Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The discovery of the CB1 cannabinoid receptor (Devane et al., 1988; Matsuda et al., 1990) resulted in considerable effort to ascertain its relevance to the pharmacological effects produced by cannabinoids. The localization of this receptor throughout the brain was found to be consistent with the pharmacological effects produced by cannabinoids. Autoradiography of cannabinoid receptors from several mammalian species, including human, revealed a conserved and unique pattern of distribution (Herkenham et al., 1990). Binding was most dense in the outflow nuclei of the basal ganglia (the substantia nigra pars reticulata and globus pallidus), the hippocampus and the cerebellum. The high densities of receptors in the forebrain and the cerebellum explain the effects of cannabinoids on cognition and movement. High levels in the hippocampus provide a role for these receptors in cannabinoid impairment of memory. Sparse densities in the brainstem areas controlling cardiovascular and respiratory functions may explain why high doses of marijuana do not suppress respiration. The distribution of cannabinoid receptors in rat brain also was determined with the cannabinoid agonists [³H]-WIN 55,212-2 and [³H]-11-OH-⁹-THC-DMH (Jansen et al., 1992; Thomas et al., 1992) and antagonist SR 141716A (Rinaldi-Carmona et al., 1996). Binding distribution was very similar between these agonists and antagonists confirming that these structurally diverse compounds bind to the same receptor.

The pattern of distribution of cannabinoid binding is consistent with that of the CB1 mRNA using in situ hybridization (Mailleux and Vanderhaegen, 1992; Matsuda et al., 1990). In the hippocampus, high levels of mRNA for the cannabinoid receptor were found in granule cells of the dentate gyrus and in cells of the pyramidal and molecular layers of the hippocampus. Message for the receptor was also prevalent within the superficial and deep layers of the cerebral cortex and amygdala. In the human brain the distribution of the mRNA encoding for the cannabinoid receptor also has been studied using in situ histochemistry and oligonucleotide probes (Mailleux et al., 1992; Westlake and Howlett, 1994).

Despite a wealth of knowledge regarding the CB1 receptor, numerous questions persist, the most noteworthy of which is whether this single receptor is responsible for all of the central actions of cannabinoids. As mentioned above, several structurally diverse cannabinoids, including the aminoalkylindoles (WIN 55,212), the bicyclic compounds (CP 55,940) and dimethylheptyl analogs of THC (11-OH-⁹-THC-DMH) appear to exert their effects at the same receptor based on similarities in localization of binding throughout brain and pharmacological profiles. The discovery of the putative endogenous cannabinoid ligand, anandamide (Devane et al., 1992), made the notion of a central cannabinoid system plausible. Anandamide produces a pharmacological profile very similar to that of ⁹-THC in several behavioral models (Fride and Mechoulam, 1993; Smith et al., 1994; Wiley et al., 1995a) even though its structure deviates dramatically from that of the other cannabinoids. Although anandamide produces many of the same effects as other psychoactive cannabinoids, differences do exist. Comparison between anandamide and ⁹-THC revealed that anandamide was 4- to 20-fold less potent and had a shorter duration of action than ⁹-THC (Smith et al., 1994). Anandamide is less efficacious than ⁹-THC at the N-type calcium channels (Mackie et al., 1993). Anandamide produces antinociception like other cannabinoids, but in contrast to ⁹-THC, anandamide is not active when administered intracerebroventricularly (Smith et al., 1994). Also, unlike other cannabinoids, anandamide's antinociception is not blocked by the kappa antagonist nor-BNI (Smith et al., 1994). The question arises as to whether anandamide is interacting with the same receptor as ⁹-THC or whether it is acting at a different receptor subtype.

The development of SR 141716A, a selective CB1 antagonist (Rinaldi-Carmona et al., 1994), has proven to be a valuable tool for identifying receptor-mediated cannabinoid action. SR 141716A has been shown to be effective in blocking the actions of cannabinoids in several mouse behavioral assays (Compton et al., 1996; Rinaldi-Carmona et al., 1994), rat drug discrimination (Wiley et al., 1995b), rat memory tasks (Lichtman and Martin, 1996), mouse vas deferens (Rinaldi-Carmona et al., 1994), adenylyl cyclase (Rinaldi-Carmona et al., 1994), long-term potentiation (Collins et al., 1995), stimulated-arachidonic acid release (Shivachar et al., 1996), and cardiovascular function (Varga et al., 1995). SR 141716A blocks the actions of anandamide in mouse vas deferens (Rinaldi-Carmona et al., 1994), cardiovascular system (Varga et al., 1995), turning behavior after intrastriatal injections (Souilhac et al., 1995), adenylyl cyclase (Felder et al., 1995) and long-term potentiation (Terranova et al., 1995). However, SR 141716A has not been evaluated for blockade of anandamide's pharmacological effects in the mouse tetrad model, a model highly correlated with CB1 receptor affinity (Compton et al., 1993).

The purpose of our investigation was to determine the extent to which the cannabinoid antagonist would block the pharmacological effects of anandamide and whether differences exist between the cannabinoid receptor population that binds CP 55,940, SR 141716A and anandamide. The pharmacological effects of anandamide were assessed in the mouse model of spontaneous activity, antinociception, body temperature and immobility, because most of the most complete characterization of anandamide and its analogs has been done in this model. Autoradiography in rat brain was used to make a direct comparison of receptor affinities of anandamide, CP 55,940 and SR 141716A in different brain regions as well as determine their maximal binding capacities throughout brain. The most complete characterization of cannabinoid receptor localization has used rat brain. Because anandamide does not have high affinity for the cannabinoid receptor, in comparison to other synthetic high-affinity THC analogs, [³H]-anandamide was not used directly to label receptors. Performing autoradiography with compounds possessing weak affinity to a receptor introduces many technical problems, the most serious of which is receptor dissociation. Therefore, the strategy was to compare the abilities of unlabeled anandamide, SR 141716A and CP 55,950 to compete with [³H]CP 55,940 binding. If these agents are binding to the CB1 receptor in brain, then the binding localization should be the same in different brain regions for all three compounds. Differences in binding densities between regions for the three compounds might suggest that a receptor subtype for the central cannabinoid receptor exists. However, these differences can only be detected if [³H]CP 55,940 is labeling more than one receptor. An additional objective was to determine cannabinoid receptor affinities for anandamide, SR 141716A and CP 55,940 and compare the affinities for each compound between brain regions. Differences in affinities may indicate that a particular drug is binding to a brain region in a different manner than to cannabinoid receptors in other regions.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Male Sprague-Dawley rats (150-200 g) from Harlan Laboratories (Dublin, VA) were maintained on a 14:10 hr light/dark schedule and freely received food and water. Male ICR mice weighing 18 to 25 g were used in all in vivo experiments. The mice were also maintained on a 14:10 hr light:dark cycle with free access to food and water.

Chemicals. [³H]CP 55,940 was purchased from Du Pont NEN (Wilmington, DE). CP 55,940 and SR 141716A were a gift from Pfizer, Inc. (Groton, CT), and anandamide and 2-methyl-2-fluoroethylanandamide were kindly provided by Dr. Raj K. Razdan of Organix, Inc. (Woburn, MA). These compounds were prepared as 1 mg/ml stock solutions in absolute ethanol and stored at 20°C for receptor binding. PMSF was dissolved in absolute ethanol as a 20 mg/ml stock solution. ⁹-THC was obtained from the National Institute on Drug Abuse. For in vivo experiments, all drugs were dissolved in 1:1:18 (emulphor-ethanol-saline) to prepare micellular suspensions of drugs suitable for in vivo administration (Olsen et al., 1973). Emulphor (EL-620, a polyoxyethylated vegetable oil, GAF Corporation, Linden, NJ) is currently available as Alkmulphor. Drug injections were administered i.v. (tail vein) at a volume of 0.1 ml/10 g of body weight.

Pharmacological evaluation in vivo. Mice were acclimated to the evaluation room overnight without interruption of food or water. Mice were pretreated with either vehicle or SR 141716A (i.v.) 10 min before a second i.v. injection of either vehicle or the test drug (anandamide or 2-methyl-2-fluoroethylanandamide). After the second i.v. injection, one group of mice was evaluated for tail-flick latency (antinociception) response at 5 min and spontaneous (locomotor) activity at 5 to 15 min, although another group was assessed for core (rectal) temperature at 5 min and ring-immobility (catalepsy) at 5 to 10 min, as described elsewhere (Adams et al., 1995; Smith et al., 1994). Inhibition of spontaneous activity was accomplished by placing mice into individual activity cages (6.5 × 11 in), and recording interruptions of the photocell beams (16 beams per chamber) for a 10-min period using a Digiscan Animal Activity Monitor (Omnitech Electronics Inc., Columbus, OH). Activity in the chamber was expressed as the total number of beam interruptions. Antinociception was assessed using the tail-flick procedure (Dewey et al., 1970). The heat lamp of the tail-flick apparatus was maintained at an intensity sufficient to produce control latencies of 2 to 3 sec. Control values for each animal were determined prior to drug administration. Mice were then reevaluated after drug administration and latency (sec) to tail-flick response was recorded. A 10-sec maximum was imposed to prevent tissue damage. The degree of antinociception was expressed as the % MPE (maximum possible effect) which was calculated as: Hypothermia was assessed by first measuring baseline core temperatures before drug treatment with a telethermometer (Yellow Springs Instrument Co., Yellow Springs, OH) and a rectal thermistor probe inserted to 25 mm. Rectal temperatures were also measured after drug administration. The temperature difference (°C) before and after drug administration was calculated for each animal. Catalepsy was determined by a ring-immobility procedure (Pertwee, 1972). Mice were placed on a ring (5.5 cm in diameter) that was attached to a stand at a height of 16 cm. The amount of time (sec) that the mouse spent motionless during a 5-min test session was recorded. The criterion for immobility was the absence of all voluntary movements (excluding respiration, but including whisker movement). The immobility index was calculated as: Mice that fell or actively jumped from the ring were allowed five such escapes. After the fifth escape, the test for that animal was terminated and immobility was calculated as a percentage of time that it remained on the ring before being discontinued. Data from mice failing to remain on the ring at least 2.5 min were not included.

Tissue preparation. After decapitation, rat brains were quickly removed and frozen in 2-methylbutane (50°C). The brains were embedded in M-1 embedding matrix and stored at 70°C until sectioning. Brains were mounted onto cryostat chucks with TFM tissue freezing medium. Consecutive coronal brain sections (16 µm) were thaw-mounted onto slides coated with 0.5% gelatin and 0.05% chromium potassium sulfate. Sections were made at the stereotaxic coordinates 1.2 mm from bregma, 0.48 mm from bregma, 5.2 mm from bregma and 12.8 mm from bregma (Paxinos and Watson, 1986). Sections were stored desiccated at 70°C before use in binding assays.

In situ cannabinoid binding assays. Assay conditions for cannabinoid binding have been described previously (Herkenham et al., 1991). Saturation experiments were first performed to determine the K_d of [³H]CP 55,940. Coronal sections containing primarily frontal cortex and caudate-putamen (1.2 mm from bregma) were used for Scatchard analysis. Slides were allowed to return to room temperature and incubated for 2 hr in slide mailers at 37°C in reaction buffer (50 mM Tris-HCl with 5% BSA, pH 7.4). Total binding was determined with seven concentrations of [³H]CP 55,940 (1.1, 2.3, 4.5, 7.5, 15, 23 and 30 nM). Nonspecific binding was determined by incubating [³H]CP 55,940 in the presence of 1 µM CP 55,940. Saturation experiments were also performed with 50 µM PMSF in the incubation buffer. After incubation, slides were washed for 4 hr at 0°C in 50 mM Tris-HCl with 1% BSA (pH 7.4). Sections were scraped from the slides with Whatman GF/C filters. The filters were placed in scintillation vials, and the tissue was solubilized overnight with 1.0 ml of TS-2. Samples were acidified with 10 µl of glacial acetic acid and counted by liquid scintillation spectrometry. Transformation of the data and calculation of K_d values was accomplished using the LIGAND computer software (Munson and Rodbard, 1980) as supplied by Biosoft Inc. (Cambridge, U.K.).

Competition for [³H]CP 55,940 binding. Due to the reported instability of anandamide (Childers et al., 1994; Deutsch and Chin, 1993), optimal binding conditions were first determined in the presence and absence of the enzyme inhibitor PMSF. In the competition experiments, reaction buffers, incubation temperatures and times were identical to the in situ binding assay described above. Sections were made from 0.48 mm from bregma, 5.2 from bregma and 12.8 mm from bregma. For CP 55,940, SR 141716A and anandamide, nonspecific binding was determined using 10 µM CP 55,940, and total binding was determined for 10 nM [³H]CP 55,940 (approximately 40% receptor occupation). Eight concentrations of CP 55,940 ranging from 0.1 to 300 nM were assayed; concentrations of anandamide ranged from 0.01 to 10 µM, and concentrations of SR 141716A were 0.001 to 10 µM. Each experiment was conducted in at least triplicate. For anandamide displacement assays, additional experiments were performed either with 50 µM PMSF in the incubation buffer or with sections pretreated for 30 min in a buffer containing 50 µM PMSF before exposure of 50 µM PMSF in the incubation buffer. Anandamide sections were wiped from the slides, solubilized and counted. Optimal conditions for anandamide competition experiments occurred when sections were pretreated for 30 min with PMSF and exposed to PMSF in the incubation buffer.

[³H]CP 55,940 autoradiography. After the wash, slides were rapidly dried with a stream of cool air and stored in a desiccator overnight at 4°C. Sections were apposed to tritium-sensitive film with [³H]-microscales for 3 wk before developing with a D-19 developer. Developed films were analyzed using the NIH Image 1.49 program. Levels of transmittance were converted to dpm/mg protein using a polynominal curve fit of the standards. Brain structures were outlined, and optical density in each area was measured. Curve-fitting of the displacement data and determination of K_i and B_max values for anandamide, CP 55,940 and SR 141716A were done using EBDA software. K_i and B_max values for anandamide in the substantia nigra and the molecular layer of the cerebellum were determined from autoradiograms apposed to film for 1 wk.

Statistical analysis. Significant differences between K_i values were determined using the ANOVA analysis (Scheffe post hoc analysis). To determine if curves were parallel, data from a representative displacement curve from each brain area for anandamide, SR 141716A and CP 55,940 were analyzed using the ALLFIT curve-fitting program. B_max values were compared by linear correlations for anandamide and CP 55,940 and SR 141716A and CP 55,940. ANOVA and Scheffe post hoc analysis was also used to determine significance in the in vivo antagonism studies.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Pharmacological evaluation in vivo. The i.v. administration of anandamide produced pharmacological effects in all four mouse pharmacological procedures similar to that described previously (Smith et al., 1994). The ED50's for producing hypomotility in the absence and presence of 3.0 and 30 mg/kg of SR141716A were 11.4, 15.7, and 15.8 mg/kg, respectively (fig. 1). The corresponding ED50's for antinociception were 3.0, 4.2, and 2.9 mg/kg, respectively. The ED50's for producing immobility in the absence and presence of 3.0 and 30 mg/kg of SR141716A were 13.7, 26.9 and 16.0 mg/kg, respectively. Anandamide's effects on rectal temperature were rather modest (maximal effect of 2.4°C decrease at 20 mg/kg) and variable. Despite anandamide's rather weak hypothermic effects, SR141716A still was unable to effectively prevent these effects.

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Failure of SR 141716A to block the effects on anandamide on spontaneous activity (A), tail-flick response (B), rectal temperature (C) and the ring test (D) in mice. The results are presented as means ± S.E. (N = 6 mice per group). Mice were pretreated (i.v.) with either vehicle (), 3 mg/kg of SR 141716A () or 30 mg/kg of SR 141716A () 10 min before the i.v. administration of the indicated doses of anandamide.

View this table: [in this window] [in a new window]   TABLE 1 Pretreatment of SR 141716A before either 9-THC or anandamide in micea

View this table: [in this window] [in a new window]   TABLE 2 SR 141716A blockade of hypomotility and antinociception induced by 2-methyl-2-fluoroethylanandamide

In vitro receptor binding. Cannabinoid receptor affinity (K_d) was determined for CP 55,940 both in the presence and absence of the enzyme inhibitor PMSF. Without PMSF a K_d value of 15.3 ± 1.2 nM (n = 5) was calculated which correlated with the value of 15 ± 3 nM reported in the literature for binding to tissue slices (Herkenham et al., 1990). In the presence of PMSF a K_d value of 12.3 ± 2.1 nM (n = 3) resulted, which is not statistically different from the K_i value obtained without PMSF. Since the presence of PMSF did not influence affinity to the cannabinoid receptor, all K_is were calculated using the K_d value of 15.3 nM.

View larger version (88K): [in this window] [in a new window]   Fig. 2.   Autoradiogram of total (A) and nonspecific (B) binding of [3H]CP 55,940 to coronal sections of rat brain.

View larger version (89K): [in this window] [in a new window]   Fig. 3.   Autoradiogram of total [3H]CP 55,940 binding to a coronal section of rat brain 0.48 mm from bregma (A); [3H]CP 55,940 displacement at 0.48 mm from bregma by SR 141716A (B, 30 nM), CP 55,940 (C, 3 nM) and anandamide (D, 300 nM).

View larger version (92K): [in this window] [in a new window]   Fig. 4.   Autoradiogram of total [3H]CP 55,940 binding to a coronal section of rat brain 5.2 mm from bregma (A); [3H]CP 55,940 displacement at 5.2 mm from bregma by SR 141716A (B, 30 nM), CP 55,940 (C, 3 nM) and anandamide (D, 300 nM).

View larger version (69K): [in this window] [in a new window]   Fig. 5.   Autoradiogram of total [3H]CP 55,940 binding to a coronal section of rat brain 12.8 mm from bregma (A); [3H]CP 55,940 displacement at 12.8 mm from bregma by SR 141716A (B, 30 nM), CP 55,940 (C, 3 nM) and anandamide (D, 300 nM).

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Anandamide, SR 141716A and CP 55,940 displacement of [3H]CP 55,940 in the lateral caudate-putamen. The curves are representative displacement curves.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Relationship between Bmax values for anandamide, SR 141716A and CP 55,940 in different brain areas.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The failure of SR 141716A to block the effects of AN in the mouse behavioral assays was quite surprising given the wide range of anandamide effects that it blocks in other systems that include smooth muscle preparations (Rinaldi-Carmona et al., 1994), the cardiovascular system of rats (Varga et al., 1995), turning behavior after intrastriatal injections (Souilhac et al., 1995), adenylyl cyclase (Felder et al., 1995) and long-term potentiation (Terranova et al., 1995). In addition, SR 121716A is very effective in blocking the behavioral effects of ⁹-THC in mice with AD50s of approximately 0.1 mg/kg (Compton et al., 1996). There is the possibility that methodological issues could account for SR 141716A's ineffectiveness against anandamide. However, the antagonism of ⁹-THC occurred rapidly after administration of SR 141716A and was of long duration so that the time course of SR141716A would not seem to be an issue. Furthermore, the doses of SR141716A are more than sufficient for producing a complete blockade of THC's effects.

The question of whether anandamide is capable of interacting with the central cannabinoid receptor is not an issue. Several laboratories have shown that anandamide competes for cannabinoid binding in brain homogenates (Adams et al., 1995; Childers et al., 1994; Devane et al., 1992; Felder et al., 1993) with an affinity that is consistent with its relatively weak pharmacological potency. Although receptor binding studies suggest that CP 55,940, SR 141716A and anandamide are all acting at the same binding site, it is possible that they bind differently to the CB1 receptor in discrete brain areas or receptor subtypes exist that have not yet been discovered. The autoradiographic examination undertaken in the present studies provided a means for systematically examining the interactions of these compounds with cannabinoid receptors throughout the brain. The importance of receptor distribution has been amply demonstrated in several mammalian species (Herkenham et al., 1990, 1991). The densest binding occurs in the basal ganglia (substantia nigra pars reticulata, globus pallidus, entropeduncular nucleus and lateral caudate putamen), and the molecular layer of the cerebellum. Binding in these regions may explain cannabinoid interference with movement. Intermediate levels of binding were found in the CA pyramidal cell layers of the hippocampus, the dentate gyrus and layers I and VI of the cortex. ⁹-THC disrupts short-term memory in humans (Chait and Pierri, 1992). Cannabinoid effects on memory and cognition are consistent with receptor localization in the hippocampus and cortex. The presence of cannabinoid receptors in regions associated with mediating brain reward (ventromedial striatum and nucleus accumbens) suggests an association with dopamine neurons. Sparse levels were detected in the brainstem, hypothalamus, corpus callosum and the deep cerebellum nuclei. Low levels of receptors in brainstem areas controlling respiratory function is also consistent with the lack of lethality of marijuana. A similar binding profile between CP 55,940 and anandamide would reinforce the premise that these two cannabinoids have similar pharmacological characteristics.

Because anandamide is degraded in homogenate binding, it was also necessary to determine optimal binding conditions for anandamide. Consistent results were obtained when slices were pretreated with PMSF and exposed to PMSF during the reaction incubation. These results confirm previous reports of anandamide's instability in a biological system. Because slices had to be both pretreated and exposed to PMSF during the entire 2-hr experiment, this finding suggests that levels of the enzyme that degrade anandamide are high in the brain. If anandamide is a neurotransmitter, then mechanisms must exist in the CNS to rapidly remove anandamide and prevent continuous stimulation. As such, one would not expect an endogenous compound to possess great stability. Development of more stable analogs would eliminate the need of exposing brain tissue to enzyme inhibitors. PMSF inhibits a wide variety of enzymes, not just amidases. Therefore, it was necessary to determine if PMSF produced an effect on cannabinoid receptor affinity, as determined from saturation experiments. Unlike the homogenate receptor binding assay, PMSF did not influence cannabinoid receptor affinity for brain slice binding. The K_d determined in the presence of PMSF was not statistically different from the K_d obtained in the absence of PMSF. It is unknown why PMSF caused a 2-fold shift in the receptor affinity in the homogenate receptor assay, but the shift is probably due to methodological differences between slice and homogenate binding.

Establishing conditions for incubation of anandamide with tissue slices enabled us to compare the abilities of anandamide, CP 55,940 and SR 141716A to bind to the cannabinoid receptor in the following brain areas: lateral and medial caudate-putamen, frontal, occipital, entorhinal and parietal cortices, dentate gyrus, substantia nigra and the molecular layer of the cerebellum. These areas were selected because cannabinoids affect the functioning of these regions as discussed above. Also, they are all large enough so that a sufficient number of consecutive 16 µm slices could be made. Several of the areas, including the molecular layer of the cerebellum and the substantia nigra, have very dense levels of receptors. The dentate gyrus, CA1, CA3 and lateral caudate-putamen also have dense receptor populations. The cortical regions have moderate levels of cannabinoid receptors.

For anandamide, CP 55,940 and SR 141716A, no statistical difference existed between their K_is in different brain regions. Because anandamide is a weak ligand, it is possible that at the lower concentrations in anandamide's displacement curves the percentage of displacement is less accurate because the amount of displacement is overshadowed by the high numbers of receptors. If any of these compounds were binding to a receptor subtype possessing either a higher or lower K_i from the other regions, a statistical difference should result. No such differences were found for CP 55,940, anandamide or SR 141716A. Therefore, these findings support the notion of a common receptor for both CP 55,940 and anandamide.

An additional objective was to analyze representative curves for parallelism from each region for anandamide, SR 141716A and CP 55,940. Representative displacement curves from each brain region for anandamide were analyzed using the statistical program ALLFIT to determine if they were parallel. Differences in parallelism would provide evidence that a compound was interacting with the cannabinoid receptor in a different manner from other regions. All 11 curves for anandamide were parallel, as were curves for CP 55,940. For an unknown reason, the entorhinal cortex curve for SR 141716A was not parallel to the other brain regions analyzed. Thus, these three compounds appear to bind to the CB1 receptor in a similar manner.

Cannabinoid receptor densities were calculated for each brain region for anandamide, SR 141716A and CP 55,940. The purpose of calculating B_max values for each region was to determine if one compound might bind more or less in one brain area than in other regions. The relationship between B_max values for SR 141716A and CP 55,940 and anandamide and CP 55,940 were compared by linear plots of the respective values, and correlation coefficients were determined. A high correlation was obtained both when comparing the B_max values of SR 141716A and anandamide to those of CP 55,940. These correlations indicate that the three compounds are fully capable of maximal binding to the same population of receptors in all of the brain regions.

The finding that SR 141716A is capable of blocking the pharmacological effects of the stable anandamide analog 2-methyl-2-fluoroethylanandamide underscores the point that the arachidonyl derivatives interact with the cannabinoid receptor. A reasonable conclusion could be that anandamide is rapidly converted to metabolites that are responsible for its pharmacological effects. Regardless of whether the effects are due to anandamide or active metabolites, the pharmacological profile is consistent with the activation of CB1 receptors, and therefore the effects should be antagonized by SR 141716A.

Several conclusions may be drawn from these results. The lack of difference between receptor affinity, receptor distribution and parallelism of the displacement curves suggests that anandamide, SR 141716A and CP 55,940 are binding to the same receptor in the same manner. No evidence of receptor subtypes in the brain was found. However, evidence is mounting that anandamide's actions may not be identical to those of THC. Our earlier studies showed that the relationship between receptor binding (CB1) and pharmacological potency for anandamide analogs (Adams et al., 1995) does not correlate to the same extent as for other cannabinoids (Compton et al., 1993). The greater variation in the correlation with anandamide analogs could be due to either pharmacokinetics factors or to differences in receptor interactions between anandamide and THC. We have also shown recently that the time courses of anandamide's behavioral effects and brain levels do not correspond (Willoughby et al., 1997). Specifically, brain levels of anandamide decrease dramatically despite the persistence of pharmacological effects. Although these latter studies are not definitive, they suggest anandamide is producing its effects in mice in an indirect manner. The ability of SR 141716A to block the effects of the stable anandamide derivative 2-methyl-2-fluoroethylanandamide (Adams et al., 1995) further suggests that metabolism may play a role in anandamide's effects. However, there has been extensive research demonstrating the relationship between the cannabinoid pharmacological effects in mice and receptor affinity for CB1 receptors so that the effects of either anandamide or its metabolites should be attenuated by the cannabinoid antagonist. Differences in the manner with which anandamide and traditional cannabinoids produce their pharmacological effects can be exploited to gain a better understanding of the endogenous cannabinoid system.