Untitled

Differential response to a selective cannabinoid receptorantagonist (SR141716: rimonabant) in female mice from linesselectively bred for high voluntary wheel-running behaviourBrooke K. Keeneya, David A. Raichlenb, Thomas H. Meeka,Rashmi S. Wijeratnea, Kevin M. Middletona,*, Gregory L. Gerdemanb,wand Theodore Garland Jra Exercise is a naturally rewarding behaviour in human not time) as compared with control females. We conclude beings and can be associated with feelings of euphoria and that altered endocannabinoid signalling plays a role in the analgesia. The endocannabinoid system may play a role in high wheel running of female HR mice. Behavioural the perception of neurobiological rewards during and after Pharmacology 19:812–820 c 2008 Wolters Kluwer Health prolonged exercise. Mice from lines that have been Lippincott Williams & Wilkins.
selectively bred for high voluntary wheel running (high runner or HR lines) may have evolved neurobiological Behavioural Pharmacology 2008, 19:812–820 mechanisms that increase the incentive salience of Keywords: artificial selection, endocannabinoids, exercise, endurance-type exercise. Here, we test the hypothesis that experimental evolution, genetics, hyperactivity, locomotor activity, mouse, endocannabinoid signalling has been altered in the four rimonabant, sex differences, wheel running replicate HR lines as compared with four nonselected aUniversity of California, Riverside, California and bUniversity of Arizona, Tucson, control lines. After 18 days of acclimation to cages with attached wheels, we injected mice with rimonabant(SR141716), a selective cannabinoid CB1 receptor Correspondence to Dr Theodore Garland Jr, Department of Biology, University of antagonist. During the time of normal peak running, each California, Riverside, CA 92521, USAE-mail: [email protected] mouse received, in a randomized order, intraperitonial injection of rimonabant (0.1 or 3.0 mg/kg) or vehicle, over 9 *Present address: California State University San Bernardino, San Bernardino, days. Drug response was quantified as wheel revolutions, wPresent address: Eckerd College, St. Petersburg, Florida, USA time and speed 10–70 min postinjection. Rimonabant decreased running in all mice; however, female HR mice Received 7 April 2008 Accepted as revised 24 September 2008 differentially decreased running speed and distance (but Demuth and Molleman, 2006). There are two primary Voluntary wheel running is a classically self-rewarding cannabinoid receptors: CB1 and CB2. The ECS is behaviour in both rats and mice (Premack, 1964; hypothesized to have a general modulatory effect on Timberlake and Wozny, 1979; Sherwin, 1998). For circuits of the reward system, and the perception of example, Sherwin and Nicol, (1996) have demonstrated neurobiological rewards associated with such behaviours that mice are willing to cross an aversive water barrier, as voluntary locomotion and food consumption (Thornton- even as the barrier was increased in size, to receive Jones et al., 2005; Maldonado et al., 2006). The ECS is also a wheel-running reward. Moreover, several operant involved in aspects of energy balance, lipid metabolism, conditioning studies have shown that rats and mice are nociception and the stress response, among other factors, highly motivated to bar-press for a wheel-running reward which are also relevant to the physiology of wheel running (Belke and Heyman, 1994; Belke, 1996; Belke and (Girard and Garland, 2002; Li et al., 2004; Rhodes et al., Garland, 2007). However, our current understanding of 2005; Pacher et al., 2006; Pagotto et al., 2006; Malisch et al., the neurobiological basis of apparently high motivation for wheel running in rodents is limited.
Endocannabinoid signalling is activated by aerobic The endocannabinoid system (ECS) is a complex exercise in human beings (Sparling et al., 2003), and is endogenous signalling system made up of transmembrane associated with analgesia and the stimulation of locomotor cannabinoid receptors (CB receptors), their ligands activity in rodents (Lichtman et al., 1996; Wiley, 2003; Hohmann and Suplita, 2006). Recent evidence indicates that CB1 signalling facilitates dopamine release in the (De Petrocellis et al., 2004; Cota and Woods, 2005; shell of the nucleus accumbens, a neurochemical effect c 2008 Wolters Kluwer Health Lippincott Williams & Wilkins Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Rimonabant affects wheel running in mice Keeney et al.
common to drugs of addiction (Kelley, 2004; Cheer et al., blockers found differential effects on wheel running in 2007), and the expression of drug-seeking, conditioned HR and control mice (Rhodes et al., 2001; Rhodes and behaviours (Cohen et al., 2005; DeVries and Schoffelmeer, Garland 2003). This differential effect was attributed to 2005; Laviolette and Grace, 2006; Xi et al., 2006; Alvarez- altered functionality in the D1 receptor system, although Jaimes et al., 2008). Dietrich and McDaniel (2004) apparently not in the D2 receptor, serotonergic or recently suggested that CB1 signalling may account for opioidergic systems (Rhodes et al., 2001, 2003, 2005; exercise addiction that has been reported for human Li et al., 2004). In addition, Fos immunohistochemistry distance runners (Morgan, 1979). When injected directly studies show a greater increase in activity in several brain into brain reward centres of the rat, low doses of cannabinoid regions implicated in reward and motivation when wheel agonists enhance locomotor behaviours and promote self- access is blocked, including the caudate–putamen administration and conditioned place preference (Zangen complex, lateral hypothalamus, prefrontal cortex, medial et al., 2006), consistent with the idea that motivational frontal cortex, NAc, piriform cortex and sensory cortex properties of the ECS may be physiologically coupled to (Rhodes et al., 2003). These differences prompt numer- behavioural activation. Accordingly, cocaine-induced hyperlo- ous questions about how reward and salience mechanisms comotion can be blunted by either genetic deletion or may have differentiated in HR mice over the course of pharmacological blockade of the CB1 receptor (Cheer et al., selective breeding. However, because previous operant- 2007; Corbille et al., 2007; Gerdeman et al., 2008). Thus, CB1 conditioning, pharmacological and brain-imaging studies signalling may ‘motivate' running behaviours, similar to have involved only females, it is not known whether sex motivation and conditioning induced by drugs of abuse.
differences in high wheel running (i.e. longer-durationrunning only in males) are related to differences in This study examined the possible links between reward or motivation.
endocannabinoids and voluntary exercise in mice fromlines that have been selectively bred for high amounts of The purpose of this study was to test the hypothesis that voluntary wheel running. A 15-year selection experiment mice from the HR and control lines would respond to increase voluntary wheel running in laboratory house differentially in voluntary wheel running when adminis- mice (Mus domesticus) provides a unique opportunity to tered a selective cannabinoid receptor antagonist/inverse understand how neural incentives may evolve in real time agonist. Moreover, we tested the hypothesis that the to affect motivation for exercise. After 10 generations of differential response would be sex specific. We used selection, mice from four replicate high runner (HR) rimonabant (SR141716), a selective CB1 receptor lines, of both sexes, ran at least 70% more than four antagonist (Carai et al., 2005), that has previously been nonselected control lines (Swallow et al., 1998). After 16 used to block both cannabinoid reward and conditioned generations, HR mice ran on average 170% more than drug-seeking or food-seeking behaviours (DeVries and controls (Rhodes et al., 2000). This differential response Schoffelmeer, 2005; Thornton-Jones et al., 2005; Zangen has been achieved primarily by an increase in running et al., 2006; Ward et al., 2007).
speed, as opposed to duration of running, particularly infemale HR mice (Swallow et al., 1998, 1999; Rhodes et al.,2000; Girard et al., 2001; Koteja and Garland, 2001; Garland, 2003). In general, over the course of selection, male and female HR mice have increased their total wheel revolu- We studied 48 female and 48 male mice from generation tions in different ways: female HR mice increased speed, 48 of a long-term selection experiment for high voluntary whereas male HR mice increased both speed and, to wheel-running behaviour (Swallow et al., 1998; Garland, a lesser degree, the amount of time spent running per day.
2003). The original progenitors were outbred Hsd: In addition to changes in locomotor behaviour, the selective Institute for Cancer Research mice (Mus domesticus) breeding regimen has led to changes in many morpholo- purchased from Harlan Sprague Dawley (Indianapolis, gical, physiological and behavioural traits (Garland, 2003).
Indiana, USA). Mice were randomly mated for two For example, HR mice exhibit reduced body mass (Swallow generations, then assigned to eight closed lines, four to et al., 1999), reduced body fat (Swallow et al., 1999), be selectively bred for HR and four to be bred without differences in open-field behaviour (Bronikowski et al., regard to wheel running, hence serving as controls for 2001), differences in thermoregulatory nest-building beha- founder effects and random genetic drift (control lines).
viour (Carter et al., 2000), increased predatory aggression In each subsequent generation, mice were paired within (Gammie et al., 2003) and higher plasma corticosterone line, and offspring were toe clipped, weighed and weaned (Girard and Garland, 2002; Malisch et al., 2008) and from dams at 21 days of age. Mice were then housed in adiponectin levels (Vaanholt et al., 2007).
same-sex groups of four until approximately 6–8 weeks ofage, at which point they began a 6-day wheel-access trial.
Rhodes et al. (2005) hypothesized that the motivation for Mice were housed in standard cages with Wahman-type voluntary endurance exercise has been altered in HR activity wheels (1.12 m circumference, 35.7 cm diameter, mice. Pharmacological studies with dopamine transporter 10-cm-wide running surface of a 10-mm mesh enclosed Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Behavioural Pharmacology 2008, Vol 19 No 8 by clear Plexiglas and stainless steel) attached through off at 16.00 h, and injections began 2 h later, which was a 5.5-cm-long stainless-steel tube inserted though a during the time of typical peak wheel-running activity 7.7-cm-diameter hole in the side of the cage, allowing (Girard et al., 2001; Girard and Garland, 2002; Rhodes the mouse to continuously access the wheel. Wheel et al., 2003; Malisch et al., 2008). Mice were then split into revolutions were recorded daily in 1-min intervals by three measurement batches for convenience, thus allow- a photocell counter attached to the wheel and compiled ing injections to be completed between 18.00 and through customized software by San Diego Instruments 20.00 h. Intraperitonial injections were administered as (San Diego, California, USA). In the four replicate an experimenter held the scruff of the neck manually to HR lines, the male and female from each family with restrain the mouse.
the most total revolutions on days 5 and 6 of the 6-daytest were chosen to propagate the lines to the next The acute locomotor response to treatment was mea- generation. In the four control lines, a male and a female sured as the total number of wheel revolutions in the were randomly chosen from each family. Within all lines, period from 10 to 70 min postinjection (Coimbra, 2001).
breeders were randomly paired, with the exception that In addition, we analysed the number of 1-min intervals sibling matings were not allowed. Throughout the with at least one revolution (time spent running), the selection experiment and for all studies described here, average running speed (revolutions/active intervals), and mice were routinely housed with free access to food and the maximum speed (revolutions in the single highest water, and maintained on a 12-h light-dark cycle.
1-min interval).
As top runners were unavailable in the HR lines (used as breeders), we also excluded the lowest-running animals Statistical analyses were performed using SAS version 9.1 in HR-line families. Of the remaining mice, one male and (SAS Institute, Cary, North Carolina, USA). Analyses female were chosen from each available family, except were first conducted separately by sex. The primary when only one sex was available from a given family.
grouping factors were linetype (HR vs. C) and dose, and As a part of the routine selection protocol (see previous replicate line was a random effect nested within linetype.
paragraph), mice were tested for voluntary wheel running Individual was the factor for repeated measures, and we over a 6-day period. Placement of mice in wheel cages assumed compound symmetry in SAS Procedure Mixed.
was randomized with respect to linetype (HR vs. control) In this mixed-model analysis of covariance, the degrees of and sex, and experimenters were blind to line and freedom for testing the effect of linetype, relative to line, is always 1 and 6. For dose and the dose  linetypeinteraction [tested relative to the dose  line (linetype)effect], degrees of freedom are 2 and 12, respectively.
Drug protocol and wheel running This interaction term is of prime interest because, if Rimonabant (SR141716) was obtained from the NIDA significant, it indicates a differential response of the HR Drug Supply Program (Baltimore, MD, USA). Several and control lines to the drug dose. Wheel freeness earlier pharmacological studies have used HR mice (a measure of how easy it is to turn each wheel) was (Rhodes et al., 2001; Rhodes and Garland, 2003; Li measured before each experiment and was included as et al., 2004). Following the design of Li et al. (2004), each a covariate in statistical analyses, as was age. During the individual mouse received vehicle injection, low-dose course of the experiments, a total of three males and rimonabant in a randomly determined order over the (one male), injection problems (one female), wheel course of 6 days, with 48 h between each injection to malfunction (one male), or because they were observed avoid carryover effects of the previous treatment. Vehicle to exhibit twirling behaviour (running in rapid, small, injections were solution of 20% DMSO, 10% Tween-80 stereotypic circles) in their cages (two females and one and 70% physiological saline. This vehicle solution has male). Thus, 45 males and 45 females were analysed been previously described for the in-vivo delivery of cannabinoid compounds including rimonabant and doesnot by itself influence open-field locomotor behaviour inmice (Gerdeman et al., 2008) or the firing of dopamine Second, we performed combined analyses of both sexes neurons (Wu and French, 2000). The doses of 0.1 and using the difference between the running values of the 3.0 mg/kg rimonabant have been previously reported to female and male within each family. In other words, we maintain a physiologically effective blockade of CB1 analysed the sex difference within each family.
receptors in both mice and rats (Carai et al., 2005).
Injection solutions were prepared fresh each day, and We also analysed wheel running of all of the mice from injection volumes adjusted for dose and body mass of the this generation that received the routine 6-day wheel test animal. Mice received treatment at approximately the (i.e. including those used in this study), with age and same time of day for each injection. Lights were turned wheel freeness as covariates.
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Rimonabant affects wheel running in mice Keeney et al.
HR males ran both faster (HR = 23.2 ± 1.1, control = Baseline wheel running 9.5 ± 1.0 RPM, P < 0.001) and more minutes per day than Considering all of the mice from generation 48 (n = 324 control (HR = 512 ± 40, control = 369 ± 39, P < 0.05).
females, 265 males) that received the 6-day wheel test,females from the HR lines (12 891 ± 285 revolutions/day; As expected, and as shown in Fig. 1 for the three days before injections, the 45 HR mice used in the drug trials more than control females (4294 revolutions/day) on days ran significantly more total revolutions than did the 45 5 + 6 (P < 0.001). For males, HR ran 11 026 (± 492) as mice from the control lines. Figure 1 also shows that compared with 3529 (± 551) revolutions/day for con- females ran more than males within both the HR and trol, yielding a 3.12-fold differential (P < 0.001). HR control lines.
females ran faster than controls (HR = 24.8 ± 0.8,control = 9.3 ± 0.9 RPM, P < 0.001) but they did notrun significantly more minutes per day (HR = 520 ± 27, control = 451 ± 28, P = 0.13). HR males ran both faster Figure 2 shows the wheel running in 10-min bins of (HR = 21.7 ± 1.0, control = 9.2 ± 1.1 RPM, P < 0.001) female (top) and male (bottom) HR and control mice and more minutes per day than control (HR = 503 ± 35, during the 10–130 min postinjection, as well as average control = 371 ± 36, P < 0.05).
revolutions per 10 min during the 30 min before injection.
Injections of both vehicle and rimonabant caused Results were similar for the subset of males and females a reduction in wheel running in all mice. For females, used in the present experiment. Females from the HR the reduction depended on dose (repeated-measures lines (13 201 ± 584 revolutions/day) ran 3.07-fold more than analyses of covariance, Table 1, all P < 0.001), and the control females (4294 ± 547 revolutions/day) on days 5 + 6 effect of dose depended on linetype for total revolutions, (P < 0.001). For males, HR ran 11 930 ( ± 615) as compared average speed, and maximum speed (all P < 0.05), but with 3631 ( ± 589) revolutions/day for control, yielding not for the amount of time spent running (P = 0.7).
a 3.29-fold differential (P < 0.001). HR females ran faster Adjusted means for females are shown in Table 2.
than controls (HR = 25.6 ± 1.2, control = 9.1 ± 1.1 RPM,P < 0.001), but they did not run significantly more minutes For males, the reduction in wheel running also depended per day (HR = 518 ± 36, control = 442 ± 35, P = 0.19).
on dose for revolutions and average speed (Table 1, HR femaleHR maleControl female Revolutions per 20 min 200 Daily pattern of wheel running (revolutions in 20-min bins) for mice from high runner (HR)(selected) and control lines during 3 days before the start ofrimonabant injections (28 April–1 May 2007). Note that females run more than males in both linetypes. Grey bars indicate lights off.
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Behavioural Pharmacology 2008, Vol 19 No 8 Control 3.0 mg/kg Control 0.1 mg/kg Revolutions per 10 min –30 –20 –10 90 100 110 120 130 Minutes postinjection Control 3.0 mg/kg Control 0.1 mg/kg Revolutions per 10 min –30 –20 –10 90 100 110 120 130 Minutes postinjection Wheel running revolutions in 10-min bins during intraperitonial rimonabant injections (low dose = 0.1 mg/kg, high dose = 3.0 mg/kg). (First 10-minperiod after injection is omitted.) Values at – 15 min are pooled revolutions in the 30-min period before injections. Values are simple means andstandard errors. Points are centered on the 5-min mid-point (i.e. the point for the 11–20 min bin is located at 15 min). Rimonabant reduced wheelrunning acutely in all mice, but for females (top panel) the reduction was significantly greater for high runner (HR) lines than for control lines (seeTable 1, P for linetype  dose interaction = 0.0227).
P < 0.05), but not for duration (Table 1, P = 0.10).
(P = 0.17), but a significant dose  linetype interaction Moreover, the dose linetype interaction was not statistically (P < 0.05). Thus, the magnitude of the sex difference significant for any measure of wheel running (all P > 0.5).
depends on the dose of rimonabant, and this effect Adjusted means for males are shown in Table 3.
depends further on linetype. For the time spent running,the analysis indicated a significant effect of dose The foregoing separate analyses of males and females (P < 0.05), but a nonsignificant effect of linetype suggest a significant interactive effect of sex and linetype (P < 0.63), and dose  linetype interaction (P = 0.28).
on the wheel-running response to rimonabant, and this is Thus, the magnitude of the sex difference in running supported by analyses of the difference in running time depends on the dose of rimonabant. For average speed, the analysis indicated no significant effect of revolutions, this analysis indicated a significant effect of dose (P = 0.19), linetype (P = 0.14) or dose  linetype dose (P < 0.05), a nonsignificant effect of linetype (P = 0.30). Finally, for maximum speed, the analysis Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Rimonabant affects wheel running in mice Keeney et al.
Repeated-measures analyses (SAS Procedure Mixed) of Least-squares (adjusted) means and standard errors from wheel running 10–70 min after injecting vehicle, low or high dose repeated-measures analyses of wheel running 10–70 min after injections, as reported in Table 1, for males linetype interaction interaction Time denotes number of 1-min intervals with at least one revolution; averagespeed is revolutions/time; maximum speed is revolutions in the single highest All values are means per 10-min intervals. Time denotes number of 1-min intervals 1-min interval. Degrees of freedom are 2 and 12 for dose, 1 and 6 for linetype, with at least one revolution; average speed is revolutions/time; maximum speed is and 2 and 12 for the dose  linetype interaction, respectively. All P values are for revolutions in the single highest 1-min interval.
two-tailed tests. All analyses also included age and wheel freeness as covariates HR, high runner.
(results not shown).
Least-squares (adjusted) means and standard errors from repeated-measures analyses of wheel running 10–70 min after It is important to emphasize that this interpretation is injections, as reported in Table 1, for females based on statistical analyses of the actual wheel-running traits measured in control line versus HR females. If posttreatment running responses are analysed as propor- tional values, relative to baseline running behaviours within each group, then differences between groups lose statistical significance. For example, an analysis of the ratio of revolutions after high-dose/sham injection indicates no significant effect of linetype for either females (P = 0.62) or males (P = 0.83). Thus, the proportional response to rimonabant does not differ between HR and control lines. However, analysis of proportional responses can be quite misleading when the groups being compared differ greatly in baseline values, as is true in the present case, where HR mice run approximately three-fold more than controls (see Figs 1 All values are means per 10-min intervals. Time denotes number of 1-min intervals and 2). We believe that the repeated-measures analysis of with at least one revolution; average speed is revolutions/time; maximum speed is the actual values – not ratios to sham-injection values – is revolutions in the single highest 1-min interval.
HR, high runner.
the most statistically sound way to analyse these databecause of physiological differences in wheel running indicated no significant effect of dose (P = 0.11), between the HR and control lines. The HR mice may run linetype (P = 0.07) or dose  linetype (P = 0.24).
voluntarily near their maximal aerobic speed (i.e. almostat their maximal rate of oxygen consumption), whereas mice from the control lines do not (Girard et al., 2001; Females from four replicate lines of mice that have been Rezende et al., 2005). This means that HR and control selectively bred (48 generations) for high voluntary mice are operating under different physiological regimens wheel-running behaviour (HR lines) showed altered during the times of peak running every night, when this responsiveness to a selective CB1 receptor antagonist as study was conducted. Thus, an increase or decrease of, compared with females from four nonselected control say, 10%, in wheel running would not mean the same lines. Males from the HR lines did not exhibit thing physiologically to an HR and control mouse.
a differential response. Thus, aspects of the ECS, ora physiological system regulated by the ECS, seem to have evolved in a sex-specific manner in response to Earlier studies of these lines of mice have documented selective breeding for high activity levels.
substantial sex differences in wheel running between HR Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Behavioural Pharmacology 2008, Vol 19 No 8 and control lines. In particular, female HR mice have (De Vries and Schoffelmeer, 2005; Maldonado et al., evolved higher daily running distances almost entirely by 2006), and may therefore motivate increased high-speed an increase in average (and maximum) running speed, running in females. If high-speed wheel running increases whereas males have shown increases in both speed and CB1 signalling, then it could lead to conditioning through duration of wheel activity (Swallow et al., 1998, 1999; neurobiological rewards. Although these possibilities Rhodes et al., 2000; Girard et al., 2001; Koteja and Garland, require further testing, they suggest that the evolution 2001; Garland, 2003; this study). Baseline wheel-running of high-speed (i.e. high-intensity) wheel running in of mice in this study (Fig. 1) is consistent with the female mice may be linked to CB1 signalling, whereas activity profiles of mice from previous generations the increased running duration observed only in HR and studies. However, this is the first study to use males may have other causes.
pharmacology as a means of understanding the neurobio-logical correlates of the sex differences. After injecting Studies of rats suggest that sex-based differences in CB1 high dose of rimonabant, female HR mice differentially signalling are common. For example, male and female rats decreased total revolutions, and this was done through show differential CB1 receptor expression in the brain a reduction in running speed, with no statistical reduction (Rodriguez de Fonseca et al., 1994; Gonzalez et al., 2000), in the amount of time spent running (Table 1).
and several cannabinoids are known to be more potent, orshow greater effects, in females than males (Cohn, 1972; One possible partial explanation for these results is that Tseng and Craft, 2001; Craft, 2005; Fattore et al., 2007). It the effects of rimonabant are stronger in females – at is also possible that the expression or function of the CB1 least those prone to high activity – compared with males.
receptor itself is not directly related to the promotion of However, there is little indication that the effects of voluntary wheel running in female HR mice. For example, rimonabant are sex dependent in other taxa. For example, HR females could be more sensitive to potential negative Foltin and Haney (2007) found few sex differences in the effects of rimonabant administration (Pacher et al., 2006), effects of rimonabant on appetite in baboons. In addition, or have alterations upstream of the CB1 receptor that human trials of rimonabant for weight loss report no sex affect its functionality, such as in the synthesis, release or differences in efficacy (Isoldi and Aronne, 2008).
degradation of endocannabinoids.
Therefore, the simplest interpretation of these results is Finally, it is important to note that our results seem to that selection for high voluntary activity has altered some differ from a study by Compton et al. (1996), in which aspect of CB1 functionality in female HR mice, but intravenous injection of rimonabant in doses upwards of apparently not in males. At a high dose of rimonabant 3 mg/kg is shown to produce locomotor stimulation for up (3.0 mg/kg), HR females had the greatest proportional to 4 h postinjection in male Institute for Cancer Research response in terms of voluntary wheel running, which mice. However, their measure of locomotion was taken indicates that they are more sensitive to that dosage than in a standard (novel) cage, during the day, 5–15 min male HR mice or controls in general. These results postinjection. It is likely that this sort of locomotion has suggest that female HR mice may differentially utilize little to do with voluntary wheel running, which occurs at CB1 signalling during wheel running.
much higher speeds and over much longer time periods.
Of course, it is possible that if we had included a higher We suggest two hypotheses to explain such enhanced dose then we may have observed locomotor stimulation in CB1 signalling in female HR mice. First, CB1 signalling our mice, given that our highest dosage level (3 mg/kg) is modulates pain perception and likely plays an important at the lower bound of the range of doses used by role in exercise-induced analgesia (Richardson, 2000; Compton et al. (1996). Our maximum dosage level was Sparling et al., 2003; Hohmann and Suplita, 2006).
chosen because it has been shown to be fully effective in blocking many cannabinoid, or presumed endocannabi- female HR mice to increase running intensity and, noid effects, specific to CB1 receptors (Carai et al., 2005; therefore, run longer distances than controls. A recent DeVries and Schoffelmeer, 2005; Pacher et al., 2006).
study indicates that exercise-induced analgesia is inten- Multiple studies have found this dosage of rimonabant to sity dependent (Hoffman et al., 2004), suggesting that have no effect on baseline exploratory behaviours in mice exercise-induced increases in CB1 signalling may be of the C57Bl6 strain (Tzavara et al., 2003; Patel and intensity dependent, and are able to influence neural Hillard, 2006; Gerdeman et al., 2008). Thus, rather than systems of nociception. Although we have evidence that stimulating behaviour, several studies have now found CB1 female HR mice do not differ in one measure of pain receptor blockade to reduce locomotor activity that is sensitivity (thermal tail-flick test: Li et al., 2004), it is thought to reflect motivation (Introduction, and DeVries possible that exercise-induced analgesia is under at least and Schoffelmeer, 2005) or emotional affect (e.g. swimming partly separate control. Second, CB1 signalling is thought in the forced swim test: Tzavara et al., 2003; Steiner et al., to mimic the action of drugs of abuse, producing 2007), suggesting in both cases a blunting of internal a rewarding sensation that can condition behaviours reward or positive emotional mechanisms.
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Rimonabant affects wheel running in mice Keeney et al.
Our results suggest an important link between the ECS Alvarez-Jaimes L, Polis I, Parsons LH (2008). Attenuation of cue-induced heroin-seeking behavior by cannabinoid CB1 antagonist infusions into the and the evolution of increased voluntary wheel-running nucleus accumbens core and prefrontal cortex, but not basolateral amygdala.
behaviour in house mice. Knowledge of the relation between CB1 receptor signalling and voluntary exercise Belke TW (1996). Investigating the reinforcing properties of running: or, running can increase the understanding of the role of cannabinoid is its own reward. In: Epling WF, Pierce WD, editors. Activity anorexia: theory,research, and treatment. Mahwah, NJ: Lawrence Erlbaum Associates; signalling in exercise, and how the neurobiological pp. 45–55.
correlates of reward incentive may change with selective Belke TW, Garland T Jr (2007). A brief opportunity to run does not function as a breeding for a voluntary behaviour. In addition, because reinforcer for mice selected for high daily wheel-running rates. J Exp AnalBehav 88:199–213.
the ECS is an important neuromodulator of neural Belke TW, Heyman GM (1994). A matching law analysis of the reinforcing systems implicated in the perception of reward (Lupica efficacy of wheel running in rats. Anim Learn Behav 22:267–274.
and Riegel, 2005; Maldonado et al., 2006; Cheer et al., Brady KT, Randall C (1999). Gender differences in substance use disorders.
Psychiatr Clin N Am 22:241–252.
2007; Mahler, 2007; Pillolla et al., 2007), our results can Bronikowski AM, Carter PA, Swallow JG, Girard IA, Rhodes JS, Garland T Jr help to elucidate aspects of the neurobiology of moti- (2001). Open-field behavior of house mice selectively bred for high voluntary vated behaviours in general.
wheel-running. Behav Genet 31:309–316.
Carai MAM, Colombo G, Gessa GL (2005). Rimonabant: the first therapeutically relevant cannabinoid antagonist. Life Sci 77:2339–2350.
Carter PA, Swallow JG, Davis SJ, Garland T Jr (2000). Nesting behavior of house Our results also supplement the growing body of work on mice (Mus domesticus) selected for increased wheel-running activity.
sex differences in the behavioural effects of cannabinoids Behav Genet 30:85–94.
and their receptors. Although the field is primarily built Cheer JF, Wassum KM, Sombers LA, Heien MLAV, Ariansen JL, Aragona BJ, et al.
(2007). Phasic dopamine release evoked by abused substances requires on studies of rats and mice (Craft, 2005; McGregor and cannabinoid receptor activation. J Neuroscience 27:791–795.
Arnold, 2007), these sex differences extend to human Cohen C, Perrault G, Griebel G, Soubrie P (2005). Nicotine-associated cues beings. Studies show that women seem to be more maintain nicotine-seeking behavior in rats several weeks after nicotinewithdrawal: reversal by the cannabinoid (CB1) receptor antagonist, sensitive to cannabinoid-induced hypotension than men rimonabant (SR141716). Neuropsychopharmacology 30:145–155.
(Mathew et al., 2003) and have greater expression of Cohn RA, Barratt E, Pirch JH (1972). Differences in behavioral responses of male CB1 receptor protein in leucocytes (Onaivi et al., 1999), and female rats to marijuana. Proc Soc Exp Biol Med 140:1136–1139.
and show age-related changes in CB1 receptor binding Coimbra NC, Castro-Souza C, Segato EN, Nora JEP, Herrero CFPS, that are distinct from men, including in brain areas related to reward and emotion (Van Laere et al., 2008).
Brain Res 888:314–320.
Makela et al. (2006) found that women were more prone Compton DR, Aceto MD, Lowe J, Martin BR (1996). In vivo characterization of a specific cannabinoid receptor antagonist (SR141716A): Inhibition of than men to exhibit deficits in a spatial span memory DELTA-9-tetrahydrocannabinol-induced responses and apparent agonist task after a low sublingual dose of the cannabinoid, activity. J Pharmacol Exp Ther 277:586–594.
tetrahydrocannabinol. Interestingly, because the tetrahy- Corbille A-G, Valjent E, Marsicano G, Ledent C, Lutz B, Herve D, Girault J-A (2007). Role of cannabinoid type 1 receptors in locomotor activity and striatal drocannabinol treatment also enhanced a spatial work- signaling in response to psychostimulants. J Neurosci 27:6937–6947.
ing memory task facilitated by dopamine, the authors Cota D, Woods SC (2005). The role of the endocannabinoid system in the speculated that both effects might be secondary to regulation of energy homeostasis. Curr Opin Endocrinol Diab 12:338–351.
Craft RM (2005). Sex differences in behavioral effects of cannabinoids. Life Sci cannabinoid-enhanced dopamine signalling (Makela et al., 2006). These observations augment clinical studies that De Petrocellis L, Cascio MG, Di Marzo V (2004). The endocannabinoid system: suggest, in general, women are more vulnerable than men a general view and latest additions. Br J Pharmacol 141:765–774.
during the transition period between opportunity to use De Vries TJ, Schoffelmeer ANM (2005). Cannabinoid CB1 receptors control conditioned drug seeking. Trends Pharmacol Sci 26:420–426.
and drug abuse (Brady et al., 1999), and that women are Demuth DG, Molleman A (2006). Cannabinoid signalling. Life Sci 78:549–563.
more responsive to the rewarding effects of addictive drugs Dietrich A, McDaniel WF (2004). Endocannabinoids and exercise. Br J Sports (Lynch, 2006). Given the therapeutic potential of canna- Med 38:536–541.
Fattore L, Spano MS, Altea S, Angius F, Fadda P, Fratta W (2007). Cannabinoid binoid-related treatments for a variety of disorders (Russo, self-administration in rats: sex differences and the influence of ovarian 2004; Pacher et al., 2006), including many that occur function. Br J Pharmacol 152:795–804.
disproportionately in women, further examination of sex Foltin RW, Haney M (2007). Effects of the cannabinoid antagonist SR141716 differences in the effect of cannabinoids are clearly (rimonabant) and d-amphetamine on palatable food and food pellet intake innon-human primates. Pharmacol Biochem Behav 86:766–773.
Gammie SC, Hasen NS, Rhodes JS, Girard I, Garland T Jr (2003). Predatory aggression, but not maternal or intermale aggression, is associated with highvoluntary wheel-running behavior in mice. Horm Behav 44:209–221.
Garland T Jr (2003). Selection experiments: an under-utilized tool in biomechanics and organismal biology. In: Bels VL, Gasc J-P, Casinos A, The authors thank W. Saltzman for suggestions on editors. Vertebrate biomechanics and evolution. Oxford, UK: BIOS ScientificPublishers; pp 23–56.
statistical analyses, two anonymous reviewers for helpful Gerdeman GL, Schechter JB, French ED (2008). Context-specific reversal of comments on the manuscript, and J. L. Wiley for advice cocaine sensitization by the CB(1) cannabinoid receptor antagonist during manuscript revision. U.S. National Science Foun- Girard I, Garland T Jr (2002). Plasma corticosterone response to acute and dation Grant IOB-0543429 to T.G. and DA14263-04 to chronic voluntary exercise in female house mice. J Appl Physiol 92: Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Behavioural Pharmacology 2008, Vol 19 No 8 Girard I, McAleer MW, Rhodes JS, Garland T Jr (2001). Selection for high Rhodes JS, Gammie SC, Garland T Jr (2005). Neurobiology of mice selected for voluntary wheel-running increases speed and intermittency in house mice high voluntary wheel-running activity. Integr Comp Biol 45:438–455.
(Mus domesticus). J Exp Biol 204:4311–4320.
Rhodes JS, Garland T Jr (2003). Differential sensitivity to acute administration of Gonzalez S, Bisogno T, Wenger T, Manzanares J, Milone A, Berrendero F, et al.
ritalin, apormorphine, SCH 23390, but not raclopride in mice selectively bred (2000). Sex steroid influence on cannabinoid CB1 receptor mRNA and for hyperactive wheel-running behavior. Psychopharmacology 167:242–250.
endocannabinoid levels in the anterior pituitary gland. Biochem Biophys Res Rhodes JS, Garland T Jr, Gammie SC (2003). Patterns of brain activity associated with variation in voluntary wheel-running behavior. Behav Hoffman MD, Shepanski MA, Ruble SB, Valic Z, Buckwalter JB, Clifford PS (2004). Intensity and duration threshold for aerobic exercise-induced Rhodes JS, Hosack GR, Girard I, Kelley AE, Mitchell GS, Garland T Jr (2001).
analgesia to pressure pain. Arch Phys Med Rehabil 85:1183–1187.
Differential sensitivity to acute administration of cocaine, GBR 12909, and Hohmann A, Suplita R (2006). Endocannabinoid mechanisms of pain modulation.
fluoxetine in mice selectively bred for hyperactive wheel-running behavior.
AAPS J 8:E693–E708.
Isoldi KK, Aronne L J (2008). The challenge of treating obesity: the endocanna- Rhodes JS, Koteja P, Swallow JG, Carter PA, Garland T Jr (2000). Body binoid system as a potential target. J Am Diet Assoc 108:823–831.
temperatures of house mice artificially selected for high voluntary wheel- Kelley AE (2004). Memory and addiction: shared neural circuitry and molecular running behavior: repeatability and effect of genetic selection. J Thermal Biol mechanisms. Neuron 44:161–179.
Koteja P, Garland T Jr (2001). Forum: response to R. Eikelboom. Anim Behav Richardson J (2000). Cannabinoids modulate pain by multiple mechanisms of action. J Pain 1:2–14.
Laviolette SR, Grace AA (2006). The roles of cannabinoid and dopamine Rodriguez de Fonseca F, Cebeira M, Ramos JA, Martin M, Fernandez-Ruiz JJ receptor systems in neural emotional learning circuits: implications for (1994). Cannabinoid receptors in rat brain areas: sexual differences, schizophrenia and addiction. Cell Mol Life Sci 63:1597–1613.
fluctuations during the estrous cycle and changes after gonadectomy and Li G, Rhodes JS, Girard I, Gammie SC, Garland T Jr (2004). Opioid-mediated sex steroid replacement. Life Sci 54:159–170.
pain sensitivity in mice bred for high voluntary wheel running. Physiol Behav Russo EB (2004). Clinical endocannabinoid deficiency (CECD): can this concept explain therapeutic benefits of cannabis in migraine, fibromyalgia, Lichtman AH, Cook SA, Martin BR (1996). Investigation of brain sites mediating cannabinoid-induced antinociception in rats: evidence supporting peri- Neuroendocrinol Lett 25:31–39.
aqueductal gray involvement. J Pharmacol Exp Ther 276:585–593.
Sherwin CM (1998). Voluntary wheel running: a review and novel interpretation.
Lupica CR, Riegel AC (2005). Endocannabinoid release from midbrain dopamine Animal Behaviour 56:11–27.
neurons: A potential substrate for cannabinoid receptor antagonist treatment Sherwin CM, Nicol CJ (1996). Reorganization of behaviour in laboratory mice, of addiction. Neuropharmacology 48:1105–1116.
Mus musculus, with varying cost of access to resources. Animal Behaviour Lynch WJ (2006). Sex differences in vulnerability to drug self-administration.
Exp Clin Psychopharmacol 14:34–41.
Sparling PB, Giuffrida A, Piomelli D, Rosskopf L, Dietrich A (2003). Exercise Mahler SV, Smith KS, Berridge KC (2007). Endocannabinoid hedonic hotspot for activates the endocannabinoid system. NeuroReport 14:2209–2211.
sensory pleasure: anandamide in nucleus accumbens shell enhances ‘liking' Steiner MA, Wanisch K, Monory K, Marsicano G, Borroni E, Bachli H, et al.
of a sweet reward. Neuropsychopharmacology 32:2267–2278.
(2007). Impaired cannabinoid receptor type 1 signaling interferes with stress- Makela P, Wakeley J, Gijsman H, Robson PJ, Bhagwagar Z, Rogers RD (2006). Low coping behavior in mice. Pharmacogenomics J 8:196–208.
doses of D-9 tetrahydrocannabinol (THC) have divergent effects on short-term Swallow JG, Carter PA, Garland T Jr (1998). Artificial selection for increased spatial memory in young, healthy adults. Neuropsychopharmacology 31: wheel-running behavior in house mice. Behav Genet 28:227–237.
Swallow JG, Koteja P, Carter PA, Garland T Jr (1999). Artificial selection for Maldonado R, Valverde O, Berrendero F (2006). Involvement of the endocanna- increased wheel-running activity in house mice results in decreased body binoid system in drug addiction. TrendsNeurosci 29:225–232.
mass at maturity. J Exp Biol 202:2513–2520.
Malisch JL, Breuner CW, Gomes FR, Chappell MA, Garland T Jr (2008).
Thornton-Jones ZD, Vickers SP, Clifton PG (2005). The cannabinoid CB1 Circadian pattern of total and free corticosterone concentrations, cortico- receptor antagonist SR141716A reduces appetitive and consummatory steroid-binding globulin, and physical activity in mice selectively bred for responses for food. Psychopharmacology 179:452–460.
high voluntary wheel-running behavior. General Comp Endocrinol 156: Timberlake W, Wozny M (1979). Reversibility of reinforcement between eating and running by schedule changes – comparison of hypotheses and models.
Mathew RJ, Wilson WH, Davis R (2003). Postural syncope after marijuana: a Anim Learn Behav 7:461–469.
transcranial Doppler study of the hemodynamics. Pharmacol, Biochem Tseng AH, Craft RM (2001). Sex differences in antinociceptive and motoric Behav 75:309–318.
effects of cannabinoids. Eur J Pharmacol 430:41–47.
McGregor IS, Arnold JC (2007). Cannabis reward: biased towards the fairer sex? Tzavara ET, Davis RJ, Perry KW, Li X, Salhoff C, Bymaster FP, et al. (2003). The Br J Pharmacol 152:562–564.
CB1 receptor antagonist SR141716A selectively increases monoaminergic Morgan WP (1979). Negative addiction in runners. Physician Sports Med 7: neurotransmission in the medial prefrontal cortex: implications for therapeutic actions. Br J Pharmacol 138:544–553.
Onaivi ES, Chaudhuri G, Abaci AS, Parker M, Manier DH, Martin PR, Hubbard JR Vaanholt LM, Meerlo P, Garland T Jr, Visser GH, van Dijk G (2007).
(1999). Expression of cannabinoid receptors and their gene transcripts in Plasma adiponectin is increased in mice selectively bred for high wheel- human blood cells. Prog Neuropsychopharmacol Biol Psychiatry 23: running activity, but not by wheel running per se. Horm Metab Res Pacher P, Batkai S, Kunos G (2006). The endocannabinoid system as an Van Laere K, Goffin K, Casteels C, Dupont P, Mortelmans L, de Hoon J, Bormans G emerging target of pharmacotherapy. Pharmacol Rev 58:389–462.
(2008). Gender-dependent increases with healthy aging of the human cerebral Pagotto U, Cervino C, Vicennati V, Marsicano G, Lutz B, Pasquali R (2006). How cannabinoid-type 1 receptor binding using [F-18]MK-9470 PET. NeuroImage many sites of action for endocannabinoids to control energy metabolism? Int J Obesity 30:S39–S43.
Ward SJ, Walker EA, Dykstra LA (2007). Effect of cannabinoid CB1 receptor Patel S, Hillard CJ (2006). Pharmacological evaluation of cannabinoid receptor antagonist SR141714A and CB1 receptor knockout on cue-induced reinstate- ligands in a mouse model of anxiety: further evidence for an anxiolytic role for ment of Ensures and corn-oil seeking in mice. Neuropsychopharmacology endogenous cannabinoid signaling. J Pharmacol Exp Ther 318:304–311.
Pil olla G, Melis M, Perra S, Muntoni AL, Gessa GL, Pistis M (2007). Medial Wiley JL (2003). Sex-dependent effects of DELTA9-tetrahydrocannabinol on forebrain bundle stimulation evokes endocannabinoid-mediated modulation of locomotor activity in mice. Neurosci Lett 352:77–80.
ventral tegmental area dopamine neuron firing in vivo. Psychopharmacology Wu X, French ED (2000). Effects of chronic DELTA 9-tetrahydrocannabinol on rat midbrain dopamine neurons: an electrophysiological assessment.
Premack D, Schaeffer RW, Hundt A (1964). Reinforcement of drinking by running: Effect of fixed ratio and reinforcement time. J Exp Anal Behav 7: Xi Z-X, Newman AH, Gilbert JG, Pak AC, Peng X-Q, Ashby CR Jr, et al. (2006).
The novel dopamine D-3 receptor antagonist NGB 2904 inhibits cocaine's Rezende EL, Chappell MA, Gomes FR, Malisch JL, Garland T Jr (2005). Maximal rewarding effects and cocaine-induced reinstatement of drug-seeking metabolic rates during voluntary exercise, forced exercise, and cold exposure behavior in rats. Neuropsychopharmacology 31:1393–1405.
in house mice selectively bred for high wheel-running. J Exp Biol 208: Zangen A, Solinas M, Ikemoto S, Goldberg SR, Wise RA (2006). Two brain sites for cannabinoid reward. J Neurosci 26:4901–4907.
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Source: http://www.raichlen.arizona.edu/DavePDF/KeeneyEtAl2008.pdf