Paper | Published:

CB1 cannabinoid receptor knockout in mice leads to leanness, resistance to diet-induced obesity and enhanced leptin sensitivity


OBJECTIVE: There is growing evidence for an implication of the CB1 receptor subtype of the endocannabinoid system in the regulation of eating and fat deposition. To further define the physiological role of these receptors in the control of energy balance, we characterized the phenotype of CB1 receptor knockout (CB1−/−) mice maintained on an obesity-prone regimen or on a standard chow.

DESIGN: CB1−/− male mice were compared to wild-type animals (CB1+/+ male mice) in two feeding paradigms: (1) with a standard laboratory regimen (3.5 kcal/g, 14.5% of energy as fat) and (2) on a free-choice paradigm consisting of offering both the standard laboratory chow and a high-fat diet (HFD) (4.9 kcal/g, 49% of energy as fat).

RESULTS: When maintained on the standard diet, CB1−/− mice are lean. At the age of 20 weeks, their body weight and adiposity are, respectively, 24 and 60% lower than that of CB1+/+ mice. They are slightly hypophagic, but when expressed as percent of body weight, their relative energy intake is similar to that of the wild-type animals. Furthermore, inactivation of CB1 receptors reduces plasma insulin and leptin levels, and enhances the response to intracerebroventricular leptin injection. The free-choice paradigm shows that the preference for a high-fat highly palatable chow is slightly delayed in onset but maintained in CB1−/− mice. However, loading CB1−/− mice with this obesity-prone diet does not result in development of obesity. Knockout mice do not display hyperphagia or reduction of their relative energy intake in contrast to CB1+/+ mice, and their feeding efficiency remains low. These data suggest an improved energetic metabolism with the high-fat regimen. Furthermore, the insulin resistance normally occurring in HFD-fed mice is not present in CB1−/− mice.

CONCLUSION: These results provide evidence that the stimulation of CB1 receptors is a key component in the development of diet-induced obesity, and that these receptors and their endogenous ligands are implicated not only in feeding control but also in peripheral metabolic regulations. The lack of effect of SR141716, a selective CB1 receptor antagonist, in CB1−/− mice further supports this hypothesis, as this compound was previously shown to display potent anti-obesity properties in diet-induced obese C57BL/6 mice.


Cannabinoid receptors and their endogenous ligands have been implicated in a variety of physiological functions, including the modulation of pain,1 emotional behavior,2 cognition,3 and feeding.4 Among these diverse actions, the stimulation of eating has been extensively investigated. This role has been supported by the demonstration that Δ9-tetrahydrocannabinol (Δ9-THC), the main active component of marijuana, as well as the endocannabinoids, anandamide and 2-arachidonoyl glycerol (2-AG), stimulate feeding in laboratory animals.5,6,7,8 In this field, the discovery of the first selective cannabinoid CB1 receptor antagonist SR1417169 has been an important tool for research on the cannabinoid system. Cannabinoid CB1 receptor blockade by SR141716 was shown to antagonize the orexigenic effect of the cannabinoid agonists Δ9-THC, anandamide and 2-AG7,8,10 and demonstrated the implication of CB1 receptors in feeding control. Furthermore, when administered alone, SR141716 suppressed food intake in rodents,11,12 which indicates that tonic endocannabinoid activity may be a key component of appetite regulation.

Interestingly, the role of the cannabinoid system in modulating food intake is more apparent with highly palatable foods. CB1 receptor agonists4 and antagonists have been reported to preferentially modify the consumption of sweet diets in rodents13 and in the marmoset.14 The mechanisms by which cannabinoids control feeding may involve an interaction between feeding behavior and the reward systems. These mechanisms remain, however, to be clarified.

In addition to regulating food intake, cannabinoid systems have been shown to modulate metabolic regulations. We previously demonstrated that in diet-induced obese mice, a chronic treatment with the CB1 receptor antagonist SR141716 induced a pronounced and sustained body weight loss, which outlasted the decrease in food intake since hypophagia declined with the repetition of treatments.15 These findings strongly suggest that cannabinoid systems are involved not only in the feeding control but also in the regulation of body weight by acting directly or indirectly via CB1 receptors. Recently, a relationship between the endocannabinoid system and leptin has been reported.16 Genetically obese rodents with defective leptin signaling exhibit elevated hypothalamic endocannabinoid levels, and these levels are reduced after an acute leptin treatment in ob/ob mice. This hormone is a key component of systems that regulate both feeding and body weight and the suggested interactions between cannabinoids and leptin open new aspects of the extremely complex pathways that control energy balance.

In view of its implication in the control of food intake and body weight, it seems justified to use cannabinoid blockade in the treatment of obesity. This growing disorder appears highly related to an increased availability of palatable foods in our modern society. Thus, by reducing food intake, modifying the attractiveness of palatable food and influencing metabolic regulations, the blockade of CB1 receptors was expected to be highly efficient in reducing obesity. This hypothesis has been confirmed in previous studies showing that treatment with CB1 receptor antagonists in diet-induced obese mice results in a marked body weight loss.15,17 Furthermore, a recent work18 demonstrated that mice with disrupted CB1 gene (CB1−/−) and maintained on standard mouse chow are lean and hypophagic, and that food intake-independent mechanisms partly contribute to their lean phenotype. To confirm the role of CB1 receptors in regulating both food intake and body weight, we characterized CB1−/− mice given an obesity-prone regimen. We used a free-choice-feeding paradigm consisting of offering mice both a standard chow and a high-fat highly palatable diet. This experimental paradigm allows the detection of phenotypic differences regarding sensitivity to obesity development and the assessment of the influence of the system on food preference. The phenotypes of CB1−/− and CB1+/+ mice given this regimen were compared to that of CB1+/+ fed the standard diet alone in order to quantify obesity and the associated insulin resistance in the two strains. A group of CB1−/− mice fed the standard diet was also included to assess the basal phenotype of these knockout animals. Finally, we confirmed the lack of effect of the CB1 receptor antagonist SR141716 in knockout mice fed a high-fat diet (HFD) not only on adiposity but also on plasma insulin levels. Our studies demonstrate a major role of CB1 receptors in controlling body weight as well as food intake, and confirm the view that the blockade of these receptors may constitute a new approach for the treatment of obesity.

Materials and methods


CB1−/− mice were generated as described previously.19 Briefly, for constructing the targeting vector, a 9-kb KpnI–KpnI fragment including the entire coding region of the CB1 receptor gene was cloned from a 129SvJ6 mouse genomic library. An ApaI–ApaI fragment containing part of the coding region was inactivated by substitution with a neomycin resistance cassette. The targeting vector was transfected into embryonic stem cells from the 129/Ola line. Then, mutated ES cells were injected into C57BL/6 blastocytes to generate chimeric founder mice. Germline transmission of the targeted allele was determined by PCR analysis of mouse tail genomic DNA. The homozygous CB1−/− and CB1+/+ mice used in this work were from a C57BL/6X129/Ola F2 genetic background. The lack of binding of the synthetic CB1 receptor agonist [3H]CP 55 940 in these mice was checked by autoradiography in various brain sections.15 Male mice were housed individually and given ad libitum food and tap water. Two rodent diets differing in fat and caloric content were used. The HFD was a 4.7 kcal/g energy density diet (TD 97366, HARLAN) with 49% energy as fat, and the standard diet for transgenic mice (STD) contained 3.5 kcal/g (A04C, UAR) with 14.5% energy as fat. The standard diet was given to the animals alone (STD) or was proposed together with the HFD in two different feeders (ST/HFD). The protocols have been approved by the Ethical Committee for Laboratory Animals of Sanofi-Synthelabo Recherche. They were carried out in accordance with the European Directive 86/609/EEC.

Feeding study

Male CB1−/− mice (9 weeks old) were maintained on a 12 h light/dark cycle (light off 21:00 hours) and were allowed free access to the standard diet (STD−/−, n=9), or to the standard and the HFD both offered in two separated hanging feeders (ST/HFD−/−, n=15). Equivalent groups of wild-type animals were maintained in the same conditions (STD+/+, ST/HFD+/+). Body weight was recorded daily for 11 weeks on each specific regimen. Food intake was daily monitored during the first 4 weeks on diet and then during week 12. When spillage was observed, the individual value was removed from data.

At the end of the feeding study, glycemia was measured using a One Touch Glucometer (Bayer Diagnostics) in 18-h fasted mice before and after (30, 60, 90, 120 and 180 min) intraperitoneal insulin injection (0.60 U/kg). The AUC(0–180 min) of glycemia was calculated to assess insulin sensitivity. After recovering from the insulin sensitivity test, the body composition was estimated using a small-animal body composition analyzer (EM-SCAN®, SA-3000) in anesthetized mice.20 Then, mice were killed by decapitation and blood was collected in EDTA-containing tubes. Insulin and leptin plasma levels were determined using radioimmunochemical kits (Amersham and Linco).

Response to leptin

Male CB1−/− mice (9 weeks old) were maintained on a reverse 12 h light/dark cycle (light on 21:00 hours) and fed the standard chow (STD). Recombinant mouse leptin (Biochemicals) was injected freehand at the beginning of the dark phase in 24-h fasted CB1−/− or wild-type animals (n=15) by intracerebroventricular route at a dose of 1 μg/mouse. Equivalent groups of each strain received the vehicle in the same conditions.

Treatment with SR141716

Male CB1−/− mice (9 weeks old) and wild-type animals were maintained on a reverse 12 h light/dark cycle (light on 21:00 hours) and fed the HFD. After 6 weeks on diet, a dose of 10 mg/kg/d of SR141716 [N-piperidino-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methylpyrazole-3-carboxamide] (Sanofi-Synthelabo Recherche, Montpellier, France) or the vehicle alone was administered to equivalent groups of CB1−/− mice and wild-type animals (n=12). Treatments were performed orally just before the beginning of the dark phase. Body weight and food intake were daily monitored. After 3 weeks of treatment, mice were killed by decapitation, lumbar abdominal fat pads were removed and weighed and blood was collected for plasma insulin determination.

Statistical analyses

The results are expressed as mean±s.e.m. A repeated two-way ANOVA (repeated measures on body weight, food intake and metabolic indicators) or one-way ANOVA (other variables) was performed using SAS version 8.2. In the feeding study, analysis compared all the groups on diet to the standard diet-fed wild-type animals, which were considered to be the reference animals. When necessary, a direct comparison between the two basal phenotypes alone (STD−/− vs STD+/+) has been added. In this case, the result of this analysis is reported in the legend of the figures. SR141716-treated mice were compared to their respective controls. A Dunnett's test was used to determine statistical significance between groups. In the response to exogenous leptin test, groups were compared using a Student–Newman–Keuls's test. When the hypotheses necessary for the application of parametric tests were not achieved, appropriate nonparametric tests were applied.


CB1−/− mice are lean and resistant to diet-induced obesity

When maintained on an STD diet, the mean body weight of male CB1−/− mice was reduced and this was first apparent at 14 weeks of age (Figure 1). At 20 weeks of age, STD−/− mice weighed 24% less than STD+/+ mice (P<0.01). The obesity-prone regimen (ST/HFD) induced a rapid increase of the growth curve in wild-type mice, which reached a 22% higher body weight than that of STD+/+ animals after 11 weeks on diet (P<0.01). In contrast, the weight of CB1−/− male mice tended to increase with the obesity-prone regimen but not higher than that of STD+/+ mice (−12%, not significant). Carcass analyses (Table 1) showed that CB1−/− male mice had reduced fat content in the two feeding conditions (P<0.05). However, differences were attenuated with the obesity-prone regimen. As expected, obesity-prone diet fed wild-type mice showed a significant increase (P<0.05) in body fat deposition. Consistent with these findings, plasma leptin levels were reduced in STD−/− mice (4.72±1.07 ng/ml, P<0.05), increased in ST/HFD+/+ animals (31.86±2.67 ng/ml, P<0.01), and were similar in the ST/HFD−/− mice (7.53±1.14 ng/ml, not significant) when compared to STD+/+ animals (9.95±1.43 ng/ml).

Figure 1

Body weight of CB1−/− and wild-type male mice maintained on a standard chow (STD) or on an obesity-prone diet (ST/HFD). The obesity-prone diet consisted of offering ad libitum both a standard chow and an HFD. Values are means±s.e.m. Statistical analyses (repeated ANOVA/Dunnett's tests) compared all groups to STD+/+ (*P<0.05, **P<0.01).

Table 1 Body composition of CB1−/− mice and wild-type animals maintained on a standard chow (STD) or on an obesity-prone diet (ST/HFD) for 12 weeks

Leanness of CB1−/− mice is related to regulation of energy intake and change in energetic metabolism

The reduced body weight and lean phenotype were associated with a slightly lower food intake level as STD−/− mice ate about 10% fewer calories than STD+/+ over a 24-h period (Figure 2). The difference was not significant when all groups were included in the analyses. However, when the two strains maintained with the standard diet were compared directly, STD−/− mice ate significantly (P<0.05) less food than STD+/+ mice. When given the obesity-prone diet, the transient hyperphagia observed in ST/HFD+/+ animals was not apparent in the knockout mice fed with the same regimen. However, like wild-type animals, ST/HFD−/− mice consumed almost exclusively the high-fat pellets rather than the standard pellets, although their choice regarding this palatable food was slightly delayed (Figure 3).

Figure 2

Mean daily energy intake (kcal) per week in CB1−/− and wild-type male mice maintained on a standard chow (STD) or on an obesity-prone diet (ST/HFD). The obesity-prone diet consisted of ad libitum offering both a standard chow and an HFD. Values are means±s.e.m. Statistical analyses (repeated ANOVA/Dunnett's tests) compared all groups to STD+/+ (**P<0.01). When the STD−/− group was compared alone with the STD+/+ group, the two groups were significantly different (P=0.0146).

Figure 3

Food intake (expressed as kcal) of CB1−/− and wild-type male mice during the first week of a free-choice feeding. A standard diet (STD) and an HFD were given ad libitum to mice. The consumption of each kind of food (a) is daily measured over a 24-h period. The total energy intake (b) represents the addition of energy values coming from the STD food and the HFD food ingested every day. Values are mean±s.e.m. Statistics have been performed on the total energy intake values (b) only (*P<0.05, **P<0.01).

Metabolic indicators were assessed during the first 4 weeks on diet (Figure 4). The feeding efficiency was slightly reduced in STD−/− mice (P<0.05), increased in ST/HFD+/+ animals (P<0.01) and similar in the ST/HFD−/− mice when compared to the STD+/+ group. The relative energy intake, expressed as daily calories per gram of mouse, was similar in STD−/− mice and in STD+/+ animals. With the obesity-prone diet, this ratio decreased as obesity developed in wild-type animals (P<0.01), while in CB1−/− mice it remained elevated.

Figure 4

Feeding efficiency (a) and relative energy intake (b) of CB1−/− and wild-type male mice maintained on a standard diet (STD) or on an obesity-prone diet (ST/HFD). The obesity-prone diet consisted of offering ad libitum both a standard chow and an HFD. Feeding efficiency and relative energy intake were calculated over the first 4 weeks on the regimen. Values are means±s.e.m. Statistical analyses (repeated ANOVA/Dunnett's tests) compared all groups to STD+/+ (*P<0.05, **P<0.01). When feeding efficiency of the STD−/− group was compared alone with the STD+/+ values, the two groups were significantly different (P=0.0011).

CB1−/− mice have lower plasma insulin level and do not develop diet-induced insulin resistance

We studied the potential effects of CB1 receptor inactivation on glucose homeostasis. Plasma insulin levels were low in STD−/− mice (2.64±0.59 ng/ml, P<0.05 compared to 5.93±0.98 ng/ml in the STD+/+ group). As expected, with the obesity-prone regimen, insulin levels tended to increase in the wild-type mice, although the difference did not reach statistical significance (8.48±1.05 ng/ml, not significant), in contrast to that observed in ST/HFD−/− mice where plasma insulin remained low (2.95±0.28 ng/ml, P<0.05 vs the STD+/+ group). The glycemia measured in 18-h fasted mice and the insulin sensitivity (Figure 5) were similar in STD−/− mice and in STD+/+ animals. The obesity-prone diet induced a significant increase of fasting glycemia in the two genotypes, but the sensitivity to insulin remained unchanged in ST/HFD−/− mice, while it was significantly reduced in the ST/HFD+/+ animals.

Figure 5

Fasting glycemia (a) and insulin sensitivity (b) in CB1−/− mice fed a standard diet or on an obesity-prone regimen for 12 weeks. The obesity-prone diet (ST/HFD) consisted of offering ad libitum both a standard chow and an HFD. Insulin sensitivity was assessed by calculating the area under the curve (AUC) of glycemia during the 3 h after intraperitoneal insulin (0.6 U/kg) injection in fasted mice. Values are the means±s.e.m. Statistical analyses (ANOVA/Dunnett's tests) compared all groups to STD+/+ (*P<0.05, **P<0.01).

CB1−/− mice display enhanced leptin sensitivity

We examined the effect of an increase in leptin levels in the context of CB1 receptor inactivation by injecting exogenous leptin in CB1−/− male mice fed a standard diet (Figure 6). Leptin injection induced a higher weight loss and food intake decrease in CB1−/− male mice than that observed in wild-type animals. While wild-type animals lost 4.7% of their body weight 24 h after leptin injection, body weight loss in CB1−/− mice reached 7.5% (P<0.01 vs leptin-injected CB1+/+ mice). The food intake reduction was also stronger in the knockout mice (−14.4% vs vehicle-injected CB1−/− mice, P<0.01) than in wild-type animals (−6.1% vs vehicle-injected CB1+/+ mice, not significant). At 48 h after leptin injection, the effect remained stronger in CB1−/− mice than in wild-type littermates.

Figure 6

Response of CB−/− mice and wild-type mice to recombinant mouse leptin, administered at a dose of 1 μg by the intracerebroventricular route. (a) Body weight loss after leptin injection expressed as percent of initial body weight. The body weight loss of wild-type and knockout mice receiving leptin was significantly different (P<0.05) from that of their respective control groups 24 and 48 h after injection. (b) Daily food intake after leptin injection. Values are the means±s.e.m. Repeated ANOVA followed by an SNK test was performed to compare all groups to each other (*P<0.05, **P<0.01).

CB1−/− mice fed a high-fat diet are insensitive to a chronic treatment with SR141716

As previously reported,15 the 3-week treatment with SR141716 (10 mg/kg/d) had no effect on the body weight and food intake of CB1−/− mice. In contrast, SR141716 induced a marked decrease of body weight in wild-type animals (after 21 treatments, 34.6±1.0 g in SR141716-treated mice vs 41.5±0.7 g in vehicle-treated group, P<0.01). This effect was accompanied by a strong reduction in energy intake (9.6±0.5 kcal/d in SR141716-treated CB1+/+ mice vs 15.2±0.4 kcal/d in the vehicle CB1+/+ group during week 1, P<0.01), which attenuated thereafter. As expected, the weight of lumbar fat masses was unchanged in CB1−/− mice (0.142±0.032 g in the vehicle group and 0.178±0.024 g in the SR141716 group, not significant) and decreased in wild-type animals (0.526±0.030 g in the vehicle group and 0.299±0.037 g in the SR141716 group, P<0.01). SR141716 also lowered the insulin plasma levels in wild-type mice fed an HFD (2.16±0.39 ng/ml in the SR141716 group vs 6.94±1.51 in the vehicle group, P<0.05), and had no effect in CB1−/− mice fed the same diet (1.34±0.11 ng/ml in the SR141716 group vs 1.97±0.33 ng/ml in the vehicle group, not significant).


The role of cannabinoids as potential orexigenic agents has been widely suggested. Furthermore, a strong anti-obesity potential has been recently demonstrated with CB1 receptor antagonists in various rodent models.15,17,21 However, the phenotype of the CB1 receptor knockout mice regarding obesity development was still unknown. In this study, we challenged the animals with a free-choice-feeding paradigm consisting in offering both a standard chow and a high-fat highly palatable diet in order to detect phenotypic differences regarding sensitivity to obesity development and to assess the influence of the cannabinoid system on food preference. In these conditions, the body weight and the food intake of CB1−/− mice offering this obesity-prone regimen remained low and, in contrast to wild-type mice, obesity did not develop. Furthermore, the preference for a high-fat highly palatable chow was maintained in CB1−/− mice. The lean phenotype of CB1−/− mice given a standard chow was also confirmed.

When mice are maintained on standard diet, the inactivation of CB1 receptors in male mice results in leanness and in a slight hypophagia. The low body weight of CB1−/− mice was associated with a reduced adiposity. Mutant mice fed a standard chow display only half the body fat storage of their wild-type littermates at the age of 20 weeks. These findings are in agreement with the results published by Cota et al.18 Interestingly, the degree of hypophagia (about 10% less energy intake than in wild-type animals) appears low. When expressed as a percentage of body weight, the relative food intake was similar in mutant and in wild-type animals, which indicates that the balance between the body weight and the energy intake remains unchanged. In contrast to the relative food intake, the feeding efficiency of the knockout mice was significantly lower than that of wild-type animals. This parameter estimates the anabolic processes relative to the food intake. Therefore, the lower feeding efficiency in CB1−/− mice suggests that additional metabolic changes could potentiate the effect of the hypophagia in reducing fat storage.

When given a high-energy regimen, the adiposity of CB1−/− mice remains low in contrast to that observed in the wild-type animals. Although fat deposition slightly increases with the diet, it still remains lower than that of control animals fed the standard diet alone. These data suggest that inactivation of CB1 receptors is capable of counterbalancing the development of the obesity induced by a high-energy regimen. In this study, mice were fed using a free-choice paradigm consisting of offering ad libitum both a standard chow and an HFD. Considering the total daily energy intake, during the first week on the regimen, wild-type mice display hyperphagia as they consumed 25% more energy than mice fed the standard chow alone. Thereafter, the hyperphagia attenuates, although body weight gain remains elevated, as has been previously reported in the diet-induced obesity model.22 In the knockout mice, the hyperphagia does not occur. This finding suggests that suppressing hyperphagia during the early phase of a high-energy feeding may be useful in preventing obesity development. However, the ability of mice to gain adiposity with a high-energy regimen cannot be limited to the amount of energy intake, as no correlation has previously been found between resistance to obesity and energy intake in several responsive or nonresponsive strains of mice.22 When analyzing the composition of the meal, the low level of fat storage in CB1−/− mice does not seem to be related to a difference in the nature of the diet ingested. During the first day of the free-choice paradigm, CB1−/− mice ate about two-thirds of their daily energy from the HFD, while the wild-type mice chose only the high-fat food. However, after the second day of exposure to the two diets, mutant mice consumed exclusively the HFD like their wild-type littermates as demonstrated by the data collected during the first 4 weeks on study and after 11 weeks on regimen (data not shown). Thus, the free-choice regimen finally results in a high-fat feeding in the two strains. Therefore, inactivation of CB1 receptors does not prevent mice from choosing the highly palatable diet, but prevents excessive eating. These results suggest that CB1 receptor activation is not crucial to the pleasure derived from the orosensory characteristics of a high-fat highly palatable food.4 The endocannabinoid systems seem rather to be associated with the incentive aspects of eating motivation. In agreement with this hypothesis are the results of a previous study showing that blockade of CB1 receptors by the CB1 antagonist SR141716 failed to affect sucrose sham feeding in the rat.23 In this model, rats ingested a palatable solution, which was recovered within seconds directly from the stomach by a surgically implanted cannula. The lack of sham-feeding disruption by SR141716 strongly supports the concept that the cannabinoid system seems not to be significantly implicated in pathways that maintain sucrose ingestion. In contrast, the role of endocannabinoids in appetitive-incentive processes was demonstrated with alcohol using an operant progressive ratio paradigm in the rat.24,25 In this model, SR141716 was found to decrease motivation while CP 55940, a CB1 receptor agonist, increased motivation to obtain beer. Nevertheless, because the cannabinoid system is widely known to be more efficient in reducing highly palatable foods, our results on free-choice feeding in CB1−/− mice remain surprising. However, the previous experiments, which described a more efficient effect of cannabinoid compounds on palatable food, were performed by offering animals only one type of diet or by using acute treatments. Therefore, free-choice-feeding studies with repeated treatments would probably provide important insights into the mechanism of action of cannabinoid compounds. Finally adaptative compensatory mechanisms, which might develop in knockout animals, could also interact with the effect of CB1 gene inactivation.

In contrast to observations in mice fed a standard chow, CB1−/− mice fed a high-energy diet display both a higher relative food intake and a lower feeding efficiency compared to the wild-type mice given the same diet. Therefore, despite a high food intake on a weight basis in CB1−/− mice, these mice were leaner than their wild-type controls. This suggests that the decreased weight gain and fat storage observed in the mutant mice are also likely to be related to metabolic changes. Such metabolic changes seem mainly apparent with the high-energy diet, which enhances the differences between the two strains. However, more detailed metabolic studies need to be performed in these knockout mice maintained on an obesity-prone diet to confirm these results.

The mechanisms by which cannabinoid systems may modulate the energetic metabolism remain to be determined. We recently reported that the CB1 receptor antagonist SR141716 not only regulates food intake of diet-induced obese mice during a chronic treatment but can also induce metabolic changes to potentiate and sustain weight loss.15 Bensaid et al26 found that adipose tissue of the Zucker rat expresses the CB1 receptor mRNA and that this expression is upregulated in the tissue of the obese (fa/fa) rat compared to the lean animals. Culture mouse 3T3 F442A adipocytes also overexpress CB1 receptors when cells are differentiated. It is tempting to speculate that endocannabinoids may act directly on adipose tissue to regulate metabolic pathways. One of their possible targets is Acrp30 (adiponectin), an adipocyte-derived hormone, which has been shown to induce free fatty acid oxidation and body weight loss in mice.27 In their recent work, Bensaid et al26 also reported that pharmacological blockade of CB1 receptors by SR141716 stimulates the expression of Acrp30 mRNA in rat adipose tissue. CB1 receptor activation has also been shown to enhance lipogenesis by increasing lipoprotein lipase activity in primary adipocyte cultures, and the CB1 selective antagonist SR141716 blocked this effect.18 Overall, the adipose tissue might be one of the peripheral targets of the cannabinoid system. In addition to their food intake-dependent effects, endocannabinoids may influence energetic processes by regulating lipogenic pathways in adipocytes via CB1 receptors.

Low plasma leptin levels accompanied the low adiposity of CB1−/− mice. Furthermore, we found that the suppressive effects of exogenous leptin on feeding behavior and body weight were enhanced in CB1−/− mice. Such an effect has been previously observed in mice lacking melanin-concentrating hormone and in mice lacking neuropeptide Y.28,29,30 The increase of leptin sensitivity in CB1−/− mice suggests that, like neuropeptide Y and melanin-concentrating hormone, endocannabinoids may directly or indirectly oppose the effect of leptin in regulating food intake and fat storage and that they probably act via CB1 receptors. A relationship between leptin and the cannabinoid system has been previously reported. Defective leptin signaling is associated with elevated hypothalamic endocannabinoid levels in obese db/db and ob/ob mice and in Zucker rats, and these levels are reduced after an acute leptin treatment in ob/ob mice.16 As leptin is widely implicated in the regulation of food intake and energetic metabolism, this hormone may also constitute a way by which CB1 receptor blockade could act in preventing obesity. However, the interaction of the cannabinoid system with the leptin pathways remains to be clarified.

Interestingly, the inactivation of CB1 receptors results in low plasma insulin levels. Insulin sensitivity of mutant mice was similar to that of wild-type mice when animals are maintained with the standard diet. However, the inactivation of CB1 receptors prevents the decrease of insulin sensitivity induced by high-energy feeding observed in wild-type animals. This finding may be related to the low adiposity of CB1−/− mice fed the high-energy diet, as insulin sensitivity was shown to correlate with the level of fat storage.31,32,33

A previous study15 showed that treatment of CB1−/− mice with the CB1 receptor antagonist SR141716 did not induce any change in body weight and food intake. The lack of effect of the compound is confirmed by the unchanged adiposity in treated CB1−/− mice. In addition to its effect on body weight and food intake, SR141716 was shown to reduce plasma insulin levels.15 We demonstrate that plasma insulin remains unchanged in treated knockout mice, while this parameter was sharply decreased by SR141716 in wild-type mice.

In conclusion, we demonstrate that CB1 receptor knockout mice are lean and resistant to diet-induced obesity. Genetic inactivation of CB1 receptors seems to reproduce closely the effects of a pharmacological blockade of these receptors in mice.15 These findings support a major implication of CB1 receptors in the regulation of both food intake and body weight, and suggest that the endocannabinoid effects probably involve leptin-regulated pathways. Therefore, this target may constitute a new efficient approach to the treatment and/or prevention of human obesity.


  1. 1

    Walker JM, Hohmann AG, Martin WJ, Strangman NM, Huang SM, Tsou K . The neurobiology of cannabinoid analgesia. Life Sci 1999; 65: 665–673.

  2. 2

    Martin M, Ledent C, Parmentier M, Maldonado R, Valverde O . Involvement of CB1 cannabinoid receptors in emotional behaviour. Psychopharmacology 2002; 159: 379–387.

  3. 3

    Terranova JP, Michaud JC, Le Fur G, Soubrié P . Inhibition of long-term potentiation in rat hippocampal slices anandamide and WIN55212-2: reversal by SR141716A, a selective antagonist of CB1 cannabinoid receptors. Naunyn Schmiedebergs Arch Pharmacol 1995; 352: 576–579.

  4. 4

    Kirkham TC, Williams CM . Endogenous cannabinoids and appetite. Nutr Res Rev 2001; 14: 65–86.

  5. 5

    Koch JE . 9-THC stimulates food intake in Lewis rats. Effects on chow, high-fat and sweet high-fat diets. Pharmacol Biochem Behav 2001; 68: 539–543.

  6. 6

    Williams CM, Kirkham TC . Observational analysis of feeding induced by Δ9-THC and anandamide. Physiol Behav 2002; 76: 241–250.

  7. 7

    Williams CM, Kirkham TC . Anandamide induces overeating: mediation by central cannabinoid (CB1) receptors. Psychopharmacology 1999; 143: 315–317.

  8. 8

    Kirkham TC, Williams CM, Fezza F, Di Marzo V . Endocannabinoid levels in rat limbic forebrain and hypothalamus in relation to fasting, feeding and satiation: stimulation of eating by 2-arachidonoyl glycerol. Br J Pharmacol 2002; 136: 550–557.

  9. 9

    Rinaldi-Carmona M, Barth F, Héaulme M, Shire D, Calandra B, Congy C, Martinez S, Maruani J, Néliat G, Caput D, Ferrara P, Soubrié P, Brelière JC, Le Fur G . SR141716A, a potent and selective antagonist of the brain cannabinoid receptor. FEBS Lett 1994; 350: 240–244.

  10. 10

    Williams CM, Kirkham TC . Reversal of Δ9-THC hyperphagia by SR141716 and naloxone but not dexfenfluramine. Pharmacol Biochem Behav 2002; 71: 341–348.

  11. 11

    Colombo G, Agabio R, Diaz G . Appetite suppression and weight loss after the cannabinoid antagonist SR141716. Life Sci 1998; 63: 113–117.

  12. 12

    Rowland NE, Mukherjee M, Robertson K . Effects of the cannabinoid receptor antagonist SR141716, alone and in combination with dexfenfluramine or naloxone, on food intake in rats. Psychopharmacology 2001; 159: 111–116.

  13. 13

    Arnone M, Jung M, Keane PE, Maffrand JP, Soubrié P . The CB1 cannabinoid receptor antagonist SR141716 reduces sucrose intake and fat diet preference in lean and obese rats. Int J Obes Relat Metab Disord 1999; 23 (Suppl 5): S63.

  14. 14

    Simiand J, Keane M, Keane PE, Soubrié P . SR141716, a CB1 cannabinoid receptor antagonist, selectively reduces sweet food intake in marmoset. Behav Pharmacol 1998; 9: 179–181.

  15. 15

    Ravinet Trillou C, Arnone M, Delgorge C, Gonalons N, Keane P, Maffrand JP, Soubrié P . Anti-obesity effect of SR141716, a CB1 receptor antagonist, in diet-induced obese mice. Am J Physiol Regul Integr Comp Physiol 2003; 284: R345–R353.

  16. 16

    Di Marzo V, Goparaju SK, Wang L, Liu J, Bátkai S, Járai Z, Fezza F, Miuras GI, Palmiter RD, Sugiura T, Kunos G . Leptin-regulated endocannabinoids are involved in maintaining food intake. Nature 2001; 410: 822–825.

  17. 17

    Hildebrandt AL, Kelly-Sullivan DM, Black SC . Antiobesity effects of chronic cannabinoid CB1 receptor antagonist treatment in diet-induced obese mice. Eur J Pharmacol 2003; 462 (1–3): 125–132.

  18. 18

    Cota D, Giovanni M, Tschöp M, Grübler Y, Flachskamm C, Schubert M, Auer D, Yassouridis A, Thöne-Reineke C, Ortmann S, Tomessoni F, Cervino C, Nisoli E, Linthorst AG, Pasquali R, Lutz B, Stella GK, Pagotto U . The endogenous cannabinoid system affects energy balance via central orexigenic drive and peripheral lipogenesis. J Clin Invest 2003; 112: 423–431.

  19. 19

    Robbe D, Kopf M, Remaury A, Bockaert J, Manzoni OJ . Endogenous cannabinoids mediate long-term synaptic depression in the nucleus accumbens. Proc Natl Acad Sci 2002; 99: 8384–8388.

  20. 20

    Dickinson K, North TJ, Telford G, Smith S, Brammer R, Jones RB, Heal DJ . Determination of body composition in conscious adult female Wistar utilising total body electrical conductivity. Physiol Behav 2001; 74: 425–433.

  21. 21

    Vickers SP, Webster LJ, Wyatt A, Dourish CT, Kennett GA . Preferential effects of the cannabinoid CB1 receptor antagonist, SR141716, on food intake and body weight gain of obese (fa/fa) compared to lean Zucker rats. Psychopharmacology 2003; 167: 103–111.

  22. 22

    West DB, Boozer CN, Moody DL, Atkinson RL . Dietary obesity in nine inbred mouse strains. Am J Physiol Regul Integr Comp Physiol 1992; 262: R1025–R1032.

  23. 23

    Kirkham TC, Williams CM . The cannabinoid receptor antagonist SR141716 fails to suppress sucrose sham feeding. J Psychopharmacol 1998; 12 (Suppl A): 37.

  24. 24

    Gallate JE, McGregor IS . The motivation for beer in rats: effects of ritanserin, naloxone and SR141716. Psychopharmacology 1999; 142: 302–308.

  25. 25

    Gallate JE, Saharov T, Mallet PE, McGregor IS . Increase motivation for beer in rats following administration of a cannabinoid CB1 receptor agonist. Eur J Pharmacol 1999; 370: 233–240.

  26. 26

    Bensaid M, Gary-Bobo M, Esclangon A, Maffrand JP, Le Fur G, Oury-Donat F, Soubrié P . The cannabinoid CB1 receptor antagonist SR141716 increases Acrp30 mRNA expression in adipose tissue of obese fa/fa rats and in cultured adipocyte cells. Mol Pharmacol 2003; 63: 908–914.

  27. 27

    Fruebis J, Tsao TS, Javorschi S, Ebbets-Reed D, Erickson MR, Yen FT, Bihain BE, Lodish HF . Proteolytic cleavage product of 30-kDa adipocyte complement-related protein increases fatty acid oxidation in muscle and causes weight loss in mice. Proc Natl Acad Sci 2001; 98: 2005–2010.

  28. 28

    Shimada M, Tritos NA, Lowell BB, Flier JS, Maratos-Flier E . Mice laking melanin-concentrating hormone are hypophagic and lean. Nature 1998; 396: 670–674.

  29. 29

    Erickson JC, Clegg KE, Palmiter RD . Sensitivity to leptin and susceptibility to seizures of mice lacking neuropeptide Y. Nature 1996; 381: 415–418.

  30. 30

    Hollopeter G, Erickson JC, Seeley RJ, Marsh DJ, Palmiter RD . Response of neuropeptide Y-deficient mice to feeding effectors. Regulat Pept 1998; 75–76: 383–389.

  31. 31

    Kim JY, Nolte LA, Hansen PA, Han DH, Ferguson K, Thompson PA, Holloszy JO . High-fat diet-induced muscle insulin resistance: relationship to visceral fat mass. Am J Physiol Regul Integr Comp Physiol 2000; 279: R2057–R2065.

  32. 32

    Lemonnier D, Suquet JP, Aubert R, De Gasquet P, Pequignot E . Metabolism of the mouse made obese by a high-fat diet. Diabetes Metab 1975; 1: 77–85.

  33. 33

    Sindelka G, Skrha J, Prazny M, Haas T . Association of obesity, diabetes, serum lipids and blood pressure regulates insulin action. Physiol Res 2002; 51: 85–91.

Download references


We thank N Boussac for help with the statistical analyses and C Dumontet for preparation of the manuscript.

Author information

Correspondence to C Ravinet Trillou.

Rights and permissions

Reprints and Permissions

About this article


  • endocannabinoid
  • food intake
  • knockout
  • leptin
  • body weight

Further reading