Untitled

Effects of rimonabant on behavior maintained by progressiveratio schedules of sucrose reinforcement in obeseZucker (fa/fa) ratsErin B. Rasmussen and Sally L. Huskinson This experiment reports on the ability of rimonabant to alter findings extend the literature that rimonabant reduces food the reinforcing properties of food in the genetically obese reinforcer efficacy, and suggest that obese Zuckers may Zucker (fa/fa) rat, a strain that exhibits higher levels exhibit a heightened sensitivity to rimonabant. The findings of endocannabinoids in brain regions that correspond also suggest that the effort required to obtain food to heightened food intake. We characterized food reinforcement may also play a role in the efficacy of reinforcement in obese and lean Zucker rats by placing rimonabant. Behavioural Pharmacology 19:735–742 behavior under progressive ratio schedules of sucrose 2008 Wolters Kluwer Health | Lippincott Williams reinforcement. Then, doses of rimonabant (1–10 mg/kg), a CB1 receptor antagonist, were administered. Obese Zuckers had slightly higher breakpoints for sucrose under Behavioural Pharmacology 2008, 19:735–742 baseline conditions compared with leans, and also Keywords: cannabinoids, food intake, food reinforcement, obesity, demonstrated significantly higher response rates than progressive ratio schedule, rat, rimonabant, sucrose, Zuckers (fa/fa) leans. Rimonabant dose-dependently decreasedbreakpoints and response rates for both groups, though Department of Psychology, Idaho State University, Idaho, USA only obese Zuckers demonstrated suppressed behaviorunder the 1 mg/kg dose. The 10 mg/kg dose of rimonabant Correspondence to Erin B. Rasmussen, PhD, Department of Psychology,Stop 8112, Idaho State University, Idaho 83204, USA reduced breakpoints equally for both groups (by about 60%). This dose of rimonabant also reduced food intake by 20% in lean Zuckers, and by 30% in obese Zuckers. These Received 13 March 2008 Accepted as revised 17 July 2008 palatable food is ingested (Melis et al., 2007), so The endocannabinoid neurotransmitter system plays a rimonabant may reduce motivation for food through this role in feeding and hyperphagia. Activation of the CB1 receptor by endogenous cannabinoids, such as 2-arachi-donyl glycerol (2-AG) or anandamide, or exogenous As is gleaned from the above studies, food intake is the compounds, like D-9 tetrahydrocannabinol, increases food most frequently used dependent variable for behavior intake (Drewnowski and Grinker, 1978; Williams et al., assessment of food reward. Food intake usually occurs 1998; Williams and Kirkham, 1999; Hao et al., 2000; in an environment where food is readily available, that DiMarzo et al., 2001; Kirkham and Williams, 2001; Berry is, available at a low response cost to the organism. For and Mechoulam, 2002; Jarrett et al., 2005). The afore- example, a rat may simply move toward a food aperture mentioned studies also demonstrate that the induction located within its home cage and eat from a large amount of eating occurs even when the organism is not food of freely available food. Although food intake is an deprived, suggesting that one behavioral mechanism informative dependent variable, it does little to assess affected by cannabinoid activity is enhancement of food the value of food at higher response costs, when the value of food may decrease (Hursh, 1984; Hursh et al., 1988).
SR141716, or rimonabant, is a cannabinoid antagonist The progressive ratio (PR) schedule of reinforcement is that blocks the CB1 receptor, thereby reducing food intake a procedure that shows the relation between response (Colombo et al., 1998; Simiand et al., 1998; McLaughlin effort and the value of a particular reinforcer, for example, et al., 2003; Vickers et al., 2003; Thornton-Jones et al., food or drugs of abuse (Hodos, 1961; see also Markou 2005; Carai et al., 2006; Gardner and Mallet, 2006; Herling et al., 1993; Stafford et al., 1998 for reviews). Under this et al., 2008; Serrano et al., 2008). This effect seems to be schedule, the initial response requirement for a reinforcer especially relevant to palatable foods, such as those that is low, then the ratio requirement increases systematically are fatty or sweet (Verty et al., 2004; Gessa et al., 2006; within a single session. The ‘breakpoint’ or strain, in Melis et al., 2007; Thornton-Jones et al., 2007). On a which the animal no longer responds, is the referent neural level, rimonabant reduces extracellular dopamine for the value of the reinforcer. Indeed, the PR schedule release in the reward areas of the brain of rodents when is used extensively in determining reinforcing efficacy c 2008 Wolters Kluwer Health | Lippincott Williams & Wilkins for drugs of abuse (see Spealman and Goldberg, 1978; hypothesized that this rat strain may exhibit altered Stafford et al., 1998 for reviews). However, its application behavioral sensitivity to rimonabant compared with lean to the conditions under which food functions as a controls. We also compared the effects of rimonabant in reinforcer, especially with food-related problems such as Zuckers using the standard dependent variable of food obesity, has been limited in recent times.
intake to determine whether this drug would reduce foodintake to the same extent as the effect on behavior Although many studies have shown how cannabinoids maintained by PR schedules of reinforcement.
influence food intake, fewer studies have examined howcannabinoids affect food reinforcement using operants that require larger response requirements. Two reports that used PR schedules, for example, confirmed that Nine male genetically obese Zucker (fa/fa) and nine male tetrahydrocannabinol increased breakpoints for food lean Zucker rats were purchased from Harlan at 4–5 under PR schedules in standard rat strains such as weeks of age, and singly housed in home cages with free Sprague–Dawley and Wistar (Higgs et al., 2005; Solinas access to standard food and water. The home cages were and Goldberg, 2005). In addition, rimonabant has been located in a temperature-controlled and humidity-con- shown to reduce breakpoints for food reinforcement in trolled room with a 12 :12 h light/dark cycle (08.00 h).
rats (Solinas and Goldberg, 2005) and mice (Ward and After 7–8 weeks of free feeding, lean controls weighed a Dykstra, 2005). These studies expand the food intake mean ( ± SEM) of 295.23 ( ± 13.23) g and obese (fa/fa) literature by providing a clearer characterization that food rats weighed a mean of 474.43 ( ± 10.34) g; these weights reinforcement may be a behavioral mechanism altered were significantly different [t(18) = – 10.68, P < 0.01].
At 12–13 weeks, subjects began a food restriction protocolin which they were allowed 2 h of free access to food atthe same time period in the early afternoon, followed Genetics may also influence the reinforcing properties of by 22 h without access to food (before an experimental food by way of the cannabinoid system. The obese Zucker session) to ensure that food would function as a rat, for example, exhibits leptin insensitivity. Leptin is reinforcer. After 2 weeks under this deprivation protocol, a peptide secreted by adipose tissue that, in a normal obese rats ate an average of 11.0 ( ± 0.64) g, or about 2.3% lean animal, blocks hunger signals, such as neuropeptide Y of their body weights, during the 2-h free-feed sessions; (Figlewicz, 2003; Sahu, 2004a, b). One manner by which leans ate an average of 7.2 ( ± 0.49) g, or about 2.4% of leptin lowers the reinforcing properties of food (Harrold their body weights. After this 2-week period, deprivation and Williams, 2003; Fulton et al., 2004; Jo et al., 2005) is by continued and all rats were trained to lever press for blocking receptors in reward areas of the brain (Figlewicz sucrose food pellets. Over the course of the experiment, et al., 2003) and by reducing endocannabinoid activity two rats (one lean and one obese) died, so eight rats from in the brain (DiMarzo et al., 2001). The Zucker rat, in each group completed the study. All rats had previous contrast, is missing the leptin receptor. As a result, this exposure to acute doses of 2-AG (0.03–3 mg/kg), an strain eats approximately three times more food than endocannabinoid, though these data will not be pre- control wild types (e.g. Zucker and Zucker, 1962; DiMarzo et al., 2001) and exhibits an obese phenotype.
Moreover, the obese Zucker has been shown to have higher 2-AG (an endocannabinoid) levels in areas of the Seven Coulbourn Habitest standard (Coulbourn Instru- brain that are relevant to food intake (Wenger and ments, Whitehall, PA, USA) rat experimental chambers Moldrich, 2002; Harrold and Williams, 2003; Jo et al., were used for data collection. Each chamber was 2005), and these higher levels seem to be mediated by an equipped with two levers on one panel situated from insensitivity to leptin (DiMarzo et al., 2001). Few studies, the bottom of a grid floor. When response criteria were have examined how these physiological changes manifest met, a cue light above the lever, and one in the collection in behavioral choice that go beyond food intake as a area (magazine) illuminated for 5 s as a 45-mg sucrose dependent variable. One exception, however, is found in pellet was delivered. During this reinforcer interval, lever a study by Glass et al. (1999), who showed that naloxone, presses has no scheduled consequences. A 28-v house- a drug that blocks the opioid receptor (which is light was situated 13 cm above the food dispenser.
also implicated in food reinforcement), decreases the A speaker placed in the upper left corner of the left reinforcing properties of food under PR schedules in wall of the chamber generated white noise. A 200 Â 200 fan was situated in the upper right corner of the left wall.
A sound-attenuating cubicle surrounded each chamber.
This study examined the degree to which the cannabi- Graphic State software (Coulbourn Instruments, White- noid antagonist rimonabant affected food reinforcer hall, PA, USA) on an IBM compatible computer, with efficacy in obese Zucker rats. As the obese Zucker 0.0100 resolution, controlled all reinforcement contingen- strain exhibits higher endocannabinoid levels, it was cies and data collection. Computers and software were Zuckers and cannabinoids Rasmussen and Huskinson stationed in a room separate from the chambers. Sessions dose–response determination. Here, lever presses pro- were conducted in the mornings at the same time duced environmental conditions that were identical to (± 15 min) from Monday to Thursday.
the PR sessions, except that no food delivery occurred.
After behavior stabilized across three sessions of extinc- tion, the 10 mg/kg dose of rimonabant for each rat was To train lever pressing, each subject was placed in administered 1 h before a subsequent extinction session.
an experimental chamber for a 3-h session, in which This was done to ensure that changes in responding responses on the right lever were reinforced under a observed in the food-delivery condition during the drug fixed ratio 1 (FR1) schedule of reinforcement. Lever phase were a result of motivation for food rather than pressing was considered trained when 90 reinforcers were another mechanism, for example, motor effect. All earned in a session. If this criterion was not met within experimental procedures were approved and in compli- six sessions, the rat was hand shaped, or reinforced for ance with the Idaho State University’s Institution for successive approximations until the lever press was sufficiently trained. Of the 18 rats, three lean Zuckersand one obese Zucker were hand shaped.
Food intakeFood intake in the home cage during the 2-h free-feed Lever pressing was placed under a PR schedule of sucrose sessions was monitored throughout the experiment. The reinforcement 2 days/week, for example, Tuesdays and amount of food placed in the home cage was measured Thursdays. Under the PR schedule, a single lever press before a 2-h session and then compared with the amount produced a food pellet, and the response requirement of food that was left after the 2-h session.
increased within the experimental session by an exponentof 0.2 multiplied by the number of reinforcers earned in the session to that point and rounded to the nearest Rimonabant (National Institute of Mental Health integer (Roberts and Bennett, 1993). This resulted in the Chemical Synthesis and Drug Supply Program) was following ratio steps: 1, 2, 4, 6, 9, 12, 15, 20, 25, 32, 40, 50, dissolved in a 1 : 1 : 18 ethanol (Sigma, Sigma-Aldrich 62, 77, and 95. When a ratio was not completed within Co., St Louis, MO, USA), Cremaphor (Sigma), and saline 20 min, the session ended. The breakpoint was the main solution (1 ml/kg) and was administered by intraperito- dependent variable of interest. A breakpoint baseline was neal injection 1 h before the start of PR sessions. A saline determined for each rat before the administration of each vehicle (1 ml/kg) was administered intraperitoneally drug by averaging the behavior across the last three stable before the beginning of some PR sessions.
sessions. Stability was defined as at least three sessionsin which the breakpoint did not deviate by more than two steps (response requirements) and there were no trends.
Dependent variables included mean breakpoints andresponse rates per session, as well as mean food intake An FR1 schedule was placed in effect on days before PR in grams during the 2-h free-feed session. Response rates schedules (e.g. Monday and Wednesday) and operated as were determined by examining only the number of a maintenance schedule. Here, every response resulted in responses per session that counted toward the response sucrose pellet. The session terminated after 25 reinfor- ratios, that is, responses made during the reinforcement interval were subtracted from the total number ofresponses across the session. The amount of time during Once baseline data were collected, rats received vehicle the delivery of reinforcers was subtracted from the total (saline) injections 1 h before PR sessions until behavioral session time. For example, if 12 pellets were earned, and stability was reached. Next, acute doses of rimonabant the reinforcement interval was 5 s, then 60 s was (dosing range: 1–10 mg/kg) were administered to each subtracted from the session duration (e.g. 45 min), such rat 1 h before the beginning of a PR session. Doses were that the net session time was 44 min. The number of given in ascending order in half-log unit increases. If responses made during the session was then divided by a particular dose reduced behavior by greater than 70% for a particular animal, the next higher dose was notattempted, to reduce the chances of overdose. No fewer A two-way analysis of variance (ANOVA) with repeated than 2 days separated injections. No injections were measures (SPSS, version 14.0, SPSS, Inc., Chicago, IL, USA) was used to analyze the data, with obese versuslean Zuckers as a between-subjects variable and dose of drug as a within-subjects variable. For each rat, the last To ensure the food pellet maintained behavior, each three baseline sessions (no vehicle administrations) were subject was exposed to a no-food delivery, or extinc- averaged into a single mean; the same process was applied tion condition, following the completed rimonabant to the vehicle (0 mg/kg) data. In other words, for each rat, a single datum represented each within-subject condi- tion (control, vehicle, and each dose of rimonabant). Inaddition, the food-delivery versus no-food delivery condition was analyzed using a repeated measures ANOVA, with food-delivery versus no-food delivery as a between-subjects variable and drug condition (baselinevs. 10 mg/kg rimonabant) as a within-subjects variable.
Effect sizes, as reported by partial eta squared (Z2p), are included. The traditional P value is used to reflect statistical significance levels, though they are often limited in that they do not describe effect sizes. Partial Z2 values refer to the proportion of variance caused bythe effect, or independent variable, which can extend information provided by a P value (Neter et al., 1996).
Partial Z2 provides similar information to an r2 value in Figure 1 shows the mean number of sessions to acquire the lever press, including data for the four animals that required hand shaping (these four animals were treated as though they completed six sessions). The lean Zuckers acquired the lever press in an average of 5.3 ( ± 0.95) sessions and the obese Zuckers acquired it in 3.9 ( ± 1.1) sessions [t(18) = 3.05, P < 0.01]. When the four animals that required hand shaping were removed from theanalysis, the result did not change [lean: 5.0 ± 0.38; obese: 3.67 ± 0.29; t(14) = 2.86, P < 0.01].
The upper panel of Fig. 2 shows mean breakpoints as a function of dose of rimonabant. The baseline and vehicle(0 mg/kg) means were compared using a two-way ANOVA with repeated measures, which yielded no differences Mean ( ± SEM) breakpoint (upper panel) and response rate per session(lower panel) as a function of dose of rimonabant. *P < 0.05 difference between lean and obese Zucker rats; **P < 0.05 difference betweenvehicle and dose of rimonabant for obese Zuckers only; ***P < 0.05difference between vehicle and dose of rimonabant for lean and obese between baseline and vehicle conditions (P = 0.52).
However, there was a marginal group difference in thesedata [F(1,17) = 3.83, P = 0.07, Z2p = 0.19], but no signi- Rimonabant dose-dependently reduced breakpoints forlean and obese Zuckers [F(3,42) = 25.98, P < 0.01, Z2p = 0.65). A significant main effect of group [F(1,14) = 4.9, P < 0.05, Z2p = 0.26] was observed, but no significant contrasts were conducted on vehicle versus each dose Mean ( ± SEM) number of sessions to acquire the lever-press responsein lean (black) and obese (grey) Zucker rats. **P < 0.01.
of rimonabant for each group. The 1 mg/kg dose ofrimonabant reduced mean breakpoints by 28% for the Zuckers and cannabinoids Rasmussen and Huskinson obese Zuckers, and this was statistically significant Figure 3 summarizes effects on breakpoint under food- [F(1,7) = 15.51, P < 0.01]. This same dose reduced delivery (PR) and no-food delivery (extinction) condi- breakpoints for the leans by only 12%, and this was tions. The left half of Fig. 3 shows baseline and 10 mg/kg not statistically significant (P = 0.24). The 3 and 10 mg/kg data from Fig. 2. The 10 mg/kg dose of rimonabant doses, however, reduced breakpoints significantly for reduced breakpoints equally (about 58% for lean controls the lean rats [3 mg/kg F(1,7) = 20.79, P < 0.01; 10 mg/kg: and obese Zuckers. The right half of Fig. 3 shows mean F(1,7) = 15.04, P < 0.01] and obese rats [F(1,7) = 12.76, breakpoint under the no-food delivery condition. Extinc- P < 0.01; F(1,7) = 27.73, P < 0.01]. Lean and obese rats’ tion reduced the mean breakpoint (from PR) to 9.33 breakpoints did not differ significantly at any dose ( ± 0.78) and 11.25 ( ± 1.1) for the lean and obese Zuckers, respectively, and this reduction was significant.
[F(1,15) = 76.54, P < 0.01, Z2p = 0.84]. The 10 mg/kg The lower panel of Fig. 2 shows mean response rates as dose of rimonabant significantly reduced behavior under a function of dose of rimonabant. Baseline and vehicle extinction for the lean (3.11 ± 0.94) and obese Zuckers conditions did not differ significantly (P = 0.76), though (6.38 ± 1.81) [F(1,15) = 22.88, P < 0.01, Zp = 0.60). In there was a significant group difference [F(1,14) = 8.1, addition, there was a significant main effect of group [F(1,15) = 4.56, P < 0.05, Zp = 0.23), but no significant decreased response rate [F(4,36) = 17.96, P < 0.01, interaction (P = 0.57). Similar effects were found with response rate, and these data are summarized in Table 1.
p = 0.66). In addition, there was a significant main effect of group [F(1,9) = 7.13, P < 0.02, Z2p = 0.44) andan interaction [F(4,36) = 3.04, P < 0.05, Z2p = 0.25].
Figure 4 shows food intake (g) in the 2-h free-feed Vehicle-dose contrasts revealed that the 1 mg/kg of session after rimonabant was administered. Rimonabant rimonabant reduced response rates for obese Zuckerssignificantly, by about 28% [F(1,7) = 159.37, P < 0.01], Response rates (responses/minute) for lean and obese but only reduced lean Zucker rates by about 14.5% Zucker rats under food-delivery (PR) and no-food delivery (P = 0.32). The 3 mg/kg dose significantly reduced rates in lean rats [F(1,7) = 7.3, P < 0.03] and obese rats [F(1,7) = 41.02, P < 0.01]. The 10 mg/kg dose reduced both groups’ rates significantly [lean F(1,7) = 15.79, P < 0.01; obese F(1, 7) = 5.83, P < 0.01]. Lean and obese rats’ rates did not differ at any dose of rimonabant Mean response rates (SEM) under baseline and the 10 mg/kg of rimonabantare shown.
a 10 mg/kg reduced breakpoints significantly for both groups (see text for details).
bResponse rate under PR vs. extinction: P < 0.01 [F(1,14) = 79.93, Z2p = 0.85; nomain effect of group or interaction (P > 0.16).
cResponse rate under extinction vs. 10 mg/kg dose of rimonabant underextinction: P < 0.01 [F(1,14) = 48.35, Z2p = 0.76]; no main effects of group or Mean ( ± SEM) breakpoint in food-delivery (left) and no-food delivery(right) conditions comparing behavior under baseline and 10 mg/kg of rimonabant. *P < 0.01 for progressive ratio (PR) baseline vs. 10 mg/kg,**P < 0.01 for PR vs. extinction, ***P < 0.01 for extinction vs. 10 mg/kg Mean ( ± SEM) food intake (g) as a function of dose of rimonabant.
dose-dependently decreased food intake for lean and Rimonabant dose-dependently reduced breakpoints and obese Zuckers [F(3,39) = 9.29, P < 0.01, Z2p = 0.42). The response rates for all rats, and the effect sizes were 10 mg/kg dose reduced food intake by about 20% (from a substantial (0.65 or greater). This is consistent with what mean of 9.9 under vehicle to 7.94) for the lean rats and others have found in standard rat strains (Solinas and by over 30% (from 11.1 to 7.65) for the obese Zuckers.
Goldberg, 2005). Despite the food reinforcement-sup- No significant main effect of group or a group  dose pressing effects of rimonabant, obese Zuckers had higher interaction (P > 0.15) was observed. Vehicle-dose con- breakpoints and response rates for sucrose pellets than trasts revealed a significant reduction from baseline in the lean controls across all doses, and this difference was 1 mg/kg [F(1,7) = 11.66, P < 0.01], 3 mg/kg [F(1,7) = significant for rate. Moreover, between 25 and 66% of 9.19, P < 0.01], and 10 mg/kg doses [F(1,7) = 128.48, the variance in the data was accounted for by group. This P < 0.01] for the obese rats. Only the 10 mg/kg dose study is the first to show that rimonabant reduced the reduced food intake significantly for the lean rats reinforcing properties of food in the obese Zucker rat, which is noteworthy because Zuckers have alteredendocannabinoid laboratory rat strains (DiMarzo et al., 2001). In addition, the 1 mg/kg dose of rimonabant reduced breakpoints and In this study, food-deprived lean and obese Zucker rats response rates by almost 30% for the obese Zuckers, and responded under a PR schedule for sucrose pellets.
did not reduce behavior significantly for the leans.
During training, obese Zucker rats acquired lever pressing Indeed, a three-fold increase in the drug dose was for sucrose pellets in fewer sessions compared with lean necessary to cause a significant decrease in the lean rats.
controls and more leans required hand shaping compared The significant reduction for Zuckers at the 1 mg/kg dose with obese Zuckers (3 vs. 1). This finding, to our may be due to an endocannabinoid-related sensitivity to knowledge, has not been reported in the literature with rimonabant. It is, however, important to note that obese Zucker rats. The mechanism for this finding is unclear; it Zuckers did exhibit higher response rates and breakpoints may reflect genetic differences in learning, for example.
under baseline conditions, so the 1 mg/kg effect may be We believe, however, in light of other effects to be due to a rate-dependent effect (Dews, 1955) or may be discussed, the mechanism may likely be due to differ- confounded with the number of contacts made with ences in sensitivity to food reinforcement, with obese the reinforcer (Nevin et al., 1983). A study that holds Zucker rats showing a higher sensitivity than lean response rate and the number of reinforcers constant between the two groups may be able to elucidatethe mechanism involved in the strain difference at the The baseline data under PR schedules suggest that obese Zuckers had higher, though not significantly higher,breakpoints for sucrose reinforcement. This finding The no-food delivery condition (extinction) confirmed supports an observation reported by Glass et al. (1999), that the lever press was maintained by the sucrose pellet.
who also found that lean and obese Zucker rats did not Extinction reduced breakpoints for sucrose pellets in differ significantly with regard to motivation for sucrose the lean and obese Zucker rats by about 70% for both under PR schedules. It is noteworthy to mention, groups, and accounted for 84% of the variance between however, that response rates under PRs in this study conditions. A group difference was found in which obese differed between groups – obese Zuckers had signifi- Zuckers responded more under extinction than controls.
cantly higher response rates compared to leans. It may This may have resulted from the degree of conditioning be the case that response rate is a more sensitive measure to cues that accompanied pellet delivery during PR than breakpoints. It is noteworthy, for example, that in sessions. Although there was no pellet delivery during this study, the mean breakpoint averaged between 25 and the 5 s reinforcer interval throughout the extinction condi- 40 across both groups. The relevant steps of the PR tion, the cues associated with pellet delivery (e.g. the cue sequence progressed from 25 to 32 to 40. Consider two light) were presented. As a food pellet was a more potent animals that exhibited ratio strain between 32 and 40.
reinforcer for the obese Zuckers, it may have conditioned Both would be assigned a breakpoint of 32. However, if more strongly to these cues (Rescorla and Wagner, 1972).
the first of those rats stopped responding immediately This may create a situation in which the cues were after the reinforcer was delivered, its response rate would stronger conditioned reinforcers for the obese Zuckers, be lower than the second rat, who may have ceased thereby resulting in higher breakpoints and response responding just before the next pellet delivery (just shy of the 40-response requirement). As there is a potentialfor more variability within each ratio step, response rate,as opposed to breakpoint, may be a more sensitive Rimonabant also significantly reduced behavior in the measure, and thus better able to reveal a group difference no-food delivery condition. This suggests that another mechanism unrelated to food directly may also be Zuckers and cannabinoids Rasmussen and Huskinson possible, for example, a motor effect. A more likely reason intake may be the nature of the food. Some studies may have to do with the properties of food that were suggest that palatable food is more sensitive to effects conditioned to the cues associated with food delivery.
of rimonabant compared with less palatable food (e.g.
In this instance, rimonabant may have suppressed the Arnone et al., 1997; Ward and Dykstra, 2005). This reinforcing efficacy of these conditioned reinforcers assertion, however, is confounded by the number of times during extinction. Wickelgreen (1997), for example, an animal makes contact with palatable versus less pala- showed that after repeated presentations, stimuli paired table food during baseline conditions (Thornton-Jones with food can activate the mesolimbic dopamine system et al., 2007 for a discussion on this topic). Moreover, more than consumption of food itself. If that is the case some studies have reported little difference in the ability here, rimonabant may act to reduce sensitivity to stimuli of rimonabant to reduce food intake in foods that differ associated with food. Regardless of the mechanism of in palatability (Verty et al., 2004). Therefore, this issue is action of rimonabant during extinction, it is clear the still unclear. We offer a third reason: the type of response majority of the drug-induced reduction in breakpoint output that is required to gain access to food in each task occurred in the food-delivery condition which suggests differs. With food intake, the rat only needed to move that the drug’s primary mechanism (in this context) is toward the food aperture once and was able to eat as reduction of the reinforcing properties of food.
much as possible. Under the PR schedule, the rat wasrequired to move toward the lever and emit a series Rimonabant also dose-dependently reduced food intake of lever presses to gain access to a single 45 mg pellet.
in both lean and obese Zuckers, and this finding Indeed, the rat had to repeat this process for each pellet, replicates what others have found in standard rat strains and the requirement increased with each reinforcer.
(e.g. Colombo et al., 1998; Simiand et al., 1998; Hence, the response to reinforcer ratio in the PR task was McLaughlin et al., 2003; Vickers et al., 2003; Carai et al., greater (and increased throughout the session) compared 2006; Gardner and Mallet, 2006; Herling et al., 2008; with the food intake task, in which the response cost was Serrano et al., 2008). The 1 and 3 mg/kg doses of small and the potential reinforcer was large. It would rimonabant did not significantly reduce food intake seem then, that effects of rimonabant may be greatest in the lean rats, but those doses did reduce food intake in when the response to reinforcer ratio is larger. Future the obese Zuckers. Moreover, the 10 mg/kg dose of research may focus on better characterizing this rela- rimonabant only reduced food intake by 20% for leans and tionship by examining food access across a range of just over 30% for obese, suggesting that the obese costs (small to large) that vary from session to session Zuckers showed stronger effects to the 10 mg/kg dose.
and hold the amount of food in each ‘eating bout’ Vickers et al. (2003) also showed that a 10 mg/kg dose of constant. The behavioral economic approach, which rimonabant reduced food intake in free-feeding Zuckers characterizes reinforcer efficacy by examining the number by 45% in obese Zuckers, but only 25% in the leans.
of reinforcers earned at different unit prices (e.g. 1–100 Hence, our data are consistent with this study in showing lever presses for access to 1–5 food pellets) between differences in Zuckers in terms of the effects of sessions, may be a good place to begin. This would allow for careful control of response requirement, reinforceramount, as well as palatability of the reinforcer.
The effects of rimonabant at the 10 mg/kg dose weresmaller for food intake (20–30% reduction; effects size0.42) compared with the PR task (B60% reduction; 0.65 In summary, rimonabant dose-dependently reduced effect size). Although it may be unfair to compare food breakpoints for obese and lean Zucker rats. Obese Zucker intake with PR performance because they are qualitatively rats exhibited effects of rimonabant at doses lower different types of tasks, a brief discussion seems necessary.
than lean Zuckers with regard to food reinforcer efficacy One reason for the discrepancy may have to do with the under a PR schedule, as well as food intake. Effects of observation that food intake was measured after operant rimonabant were also revealed in the extinction condi- sessions were conducted, so motivation for food may have tion, by possible disruption of conditioned reinforcement been weaker during food intake assessment. However, the associated with cues paired with food. These data have typical number of food pellets earned by lean and obese implications for rimonabant as an antiobesity treatment.
Zucker rats was 10–12 per session. Ten 45-mg food pellets Some individuals who may exhibit genetically induced would equal 450 mg. The Zuckers ate a range of 8–15 g or environmentally induced leptin insensitivity (Sahu, of food during 2 h free-feed sessions, so the 450 mg of food 2004a, b), a common observation with obese individuals, probably did not subtract much from the reinforcing may exhibit a sensitivity to this drug, though the mecha- properties of food during these 2 h free-feed sessions.
nism for this sensitivity (e.g. leptin-mediated endocan-nabinoid sensitivies, higher contacts with food, etc.) is A second reason why rimonabant may be more effective unclear at this point. Further research on interactions for food gained under PR schedules versus free food of rimonabant with behavior is needed.
Kirkham T, Williams C (2001). Endogenous cannabinoids and appetite. Nutr Res This research was supported by a grant from the Markou A, Weiss F, Gold LH, Caine SB, Schulteis G, Koob GF (1993). Animal WeLEAD project which is funded by NSF Grant models of drug craving. Psychopharmacology (Berl) 112:163–182.
# SBE-0620073, Idaho State University, Pocatello, ID.
McLaughlin PJ, Winston K, Swezey L, Wisniecki A, Aberman J, Tardif DJ, et al.
(2003). The cannabinoid CB1 antagonists SR141716A and AM 251suppress food intake and food-reinforced behavior in a variety of tasks inrats. Behav Pharmacol 14:583–588.
Melis T, Succu S, Sanna F, Boi A, Argiolas A, Melis MR (2007). The cannabinoid Arnone M, Maruani J, Chaperon F, Thiebot M, Poncet M, Soubrie P, et al. (1997).
antagonist SR141716A (rimonabant) reduces the increase of extra-cellular Selective inhibition of sucrose and ethanol intake by SR141716, an dopamine release in the rat nucleus accumbens induced by a novel high antagonist of central cannabinoid (CB1) receptors. Psychopharmacology palatable food. Neurosci Lett 419:231–235.
Neter J, Kutner M, Nachtsheim C, Wasserman W (1996). Applied linear statistical Berry E, Mechoulam R (2002). Tetrahydrocannibol and endocannabinoids in feeding and appetite. Pharmacol Therapeu 95:185–190.
Nevin J, Mandell C, Atak J (1983). The analysis of behavioral momentum. J Exper Carai MA, Colombo G, Maccioni P, Gessa GL (2006). Efficacy of rimonabant and other cannabinoid CB1 receptor antagonists in reducing food intake and Rescorla R, Wagner AR (1972). A theory of Pavlovian conditioning: variations in body weight: preclinical and clinical data. CNS Drug Rev 12:91–99.
the effectiveness of reinforcement and nonreinforcement. In: Black AH, Colombo G, Agabio R, Diaz G, Lobina C, Reali R, Gessa GL (1998). Appetite Porkasy WF, editors. Classical conditioning II: current research and theory.
suppression and weight loss after the cannabinoid antagonist SR141716.
Roberts DC, Bennett SA (1993). Heroin self-administration in rats under a Dews P (1955). Studies on behavior I: differential sensitivity to pentobarbital of progressive ratio schedule. Psychopharmacology (Berl) 111:215–218.
pecking performance in pigeons depending on the schedule of reward.
Sahu A (2004a). Leptin signaling in the hypothalamus: emphasis on energy J Pharmacol Exp Therapeutics 113:393–401.
homeostasis and leptin resistance. Front Neuroendrinol 24:225–253.
DiMarzo V, Goparaju S, Wang L, Liu J, Batkal S, Jaral Z, et al. (2001). Leptin Sahu A (2004b). Mini review: a hypothalamic role in energy balance with special regulated endocannabinoids are involved in maintaining food intake. Nature emphasis on leptin. Endocrinology 145:2613–2620.
Serrano A, Del Arco I, Javier Pavon F, Macias M, Perez-Valero V, Rodriguez de Drewnowski A, Grinker JA (1978). Temporal effects of delta 9-tetrahydrocanna- Fonseca F (2008). The cannabinoid CB1 receptor antagonist SR141716A binol on feeding patterns and activity of obese and lean Zucker rats. Behav (rimonabant) enhances the metabolic benefits of long-term treatment with oleoylethanolamide in Zucker rats. Neuropharmacology 54:226–234.
Figlewicz D (2003). Adiposity and food reward: expanding the CNS roles Simiand J, Keane M, Keane P, Soubrie P (1998). SR141716, a CB1 cannabinoid of insulin and leptin. Am J Physiol Regul Integr Comp Physiol 284: receptor antagonist, selectively reduces sweet food intake in marmoset.
Figlewicz D, Evans S, Murphy J, Hoen M, Baskin D (2003). Expression of Solinas M, Goldberg S (2005). Motivational effects of cannabinoids and opioids receptors for insulin and leptin in the ventral tegmental area/substantia nigra on food reinforcement depend on simutaneous activation of cannabinoid and (VTA/SN) of the rat. Brain Res 964:107–115.
opioid systems. Neuropsychopharmacology 11:2035–2045.
Fulton S, Richard D, Woodside B, Sizgal P (2004). Food restriction and leptin Spealman RD, Goldberg SR (1978). Drug self-administration by laboratory impact brain reward circuitry in lean and obese Zucker rats. Behav Brain Res animals: control by schedules of reinforcement. Annu Rev Pharmacol Toxicol Gardner A, Mallet PE (2006). Suppression of feeding, drinking, and locomotion Stafford D, LeSage MG, Glowa JR (1998). Progressive-ratio schedules of drug by a putative cannabinoid receptor ‘silent antagonist’. Eur J Pharmacol delivery in the analysis of drug self-administration: a review. Psychopharma- Gessa GL, Orru A, Lai P, Maccioni P, Lecca R, Lobina C, et al. (2006). Lack of Thornton-Jones ZD, Vickers SP, Clifton PG (2005). The cannabinoid CB1 tolerance to the suppressing effect of rimonabant on chocolate intake in rats.
receptor antagonist SR141716A reduces appetitive and consummatory Psychopharmacology (Berl) 185:248–254.
responses for food. Psychopharmacology (Berl) 179:452–460.
Glass MJ, O’Hare E, Cleary JP, Billington CJ, Levine AS (1999). The effect of Thornton-Jones ZD, Kennett GA, Vickers SP, Clifton PG (2007). A comparison of naloxone on food-motivated behavior in the obese Zucker rat. Psychopharma- the effects of the CB(1) receptor antagonist SR141716A, pre-feeding and changed palatability on the microstructure of ingestive behaviour. Psycho- Hao S, Avraham Y, Mechoulam R, Berry E (2000). Low dose anandamide affects food intake, cognitive function, neurotransmitter and corticosterone levels in Verty AN, McGregor IS, Mallet PE (2004). Consumption of high carbohydrate, diet restricted mice. Eur J Pharmacol 392:147–156.
high fat, and normal chow is equally suppressed by a cannabinoid receptor Harrold J, Williams G (2003). The cannabinoid system: a role in both the antagonist in non-deprived rats. Neurosci Lett 354:217–220.
homeostatic and hedonic control of eating? Br J Nutr 90:729–734.
Vickers SP, Webster LJ, Wyatt A, Dourish CT, Kennett GA (2003). Preferential Herling AW, Kilp S, Elvert R, Haschke G, Kramer W (2008). Increased energy effects of the cannabinoid CB-sub-1 receptor antagonist, SR141716, on expenditure contributes more to the body weight reducing effect of food intake and body weight gain of obese (fa/fa) compared to lean Zucker rimonabant than reduced food intake in candy-fed Wistar rats. Endocrinology.
rats. Psychopharmacology 167:103–111.
Ward SJ, Dykstra LA (2005). The role of CB1 receptors in sweet versus fat Higgs S, Barber D, Cooper A, Terry P (2005). Differential effects of two reinforcement: effect of CB1 receptor deletion, CB1 receptor antagonism cannabinoid receptor agonists on progressive ratio responding for food and (SR141716A) and CB1 receptor agonism (CP-55940). Behav Pharmacol free-feeding in rats. Behav Pharmacol 16:389–393.
Hodos W (1961). Progressive ratio as a measure of reward strengh. Science Wenger T, Moldrich G (2002). The role of endocannabinoids in the hypothalamic regulation of visceral function. Prostaglandins Leukot Essent Fatty Acids Hursh S (1984). Behavioral economics. J Exp Anal Behav 42:435–452.
Hursh S, Raslear T, Shurtleff D, Bauman R (1988). A cost-benefit analysis of Wickelgreen I (1997). Getting the brain’s attention. Science 278:35–37.
demand for food. J Exp Anal Behav 50:419–440.
Williams C, Kirkham T (1999). Anandamide induces overeating: mediation by Jarrett MM, Limebeer CL, Parker LA (2005). Effect of delta9-tetrahydrocanna- central cannabinoid receptors. Psychopharmacology 143:315–317.
binol on sucrose palatability as measured by the taste reactivity test. Physiol Williams C, Rogers P, Kirkham T (1998). Hyperphagia in pre-fed rats following ordal delta-9 THC. Physiol Behav 65:343–346.
Jo Y, Chen Y, Chua S, Talmage D, Role L (2005). Integration on endocannabinoid Zucker LM, Zucker TF (1962). Fatty, a new mutation in the rat. J Hered 52: and leptin signaling in an appetite-related neural circuit. Neuron 48:

Source: http://www.isu.edu/psych/Articles/Rasmussen/Rasmussen%20&%20Huskinson%202008.pdf