Cannabinoid (CB)1 receptor inverse agonists inhibit food intake in animals and humans but also potentiate emesis. It is not clear whether these effects result from inverse agonist properties or from the blockade of endogenous cannabinoid signaling. Here, we examine the effect of a neutral CB1 antagonist, AM4113, on food intake, weight gain, and emesis. Neutral antagonist and binding properties were confirmed in HEK-293 cells transfected with human CB1 or CB2 receptors. AM4113 had no effect on forskolin-stimulated cAMP production at concentrations up to 630 nM. The Ki value of AM4113 (0.80 ± 0.44 nM) in competitive binding assays with the CB1/2 agonist [3H]CP55,940 was 100-fold more selective for CB1 over CB2 receptors. We determined that AM4113 antagonized CB1 receptors in brain by blocking hypothermia induced by CP55,940. AM4113 (0–20 mg/kg) significantly reduced food intake and weight gain in rat. Compared with AM251, higher doses of AM4113 were needed to produce similar effects on food intake and body weight. Unlike AM251 (5 mg/kg), a highly anorectic dose of AM4113 (10 mg/kg) did not significantly potentiate vomiting induced by the emetic morphine-6-glucoronide. We show that a centrally active neutral CB1 receptor antagonist shares the appetite suppressant and weight loss effects of inverse agonists. If these compounds display similar properties in humans, they could be developed into a new class of antiobesity agents.
- cannabinoid receptor
- food intake
- inverse agonist
cannabinoid (CB)1 G protein-coupled receptors, along with endogenous cannabinoids (endocannabinoids), including anandamide and 2-arachidonylglycerol and the biosynthetic and degradative enzymes for these ligands are found in pathways involved in feeding and metabolism (33). These pathways include vagal afferents (5), area postrema and brain stem nuclei (40, 56), hypothalamus (11, 52), and reward pathways (29, 50). In the periphery, CB1 receptors and their ligands are also found in the gut (20), liver (43), and adipose tissue (2, 11, 28). Vagal afferents, area postrema, and brain stem are also sites involved in emesis (26, 55, 56).
Therapeutically, CB1 receptor agonists are used as antiemetics and to promote appetite during serious illness, such as cancer (14) and acquired immune deficiency syndrome (58). In the case of cancer, nausea and emesis induced by chemo- and radiation therapy is significantly attenuated by the active ingredient in marijuana, Δ-9-tetrahydrocannbinol, and other CB1 receptor agonists (14, 58). CB1 receptor inverse agonists, such as SR141716A and the structurally and pharmacologically similar compound N-(piperidin-1-yl)-5-(4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide (AM251), are also therapeutically relevant. These agents suppress appetite, promote weight loss, and improve cardiometabolic risk factors in humans (17, 46, 54) and rodents (7, 19, 22, 23, 25, 28, 47, 53, 57). However, significant psychological and physiological concerns in patients treated with CB1 receptor inverse agonists are also present, including increased risks of nausea and vomiting (17, 46, 54). Previously, CB1 receptor inverse agonists were shown to promote emesis in ferret (56) and shrew (13) and nausea in rat (37).
At CB1 receptors, agonist activity decreases adenylyl cyclase and the conversion of ATP to cAMP (18, 45). Like an antagonist, inverse agonists block receptor binding and activation by a competitive agonist, but in addition, they also inhibit intrinsic or spontaneous receptor signaling (42). The result is an effect opposite to that produced by a given agonist and is termed inverse agonism. Although no valid method for assessing inverse agonist activity in vivo exists (42), the inhibition of intrinsic CB1 receptor transduction can be shown in vitro by the increased expression of adenylyl cyclase and cAMP in cultured cells transfected with CB1 receptors (35, 39, 51). Whether the physiological and behavioral effects of CB1 receptor inverse agonists result from the ability of these compounds to inhibit intrinsic receptor activity or from the pharmacological blockade of endogenous cannabinoid signaling or both is not known. However, a recent study by Sink et al. (51) describing a novel CB1 receptor antagonist, N-piperidin-1-yl-2,4-dichlorophenyl-1h-pyrazole-3-carboxamide analog (AM4113), with no inverse agonist properties shows that such determinations are now possible. These authors demonstrated that AM4113 reduced food intake in rat with no effect on conditioned gaping, a specific marker of nausea in that species (51), suggesting that pharmacological blockade of CB1 receptors alone is sufficient to reduce food intake independent of nausea.
However, in that study, food intake was only examined during a 30-min observation period (51). The effect of a neutral CB1 receptor antagonist on daily food intake and body weight is not known. Given that in some cases (2, 10, 23, 28, 47, 49) but not others (7, 53, 57) weight loss induced by CB1 receptor inverse agonists is unrelated to their effect on feeding, studies that examine the effect of neutral CB1 receptor antagonism on body weight are crucial. In addition, rats lack the neuronal pathways required for vomiting, and the effect of a neutral CB1 receptor on emesis, which is distinct from nausea, is also unknown. In the present study, we confirm the neutral antagonist and binding properties of AM4113. We then test the ability of AM4113 to antagonize hypothermia induced by the nonselective cannabinoid (−)-cis-3-[2-hydroxy-4-(1,1-dimethylheptyl)phenyl] trans-4-(3-hydroxypropyl)cyclohexanol (CP55,940) in rat. We chose this assay because the mild hypothermia caused by cannabinoid agonists is mediated by CB1 receptors in brain (4, 21), enabling us to determine whether AM4113 crossed the blood-brain barrier. We also examine the effect of AM4113 on food intake and body weight in rats treated daily for 5 days and compare the acute effect of the antagonist on food intake with the inverse agonist AM251 directly. Finally, we examine the effect of AM4113 on emesis in ferrets treated with the emetic morphine-6-glucoronide (M6G) by using vehicle- and AM251-treated ferrets for comparison.
MATERIALS AND METHODS
Forskolin-Stimulated cAMP Assay
Intracellular cAMP levels were measured with a competitive protein-binding assay using intact HEK-293 cells expressing hCB1 or hCB2 receptors and a cAMP immunoassay kit from Sigma (St. Louis, MO). In short, forskolin-stimulated cells were incubated with various concentrations of AM4113 or AM251, cAMP antibody, and cAMP conjugate for 2 h at ambient temperature. The reaction was stopped by emptying the wells followed by the addition of p-NPP substrate and incubation for 1 h. Absorbance intensity detected at 405 nm was inversely proportional to the concentration of cAMP produced by the cells. The results are expressed as the percent stimulation of forskolin-stimulated cAMP accumulation.
[3H]CP55,940 Competitive Binding Assay
Compounds were tested for their CB1 and CB2 receptor affinity using membrane preparations from rat brain (CB1 receptors) or HEK-293 cells expressing human CB2 (hCB2) receptor, respectively, and [3H]CP55,940, as previously described (30, 32, 35, 41). Stock solutions of the compounds (10 mM in DMSO) were diluted in buffer (50 mM Tris·HCL, 3 mM MgCl2, 100 mM NaCl, 0.2 mM EDTA, pH 7.4) with 0.1% BSA and transferred to 96-well plates containing [3H]CP55,940 (specific activity 128 Ci/mmol; National Insitute of Drug Abuse) at a final concentration of 0.76 nM. Nonspecific binding was assessed in the presence of 100 nM CP55,940. The binding reaction was initiated with the addition of the respective membrane suspension (∼50 μg membrane protein) followed by incubation at 30°C with gentle agitation in a shaking water bath for 60 min. Binding was terminated by rapid filtration of the membrane suspension over Unifilter GF/B-96 Well Filter Plates (Packard Instruments) using a Packard Filtermate-196 Cell Harvester. The filter plates were washed with ice-cold wash buffer (50 mM Tris-base, 5 mM MgCl2 with 0.5% BSA) and bound radioactivity was determined using a Packard TopCount Scintillation Counter. All data were in duplicate with IC50 and Ki values determined from at least two independent experiments.
All experimental protocols were approved by the University of Calgary Animal Care Committee and were carried out in accordance with the guidelines of the Canadian Council on Animal Care. Different concentrations of drugs were administered intraperitoneally in a 1 ml/kg volume, except M6G, which was given subcutaneously in a 0.1 ml/kg volume.
Antagonism by AM4113 at CB1 Receptors in Brain
We examined hypothermia induced by the CB1 agonist CP55,940 (0.3 mg/kg) (15) in male Sprague-Dawley rats (Charles River, Montreal, Quebec, Canada) pretreated 45 min earlier with either vehicle, AM251 (5 mg/kg, n = 5) (7), or AM4113 (5 mg/kg, n = 5). Briefly, silicone-coated temperature data loggers (SubCue, Calgary, Alberta, Canada) were surgically implanted into the abdominal cavity of rats (450–500 g) under isoflurane anesthesia (4% induction; 2–2.5% maintenance). Rats recovered for 3 days before being acclimatized to testing and handling procedures for an additional 7 days. During the experiment, core body temperature readings were sampled every 5 min for 300 min. Apart from the injection procedure, the rats were not handled during the experiment.
Food Intake Studies
Male Sprague-Dawley rats weighing between 330 and 380 g at the start of the study were used to examine the effect of AM4113 on food intake and body weight. Animals were fed strawberry flavored Ensure Plus liquid diet (53.3% carbohydrate, 29% fat, and 16.7% protein; 1.41 kcal/g) (Abbott Laboratories, Abbott Park, IL) to promote food intake (1) and control for spillage. Rats were habituated to testing and handling procedures daily for 7 days prior to testing. Food and water were presented in drip-free inverted glass bottles attached to the outside of the cage. Food was available for 18 h each day starting at 16:00 (12:12-h light-dark cycle; lights off 16:00). Bottles were removed and washed at 09:00, a time during the light-dark cycle when rats are generally inactive.
In experiment 1 we examined the effect of AM4113 (1 and 5 mg/kg) in rats treated daily for 5 days with the neutral antagonist. Doses were chosen based on our thermoregulation data and previous work by our group with the inverse agonist AM251 (7, 8, 36, 37). The day before the experiment, rats were assigned to one of three treatment groups: vehicle [mean body weight ± SE; (363 ± 10 g, n = 5)], AM4113 1 mg/kg (365 ± 5 g, n = 5), or AM4113 5 mg/kg (361 ± 9 g, n = 5). Food intake and body weight were monitored daily for 5 days.
In experiment 2 the effect of AM4113 (10 and 20 mg/kg) on food intake and body weight were compared with rats treated with vehicle or the inverse agonist AM251 (5 mg/kg). On the day before the experiment, rats were randomly assigned to one of four treatment groups: vehicle (492 ± 8 g, n = 6), AM251 5 mg/kg (496 ± 11 g, n = 6), AM4113 10 mg/kg (491 ± 7 g, n = 6), or AM4113 20 mg/kg (489 ± 9 g, n = 6). Rats were injected between 15:00 and 16:00 on the day of the experiment. Treatments were counterbalanced to avoid order effects. After treatment, food was made available as described in experiment 1. Food intake was measured 1.5, 3, 5, and 18 h after treatment. After the experiment, food intake and body weight were monitored for another 5 days to assess the sustained effect of each compound after a single treatment.
We examined how AM4113 (10 mg kg; n = 5) affected vomiting in ferrets (Marshall Farms, North Rose, NY) using vehicle (n = 6)- and AM251 (5 mg/kg)-treated ferrets for comparison. Each treatment was given 15 min prior to receiving the emetic M6G (0.05 mg/kg) (56). Ferrets were lightly anesthetized with halothane for each injection. Data were videotaped and analyzed by an observer blinded to the conditions of each treatment group. During 60 min of observation, the number of vomiting (emetic) episodes were counted, and activity or sleeping time was noted. In a second experiment, the effect of two additional doses of AM4113 (5 and 20 mg/kg; n = 5/5) on M6G-induced emesis were examined. An additional group of vehicle (n = 6)treated ferrets were used for comparison. Experiments were carried out using a counterbalance design to avoid order effects.
EC50 curves were generated from the forskolin-stimulated cAMP assay data by nonlinear regression with the use of GraphPad Prism software (GraphPad Prism version 3.00 for Windows, GraphPad Software, San Diego CA). Data are expressed as mean %change in forskolin-stimulated cAMP accumulation ± SE. Results from the competitive binding assay were analyzed by using nonlinear regression to determine the actual IC50 and the Ki values of the ligand (GraphPad Prism) (9). Data are expressed as mean Ki in nM ± SD. Food intake and thermoregulation data were analyzed using a two-way mixed design ANOVA with time as the repeated measure. Food intake data are expressed as mean food intake in kcal ± SE. Body temperature data are expressed in °C ± SE. Significant differences were followed up with one-way independent-measures ANOVA at each time point. Significant differences between treatments were further analyzed using Newman-Keuls Multiple Comparisons Test. The emesis data comparisons between vehicle, AM4113 (10 mg/kg), and AM251 (5 mg/kg) were made using a one-way ANOVA. Significant differences were followed up using Newman-Keuls Multiple Comparisons Test. Data examining the effect of additional doses of AM4113 (0–20 mg/kg) on emesis were analyzed using a one-way ANOVA and then by linear regression analysis. Emesis data are expressed as mean number of vomiting episodes ± SE.
AM4113 (partial designation).
AM4113 is a pyrazole congener of AM251 with a very similar molecular weight. AM251 and AM4113 were synthesized at Northeastern University. CP55,940 (Tocris Cookson, Ellisville, MO or National Institute of Drug Abuse), and M6G (Lipomed, Arlesheim, BL, Switzerland). All compounds were dissolved in DMSO using gentle heating and sonication before being diluted with Tween 80 and saline (4% DMSO; 1% Tween 80; 95% saline), and given in a volume of 1 ml/kg. Differences between baseline and vehicle (4% DMSO; 1% Tween 80; 95% saline) conditions were nonsignificant at all times.
Forskolin-Stimulated cAMP Assay
Consistent with results reported by others (51), AM4113 did not change the forskolin-stimulated cAMP accumulation in CB1-transfected HEK cells at concentrations up to 630 nM and is therefore considered to be a CB1 neutral antagonist (data not shown). In comparison, the inverse agonist AM251 increased forskolin-stimulated cAMP production in a concentration-dependent manner (EC50 56.4 nM; 95% confidence interval 5.5–573.8 nM, R2 = 0.49). The results are from one assay performed in duplicate.
[3H]CP55,940 Competitive Binding Assay
In competitive receptor binding assays against [3H]CP55,940, AM4113 showed selectivity for the CB1 receptor compared with CB2. AM4113 bound to cannabinoid receptors with a CB1 Ki of 0.80 nM and a CB2 Ki of 97 nM, indicating that AM4113 exhibits at least a 100-fold selectivity for CB1 vs. CB2 receptors consistent with data by others (51). In comparison, AM251 bound to cannabinoid receptors with a CB1 Ki of 3.43 nM and a CB2 Ki of 124 nM (data not shown).
AM4113 in Brain: Thermoregulation Assay
To determine whether AM4113 antagonized CB1 receptors in brain we examined hypothermia induced by CP55,940 in rats pretreated with either vehicle, AM251 (5 mg/kg) or AM4113 (5 mg/kg). Figure 1A shows changes in body temperature over time. A two-Way ANOVA performed on the body temperature data showed a significant treatment by time interaction (F = 4.4, P < 0.001). Differences between treatments were analyzed using a one-way ANOVA at selected time points (Fig. 1B). Prior to treatment with CP55,940, differences in body temperature between groups were nonsignificant (P > 0.05), indicating that AM4113 and AM251 alone had no effect on core body temperature. As expected, body temperature rapidly fell in vehicle-treated rats after the administration of CP55,940 (15). In comparison, the hypothermic effect of CP55,940 was completely blocked in rats treated with the neutral antagonist AM4113 (P < 0.05) and significantly attenuated in rats pretreated with inverse agonist AM251 (P < 0.05).
Food Intake and Body Weight Studies
In experiment 1 we examined how daily treatment with AM4113 (1 and 5 mg/kg) affected food intake over 5 days (Fig. 2A). A two-way ANOVA performed on the food intake data showed that there was a significant treatment by time interaction (F = 3.3, P < 0.001). There was also a significant effect of dose (F = 3.6, P < 0.001). One-way ANOVA performed between treatments showed that food intake was significantly reduced in rats treated with 5 mg/kg of AM4113 on days 3, 4, and 5 compared with vehicle and 1 mg/kg-treated rats (P < 0.05, Newman-Keuls Multiple Comparison Test).
Reductions in food intake by AM4113 (5 mg/kg) were accompanied by significant reductions in weight gain (Fig. 2B). A two-way ANOVA revealed a significant effect of dose (F = 15.2, P < 0.001) and time (F = 15.4, P < 0.001). Rats treated with 5 mg/kg of AM4113 gained significantly less weight than rats treated with 1 mg/kg of the neutral antagonist on day 4 (−16.9 kcal; P < 0.05), and significantly less weight than vehicle-treated rats on day 4 (−22.0 kcal), day 5 (−30.7 kcal), and day 6 (−33.4 kcal). Although rats treated daily with the 1-mg/kg dose of AM4113 tended to eat less and gain less weight than vehicle-treated rats, such differences were not significant (P > 0.05). The results demonstrate that repeated administrations of a 5-mg/kg dose of the neutral antagonist AM4113 significantly reduced food intake and weight gain in rat.
In experiment 2 we examined the acute effect of AM4113 on food intake by evaluating two additional doses of the neutral antagonist (10 and 20 mg/kg) compared with vehicle-treated rats, and rats treated with the inverse agonist AM251 (5 mg/kg). Figure 3A shows the early effect of AM4113 and AM251 on food intake. Results from a two-way ANOVA performed on the data revealed a significant treatment by time interaction (F = 18.1, P < 0.001). At each time point, food intake was reduced in rats treated with AM4113 (10 and 20 mg/kg) compared with vehicle-treated rats; 1.5 h (−20.2 kcal, −21.2 kcal), 3 h (−28.2 kcal, −30.0 kcal), 5 h (−32.1 kcal, −42.4 kcal), and 18 h (−59.2 kcal, −85.4 kcal) (P < 0.05). As expected (8), AM251 produced a similar effect on feeding (P < 0.05). Differences in food intake between the 10 and 20 mg/kg treatment groups were significant 18 h after treatment (P < 0.001). Differences between rats treated with AM251 and AM4113 were significant at 3, 5, and 18 h time points (P < 0.05). The results demonstrate that the neutral receptor antagonist AM4113 produced early, and dose-dependent, reductions in food intake in rat.
Furthermore, continued monitoring revealed that the anorectic effect of the neutral antagonist AM4113 (10 and 20 mg/kg) was sustained for several days after treatment. Figure 3B shows the mean daily food intake in each group before and after the experiment. A two-way ANOVA performed on the daily food intake data revealed a significant treatment by time interaction (F = 12.7, P < 0.001). Prior to treatment, differences in food intake between groups were nonsignificant (P < 0.05). After treatment, food intake was reduced in rats given AM251 or AM4113 relative to vehicle-treated rats (P < 0.05). The sustained effect of an acute administration of AM251 has been reported previously by our group (7, 8). Here we report that reductions in food intake by the neutral antagonist AM4113 were similar in duration to the anorectic effect of AM251.
Figure 3C shows that reductions in food intake by AM251 and the neutral antagonist AM4113 (10 and 20 mg/kg) were also associated with significant reductions in weight gain. Results from a two-way ANOVA performed on the weight change data showed that there was a significant treatment by time interaction (F = 10.9, P < 0.001). Weight gain was significantly reduced in rats treated with AM4113 or AM251 for several days after the experiment (P < 0.05, Newman-Keuls Multiple Comparisons Test). A two-way ANOVA performed on the weight change data also showed a significant effect of dose (F = 27.9, P < 0.001). Differences in weight gain between 10 and 20 mg/kg AM4113 treatment groups were significant on day 5 (−12.3 g; P < 0.05). The results demonstrate that a single administration of the neutral antagonist AM4113 (10 and 20 mg/kg) produced a dose-dependent reduction in weight gain and food intake.
We examined how AM4113 (10 mg kg; n = 5) affected emesis in ferrets compared with vehicle (n = 6)-treated ferrets or ferrets treated with the inverse agonist AM251 (5 mg/kg). Each ferret was given M6G 15 min after treatment to induce emesis (Fig. 4A). Doses were chosen based on work from our feeding studies showing that a 10-mg/kg dose of AM4113 produced roughly the same effect on food intake as a 5-mg/kg dose of AM251 (Fig. 3). A one-way ANOVA performed on the data showed that there was a significant effect of treatment (F = 4.1, P < 0.05) caused by increased vomiting in AM251-treated ferrets compared with vehicle treatment (P < 0.05, Newman-Keuls Multiple Comparison Test). No significant difference in vomiting episodes between AM4113 and vehicle-treated ferrets was observed. However, there was also no significant difference between ferrets treated with AM4113 and AM251 (P > 0.05). To test the effect of AM4113 on emesis more rigorously, two additional doses of AM4113 (5 and 20 mg/kg) were examined and compared with a new group of vehicle-treated ferrets (Fig. 4B). A one-way ANOVA performed on the data revealed no significant effect of treatment (P > 0.05). However, when the data were analyzed using linear regression analysis, a significant relationship between the dose of AM4113 and number of vomiting episodes was shown (r2 = 0.96, P = 0.04). The results suggest that although AM4113 is less proemetic than the inverse agonist AM251, pharmacological blockade of endogenous cannabinoid signaling is capable of affecting emesis in ferret.
We compared the effects of a novel CB1 receptor neutral antagonist (AM4113) on food intake and body weight with those of the well-characterized inverse agonist AM251. Neutral antagonist properties were established in HEK-293 cells transfected with human CB1 receptors. We confirmed that AM4113 acted as a neutral antagonist in vitro in HEK cells transfected with human CB1 receptors. Consistent with results by others (51), AM4113 had no effect on cAMP at concentrations up to 630 nM and displayed low Ki values in competitive binding assays with the CB1 agonist [3H]CP55,940. In vivo, AM4113 antagonized the hypothermic effect of the CB1 receptor agonist CP55,940 and significantly reduced food intake and weight gain in rat. Compared with AM251, higher doses of the neutral antagonist AM4113 were needed to produce similar effects on food intake and weight gain. Unlike AM251, AM4113 did not significantly increase vomiting induced by the emetic M6G, however, there was a positive correlation between dose and vomiting in ferrets treated with the neutral antagonist.
The fact that AM4113 blocked the hypothermic effect of a CB1 agonist demonstrates that the neutral antagonist crosses the blood-brain barrier, because the hypothermic effect of cannabinoid agonists is specific to activity at CB1 receptors in the anterior hypothalamus (4, 21). Consistent with these observations, preliminary studies performed on mice in our lab show that 0.4–0.7% (initial dose/g brain) of an intravenously injected 2-mg/kg dose is found in brain 15 min postinjection (data not shown). Together with a high brain (μg/g)-to-plasma (μg/ml) ratio (0.49 ± 0.029), these data show that AM4113 crosses the blood-brain barrier to affect CB1 receptors in the central nervous system. Moreover, both AM4113 and AM251 alone had no effect on core body temperature in the 45 min prior to treatment with the CB1 agonist CP55,940. These data demonstrate that endogenous cannabinoid tone is not involved in thermoregulation under the conditions we described, even with the inhibition of intrinsic CB1 receptor activity by AM251. Similar effects on temperature with CB1 receptor inverse agonists were shown previously (15, 38).
In contrast to our thermoregulation assay, larger doses of the neutral antagonist AM4113 were needed to elicit the same effect on food intake as AM251. AM4113 only affected daily food intake and body weight in rat at a 5-mg/kg dose after several treatments, whereas the same dose of the inverse agonist AM251 produced an immediate and sustained anorectic effect that lasted for several days. If reductions in food intake by these agents were produced solely from pharmacological blockade at CB1 receptors alone, then matching doses of each compound should produce similar effects on feeding. This was not the case in our study. Instead, our data suggest that the inverse agonist properties of AM251 may contribute to the ability of this compound to affect feeding pathways either directly or, possibly indirectly through feelings of nausea and malaise (37, 51). However, we previously showed that rats do not avoid flavored food pellets associated with the effects of AM251 at this dose (7) even when alternative flavors paired with vehicle treatment are available (conditioned taste aversion). However, the fact that AM251 increases conditioned gaping (nausea) in rat (37, 51), combined with higher instances of nausea in humans treated with the CB1 receptor inverse agonist SR141716A (17, 46, 54), shows that we cannot completely rule this out as a possibility. Both AM251 (37) and the inverse agonist SR141716A (16) are capable of producing conditioned taste aversion. Regardless, the fact that a neutral antagonist is capable of producing reductions in food intake and body weight, independent of nausea and emesis, demonstrates that energy balance is controlled, in part, by the release and action of endocannabinoids. Because of the inverse agonist properties of AM251, it has been previously impossible to draw this conclusion. We note that the effect of another potential CB1 antagonist, LH-21, on food intake and body weight has recently been reported (44) but that the neutral antagonist properties of this compound have yet to be established (27).
The sustained effect of AM251 on food intake reported previously (7, 8) and in the present study may result from the long half-life (∼22 h) of this compound in rat (36). One explanation for the delayed onset of the anorectic effect of AM4113 in our 5-day feeding study is that sufficient quantities of the compound need to accumulate to become effective. If reductions in food intake by AM4113 resulted from the build up of the neutral antagonist over successive treatments, then rats treated acutely with higher doses of the compound should show an immediate and sustained reduction in feeding. Consistent with this hypothesis, results from experiment 2 show that higher doses of AM4113 produced an immediate reduction in food intake and later weight gain that persisted for several days after treatment. In our study, the time course with which AM4113 reduced food intake and body weight was similar to that seen with the inverse agonist AM251. Currently, there are no data available regarding the half-life of AM4113.
CB1 receptor inverse agonists may cause nausea and are proemetic in humans (17, 46, 54) and animals (13, 37, 56). We induced emesis in ferrets pretreated with the neutral antagonist AM4113 and found that there was a positive correlation between the dose and the number of vomiting episodes produced by the emetic. These and other data (26, 55, 56) show that emesis likely stimulates the release of endocannabinoids, which in turn serve to inhibit emesis in a negative feedback manner. We also show that the inverse agonist properties of AM251 likely contribute to its effect on emesis. AM4113 antagonized the hypothermic effect of CP55,940 as effectively as AM251 in rat, but was substantially less proemetic than the inverse agonist in so far as AM4113 did not significantly increase the mean number of vomiting episodes relative to vehicle treatment. These data suggest that the inverse agonist properties of AM251 do, in fact, contribute to its effect on emesis and are consistent with results of Sink et al. (51) who show a similar effect on nausea in rat. In this way, a neutral antagonist, such as AM4113, may be better tolerated than inverse agonists as antiobesity agents.
The CB1 receptor inverse agonist SR141716A (Acomplia/Rimonabant) has already been shown to significantly improve central obesity, cholesterol profiles, circulating triglyceride levels, and insulin resistance in overweight and obese humans (17, 46, 54). In humans, circulating levels of the endogenous cannabinoid 2-arachidonylglycerol are positively correlated with intra-abdominal obesity, (3, 12), and CB1 receptor message is downregulated in visceral adipose tissue of obese subjects (3). These and other studies (6, 22–25, 28, 29, 31, 34, 43, 46, 48, 53, 54, 57) support the view that changes in endogenous cannabinoid signaling may in fact be a fundamental aspect of obesity and the metabolic syndrome and suggest that treatment with a neutral CB1 receptor antagonist will likely improve the metabolic consequences of obesity in addition to reducing body weight. Future studies designed to assess the effect of neutral CB1 receptor antagonists on other parameters of the metabolic syndrome, such as cholesterol levels, insulin resistance, and circulating triglycerides, are needed to confirm this hypothesis.
In conclusion, we show that AM4113 is a neutral, high-affinity, CB1 receptor antagonist. Our thermoregulation assay demonstrates that this compound acts centrally to antagonize the effects of a CB1 receptor agonist. The fact that AM4113 dose-dependently reduces food intake and body weight in rat supports the view that the endogenous cannabinoid system plays a physiological role in energy balance. Furthermore, our results suggest that AM4113 may be substantially less proemetic than the CB1 receptor inverse agonist AM251.
Perspectives and Significance
In mammals, several overlapping systems ensure survival by providing the drive to maintain adequate food intake. These systems converge in the brain, although many of them lie outside of the central nervous system in the gut, liver, and adipose tissue. Endocannabinoids act through central and peripheral CB1 receptors to coordinate food intake, metabolism, and energy expenditure. The lean phenotype of the CB1 receptor-deficient mouse (11) supports the idea that this receptor system plays a central role in these processes. These findings have led to the development of CB1 receptor compounds for the treatment of obesity. An unresolved issue was whether activation of the receptor by an inverse agonist contributes to the action of these compounds. Here we show that a compound with no intrinsic activity also reduces food intake and body weight in rats. The relative contributions of central vs. peripheral CB1 receptors to these responses remains to be determined, as does the role of specific neuronal populations possessing central CB1 receptors. Further development of novel cannabinoid receptor molecules will allow these questions to be addressed to advance our understating of the physiology of endocannabinoids and their role in food intake and metabolism.
These studies were supported by grants from Canadian Institutes of Health Research (to Q. J. Pittman and K. A. Sharkey) and National Institute of Drug Abuse Grants DA-09158, DA-7215, DA-3801 (to A. Makriyannis). A. P. Chambers is an Alberta Heritage Foundation for Medical Research (AHFMR) graduate student. Q. J. Pittman and K. A. Sharkey are AHFMR medical scientists. K. A. Sharkey is the Crohn's and Colitis Foundation of Canada Chair in Inflammatory Bowel Disease Research, and Q. J. Pittman holds a University Professorship.
AM4113 is currently being patent protected by A. Makriyannis.
We thank Lorraine Oland for her excellent technical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society