The endogenous cannabinoid system modulates nicotine reward and dependence

Abstract

A growing body of evidence suggests that the endogenous cannabinoid system modulates the addictive properties of nicotine, the main component of tobacco that produces rewarding effects. In our study, complementary transgenic and pharmacological approaches were used to test the hypothesis that the endocannabinoid system modulates nicotine reward and dependence. An acute injection of nicotine elicited normal analgesic and hypothermic effects in cannabinoid receptor (CB)(1) knockout (KO) mice and mice treated with the CB(1) antagonist rimonabant. However, disruption of CB(1) receptor signaling blocked nicotine reward, as assessed in the conditioned place preference (CPP) paradigm. In contrast, genetic deletion, or pharmacological inhibition of fatty acid amide hydrolase (FAAH), the enzyme responsible for catabolism of the endocannabinoid anandamide, enhanced the expression of nicotine CPP. Although the expression of spontaneous nicotine withdrawal (14 days, 24 mg/kg/day nicotine) was unaffected in CB(1) KO mice, acute administration of rimonabant (3 mg/kg) ameliorated somatic withdrawal signs in wild-type mice. Increasing endogenous levels of anandamide through genetic or pharmacological approaches exacerbated the physical somatic signs of spontaneous nicotine withdrawal in a milder withdrawal model (7 days, 24 mg/kg/day nicotine). Moreover, FAAH-compromised mice displayed increased conditioned place aversion in a mecamylamine-precipitated model of nicotine withdrawal. These findings indicate that endocannabinoids play a role in the rewarding properties of nicotine as well as nicotine dependence liability. Specifically, increasing endogenous cannabinoid levels magnifies, although disrupting CB(1) receptor signaling, attenuates nicotine reward and withdrawal. Taken together, these results support the hypothesis that cannabinoid receptor antagonists may offer therapeutic advantages to treat tobacco dependence.

Fig. 1

Disruption of CB₁ receptor signaling blocks nicotine place preference. a, CB₁ KO mice fail to exhibit normal nicotine CPP compared with wild-type mice at 0.5 and 0.7 mg/kg nicotine s.c. b, nicotine preference is observed in the control group given vehicle (i.p.) 30 min before nicotine 0.5 mg/kg s.c. Thirty-minute pretreatment with the CB₁ antagonist, rimonabant (1 and 3 mg/kg i.p.), blocks nicotine preference.*, p < 0.05 compared with saline controls (a and b). #, p < 0.05 compared with respective wild-type controls (b). Data are expressed as means ± S.E.M. of six to eight mice per group.

Fig. 2

FAAH deletion enhances nicotine preference. This figure shows nicotine CPP at different doses after s.c. administration in male and female FAAH KO and wild-type mice. a, FAAH KO mice exhibit an increase in 0.1 mg/kg nicotine CPP and do not show any further differences at higher doses of nicotine. b, enhancement of nicotine CPP (0.1 mg/kg) is blocked by pretreatment of rimonabant (3 mg/kg). *, p < 0.05 compared with respective saline controls. #, p < 0.05 compared with respective wild-type controls (a) or compared with rimonabant/nicotine treatment (b). Data are expressed as means ± S.E.M. of 10 to 12 mice per group. Sal, saline; Veh, vehicle; Nic, nicotine.

Fig. 3

URB597 enhances nicotine CPP in mice. Pretreatment with the FAAH enzyme inhibitor URB597 or vehicle, given 30 min before an active dose of 0.1 mg/kg nicotine s.c., significantly enhances nicotine preference at 0.3 and 3.0 mg/kg URB597 i.p. *, p < 0.05 compared with respective saline control and vehicle/nicotine control. #, p < 0.05 compared with all treatment groups. Data are expressed as means ± S.E.M. of 10 to 12 mice per group.

Fig. 4

Effects of acute nicotine administration in CB₁ KO mice. This figure displays dose-response curves for nicotine-induced antinociception in the tail-flick test (a), the hot-plate test (b), and hypothermia (c) in wild-type (-●-) and CB₁ KO (-□-) mice. Mice were treated with acute doses of nicotine (s.c.) and tested 5 min later for antinociception and 30 min after injection for body temperature measures. Data are expressed as means ± S.E.M. of six to eight mice per group.

Fig. 5

Rimonabant does not block acute nicotine responses in mice. Pretreatment with mecamylamine (1 mg/kg s.c., 10 min) significantly blocks acute nicotine-induced antinociception (2.5 mg/kg s.c.) as measured in the tail-flick (a) and hot-plate (b) tests in C57BL/6J mice. Rimonabant (3 mg/kg s.c.) given 30 min before acute nicotine does not alter the analgesic response of nicotine in the tail-flick (a) and hot-plate (b) tests. *, p < 0.05 compared with respective saline control. #, p < 0.05 compared with other nicotine groups. Data are expressed as means ± S.E.M. of six to eight mice per group (VEH, vehicle; SAL, saline; MEC, mecamylamine).

Fig. 6

Effects of CB₁ deletion and antagonism in nicotine withdrawal. a, male CB₁ KO and wild-type mice are implanted with saline or nicotine minipumps (24 mg/kg/day) for 14 days and then tested for somatic signs of spontaneous withdrawal. b, effects of an acute dose of vehicle or rimonabant (3 mg/kg) on the expression of somatic signs are shown in graph b. Minipumps are removed at day 14; 15 to 20 h later, mice received injections with vehicle or rimonabant. They are observed for somatic signs of withdrawal 10 min after injections. Rimonabant significantly decreases the total withdrawal score of somatic signs compared with the vehicle/nicotine control group. *, p < 0.05 compared with respective saline controls. #, p < 0.05 compared with nicotine controls. Data are expressed as means ± S.E.M., eight to 10 mice per group.

Fig. 7

Somatic signs of nicotine withdrawal in FAAH KO mice. FAAH KOs and wild-type controls receive saline or nicotine minipumps (24 mg/kg/day) for 7 days. Minipumps are removed at day 7; 15 to 20 h later, the mice are observed for somatic signs of withdrawal. a, FAAH KOs show a 2-fold increase in the expression of withdrawal signs, compared with wild type. b, FAAH KOs do not elicit new somatic signs but showed significant increases in paw tremors, head shakes, and body tremors. *, p < 0.05 compared with the saline controls. #, p < 0.05 compared with wild-type controls. Data are expressed as means ± S.E.M., 10 to 15 mice per group.

Fig. 8

Effects of FAAH inhibition by URB597 on somatic signs of nicotine withdrawal. Somatic signs are observed in spontaneous withdrawal model mice exposed to saline or nicotine minipumps (24 mg/kg/day) for 7 days. Minipumps are removed at day 7; 15 to 20 h later, mice received injections with vehicle or URB597 (3, 5, or 10 mg/kg). They are observed for somatic signs of withdrawal 10 min after injections. All nicotine groups shown are significantly higher than their saline controls. Groups treated with 3 and 5 mg/kg URB597 are not significantly different from the vehicle/nicotine control. *, p < 0.05 compared with respective saline controls. #, p < 0.05 compared with vehicle controls groups. Data are expressed as means ± S.E.M., 10 to 15 mice per group.

Fig. 9

FAAH inhibition and deletion potentiates the affective signs of aversion in nicotine withdrawal. Male mice are implanted with saline or nicotine minipumps (36 mg/kg/day) for 7 days and then conditioned with vehicle or mecamylamine (3.5 mg/kg) for 2 days. a, FAAH KO mice exhibit higher aversion in nicotine withdrawal compared with wild-type controls. b, C57BL/6J male mice treated with URB597 (10 mg/kg) 30 min before mecamylamine fail to enhance nicotine/mecamylamine controls. *, p < 0.05 compared with the saline controls (a and b). #, p < 0.05 compared with wild-type controls (b). Data are expressed as means ± S.E.M., five to 10 mice per group. SAL, saline; NIC, nicotine; M.P., minipump.