Truth Seeker
New Member
Abstract
The cannabinoid 1 (CB1) receptor has been implicated in the regulation of food intake. Here, we examine the effect of the CB1 receptor antagonist AM 251 on food intake and body weight over a prolonged period. Further, we examine whether AM 251 produces conditioned taste aversion (CTA) and if sustained antagonism at central receptors contributes to its anorectic effect.
The effect of AM 251 of food intake and body weight was examined in daily (1mgkg−1) and 5-day (5mgkg−1) dosing schedules. Matching reductions in food intake and body weight were observed in both paradigms. A single administration of AM 251 (5mgkg−1) significantly reduced food intake for 4 days. Tolerance to the anorectic effects of AM 251 did not develop in either dosing strategy.
Active avoidance of AM 251 (3; 5mgkg−1, i.p.) was examined using a CTA assay. Rats showed no evidence of CTA associated with AM 251.
We investigated the sustained effect of AM 251 (5mgkg−1, i.p.) on CB1 receptors in the hypothalamus using Δ9-tetrahydrocannabinol (8mgkg−1, i.p.) induced hypothermia. AM 251 initially blocked hypothermia, but this effect was not seen 2 or 4 days later.
The results demonstrate that smaller, or infrequent, administrations of AM 251 can produce sustained reductions in food intake and body weight in rat. Reductions in food intake were sustained longer than AM 251 antagonized the effects of a CB1 receptor agonist in the hypothalamus, and occurred independently of CTA.
Introduction
The cannabinoid type 1 (CB1) receptor has been strongly implicated in the regulation of food intake (Di Marzo & Matias, 2005). CB1 receptors are present in the brain and periphery. In the brain, CB1 receptors have been identified in pathways responsible for reward and energy balance (Cota et al., 2003; Robbe et al., 2003; Pickel et al., 2004). In the periphery, CB1 receptors have been identified in the gut (Croci et al., 1998; Kulkarni-Narla & Brown, 2000; Coutts et al., 2002), as well as on hepatocytes (Osei-Hyiaman et al., 2005) and adipocytes (Bensaid et al., 2003; Cota et al., 2003). CB1 receptor agonists such as Δ9-tetrahydrocannabinol (THC) and the endocannabinoids, anandamide and 2-arachidonoylglycerol, increase food intake in both humans (Hart et al., 2002) and animals (Kirkham et al., 2002). This action is blocked or reversed by the selective CB1 receptor antagonist SR 141716A (Kirkham & Williams, 2004) (Rimonabant) and by the structurally and pharmacologically similar compound AM 251 (Hildebrandt et al., 2003; Chambers et al., 2004). In rodent hypothalamus, endogenous cannabinoid levels increase when animals are food deprived and decrease during feeding (Kirkham et al., 2002). Endogenous cannabinoid levels are also under negative control by the anorectic hormone leptin (Di Marzo et al., 2001). Currently, CB1 receptor knockout mice are the only genetically engineered rodent to express a lean phenotype. Weight loss in these animals is associated with reductions in daily food intake as well as reduced fat synthesis and storage (Cota et al., 2003; Ravinet-Trillou et al., 2004; Jbilo et al., 2005; Osei-Hyiaman et al., 2005).
In a previous study, we showed that a single administration of the CB1 antagonist AM 251 led to a dose-dependent reduction in food intake and body weight for up to 6 days (Chambers et al., 2004). Others have reported that chronic treatment with CB1 antagonists leads to sustained weight loss, but only to a transient reduction in daily food intake (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003; Liu et al., 2005). We examined the effect of AM 251 on food intake and body weight over a prolonged period of treatment to see if the effects we previously observed waned over time. For comparison, we studied the same dose of AM 251 given repeatedly using a daily dosing regimen to assess whether the total amount of the drug or whether the frequency of administration altered the effects on body weight and food intake.
The sustained effect of AM 251 on food intake could be due to adverse effects of the CB1 receptor antagonist. Speculation that reductions in food intake by CB1 receptor antagonists, in part or in whole, result from nausea and/or malaise (De Vry et al., 2004b; McLaughlin et al., 2005) is supported anatomically and behaviourally. Endogenous cannabinoids (endocannabinoids) and CB1 receptors are found in the brainstem in areas involved in emesis, including the area postrema (Van Sickle et al., 2001). In ferrets (Van Sickle et al., 2001) and shrews (Darmani, 2001), CB1 receptor antagonists potentiate emesis, and in rats, CB1 antagonists can induce active avoidance of a food item associated with treatment, also known as a conditioned taste aversion (CTA) (De Vry et al., 2004b; McLaughlin et al., 2005). In these studies, reductions in the consumption of flavoured water associated with the effects of SR 141716A or AM 251 were significant at doses often used in studies examining feeding behaviour (∼8—10mgkg−1) (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003; Liu et al., 2005). The estimated ED50 for AM 251 in the CTA assay was approximately 3.6mgkg−1 (McLaughlin et al., 2005), and 6.5mgkg−1 for SR 141716A (De Vry et al., 2004b). As this is within the dose range we previously employed, we investigated whether AM 251 caused CTA under our dosing regimen, using a previously described, and highly sensitive, paradigm for this assessment (Maggio & Koopmans, 1982).
The sustained action of AM 251 over 5 days (Chambers et al., 2004) could also be due to continued antagonism at the CB1 receptor. The half-life of this drug injected intraperitoneally (i.p.) is approximately 22h in rat (McLaughlin et al., 2003). Based on dose—response curves obtained in previous studies (Chambers et al., 2004; Rutkowska 2004) there could be enough AM 251 from a 5mgkg−1 dose of the receptor antagonist to sustain a significant anorectic effect over 5 days. We chose to examine the extent of CB1 receptor antagonism in the central nervous system following a single administration of AM 251 over 5 days. We examined the effect of AM 251 on agonist-induced reductions in thermoregulation, because the effects of CB1 agonists on thermoregulation are centrally mediated (Fitton & Pertwee, 1982; Rawls et al., 2002), and because CB1 antagonists alone have no effect on body temperature (De Vry et al., 2004a).
Methods
Animals
In all experiments, male Lewis strain rats were individually housed in transparent plastic cages between 20 and 22°C under a 12h light—dark cycle (lights off 19:30h). Standard rat chow (Canadian Lab Diets, Inc., Leduc, AB, Canada) and water were freely available except where indicated. All rats were habituated to testing and handling procedures for at least 7 days prior to testing. All methods used in this study 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.
Drugs
AM 251 (N-(piperidin-1-yl)-5-(4-iodophynyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide; Tocris Cookson Inc., Ellisville, MO, U.S.A.) was dissolved in dimethyl sulphoxide (DMSO) using gentle sonication before being diluted with Tween 80 and saline (2% DMSO, 1% Tween 80, 97% saline) to final concentrations of 1, 3, and 5mgml−1 delivered i.p. (Chambers et al., 2004). THC (Lipomed, Arlesheim, Switzerland) was supplied predissolved in anhydrous ethanol solution, spun into a viscous concentrate using a Speed Vac concentrator (Savant Instruments Inc., Farmingdale, NY, U.S.A.), and resuspended into vehicle consisting of DMSO (44%), Tween 80 (1%), and saline (55%) at a final concentration of 8mgml−1 delivered i.p. (Pertwee & Tavendale, 1977). A 0.1M solution of LiCl was made up in sterile distilled water. AM 251, THC, and vehicle treatments were delivered i.p. in a volume of 1mlkg−1, LiCl and saline were delivered i.p. in a volume of 10mlkg−1 (Benoit, Air, Wilmer et al., 2003). All chemicals were supplied by Fisher Scientific (Fairlawn, NJ, U.S.A.), except where indicated.
AM 251 and food intake
To examine the effect of AM 251 given daily, or every 5 days, on food intake and body weight, 17 rats weighing between 440 and 540g were fed vanilla flavoured Ensure Plus® liquid diet (Abbott Laboratories, Abbott Park, IL, U.S.A.). The Ensure is a highly palatable, nutrient dense, complete meal replacement composed of 53.3% carbohydrate, 29% fat, and 16.7% protein (1.41kcalg−1). Food was available for 17h each day starting at 16:00h. Food and water were presented in inverted glass bottles that were attached to the outside of the cage in order to minimize spillage. Rats were injected daily at 15:45h with vehicle 1mlkg−1 i.p. (mean bodyweight±standard error of the mean (s.e.m.); 461±7.3g, n=5), or AM 251 (1mgkg−1) (462±8.5g, n=6), or every 5 days with AM 251 (5mgkg−1) (465.3±9.1g, n=6). Food intake and body weight measurements were recorded daily.
AM 251 and CTA
To determine whether two highly anorectic doses of AM 251 produced active avoidance of a food item associated with the effects of AM 251, 15 male Lewis rats (315—368g) were assigned to either AM 251 (3 and 5mgkg−1) and vehicle treatment, or LiCl and saline (n=5) treatment, conditions (Hildebrandt et al., 2003; McLaughlin et al., 2003; Shearman et al., 2003; Chambers et al., 2004; Chen et al., 2004; Rutkowska, 2004; Zhou & Shearman, 2004). During training trials, rats learned to associate the effects of AM 251 with one flavour and the effects of vehicle with another flavour. During test trials, rats chose between AM 251 and vehicle-associated flavours. A similar protocol was used with LiCl and saline to demonstrate that this procedure results in CTA to flavours associated with the effects of a noxious substance (Garcia et al., 1985). Avoidance is shown by the selection of one flavour over another.
Rats were weighed each day between 10:00 and 11:00h. Food intakes were adjusted on a daily basis to maintain each rat's original body weight and increase the incentive to eat food pellets. Each rat was habituated to training and handling procedures for at least 7 days prior to the beginning of the study. During habituation, rats had access to 20 unflavoured pellets. Rats were permitted to remain in the chamber for up to 20min, or until they had eaten at least five of 20 pellets, which ever came first. All rats consumed at least 10 unflavoured pellets during the last 2 days of habituation, and most rats ate most of their pellets.
CTA assay: The experimental design was adapted from a previous study (Maggio & Koopmans, 1982) and carried out in specially designed chambers. The outer wall of the chamber consisted of a transparent plastic cylinder 24cm h × 11cm r. Rats were able to access and consume individual pellets placed on a circular shelf within the cylinder from a well located in the centre of the chamber (8cm h × 12cm diameter).
During training trials, opposing flavours were paired with the effects of AM 251 or its vehicle via a counter balance design. Rats were placed in the chamber at approximately 13:00h where they had access to 20 raspberry- or banana-flavoured food pellets (Research Diet, New Brunswick, NJ, U.S.A.). Injections were given 30min after the last pellet was eaten. If no pellets were eaten, no treatment was given. The number of pellets eaten and the length of time in minutes before the first pellet was eaten (first pellet latency) were recorded during each training trial. Exposure to the effects of AM 251 and its vehicle was separated by 24h and followed by a single test trial 24h later in which rats were able to choose between vehicle- and AM251-associated flavours. A similar protocol was used for LiCl- and saline-treated rats. Each animal completed four training trails and four testing trials over 12 days.
During testing, 12 pellets from each flavour were arranged in an alternating fashion in order to provide a choice between treatment- and vehicle-associated flavours. When no aversion is present, vehicle- and treatment-associated flavours have an equal probability of being selected. To ensure that there was always a choice between vehicle- and treatment-associated flavours, rats were only permitted to eat 10 out of a possible 24 pellets. The probability of a single animal choosing the same flavour 10 times in a row solely due to chance alone is less than 1 × 10−4. Rats in the 3mgkg−1 treatment and LiCl treatment groups were given 20% of the full dose for each LiCl- and AM 251-associated flavoured pellet eaten to avoid the possibility of extinction of a conditioned response. To avoid the possibility that AM 251 given in the presence of both flavours would impede the ability of animals to discriminate between the effects of treatment and vehicle conditions with their respective flavours, animals in the 5mgkg−1 treatment group were not injected on the day of testing. All animals completed each test trial within 20min.
AM 251 and THC-induced hypothermia
To determine how long AM 251 continued to antagonize CB1 receptors in the hypothalamus, we examined the extent of THC-induced hypothermia in rats treated with AM 251 45min, 48h, and 96h after administration. Briefly, silicone-coated temperature data loggers (SubCue Inc., Calgary, Canada) were surgically implanted into the abdominal cavity of male Lewis rats (300—350g) under halothane anaesthesia (4% induction; 2—2.5% maintenance). Rats were allowed to recover for 3 days before being acclimatized to testing and handling procedures. Core body temperature readings were sampled every 15min for 75min during the course of each experiment.
Rats were randomly assigned into three independent groups. On day 1, at 10:45h, all rats received an injection of AM 251 (5mgkg−1). After 45min, group 1 was injected i.p. with THC (n=4) to assess the hypothermic effects of the CB1 receptor agonist in the presence of AM 251 (Costa et al., 1999; Rawls et al., 2002). The duration of AM 251's effect was tested in group 2 (n=5) 48h later, and group 3 (n=5) 96h later. Baseline temperature readings were taken 15min prior to the onset of each experiment. Differences in temperature are shown as area under the curve over 75min after administration.
Statistics
Data were analysed with two-way independent measures ANOVA, two-way mixed design ANOVA, and one-way ANOVA where appropriate. Comparisons were made using paired and un-paired t-tests (two tailed), Newman—Keuls multiple comparison test, and Bonferroni post-test where appropriate.
Results
Food intake
AM 251 significantly reduced food intake in both daily and 5-day treated rats compared with vehicle-treated controls, F(2,15)=24.33, P<0.0001. There was also a significant treatment by time interaction, F(58,15)=5.93, P<0.0001 (Figure 1a). The source of the interaction was investigated by comparing differences between vehicle- and AM 251-treated rats at each time point using a one-way analysis of variance, and Newman—Keuls multiple comparison tests.
Reductions in food intake were significant for 4 days after the first dose of AM 251 (5mgkg−1) in the 5-day dosing schedule, compared with vehicle-treated controls, P<0.05. Reductions in food intake tended to be greater in rats given the 5mgkg−1 dose compared with the 1mgkg−1 dose of AM 251; however, the extent of that reduction diminished between treatments, whereas reductions in daily food intake in rats treated daily with AM 251 (1mgkg−1) were consistent from treatment to treatment. Differences in food intake between 5 and 1mgkg−1 treatments were nonsignificant at all times, P>0.05. Contrary to previous findings (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003) tolerance to the anorectic effects of AM 251 was not observed in either dosing paradigm. Consistent with previous findings (Hildebrandt et al., 2003; Vickers et al., 2003), rats treated with AM 251 became significantly hyperphagic shortly after treatment ended, F(58,15)=5.93, P<0.0001. Overeating relative to vehicle-treated rats was significant on day 21, as well as during days 23—28 in both daily and 5-day dosing schedules, P<0.05.
To determine whether there was a significant difference in overall food intake between daily and 5-day dosing schedules during and after treatment, the cumulative food intake data were analysed using a one-way independent measures ANOVA. Similar overall intakes were observed between rats treated daily (762±31g) and rats treated every 5 days (750±24g) over 15 days of treatment, P>0.05, Newman—Keuls multiple comparison tests (Figure 1b). During treatment, rats in both dosing paradigms ate significantly less than vehicle-treated rats over the same period of time (967±47g, P<0.001). Differences in cumulative food intake between rats treated daily (499±15g) and rats treated every 5 days (512±8g) were also nonsignificant during days 21—28 after treatment ended, P>0.05 (Figure 1a). Vehicle-treated rats (427±17g) ate significantly less than AM 251-treated rats during the same period of time, P<0.01. Daily and 5-day treated rats were given the same amount of AM 251 over 15 days. The results demonstrate that the major effect on food intake by AM 251 occurred independently of the dosing strategy employed.
Body weight
Reductions in body weight were very similar between daily and 5-day treated rats. AM 251 prevented weight gain in both dosing strategies. Figure 1c shows that differences in weight change between vehicle- and AM 251-treated rats were significant from day 1 to day 24, P<0.05, Newman—Keuls multiple comparison test. A two-way ANOVA performed on the cumulative weight change data showed that differences in weight change between vehicle- and AM 251-treated rats were significant between groups (F(2,15)=28.4, P<0.0001) and over time (F(29,15)=38.2, P<0.0001). The difference in weight gain in vehicle-treated rats during the first 15 days of the study also created a significant treatment by time interaction, F(58,15)=5.85, P<0.01.
Feeding efficiency
Rats treated with AM 251 tended to gain significantly less weight than vehicle-treated rats for every gram of food eaten, F(2,15)=32.5, P<0.0001 (Figure 1d). Feeding efficiency data are presented as the change in body weight over total food intake (g) × 100 during 5-day intervals, starting 5 days prior to treatment. Although AM 251 tended to reduce feeding efficiency such differences were only statistically significant in rats treated daily with the CB1 receptor antagonist during days 5—10 relative to vehicle-treated controls, P<0.05. After the end of the treatment, feeding efficiencies were significantly increased in both daily and 5-day treated rats from days 21 to 30 creating a significant treatment by time interaction, F(12,15)=13.1, P<0.0001. The results demonstrate that rats were able to compensate for reductions in feeding efficiency caused by AM 251 after treatment ended.
AM 251 and CTA
Rats did not demonstrate any aversion to flavours associated with two highly anorectic doses of AM 251 (3; 5mgkg−1). In a free-choice test trial, the ratio between pellets selected from vehicle- and treatment-associated flavours is roughly 50/50 when no aversion is present. Figure 2a shows that rats were as likely to eat flavours associated with AM 251 as they were vehicle-associated flavours. In contrast, rats showed a robust aversion to flavours associated with LiCl after the very first training trial (single-trial learning), as previously observed (Garcia et al., 1976, 1985; Maggio & Koopmans, 1982), but no aversion to flavours associated with saline treatment. A two-way mixed design ANOVA performed on the percentage of flavoured pellets eaten showed that differences between treatments were highly significant, F(2,12)=96.2, P<0.0001, whereas differences between trials were nonsignificant, F(3,12)=0.02, P=0.96.
Significant differences between LiCl- and AM 251-treated rats were followed up using paired and unpaired two-tailed t-tests. Differences between AM 251- and LiCl-associated flavours were significant during each trial, P<0.0001, whereas differences between 3 and 5mgkg−1 treatments were nonsignificant at all times, P>0.28. The combined average of AM 251-associated flavours in the percentage of pellets eaten data was 49.8±0.3%. In comparison, LiCl-associated flavours were selected only 1.0±1.0% of the time over four trials. In the last three test trials, no flavoured pellets associated with LiCl treatment were eaten.
The lack of aversion to flavours associated with AM 251 was also evident in the data gathered during training trials. Figure 2b shows the average latency to eat the first pellet in minutes during trials 2—4. The results show that first pellet latency was significantly longer to flavours associated with LiCl treatment (16.4±1.8min) compared with saline control flavours, (3.2±1.1min, P<0.01, two-tailed paired t-test). In comparison, rats showed similar latencies between vehicle- (1.6±0.4min) and AM 251 (3mgkg−1, 1.4±0.4min, 5mgkg−1, 1.2±0.1min)-associated flavours, P=0.52. The results demonstrate a lack of aversion to flavours paired with the effects of AM 251. Data from the first training trial were excluded from this analysis on the basis that rats were naïve to their respective treatments at this time.
AM 251 and thermogenesis
We tested the ability of AM 251 to block the hypothermic effects of THC given 45min, 48h, and 96h after administration of the CB1 antagonist. A two-way independent measures ANOVA and follow-up analysis performed on the cumulative change in temperature data showed that while AM 251 was highly effective at blocking the hypothermic effect of THC given 45min later, P<0.01 (unpaired t-test), it no longer antagonized the hyperthermic action of THC given 48 or 96h later, creating a significant treatment × time interaction F(2,2)=5.41, P=0.01 (Figure 3). The results demonstrate that a 5mgkg−1 dose of AM 251 no longer antagonized the hypothermic effects of a CB1 receptor agonist after 48h.
Discussion
We show that a 1mgkg−1 dose of AM 251 given daily, or a 5mgkg−1 dose given every 5 days, produced matching reductions in food intake and body weight. Contrary to previous findings, reductions in food intake continued to be significant throughout treatment with no evidence of tolerance to the anorectic effect of AM 251 in either dosing strategy. The fact that a 5mgkg−1 dose of AM 251 did not antagonize the central effects of THC after 48h suggests that the effect on food intake in the 5-day dosing strategy was not mediated by continued antagonism of CB1 receptors in the central nervous system. The same dose of AM 251 significantly reduced food intake for 4 days without causing CTA, even after multiple training and testing trials. Consistent with findings by others, AM 251 also tended to reduce feeding efficiency, suggesting that reductions in body weight were related to changes in metabolism and/or increased energy expenditure.
In comparison, the majority of studies investigating the effect of CB1 receptor antagonists on feeding behaviour report that decreases in daily food intake caused by chronic-daily administration of AM 251 or SR 141716A diminish over time, often to the point that differences between vehicle-and antagonist-treated animals are no longer significant after the first week of treatment (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003). For example, Wistar rats, weighing between 280 and 330g, developed tolerance to the anorectic effect of daily SR 141716A (2.5mgkg−1, i.p.) after only 3 days (Colombo et al., 1998), and obese mice treated orally with 3.0mgkg−1 of AM 251 developed tolerance to the anorectic effect of that compound after only 8 days (Hildebrandt et al., 2003).
Remarkably, in those studies and others (Ravinet-Trillou et al., 2003; Vickers et al., 2003), reductions in body weight relative to vehicle-treated animals continued to be significant throughout treatment, even though the daily food intakes of antagonist-treated animals were similar to those of vehicle-treated controls. Interestingly, tolerance also develops faster in lean, compared with obese, animals (Vickers et al., 2003). The observation that tolerance is linked to existing energy stores strongly suggests that peripheral factors are also involved in the development of resistance to the anorectic effect of CB1 receptor antagonists. Consistent with this hypothesis, activity at CB1 receptors increases de novo lipogenesis in the liver (Osei-Hyiaman et al., 2005) as well as in the adipose tissue (Cota et al., 2003; Ravinet-Trillou et al., 2004), possibly by inhibiting adenosine monophosphate-activated protein kinase (Kola et al., 2005), or through the induction of enzymes of the β-oxidation and tricarboxylic acid cycle (Jbilo et al., 2005). By inhibiting lipolysis, CB1 receptor antagonists could increase the number of calories required to replace and maintain existing energy stores, thereby creating conditions in which increases in food intake fail to produce additional growth. Increases in adiponectin mRNA, a circulating hormone that leads to weight loss through increased oxygen consumption (thermogenesis), could give rise to similar conditions (Bensaid et al., 2003; Liu et al., 2005).
Consistent with the hypothesis that CB1 receptor antagonists affect metabolic pathways in addition to the pathways that regulate feeding behaviour directly, treatment with SR 141716A (10mgkg−1), or genetic ablation of the CB1 receptor, also decreases feeding efficiency (Ravinet-Trillou et al., 2004; Jbilo et al., 2005). Feeding efficiency is a ratio that compares changes in body weight to the total number of calories ingested thereby providing an index of how well an organism is able to assimilate nutrients into body mass. Although it is likely that changes in the metabolism and energy expenditure also contributed to reductions in body weight in our study, the fact that feeding efficiency was only moderately reduced during both daily and 5-day dosing strategies suggests that the major effect of AM 251 on body weight in our study resulted from decreases in daily food intake. The reason that tolerance to the anorectic effect of AM 251 was not observed in our study was likely due to differences in the protocol. We employed a smaller dose of the CB1 receptor antagonist (1mgkg−1 per day) than is typically used in other studies (3—10mgkg−1 per day) (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003; Liu et al., 2005), and in relatively large rats (480—580g). Our results demonstrate that reductions in daily food intake by smaller, or less frequent, doses of AM 251 can be maintained in rats sustained on a highly palatable and nutrient-rich diet, possibly because of limited activity of the antagonist at CB1 receptors in the pathways directly affecting lipolysis and energy expenditure.
We also found that reductions in food intake by AM 251 could not be attributed to feelings of malaise as measured by a highly sensitive, two-flavour CTA procedure. Given the sustained effect a single dose of AM 251 had on food intake, and based on work carried out by others (De Vry et al., 2004b; McLaughlin et al., 2005), one might expect that rats repeatedly trained to associate the physiological effects of AM 251 with a specific flavour would avoid that flavour in a free-choice test trial. Instead, we found that rats were as likely to select flavours associated with AM 251 as they were flavours associated with vehicle treatment. We also found no difference in the way rats approached AM 251- and vehicle-associated flavours, based on the time each rat took to begin eating over four training and testing trials. The results demonstrate that the anorectic effect of AM 251 in our study was not associated with malaise as measured by CTA at doses 5mgkg−1, even though CTA can be induced by CB1 receptor antagonists when higher doses are used (De Vry et al., 2004b; McLaughlin et al., 2005).
To determine whether AM 251 was still acting on CB1 receptors in the brain over a 5-day period, we examined how long AM 251 antagonized the effects of THC using a thermoregulation assay, and found that AM 251 had no effect on hypothermia produced by an i.p. administration of the CB1 agonist given 48 and 96h later. CB1 receptors are densely distributed in the anterior nucleus of the hypothalamus or preoptic area (Herkenham et al., 1991; Moldrich & Wenger, 2000), and thermosensitive neurons in the preoptic area are thought to be critical for thermoregulation (Boulant, 1981; Boulant, 2000). Because tolerance to the hypothermic effects of peripherally administered THC can be induced with intrahypothalamic administrations of the CB1 agonist (Fitton & Pertwee, 1982), the major effect of CB1 agonists on thermoregulation is thought to be mediated by CB1 receptors in the hypothalamus. In our study, AM 251 had no significant impact on the effects of THC given 48h later. Therefore, it is likely that CB1 receptors in the central nervous system are no longer significantly antagonized by a 5mgkg−1 dose of AM 251 48h after administration.
The observation that intermittent administration of AM 251 inhibits daily food intake and weight gain at least as well as daily dosing regimes in rats may have significant clinical implications for the use of CB1 receptor antagonists in the treatment of human obesity. The advantages of an intermittent dosing strategy include targeting the peak anorectic effect of CB1 receptor antagonist to times in which people are likely to overeat, and limiting the pharmacological dose and blockade of the CB1 receptor to the shortest duration possible to limit unwanted side effects. It has long been known that people eat 8—13% more during weekends (St Jeor et al., 1983; de Castro, 2000). Data from our study suggest that weekly treatment with a CB1 receptor antagonist given just prior to the weekend could enhance the efficacy of these agents by targeting their peak anorectic effect to times in which people are likely to overeat. Our data also demonstrate that only intermittent blockade of the CB1 receptor is necessary to produce a prolonged inhibition of food intake. Mounting evidence suggests that the endogenous cannabinoid system plays important roles in gut function, suggesting that continued antagonism may lead to unwanted effects in this system. For example, daily administrations of CB1 receptor antagonists have been associated with symptoms of nausea and diarrhoea in humans (Van Gaal et al., 2005) and endogenous activity at CB1 receptors has been hypothesized to protect the gastrointestinal tract during inflammation as well as during abnormally high gastric and enteric secretions (Massa et al., 2004; Duncan et al., 2005). Minimizing the duration of antagonism at CB1 receptors with a once or twice a week dosing strategy may result in fewer side effects and better patient compliance. However, additional studies will be needed to confirm these findings under experimental circumstances more closely related to those used in the treatment of human patients.
In summary, the matching reductions in food intake and bodyweight in rats treated either daily or infrequently with AM 251 demonstrate that a relatively low dose of the receptor antagonist is capable of producing significant changes in energy balance that are resistant to the acquisition of tolerance. These reductions are independent from noxious effects as measured by CTA and last longer than the pharmacological blockade of CB1 receptors in the brain by AM 251.
Source, Graphs and Figures: AM 251 produces sustained reductions in food intake and body weight that are resistant to tolerance and conditioned taste aversion
The cannabinoid 1 (CB1) receptor has been implicated in the regulation of food intake. Here, we examine the effect of the CB1 receptor antagonist AM 251 on food intake and body weight over a prolonged period. Further, we examine whether AM 251 produces conditioned taste aversion (CTA) and if sustained antagonism at central receptors contributes to its anorectic effect.
The effect of AM 251 of food intake and body weight was examined in daily (1mgkg−1) and 5-day (5mgkg−1) dosing schedules. Matching reductions in food intake and body weight were observed in both paradigms. A single administration of AM 251 (5mgkg−1) significantly reduced food intake for 4 days. Tolerance to the anorectic effects of AM 251 did not develop in either dosing strategy.
Active avoidance of AM 251 (3; 5mgkg−1, i.p.) was examined using a CTA assay. Rats showed no evidence of CTA associated with AM 251.
We investigated the sustained effect of AM 251 (5mgkg−1, i.p.) on CB1 receptors in the hypothalamus using Δ9-tetrahydrocannabinol (8mgkg−1, i.p.) induced hypothermia. AM 251 initially blocked hypothermia, but this effect was not seen 2 or 4 days later.
The results demonstrate that smaller, or infrequent, administrations of AM 251 can produce sustained reductions in food intake and body weight in rat. Reductions in food intake were sustained longer than AM 251 antagonized the effects of a CB1 receptor agonist in the hypothalamus, and occurred independently of CTA.
Introduction
The cannabinoid type 1 (CB1) receptor has been strongly implicated in the regulation of food intake (Di Marzo & Matias, 2005). CB1 receptors are present in the brain and periphery. In the brain, CB1 receptors have been identified in pathways responsible for reward and energy balance (Cota et al., 2003; Robbe et al., 2003; Pickel et al., 2004). In the periphery, CB1 receptors have been identified in the gut (Croci et al., 1998; Kulkarni-Narla & Brown, 2000; Coutts et al., 2002), as well as on hepatocytes (Osei-Hyiaman et al., 2005) and adipocytes (Bensaid et al., 2003; Cota et al., 2003). CB1 receptor agonists such as Δ9-tetrahydrocannabinol (THC) and the endocannabinoids, anandamide and 2-arachidonoylglycerol, increase food intake in both humans (Hart et al., 2002) and animals (Kirkham et al., 2002). This action is blocked or reversed by the selective CB1 receptor antagonist SR 141716A (Kirkham & Williams, 2004) (Rimonabant) and by the structurally and pharmacologically similar compound AM 251 (Hildebrandt et al., 2003; Chambers et al., 2004). In rodent hypothalamus, endogenous cannabinoid levels increase when animals are food deprived and decrease during feeding (Kirkham et al., 2002). Endogenous cannabinoid levels are also under negative control by the anorectic hormone leptin (Di Marzo et al., 2001). Currently, CB1 receptor knockout mice are the only genetically engineered rodent to express a lean phenotype. Weight loss in these animals is associated with reductions in daily food intake as well as reduced fat synthesis and storage (Cota et al., 2003; Ravinet-Trillou et al., 2004; Jbilo et al., 2005; Osei-Hyiaman et al., 2005).
In a previous study, we showed that a single administration of the CB1 antagonist AM 251 led to a dose-dependent reduction in food intake and body weight for up to 6 days (Chambers et al., 2004). Others have reported that chronic treatment with CB1 antagonists leads to sustained weight loss, but only to a transient reduction in daily food intake (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003; Liu et al., 2005). We examined the effect of AM 251 on food intake and body weight over a prolonged period of treatment to see if the effects we previously observed waned over time. For comparison, we studied the same dose of AM 251 given repeatedly using a daily dosing regimen to assess whether the total amount of the drug or whether the frequency of administration altered the effects on body weight and food intake.
The sustained effect of AM 251 on food intake could be due to adverse effects of the CB1 receptor antagonist. Speculation that reductions in food intake by CB1 receptor antagonists, in part or in whole, result from nausea and/or malaise (De Vry et al., 2004b; McLaughlin et al., 2005) is supported anatomically and behaviourally. Endogenous cannabinoids (endocannabinoids) and CB1 receptors are found in the brainstem in areas involved in emesis, including the area postrema (Van Sickle et al., 2001). In ferrets (Van Sickle et al., 2001) and shrews (Darmani, 2001), CB1 receptor antagonists potentiate emesis, and in rats, CB1 antagonists can induce active avoidance of a food item associated with treatment, also known as a conditioned taste aversion (CTA) (De Vry et al., 2004b; McLaughlin et al., 2005). In these studies, reductions in the consumption of flavoured water associated with the effects of SR 141716A or AM 251 were significant at doses often used in studies examining feeding behaviour (∼8—10mgkg−1) (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003; Liu et al., 2005). The estimated ED50 for AM 251 in the CTA assay was approximately 3.6mgkg−1 (McLaughlin et al., 2005), and 6.5mgkg−1 for SR 141716A (De Vry et al., 2004b). As this is within the dose range we previously employed, we investigated whether AM 251 caused CTA under our dosing regimen, using a previously described, and highly sensitive, paradigm for this assessment (Maggio & Koopmans, 1982).
The sustained action of AM 251 over 5 days (Chambers et al., 2004) could also be due to continued antagonism at the CB1 receptor. The half-life of this drug injected intraperitoneally (i.p.) is approximately 22h in rat (McLaughlin et al., 2003). Based on dose—response curves obtained in previous studies (Chambers et al., 2004; Rutkowska 2004) there could be enough AM 251 from a 5mgkg−1 dose of the receptor antagonist to sustain a significant anorectic effect over 5 days. We chose to examine the extent of CB1 receptor antagonism in the central nervous system following a single administration of AM 251 over 5 days. We examined the effect of AM 251 on agonist-induced reductions in thermoregulation, because the effects of CB1 agonists on thermoregulation are centrally mediated (Fitton & Pertwee, 1982; Rawls et al., 2002), and because CB1 antagonists alone have no effect on body temperature (De Vry et al., 2004a).
Methods
Animals
In all experiments, male Lewis strain rats were individually housed in transparent plastic cages between 20 and 22°C under a 12h light—dark cycle (lights off 19:30h). Standard rat chow (Canadian Lab Diets, Inc., Leduc, AB, Canada) and water were freely available except where indicated. All rats were habituated to testing and handling procedures for at least 7 days prior to testing. All methods used in this study 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.
Drugs
AM 251 (N-(piperidin-1-yl)-5-(4-iodophynyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide; Tocris Cookson Inc., Ellisville, MO, U.S.A.) was dissolved in dimethyl sulphoxide (DMSO) using gentle sonication before being diluted with Tween 80 and saline (2% DMSO, 1% Tween 80, 97% saline) to final concentrations of 1, 3, and 5mgml−1 delivered i.p. (Chambers et al., 2004). THC (Lipomed, Arlesheim, Switzerland) was supplied predissolved in anhydrous ethanol solution, spun into a viscous concentrate using a Speed Vac concentrator (Savant Instruments Inc., Farmingdale, NY, U.S.A.), and resuspended into vehicle consisting of DMSO (44%), Tween 80 (1%), and saline (55%) at a final concentration of 8mgml−1 delivered i.p. (Pertwee & Tavendale, 1977). A 0.1M solution of LiCl was made up in sterile distilled water. AM 251, THC, and vehicle treatments were delivered i.p. in a volume of 1mlkg−1, LiCl and saline were delivered i.p. in a volume of 10mlkg−1 (Benoit, Air, Wilmer et al., 2003). All chemicals were supplied by Fisher Scientific (Fairlawn, NJ, U.S.A.), except where indicated.
AM 251 and food intake
To examine the effect of AM 251 given daily, or every 5 days, on food intake and body weight, 17 rats weighing between 440 and 540g were fed vanilla flavoured Ensure Plus® liquid diet (Abbott Laboratories, Abbott Park, IL, U.S.A.). The Ensure is a highly palatable, nutrient dense, complete meal replacement composed of 53.3% carbohydrate, 29% fat, and 16.7% protein (1.41kcalg−1). Food was available for 17h each day starting at 16:00h. Food and water were presented in inverted glass bottles that were attached to the outside of the cage in order to minimize spillage. Rats were injected daily at 15:45h with vehicle 1mlkg−1 i.p. (mean bodyweight±standard error of the mean (s.e.m.); 461±7.3g, n=5), or AM 251 (1mgkg−1) (462±8.5g, n=6), or every 5 days with AM 251 (5mgkg−1) (465.3±9.1g, n=6). Food intake and body weight measurements were recorded daily.
AM 251 and CTA
To determine whether two highly anorectic doses of AM 251 produced active avoidance of a food item associated with the effects of AM 251, 15 male Lewis rats (315—368g) were assigned to either AM 251 (3 and 5mgkg−1) and vehicle treatment, or LiCl and saline (n=5) treatment, conditions (Hildebrandt et al., 2003; McLaughlin et al., 2003; Shearman et al., 2003; Chambers et al., 2004; Chen et al., 2004; Rutkowska, 2004; Zhou & Shearman, 2004). During training trials, rats learned to associate the effects of AM 251 with one flavour and the effects of vehicle with another flavour. During test trials, rats chose between AM 251 and vehicle-associated flavours. A similar protocol was used with LiCl and saline to demonstrate that this procedure results in CTA to flavours associated with the effects of a noxious substance (Garcia et al., 1985). Avoidance is shown by the selection of one flavour over another.
Rats were weighed each day between 10:00 and 11:00h. Food intakes were adjusted on a daily basis to maintain each rat's original body weight and increase the incentive to eat food pellets. Each rat was habituated to training and handling procedures for at least 7 days prior to the beginning of the study. During habituation, rats had access to 20 unflavoured pellets. Rats were permitted to remain in the chamber for up to 20min, or until they had eaten at least five of 20 pellets, which ever came first. All rats consumed at least 10 unflavoured pellets during the last 2 days of habituation, and most rats ate most of their pellets.
CTA assay: The experimental design was adapted from a previous study (Maggio & Koopmans, 1982) and carried out in specially designed chambers. The outer wall of the chamber consisted of a transparent plastic cylinder 24cm h × 11cm r. Rats were able to access and consume individual pellets placed on a circular shelf within the cylinder from a well located in the centre of the chamber (8cm h × 12cm diameter).
During training trials, opposing flavours were paired with the effects of AM 251 or its vehicle via a counter balance design. Rats were placed in the chamber at approximately 13:00h where they had access to 20 raspberry- or banana-flavoured food pellets (Research Diet, New Brunswick, NJ, U.S.A.). Injections were given 30min after the last pellet was eaten. If no pellets were eaten, no treatment was given. The number of pellets eaten and the length of time in minutes before the first pellet was eaten (first pellet latency) were recorded during each training trial. Exposure to the effects of AM 251 and its vehicle was separated by 24h and followed by a single test trial 24h later in which rats were able to choose between vehicle- and AM251-associated flavours. A similar protocol was used for LiCl- and saline-treated rats. Each animal completed four training trails and four testing trials over 12 days.
During testing, 12 pellets from each flavour were arranged in an alternating fashion in order to provide a choice between treatment- and vehicle-associated flavours. When no aversion is present, vehicle- and treatment-associated flavours have an equal probability of being selected. To ensure that there was always a choice between vehicle- and treatment-associated flavours, rats were only permitted to eat 10 out of a possible 24 pellets. The probability of a single animal choosing the same flavour 10 times in a row solely due to chance alone is less than 1 × 10−4. Rats in the 3mgkg−1 treatment and LiCl treatment groups were given 20% of the full dose for each LiCl- and AM 251-associated flavoured pellet eaten to avoid the possibility of extinction of a conditioned response. To avoid the possibility that AM 251 given in the presence of both flavours would impede the ability of animals to discriminate between the effects of treatment and vehicle conditions with their respective flavours, animals in the 5mgkg−1 treatment group were not injected on the day of testing. All animals completed each test trial within 20min.
AM 251 and THC-induced hypothermia
To determine how long AM 251 continued to antagonize CB1 receptors in the hypothalamus, we examined the extent of THC-induced hypothermia in rats treated with AM 251 45min, 48h, and 96h after administration. Briefly, silicone-coated temperature data loggers (SubCue Inc., Calgary, Canada) were surgically implanted into the abdominal cavity of male Lewis rats (300—350g) under halothane anaesthesia (4% induction; 2—2.5% maintenance). Rats were allowed to recover for 3 days before being acclimatized to testing and handling procedures. Core body temperature readings were sampled every 15min for 75min during the course of each experiment.
Rats were randomly assigned into three independent groups. On day 1, at 10:45h, all rats received an injection of AM 251 (5mgkg−1). After 45min, group 1 was injected i.p. with THC (n=4) to assess the hypothermic effects of the CB1 receptor agonist in the presence of AM 251 (Costa et al., 1999; Rawls et al., 2002). The duration of AM 251's effect was tested in group 2 (n=5) 48h later, and group 3 (n=5) 96h later. Baseline temperature readings were taken 15min prior to the onset of each experiment. Differences in temperature are shown as area under the curve over 75min after administration.
Statistics
Data were analysed with two-way independent measures ANOVA, two-way mixed design ANOVA, and one-way ANOVA where appropriate. Comparisons were made using paired and un-paired t-tests (two tailed), Newman—Keuls multiple comparison test, and Bonferroni post-test where appropriate.
Results
Food intake
AM 251 significantly reduced food intake in both daily and 5-day treated rats compared with vehicle-treated controls, F(2,15)=24.33, P<0.0001. There was also a significant treatment by time interaction, F(58,15)=5.93, P<0.0001 (Figure 1a). The source of the interaction was investigated by comparing differences between vehicle- and AM 251-treated rats at each time point using a one-way analysis of variance, and Newman—Keuls multiple comparison tests.
Reductions in food intake were significant for 4 days after the first dose of AM 251 (5mgkg−1) in the 5-day dosing schedule, compared with vehicle-treated controls, P<0.05. Reductions in food intake tended to be greater in rats given the 5mgkg−1 dose compared with the 1mgkg−1 dose of AM 251; however, the extent of that reduction diminished between treatments, whereas reductions in daily food intake in rats treated daily with AM 251 (1mgkg−1) were consistent from treatment to treatment. Differences in food intake between 5 and 1mgkg−1 treatments were nonsignificant at all times, P>0.05. Contrary to previous findings (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003) tolerance to the anorectic effects of AM 251 was not observed in either dosing paradigm. Consistent with previous findings (Hildebrandt et al., 2003; Vickers et al., 2003), rats treated with AM 251 became significantly hyperphagic shortly after treatment ended, F(58,15)=5.93, P<0.0001. Overeating relative to vehicle-treated rats was significant on day 21, as well as during days 23—28 in both daily and 5-day dosing schedules, P<0.05.
To determine whether there was a significant difference in overall food intake between daily and 5-day dosing schedules during and after treatment, the cumulative food intake data were analysed using a one-way independent measures ANOVA. Similar overall intakes were observed between rats treated daily (762±31g) and rats treated every 5 days (750±24g) over 15 days of treatment, P>0.05, Newman—Keuls multiple comparison tests (Figure 1b). During treatment, rats in both dosing paradigms ate significantly less than vehicle-treated rats over the same period of time (967±47g, P<0.001). Differences in cumulative food intake between rats treated daily (499±15g) and rats treated every 5 days (512±8g) were also nonsignificant during days 21—28 after treatment ended, P>0.05 (Figure 1a). Vehicle-treated rats (427±17g) ate significantly less than AM 251-treated rats during the same period of time, P<0.01. Daily and 5-day treated rats were given the same amount of AM 251 over 15 days. The results demonstrate that the major effect on food intake by AM 251 occurred independently of the dosing strategy employed.
Body weight
Reductions in body weight were very similar between daily and 5-day treated rats. AM 251 prevented weight gain in both dosing strategies. Figure 1c shows that differences in weight change between vehicle- and AM 251-treated rats were significant from day 1 to day 24, P<0.05, Newman—Keuls multiple comparison test. A two-way ANOVA performed on the cumulative weight change data showed that differences in weight change between vehicle- and AM 251-treated rats were significant between groups (F(2,15)=28.4, P<0.0001) and over time (F(29,15)=38.2, P<0.0001). The difference in weight gain in vehicle-treated rats during the first 15 days of the study also created a significant treatment by time interaction, F(58,15)=5.85, P<0.01.
Feeding efficiency
Rats treated with AM 251 tended to gain significantly less weight than vehicle-treated rats for every gram of food eaten, F(2,15)=32.5, P<0.0001 (Figure 1d). Feeding efficiency data are presented as the change in body weight over total food intake (g) × 100 during 5-day intervals, starting 5 days prior to treatment. Although AM 251 tended to reduce feeding efficiency such differences were only statistically significant in rats treated daily with the CB1 receptor antagonist during days 5—10 relative to vehicle-treated controls, P<0.05. After the end of the treatment, feeding efficiencies were significantly increased in both daily and 5-day treated rats from days 21 to 30 creating a significant treatment by time interaction, F(12,15)=13.1, P<0.0001. The results demonstrate that rats were able to compensate for reductions in feeding efficiency caused by AM 251 after treatment ended.
AM 251 and CTA
Rats did not demonstrate any aversion to flavours associated with two highly anorectic doses of AM 251 (3; 5mgkg−1). In a free-choice test trial, the ratio between pellets selected from vehicle- and treatment-associated flavours is roughly 50/50 when no aversion is present. Figure 2a shows that rats were as likely to eat flavours associated with AM 251 as they were vehicle-associated flavours. In contrast, rats showed a robust aversion to flavours associated with LiCl after the very first training trial (single-trial learning), as previously observed (Garcia et al., 1976, 1985; Maggio & Koopmans, 1982), but no aversion to flavours associated with saline treatment. A two-way mixed design ANOVA performed on the percentage of flavoured pellets eaten showed that differences between treatments were highly significant, F(2,12)=96.2, P<0.0001, whereas differences between trials were nonsignificant, F(3,12)=0.02, P=0.96.
Significant differences between LiCl- and AM 251-treated rats were followed up using paired and unpaired two-tailed t-tests. Differences between AM 251- and LiCl-associated flavours were significant during each trial, P<0.0001, whereas differences between 3 and 5mgkg−1 treatments were nonsignificant at all times, P>0.28. The combined average of AM 251-associated flavours in the percentage of pellets eaten data was 49.8±0.3%. In comparison, LiCl-associated flavours were selected only 1.0±1.0% of the time over four trials. In the last three test trials, no flavoured pellets associated with LiCl treatment were eaten.
The lack of aversion to flavours associated with AM 251 was also evident in the data gathered during training trials. Figure 2b shows the average latency to eat the first pellet in minutes during trials 2—4. The results show that first pellet latency was significantly longer to flavours associated with LiCl treatment (16.4±1.8min) compared with saline control flavours, (3.2±1.1min, P<0.01, two-tailed paired t-test). In comparison, rats showed similar latencies between vehicle- (1.6±0.4min) and AM 251 (3mgkg−1, 1.4±0.4min, 5mgkg−1, 1.2±0.1min)-associated flavours, P=0.52. The results demonstrate a lack of aversion to flavours paired with the effects of AM 251. Data from the first training trial were excluded from this analysis on the basis that rats were naïve to their respective treatments at this time.
AM 251 and thermogenesis
We tested the ability of AM 251 to block the hypothermic effects of THC given 45min, 48h, and 96h after administration of the CB1 antagonist. A two-way independent measures ANOVA and follow-up analysis performed on the cumulative change in temperature data showed that while AM 251 was highly effective at blocking the hypothermic effect of THC given 45min later, P<0.01 (unpaired t-test), it no longer antagonized the hyperthermic action of THC given 48 or 96h later, creating a significant treatment × time interaction F(2,2)=5.41, P=0.01 (Figure 3). The results demonstrate that a 5mgkg−1 dose of AM 251 no longer antagonized the hypothermic effects of a CB1 receptor agonist after 48h.
Discussion
We show that a 1mgkg−1 dose of AM 251 given daily, or a 5mgkg−1 dose given every 5 days, produced matching reductions in food intake and body weight. Contrary to previous findings, reductions in food intake continued to be significant throughout treatment with no evidence of tolerance to the anorectic effect of AM 251 in either dosing strategy. The fact that a 5mgkg−1 dose of AM 251 did not antagonize the central effects of THC after 48h suggests that the effect on food intake in the 5-day dosing strategy was not mediated by continued antagonism of CB1 receptors in the central nervous system. The same dose of AM 251 significantly reduced food intake for 4 days without causing CTA, even after multiple training and testing trials. Consistent with findings by others, AM 251 also tended to reduce feeding efficiency, suggesting that reductions in body weight were related to changes in metabolism and/or increased energy expenditure.
In comparison, the majority of studies investigating the effect of CB1 receptor antagonists on feeding behaviour report that decreases in daily food intake caused by chronic-daily administration of AM 251 or SR 141716A diminish over time, often to the point that differences between vehicle-and antagonist-treated animals are no longer significant after the first week of treatment (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003). For example, Wistar rats, weighing between 280 and 330g, developed tolerance to the anorectic effect of daily SR 141716A (2.5mgkg−1, i.p.) after only 3 days (Colombo et al., 1998), and obese mice treated orally with 3.0mgkg−1 of AM 251 developed tolerance to the anorectic effect of that compound after only 8 days (Hildebrandt et al., 2003).
Remarkably, in those studies and others (Ravinet-Trillou et al., 2003; Vickers et al., 2003), reductions in body weight relative to vehicle-treated animals continued to be significant throughout treatment, even though the daily food intakes of antagonist-treated animals were similar to those of vehicle-treated controls. Interestingly, tolerance also develops faster in lean, compared with obese, animals (Vickers et al., 2003). The observation that tolerance is linked to existing energy stores strongly suggests that peripheral factors are also involved in the development of resistance to the anorectic effect of CB1 receptor antagonists. Consistent with this hypothesis, activity at CB1 receptors increases de novo lipogenesis in the liver (Osei-Hyiaman et al., 2005) as well as in the adipose tissue (Cota et al., 2003; Ravinet-Trillou et al., 2004), possibly by inhibiting adenosine monophosphate-activated protein kinase (Kola et al., 2005), or through the induction of enzymes of the β-oxidation and tricarboxylic acid cycle (Jbilo et al., 2005). By inhibiting lipolysis, CB1 receptor antagonists could increase the number of calories required to replace and maintain existing energy stores, thereby creating conditions in which increases in food intake fail to produce additional growth. Increases in adiponectin mRNA, a circulating hormone that leads to weight loss through increased oxygen consumption (thermogenesis), could give rise to similar conditions (Bensaid et al., 2003; Liu et al., 2005).
Consistent with the hypothesis that CB1 receptor antagonists affect metabolic pathways in addition to the pathways that regulate feeding behaviour directly, treatment with SR 141716A (10mgkg−1), or genetic ablation of the CB1 receptor, also decreases feeding efficiency (Ravinet-Trillou et al., 2004; Jbilo et al., 2005). Feeding efficiency is a ratio that compares changes in body weight to the total number of calories ingested thereby providing an index of how well an organism is able to assimilate nutrients into body mass. Although it is likely that changes in the metabolism and energy expenditure also contributed to reductions in body weight in our study, the fact that feeding efficiency was only moderately reduced during both daily and 5-day dosing strategies suggests that the major effect of AM 251 on body weight in our study resulted from decreases in daily food intake. The reason that tolerance to the anorectic effect of AM 251 was not observed in our study was likely due to differences in the protocol. We employed a smaller dose of the CB1 receptor antagonist (1mgkg−1 per day) than is typically used in other studies (3—10mgkg−1 per day) (Colombo et al., 1998; Bensaid et al., 2003; Hildebrandt et al., 2003; Ravinet-Trillou et al., 2003; Vickers et al., 2003; Liu et al., 2005), and in relatively large rats (480—580g). Our results demonstrate that reductions in daily food intake by smaller, or less frequent, doses of AM 251 can be maintained in rats sustained on a highly palatable and nutrient-rich diet, possibly because of limited activity of the antagonist at CB1 receptors in the pathways directly affecting lipolysis and energy expenditure.
We also found that reductions in food intake by AM 251 could not be attributed to feelings of malaise as measured by a highly sensitive, two-flavour CTA procedure. Given the sustained effect a single dose of AM 251 had on food intake, and based on work carried out by others (De Vry et al., 2004b; McLaughlin et al., 2005), one might expect that rats repeatedly trained to associate the physiological effects of AM 251 with a specific flavour would avoid that flavour in a free-choice test trial. Instead, we found that rats were as likely to select flavours associated with AM 251 as they were flavours associated with vehicle treatment. We also found no difference in the way rats approached AM 251- and vehicle-associated flavours, based on the time each rat took to begin eating over four training and testing trials. The results demonstrate that the anorectic effect of AM 251 in our study was not associated with malaise as measured by CTA at doses 5mgkg−1, even though CTA can be induced by CB1 receptor antagonists when higher doses are used (De Vry et al., 2004b; McLaughlin et al., 2005).
To determine whether AM 251 was still acting on CB1 receptors in the brain over a 5-day period, we examined how long AM 251 antagonized the effects of THC using a thermoregulation assay, and found that AM 251 had no effect on hypothermia produced by an i.p. administration of the CB1 agonist given 48 and 96h later. CB1 receptors are densely distributed in the anterior nucleus of the hypothalamus or preoptic area (Herkenham et al., 1991; Moldrich & Wenger, 2000), and thermosensitive neurons in the preoptic area are thought to be critical for thermoregulation (Boulant, 1981; Boulant, 2000). Because tolerance to the hypothermic effects of peripherally administered THC can be induced with intrahypothalamic administrations of the CB1 agonist (Fitton & Pertwee, 1982), the major effect of CB1 agonists on thermoregulation is thought to be mediated by CB1 receptors in the hypothalamus. In our study, AM 251 had no significant impact on the effects of THC given 48h later. Therefore, it is likely that CB1 receptors in the central nervous system are no longer significantly antagonized by a 5mgkg−1 dose of AM 251 48h after administration.
The observation that intermittent administration of AM 251 inhibits daily food intake and weight gain at least as well as daily dosing regimes in rats may have significant clinical implications for the use of CB1 receptor antagonists in the treatment of human obesity. The advantages of an intermittent dosing strategy include targeting the peak anorectic effect of CB1 receptor antagonist to times in which people are likely to overeat, and limiting the pharmacological dose and blockade of the CB1 receptor to the shortest duration possible to limit unwanted side effects. It has long been known that people eat 8—13% more during weekends (St Jeor et al., 1983; de Castro, 2000). Data from our study suggest that weekly treatment with a CB1 receptor antagonist given just prior to the weekend could enhance the efficacy of these agents by targeting their peak anorectic effect to times in which people are likely to overeat. Our data also demonstrate that only intermittent blockade of the CB1 receptor is necessary to produce a prolonged inhibition of food intake. Mounting evidence suggests that the endogenous cannabinoid system plays important roles in gut function, suggesting that continued antagonism may lead to unwanted effects in this system. For example, daily administrations of CB1 receptor antagonists have been associated with symptoms of nausea and diarrhoea in humans (Van Gaal et al., 2005) and endogenous activity at CB1 receptors has been hypothesized to protect the gastrointestinal tract during inflammation as well as during abnormally high gastric and enteric secretions (Massa et al., 2004; Duncan et al., 2005). Minimizing the duration of antagonism at CB1 receptors with a once or twice a week dosing strategy may result in fewer side effects and better patient compliance. However, additional studies will be needed to confirm these findings under experimental circumstances more closely related to those used in the treatment of human patients.
In summary, the matching reductions in food intake and bodyweight in rats treated either daily or infrequently with AM 251 demonstrate that a relatively low dose of the receptor antagonist is capable of producing significant changes in energy balance that are resistant to the acquisition of tolerance. These reductions are independent from noxious effects as measured by CTA and last longer than the pharmacological blockade of CB1 receptors in the brain by AM 251.
Source, Graphs and Figures: AM 251 produces sustained reductions in food intake and body weight that are resistant to tolerance and conditioned taste aversion
