FAAH reaction was run with 5'-CCTTCTTACCAAACAACATACC-3' and 5'-CCTTTTTCATGCCCTTCTTC-3' primers that produced a product of approximately 518 bp spanning nucleotides 1091–1609 of the mouse sequence (accession no. MMU82536). The phospholipase D (PLD) primers included 5'-GCTGTAAGGATTCACATTGAC-3' and 5'-AAAAGCTCTATCATCGGTATTC-3' that yielded a product of approximately 201 bp that spanned nucleotides 1136–1315 of the mouse sequence (accession no. NM178728). Mouse hypothalamus, pituitary, and spleen served as positive controls; negative controls included samples run without primers or with RNA from GT1–7 cells that had not undergone first-strand synthesis. The conditions for the PCR consisted of an initial denaturation step at 94 C for 90 sec, followed by 40 cycles at 94 C for 40 sec, 52–58 C for 40 sec, and 72 C for 45 sec. The PCR products were separated on 2% agarose gel, verified by sequencing at the Duke University Medical Center facility, and subcloned into the pCR 2.1 plasmid using the TA cloning kit (Invitrogen).
For Northern blot analyses, total RNA from GT1–1, GT1–3, and GT1–7 cells and various mouse tissues was purified using the mRNA Purification kit (Amersham) as described (31). Ten micrograms of mRNA were fractionated on 1.2% agarose gel and transferred to nylon membrane (Biotrans, Aurora, OH). The cDNA probes for the CB1 and CB2 receptors were radiolabeled with [-32P]-deoxycytidine triphosphate (PerkinElmer) by random primer and diluted to 1.2 x 107 dpm/μg DNA. Blots were hybridized, washed, and exposed to Storage Phophor Plates for visualization using a Typhoon 9200 Variable Mode Imager (Molecular Dynamics, Inc., Sunnyvale, CA).
For in situ hybridization histochemistry (ISHH), mouse brains were collected and frozen in powdered dry ice, then wrapped in Parafilm and stored at –80 C in sealed tubes. Twelve-micrometer coronal cryosections were obtained from the rostral preoptic area encompassing the organum vasculosum of the lamina terminalis region through the medial preoptic area (27). Sections were affixed to gelatin-coated slides and stored at –80 C until use. The cDNA probe for the CB1 receptor was made by PCR as described above. The cDNA template for mouse GnRH (mGnRH) was a 246-bp fragment corresponding to bases 39–285 of the mGnRH cDNA (accession no. AY39940) prepared using a forward primer of 5'-CGGCATTCTACTGCTGACTG-3' and a reverse primer sequence of 5'-CTTCTTCTGCCTGGCTTCCT-3'. Fragments were cloned into a TOPO-TA vector (Invitrogen) and sequenced to verify identity. Standard in vitro transcription was used to prepare 35S-labeled cRNA probe to the CB1 receptor and digoxigenin-labeled probes to mGnRH mRNA. Dual-label ISHH was performed as described previously (32). Sections were thawed, fixed, and prehybridized, before applying a mixture of 35S-labeled cRNA probes for the CB1 receptor (1 x 106 cpm) and 0.5 μl of the digoxigenin-labeled mGnRH probes in 25 μl hybridization buffer. To verify specificity, we hybridized representative sections to 35S-labeled sense strand probes in buffer with or without digoxigenin-labeled cRNA probes for mGnRH mRNA. Sections were hybridized at 52 C overnight under glass coverslips, washed, and processed for immunocytochemical detection of digoxigenin-labeled probes for mGnRH mRNAs as described (32). After detection of the dioxigenin-labeled probes, we used standard emulsion autoradiographical procedures (NTB3 emulsion; Kodak) to visualize radiolabeled probes for the CB1 receptor mRNA (7-d exposure). To analyze dual-label ISHH studies, we used BioQuant Windows (R and M Biometrics, Nashville, TN) interfaced to a Leitz Laborlux microscope through a 3CCD color video camera (Hitachi Denshi America, Ltd., Woodbury, NY). Sections from the preoptic region containing GnRH neurons (four to six sections/animal) were examined in five animals, and the number of GnRH neurons with and without CB1 receptor transcripts was determined.
Statistics
The data are presented as means and SEM. Student’s t tests were used to compare two groups; other comparisons were performed by ANOVA with subsequent Dunnett t or Bonferroni post hoc tests. P < 0.05 was considered significant.
Pulsatile GnRH release was analyzed by several different procedures (26). First, the threshold for pulse detection was set at three times the coefficient of variation for assay for the concentration of that particular sample (33). Patterns of secretion were also examined using cluster analyses (34) and the DETECT program (35). Because these procedures gave similar findings, the present results are depicted from the DETECT program.
Results
Immortalized GnRH neurons release 2-AMG
Because endocannabinoids are synthesized by enzymes that are activated by Ca2+-dependent processes, many investigators use ionomycin to stimulate their production (see Ref.20). The immortalized neurons were radiolabeled overnight with [3H]AA and stimulated the next day with 5 μM ionomycin. This agent produced a time-dependent increase in 2-AMG accumulation with an enhancement observed within the first 10 min of stimulation (Fig. 1A). AA and diacylglycerol (DAG) release was also augmented over the same time (Fig. 1B). To ascertain whether endocannabinoid production was responsive to other agents, such as neurotransmitters, GT1–7 neurons were stimulated with 100 μM NE. The NE also released 2-AMG, DAG, and AA (Fig. 1, C and D). Collectively, these data demonstrate that the immortalized GnRH neurons synthesize an endocannabinoid and that NE and other agents that increase entry of Ca2+ into cells can stimulate release and accumulation of 2-AMG in the medium.
Immortalized GnRH neurons release and transport ANA
Because the neurons produce 2-AMG, we wanted to determine whether they could make other endocannabinoids. Recently, a PLD species has been identified that generates the endocannabinoid, ANA (36). RT-PCR analyses and subsequent sequencing reveal that the mouse hypothalamus and GT1 cells contain transcripts for this form of PLD (Fig. 1E). No products were found when the PCR was run with RNA alone or without primers.
To ascertain whether the immortalized neurons synthesized ANA, GT1–7 neurons were radiolabeled overnight with [3H]ethanolamine and incubated the next day for 30 min with 100 μM AM 404, an ANA transporter inhibitor. The presence of the inhibitor alone increased ANA accumulation in medium by more than 500% (Fig. 1F). ANA was also identified by TLC and autoradiography after stimulation with 5 μM ionomycin (Fig. 1G). To determine whether ANA could be transported into immortalized GnRH neurons, [3H]ANA was added to the medium, and uptake was examined in the absence or presence of AM 404. In the absence of the inhibitor, neurons rapidly removed the radioendocannabinoid from the medium, with half-maximal uptake occurring within approximately 4 min after its addition (Fig. 1H). By comparison, AM 404 suppressed intracellular uptake of [3H]ANA from the medium. Together, these findings show that the immortalized GnRH neurons synthesize ANA and can rapidly remove it from the medium.
Immortalized GnRH neurons degrade endocannabinoids
FAAH degrades endocannabinoids in many different cells (37). To investigate whether the immortalized GnRH neurons contained FAAH transcripts, RT-PCR was performed (Fig. 2A). PCR products of appropriate size were detected not only in mouse pituitary but also in hypothalamus (a brain region containing GnRH neurons in vivo) (24). Products of the same size were also present in the GT1 neurons. No products were found when the PCR was run with RNA alone or without primers.
To study whether the FAAH was functional, GT1–7 cells were radiolabeled overnight with [3H]ethanolamine and incubated the next day for 30 min with KB in the presence or absence of the enzyme inhibitors oleylethanolamide or arachidonyl trifluoromethyl ketone. Basal release of ANA was enhanced more than 180-fold by the inhibitors (Fig. 2B). To evaluate whether the enzyme could degrade 2-AMG, GT1–7 neurons were labeled with [3H]AA and incubated with KB or with 5 μM ionomycin in the absence or presence of each inhibitor. Ionomycin significantly stimulated release of 2-AMG, and the accumulation of the endocannabinoid was further enhanced by addition of FAAH inhibitors (Fig. 2C). Because FAAH is an enzyme found within cells, these findings show that the GT1 neurons can remove ANA and 2-AMG from the medium and degrade these endocannabinoids.
Immortalized GnRH neurons contain CB1 and CB2 receptor transcripts
Because the immortalized GnRH neurons produced, transported, and degraded endocannabinoids, we next examined whether they contained receptors for these bioactive lipids. Northern blots were run for the CB1 receptor with lysates from mouse cerebral cortex and hypothalamus, as well as from the three immortalized neuronal cell lines. The transcript for the CB1 receptor was approximately 6 kb in all samples (Fig. 3A), a size consistent with the CB1 receptor in rodents (18). By contrast, whereas transcripts for the CB2 receptor could be visualized in mouse spleen by Northern blot, they were not observable in mouse hypothalamus or in the three immortalized GnRH neuronal cell lines (data not shown).
Because levels of the CB1 receptor were low and concentrations of the CB2 receptor were undetectable by Northern blot, we used RT-PCR to further examine expression of these transcripts in the immortalized GnRH neurons. This more sensitive method more readily identified CB1 and CB2 receptor RT-PCR products in the GT1–1 and GT1–7 cells, as well as in mouse hypothalamus (Fig. 3, B and C). Subsequent sequencing confirmed the identity of these products. Hence, the immortalized GnRH neurons contain transcripts for both CB receptors, albeit at low levels.
CBs inhibit GnRH secretion from immortalized GnRH neurons
To determine whether the CB receptors were functional in the immortalized GnRH neurons, GT1–7 neurons were incubated in KB alone or were depolarized with 56 mM [K+] in the absence or presence of a CB receptor agonist, WIN 55,212-2. The [K+] stimulus augmented GnRH secretion more than 3-fold over baseline, so that any inhibitory responses to CBs could be clearly visualized (Fig. 4A). WIN 55,212-2 reduced this response in a dose-dependent manner. An identical effect was observed with another CB receptor agonist, CP 55,940 (data not shown). Parenthetically, neither agonist influenced basal GnRH release (data not shown). To evaluate response specificity, the experiment was repeated in the presence of the selective CB1 receptor antagonist, AM 251. Again, 56 mM [K+] stimulated GnRH secretion, and this was reduced by more than 2-fold with 50 μM WIN 55,212-2 (Fig. 4B). The reduction in [K+]-stimulated GnRH secretion by the agonist was blocked with 100 μM AM 251. By comparison, the CB1 receptor antagonist alone exerted no effects on [K+]-stimulated GnRH release (Fig. 4B) or on basal GnRH release (data not shown). To ascertain whether the WIN 55,212-2 response was mediated by Gi/Go proteins, GT1–7 neurons were incubated in medium alone or were treated overnight with 0.1 μg/ml pertussus toxin. Although the toxin did not significantly influence GnRH secretion under basal conditions or in response to [K+] stimulation (data not shown), inhibition of GnRH release by the CB receptor agonist was abrogated by the toxin (Fig. 4C). Collectively, these data demonstrate that agonist activation of CB receptors reduces [K+]-stimulated GnRH secretion through a Gi/Go-mediated mechanism.
An important characteristic of immortalized GT1 neurons and GnRH neurons in vivo is their pulsatile release of GnRH (26, 38, 39, 40). To determine whether CBs alter pulsatility, GT1–7 cells were perifused for 5 h. After 1 h of stabilization, medium was collected over the next 2 h to establish baseline. During this period, GnRH was secreted in a pulsatile manner, with a frequency of 1.41 ± 0.21 pulses/h, duration of 17.4 ± 2.2 min, and interpulse interval of 25.3 ± 2.8 min (Fig. 4, D and E). After exposure to 50 μM WIN 55,212-2 for 30 min, pulsatile activity ceased over the next 1 h (Fig. 4E); responses to vehicle did not affect the pattern of GnRH release (Fig. 4D). GnRH release remained very low, and pulsatility was suppressed in the CB agonist-treated cultures even when they were returned to perifusion with KB (Fig. 4E). All samples responded to 56 mM [K+] depolarization at the end of the experiment. Hence, the previously CB-treated cells can respond to depolarization; albeit not as robustly as cells treated with KB alone. These data show that a CB receptor agonist can perturb pulsatile GnRH secretion from the immortalized GnRH neurons.
CB effects in vivo
To determine whether CBs influence GnRH function in vivo, a CB receptor agonist was administered to estrous cycling female mice. Animals were surgically implanted on metestrus with osmotic minipumps that delivered vehicle or low levels of the CB receptor agonist (CP 55,940) into the third ventricle of the hypothalamus, a region adjacent to the perikarya of GnRH neurons. Before cannulation, estrous cycles were monitored over at least 2 wk, and they were found to be approximately the same duration for both groups (vehicle controls, 5.0 ± 0.41 d; agonist-treated mice, 4.9 ± 0.32 d; n = 7–9 females). After cannulation, vehicle controls maintained a 5-d estrous cycle, whereas proestrus was delayed by at least 2 d in the agonist-treated mice (Fig. 4F). These findings show that CBs can exert effects on reproduction at the level of the hypothalamus.
The effects of CBs on estrous cycle duration could be exerted on GnRH neurons themselves and/or on neurons that control GnRH function. To distinguish between these possibilities, dual-label ISHH experiments were conducted with female C57BL/6J mice. Animals were euthanized without regard to the stage of estrous cycle. An examination of a total of 117 GnRH neurons revealed that most neurons were devoid of autoradiographic grains, and few grains were seen in the remaining neurons (Fig. 5, A–C). Despite this fact, autoradiographic grains were prominent is areas surrounding GnRH neurons. An absence of either a GnRH or CB1 receptor signal in the sense-strand controls verified the specificity of the probes (data not shown). Hence, these findings suggest that if the CB1 receptor is expressed in GnRH neurons, it is at a very low level and most CB1 receptors are localized in neighboring cells.
Discussion
CBs have long been known to exert potent effects on reproduction (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17). Although some of these influences may be mediated at the levels of the pituitary and gonads (41, 42, 43), the primary effects have been ascribed to hypothalamic action because synthetic GnRH can rescue ovulation in CB-treated animals (1, 8, 13, 14, 17). Despite this fact, the mechanism of action of CBs on GnRH function is obscure. Although ANA has been identified in hypothalamus (21), the present studies demonstrate that the immortalized hypothalamic GnRH neurons contain a complete and functional CB system. The presence of this system in the immortalized neurons, coupled with the reported in vivo effects of CB agonists on LH secretion and/or ovulation in female rodents and humans (1, 5, 6, 10, 13, 16, 17), suggests that endocannabinoids may play an important role in GnRH physiology and reproduction.
In the present report, we have found that the immortalized GnRH neurons synthesize at least two different endocannabinoids, 2-AMG and ANA. Besides endocannabinoids, Ca2+-ionophore and NE produce corresponding increases in DAG and free AA. These latter findings are consistent with a report showing calcium-dependent hydrolysis of phosphoinositides to be associated with liberation of inositol phosphates, as well as DAG and AA in GT1–7 cells (25). In the present experiment the delayed accumulation of 2-AMG, with respect to DAG, suggests that this endocannabinoid is derived from phospholipid hydrolysis and is modified subsequently by DAG lipase to produce 2-AMG (44). By contrast, ANA production may proceed through the action of PLD (38, 45). In this regard, it is noteworthy that the immortalized GnRH neurons possess transcripts for a PLD form that has been reported to generate ANA (36). Recently, PLD has been shown to be activated by GnRH (46). Because GnRH can inhibit its own secretion from the immortalized neurons (47) and because an ultrashort feedback system resides on GnRH neurons in vitro and in vivo (48, 49, 50), our results suggest that endocannabinoids may play a role in this process by autocrine; paracrine; or, more likely, juxtacrine interactions with cells in close apposition to GnRH neurons.
Besides synthesizing and releasing endocannabinoids, the immortalized GnRH neurons can also remove them from the medium. Under normal conditions, 2-AMG and ANA are transiently active and reenter the cell by passive diffusion and/or active transport (51, 52, 53). Although the existence of an endocannabinoid transporter is controversial (51, 52, 53), the present experiments show that GT1–7 cells can rapidly remove ANA from the medium and this action is blocked by a putative transporter inhibitor. Together, these data suggest that the immortalized GnRH neurons contain a CB transporter that can remove 2-AMG and/or ANA from the medium to rapidly curtail their biological actions.
Aside from removing endocannabinoids from the medium, the immortalized GnRH neurons also degrade these lipids. Two different inhibitors of FAAH, oleylethanolamide and arachidonyl trifluoromethyl ketone, enhance 2-AMG and ANA accumulation. Interestingly, in some systems, ANA is taken up and rapidly converted to other eicosanoids or is degraded by FAAH to AA and ethanolamine (54). The 2-AMG can be hydrolyzed to AA, by either a monoacylglycerol lipase or FAAH, and reesterified into phospholipid (55, 56). Furthermore, some in situ studies reveal that endocannabinoids recycled into AA can be converted into eicosanoids such as thromboxanes and prostaglandins (57). Because prostaglandin E2 can stimulate GnRH secretion from median eminence tissue fragments in vitro (58) and from the immortalized GnRH neurons (59), their potential synthesis from endocannabinoids suggests that CBs may be intimately linked with various eicosanoid signaling pathways to modulate GnRH release.
In addition to synthesizing endocannabinoids, the immortalized GnRH neurons also contain transcripts for the CB1 and CB2 receptors, and activation of these receptors inhibits GnRH secretion. In perifusion experiments, a CB receptor agonist completely disrupts pulsatile secretion of GnRH. It is noteworthy that, whereas GnRH secretion can be stimulated with [K+] depolarization subsequent to treatment with a CB agonist, the response is less robust than in cells previously exposed to KB alone. A similar relationship is seen in vivo, where THC is more efficient than atropine in maintaining blockade of LH responses to electrical stimulation of the preoptic area (60). Although a CB receptor agonist suppresses [K+]-stimulated GnRH secretion and this response is blocked by a CB receptor antagonist in the immortalized neurons, it should be emphasized that the levels of the CB1 and CB2 receptor transcripts in the immortalized cells are low. Moreover, [K+]-stimulated GnRH secretion is only partially inhibited by the highest concentration of the receptor agonist. Thus, CBs appear to exert modest effects on GnRH secretion from the immortalized neurons.
Effects of CBs were next evaluated in adult female C57BL/6J mice. A CB receptor agonist was found to prolong the estrus cycle in the mice by at least 2 d. These data are in agreement with other investigations where peripheral administration of CBs is reported to delay ovulation in a number of species, including the human (1, 5, 6, 10, 13, 16, 17). There are a number of different reasons to suspect that the reproductive effects of CBs are exerted primarily at the level of the hypothalamus. First, in the present study, delivery of the agonist into the third ventricle of the hypothalamus is sufficient to delay the estrous cycle in mice by several days. Second, peripherally administered GnRH restores LH secretion in CB-treated animals (8, 13, 14). Third, electrical stimulation of the medial preoptic area alleviates the THC-induced blockade of the preovulatory LH surge (61). Finally, ANA is reported to depress GnRH release from the medial basal hypothalamus in vitro (62). Although CBs influence reproduction through hypothalamic mechanisms, it has not been clear whether these lipids exert their effects directly on GnRH neurons or on neighboring cells that control GnRH release. Importantly, CB1 receptors are localized in the preoptic area and hypothalamus (63, 64), brain areas where the GnRH neurons reside. Results from our dual-label ISHH experiments suggest that few hypothalamic GnRH neurons contain CB1 receptor transcripts. By comparison, the distribution of autoradiographic grains indicates that many cells in the vicinity of GnRH neurons possess substantial levels of CB1 receptor transcripts. Importantly, a recent report by Herbison’s group (65) using single-cell PCR from transgenic mice has shown that CB2 receptors are expressed in approximately 25% of native hypothalamic GnRH neurons. Thus, these and the present findings suggest that CBs may exert their effects on reproduction not only through cells that control GnRH function but also through GnRH neurons themselves.
GnRH neurons in vivo have long been known to respond to ligands that stimulate phosphoinositide hydrolysis and increased intracellular calcium and DAG production (see Ref.25). This cascade may also lead to the synthesis of endocannabinoids, such as 2-AMG and ANA. These lipids, in turn, can bind to CB receptors on neighboring cells and GnRH neurons, thereby exerting some feedback control on GnRH function. In this regard, it is instructive that recent electrophysiological evidence from hippocampal and cerebellar neurons indicates that CBs can send retrograde signals back to presynaptic terminals to control neurotransmitter release (66, 67, 68). If the same processes occur in GnRH neurons in vivo, then the generation of endocannabinoids and their degradation to various eicosanoid products could serve as a novel mechanism for regulating GnRH secretion and, ultimately, reproduction in mammals.
Acknowledgments
We thank Dr. P. Mellon (University of California-San Diego, La Jolla, CA) for providing the GT1 cells, Dr. A. Arimura (Tulane University, New Orleans, LA) for the A772 antiserum, Dr. Nancy E. Buckley (National Institute of Neurological Disorders and Stroke, Bethesda, MD) for the mouse CB2 receptor cDNA, Dr. Clifford D. Carpenter (University of Massachusetts, Amherst, MA) for the dual labeling of cells for ISHH, and Ms. Jiechun Zhou (Duke University Medical Center, Durham, NC) for providing the RT-PCR results for PLD.
Footnotes
This work was partially supported by the Elizabeth and Ruth Gammon Research Fund (to C.M.G.); a Summer Research Fellowship from The Endocrine Society, a Howard Hughes Forum in Neuroscience Research Fellowship, and a Sigma Xi Grants-in-Aid Award (to G.M.F.); National Institutes of Health Grant HD27305 (to S.L.P.); and Unrestricted Research Funds (to W.C.W.).
Abbreviations: AA, [3H]Arachidonic acid; 2-AMG, 2-arachidonyl monoacylglycerol; ANA, anandamide; CB, cannabinoid; DAG, diacylglycerol; FAAH, fatty-acid amide hydrolase; ISHH, in situ hybridization histochemistry; KB, Krebs-Ringer bicarbonate glucose buffer; mGnRH, mouse GnRH; NE, norepinephrine; PLD, phospholipase D; THC, 9-tetrahydrocannabinol; TLC, thin-layer chromatography.
References
Bloch E, Thysen B, Morrill GA, Gardner E, Fugimoto G 1978 Effects of cannabinoids on reproduction and development. Vitam Horm. 36:203–259
Kolodny RC, Masters WH, Kolodner RM, Toro G 1974 Depression of plasma testosterone levels after chronic intensive marihuana use. N Engl J Med. 290:872–874
Collu R, Letarte J, Leboeuf G, Ducharme JR 1975 Endocrine effects of chronic administration of psychoactive drugs to prepuberal male rats. I: 9-tetrahydrocannabinol. Life Sci. 16:533–542
Solomon J, Cocchia AM, DiMartino R 1977 Effect of -9-tetracannabinol on uterine and vaginal cytology of ovariectomized rats. Science 195:875–877
Smith CG, Almirez RG, Berenberg J, Asch RH 1983 Tolerance develops to the disruptive effects of 9-tetrahydrocannabinol on primate menstrual cycle. Science 219:1453–1455
Wenger T, Croix D, Tramu G 1988 The effects of chronic prepubertal administration of marihuana (-9-tetrahydrocannabinol) on the onset of puberty and the postpubertal reproductive functions in female rats. Biol Reprod. 39:540–545
Chakravarty I, Shah PG, Sheth AR, Ghosh JJ 1979 Mode of action of -9-tetrahydrocannabinol on hypothalamo-pituitary function in adult female rats. J Reprod Fertil. 57:113–115
Tyrey L 1978 -9-Tetrahydrocannabinol suppression of episodic luteinizing hormone secretion in ovariectomized rats. Endocrinology 102:1808–1814
Steger RW, Silverman AY, Siler-Khodr TM, Asch RH 1980 The effects of 9-tetrahydrocannabinol on the positive and negative feedback control of luteinizing hormone release. Life Sci. 27:1911–1916
Mendelson JH, Mello NK, Ellingboe J, Skupny AST, Lex BW, Griffin M 1986 Marihuana smoking suppresses luteinizing hormone in women. J Pharmacol Exp Ther. 237:862–866
Kramer J, Ben-David M 1974 Suppression of prolactin secretion by acute administration of 9-tetrahydrocannabinol in rats. Proc Soc Exp Biol Med. 147:482–484
Daley JD, Branda LA, Rosenfeld J, Younglai EV 1974 Increase of serum prolactin in male rats by (-)-trans-9-tetrahydrocannabinol. J Endocr. 63:415–416
Ayalon D, Nir I, Cordova T, Bauminger S, Puder M, Naor Z, Kashi B, Zor U, Harrell A, Lindner HR 1977 Acute effect of 1-tetrahydrocannabinol on the hypothalamo-pituitary-ovarian axis in the rat. Neuroendocrinology 23:31–42
Smith CG, Besch NF, Smith RG, Besch PK 1979 Effect of tetrahydrocannabinol on the hypothalamic-pituitary axis in the ovariectomized rhesus monkey. Fertil Steril. 31:335–339
Wenger T, Rettori V, Snyder GD, Dalterio S, McCann SM 1987 Effects of -9-tetrahydrocannabinol on the hypothalamic-pituitary control of luteinizing hormone and follicle-stimulating hormone secretion in adult male rats. Neuroendocrinology 46:488–493
Nir I, Ayalon D, Tsafriri A, Cordova T, Lindner HR 1973 Suppression of the cyclic surge of luteinizing hormone secretion and of ovulation in the rat by 1-tetrahydrocannabinol. Nature 243:470–471
Asch RH, Fernandez EO, Smith CG, Pauerstein CJ 1979 Precoital single doses of 9-tetrahydrocannabinol block ovulation in the rabbit. Fertil Steril. 31:331–334
Matsuda LA, Lolait SJ, Brownstein MJ, Young AC, Bonner TI 1990 Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature 346:561–564
Munro S, Thomas KL, Abu-Shaar M 1993 Molecular characterization of a peripheral receptor for cannabinoids. Nature 365:61–65
Piomelli S 2003 The molecular logic of endocannabinoid signaling. Nat Rev Neurosci. 4:873–884
Wenger T, Gerendai I, Fezza F, González S, Bisogno T, Fernandez-Ruiz J, Di Marzo V 2002 The hypothalamic levels of the endocannabinoid, anandamide, peak immediately before the onset of puberty in female rats. Life Sci. 70:1407–1414
Mailleux P, Vanderhaeghen J-J 1992 Distribution of neuronal cannabinoid receptor in the adult rat brain: a comparative receptor binding radiography and in situ hybridization histochemistry. Neuroscience 48:655–668
Moldrich G, Wenger T 2000 Localization of the CB1 cannabinoid receptor in the rat brain. An immunohistochemical study. Peptides 21:1735–1742
Merchenthaler I, Grcs T, Sétàlò G, Petrusz P, Flerkò B 1984 Gonadotropin-releasing hormone (GnRH) neurons and pathways in the rat brain. Cell Tissue Res. 237:15–29
Kreda SM, Sumner M, Fillo S, Ribeiro CM, Luo GX, Xie W, Daniel KW, Shears S, Collins S, Wetsel WC 2001 1-Adrenergic receptors mediate LH-releasing hormone secretion through phospholipases C and A2 in immortalized hypothalamic neurons. Endocrinology 142:4839–4851
Wetsel WC, Valenca MM, Merchenthaler I, Liposits Z, Lopez F, Weiner RI, Mellon P, Negro-Vilar A 1992 Intrinsic pulsatile secretory activity of immortalized luteinizing hormone-releasing hormone-secreting neurons. Proc Natl Acad Sci USA. 87:4149–4153
Franklin KBJ, Paxinos G 1997 The mouse brain in stereotaxic coordinates. San Diego: Academic Press; 1–186
Gammon CM, Allen AC, Morell P 1989 Bradykinin stimulates phosphoinositide hydrolysis and mobilization of arachidonic acid in dorsal root ganglion neurons. J Neurochem. 53:95–101
Sugiura T, Kondo S, Sukagawa A, Nakane S, Shinoda A, Itoh K, Yamashita A, Waku K 1995 2-Arachidonoylgylcerol: a possible endogenous cannabinoid receptor ligand in brain. Biochem Biophys Res Commun. 215:89–97
Fontana A, Di Marzo V, Cadas H, Piomelli D 1995 Analysis of anandamide, an endogenous cannabinoid substance, and of other natural N-acylethanolamines. Prostaglandins Leukot Essent Fatty Acids 53:301–308
Wetsel WC, Liposits Z, Seidah NG, Collins S 1995 Expression of candidate pro-LHRH processing enzymes in rat hypothalamus and an immortalized hypothalamic neuronal cell line. Neuroendocrinology 62:166–177
Ottem EN, Godwin JG, Krishnan S, Petersen SL 2004 Dual-phenotype GABA/glutamate neurons in adult preoptic area: sexual dimorphism and function. J Neurosci. 24:8097–8105
Lumpkin MD, DePaolo LV, Negro-Vilar A 1984 Pulsatile release of follicle-stimulating hormone in ovariectomized rats is inhibited by porcine follicular fluid (inhibin). Endocrinology 114:201–206
Veldhuis J, Johnson M 1986 Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol. 250:E486–E493
Oerter KE., Guardabasso V, Rodbard D 1986 Detection and characterization of peaks and estimation of instantaneous secretory rate for episodic pulsatile hormone secretion. Comput Biomed Res. 19:170–190
Okamoto Y, Morishita J, Tsuboi K, Tonai T, Ueda N 2004 Molecular characterization of a phospholipase D generating anandamide and its congers. J Biol Chem. 279:5298–5305
Katayama K, Ueda N, Kurahashi Y, Suzuki H, Yamamoto S, Kato I 1997 Distribution of anandamide aminohydrolase in rat tissues with special reference to small intestine. Biochim Biophys Acta. 1347:212–218
Martinez de la Escalera G, Choi ALH, Weiner RI 1992 Generation and synchronization of gonadotropin-releasing hormone (GnRH) pulses: intrinsic properties of the GT1–1 GnRH neuronal cell line. Proc Natl Acad Sci USA. 89:1852–1855
Krsmanovic LZ, Stojilkovic S, Merelli F, Dufour S, Virmani M, Catt KJ 1992 Calcium signaling and episodic secretion of gonadotropin-releasing hormone in hypothalamic neurons. Proc Natl Acad Sci USA. 89:8462–8466
Knobil E 1980 The neuroendocrine control of the menstrual cycle. Recent Prog Horm Res. 36:52–88
Burstein S, Hunter SA, Shoupe TS 1979 Cannabinoid inhibition of rat luteal cell progesterone synthesis. Res Commun Chem Pathol Pharmacol. 24:413–416
Gerard CM, Mollereau C, Vassart G, Parmentier M 1991 Molecular cloning of a human cannabinoid receptor which is also expressed in testis. Biochem J. 279:129–134
Pagotto U, Marsicano G, Fezza F, Theodoropoulou M, Gruber Y, Stalla J, Arzberger T, Milone A, Losa M, Di Marzo V, Lutz B, Stalla GK 2001 Normal human pituitary gland and pituitary adenomas express cannabinoid receptor type 1 and synthesize endogenous cannabinoids: first evidence for a direct role of cannabinoids on hormone modulation at the human pituitary level. J Clin Endocrinol Metab. 86:2687–2696
Allen AC, Gammon CM, Ousley AH, McCarthy KD, Morell P 1992 Bradykinin stimulates arachidonic acid release through the sequential actions of an sn-1 diacylglycerol lipase and a monoacylglycerol lipase. J Neurochem. 58:1130–1139
Di Marzo V, Fontana A, Cadas H, Schinelli S, Cimino G, Schwartz JC, Piomelli D 1994 Formation and inactivation of endogenous cannabinoid anandamide in central neurons. Nature 372:686–691
Zheng L, Krsmanovic LZ, Vergara LA, Catt KJ, Stojilkovic SS 1997 Dependence of intracellular signaling and neurosecretion on phospholipase D activation in immortalized gonadotropin-releasing hormone neurons. Proc Natl Acad Sci USA. 94:1573–1578
Krazmanovic LZ, Stojilkovic SS, Mertz LM, Tomic M, Catt KJ 1993 Expression of gonadotropin-releasing hormone receptors and autocrine regulation of neuropeptide release in immortalized hypothalamic neurons. Proc Natl Acad Sci USA. 90:3908–3912
Hyyppa M, Motta M, Martini L 1971 ‘Ultrashort’ feedback control of follicle-stimulating hormone-releasing factor secretion. Neuroendocrinology 7:227–235
Castro JCB, Khorram O, McCann SM 1985 Possible negative ultra-short loop feedback of luteinizing hormone releasing hormone (LHRH) in the ovariectomized rat. Proc Soc Exp Biol Med. 179:132–135
Valenca MM, Johnston CA, Ching M, Negro-Vilar A 1987 Evidence for a negative ultrashort loop feedback mechanism operating on the luteinizing hormone-releasing hormone neuronal system. Endocrinology 121:2256–2259
Beltramo M, Stella N, Calignano A, Lin SY, Makriyannis A, Piomelli D 1997 Functional role of high-affinity anandamide transport, as revealed by selective inhibition. Science 277:1094–1097
Glaser ST, Abumrad NA, Fatade F, Kaczocha M, Studholme KM, Deutsch DG 2003 Evidence against the presence of an anandamide transporter. Proc Natl Acad Sci USA. 100:4269–4274
Fegley D, Kathuria S, Mercier R, Li C, Goutopoulos A, Makriyannis A, Piomelli D 2004 Anandamide transport is independent of fatty-acid amide hydrolase activity and is blocked by the hydrolysis-resistant inhibitor AM1172. Proc Natl Acad Sci USA. 101:8756–8761
Fowler CJ, Jonsson K, Tiger G 2001 Fatty acid amide hydrolase: biochemistry, pharmacology, and therapeutic possibilities for an enzyme hydrolyzing anandamide, 2-arachidonylglycerol, palmitoylethanolamine, and oleamide. Biochem Pharmacol. 62:517–526
Cravatt B, Giang D, Mayfield S, Bolger D, Lerner R, Guila N 1996 Molecular characterization of an enzyme that degrades neuromodulatory fatty acid amides. Nature 384:84–87
Di Marzo V 1999 Biosynthesis and inactivation of endocannabinoids: relevance to their proposed role as neuromodulators. Life Sci. 65:645–655
Kozak KR, Crews BC, Morrow JD, Wang L, Ma YH, Weinander R, Jakobsson P-J, Marnett LJ 2002 Metabolism of the endocannabinoids, 2-arachidonylglycerol and anandamide, into prostaglandin, thromboxane, and prostacyclin glycerol esters and ethanolamides. J Biol Chem. 277:44877–44885
Ojeda SR, Negro-Vilar A, McCann SM 1979 Release of prostaglandin E2 by hypothalamic tissue: evidence for their involvement in catecholamine-induced luteinizing hormone-releasing hormone release. Endocrinology 104:617–624
Ma YJ, Berg-von der Emde K, Rage F, Wetsel WC, Ojeda S 1997 Hypothalamic astrocytes respond to transforming growth-factor with the secretion of neuroactive substances that stimulate the release of luteinizing hormone-releasing hormone. Endocrinology 138:19–25
Tyrey L 1992 -9-Tetrahydrocannabinol attenuates luteinizing hormone release induced by electrochemical stimulation of the medial preoptic area. Biol Reprod. 47:262–267
Murphy LL, Tyrey L 1986 Induction of luteinizing hormone release by electrical stimulation of the medial preoptic area in 9-tetrahydrocannabinol-blocked proestrus rats. Neuroendocrinology 43:471–475
Fernández-Solari J, Scorticati C, Mohn C, De Laurentiis A, Billi S, Franchi A, McCann SM, Rettori V 2004 Alcohol inhibits luteinizing hormone-releasing hormone release by activating the endocannabinoid system. Proc Natl Acad Sci USA. 101:3264–3268
Herkenham M, Lynn AB, Johnson MR, Melvin LS, de Costa BR, Rice KC 1991 Characterization and localization of cannabinoid receptors in rat brain: a quantitative in vitro autoradiographic study. J Neurosci. 11:563–583
Matsuda LA, Bonner TI, Lolait SJ 1993 Localization of cannabinoid receptor mRNA in rat brain. J Comp Neurol. 327:535–550
Todman MG, Han S-K, Herbison AE 2005 Profiling neurotransmitter receptor expression in mouse gonadotropin-releasing hormone neurons using green fluorescent protein-promoter transgenics and microarrays. Neuroscience 132:705–712
Kreitzer AC, Regehr WG 2001 Retrograde inhibition of presynaptic calcium influx by endogenous cannabinoids at excitatory synapses onto purkinje cells. Neuron 29:717–727
Ohno-Shosaku T, Maejima T, Kano M 2001 Endogenous cannabinoids mediate retrograde signals from depolarized postsynaptic neurons to presynaptic terminals. Neuron 29:729–738
Wilson RI, Nicoll RA 2001 Endogenous cannabinoids mediate retrograde signaling at hippocampal synapses. Nature 410:588–592(C. Michael Gammon, G. Mar)