Supersensitivity to anandamide and enhanced endogenous cannabinoid signaling in mice lacking fatty acid amide hydrolase

Abstract

The medicinal properties of marijuana have been recognized for centuries, but clinical and societal acceptance of this drug of abuse as a potential therapeutic agent remains fiercely debated. An attractive alternative to marijuana-based therapeutics would be to target the molecular pathways that mediate the effects of this drug. To date, these neural signaling pathways have been shown to comprise a cannabinoid receptor (CB(1)) that binds the active constituent of marijuana, tetrahydrocannabinol (THC), and a postulated endogenous CB(1) ligand anandamide. Although anandamide binds and activates the CB(1) receptor in vitro, this compound induces only weak and transient cannabinoid behavioral effects in vivo, possibly a result of its rapid catabolism. Here we show that mice lacking the enzyme fatty acid amide hydrolase (FAAH(-/-)) are severely impaired in their ability to degrade anandamide and when treated with this compound, exhibit an array of intense CB(1)-dependent behavioral responses, including hypomotility, analgesia, catalepsy, and hypothermia. FAAH(-/-)-mice possess 15-fold augmented endogenous brain levels of anandamide and display reduced pain sensation that is reversed by the CB(1) antagonist SR141716A. Collectively, these results indicate that FAAH is a key regulator of anandamide signaling in vivo, setting an endogenous cannabinoid tone that modulates pain perception. FAAH may therefore represent an attractive pharmaceutical target for the treatment of pain and neuropsychiatric disorders.

Figure 1

Generation and biochemical characterization of FAAH^−/− mice. (A) The genomic structure surrounding the deleted FAAH exon 1 (E1). Only relevant restriction sites are designated. The deleted E1 exon encodes amino acids 1–65 of the FAAH protein. (B) Southern blot analysis of EcoRV-digested genomic DNA by using the indicated probe (External probe in A), where 4.5- and 14-kb bands correspond to FAAH^−/− and FAAH^+/+ genotypes, respectively. (C) Western blot analysis of tissues from FAAH^+/+, FAAH^+/−, and FAAH^−/− mice demonstrating the selective absence of FAAH protein in FAAH^−/− animals. (D) Confocal microscopy immunofluorescence images of cerebellar sections of FAAH^+/+ (Left) and FAAH^−/− (Right) mice. Green signal, anti-FAAH; red signal, propidium iodide (stains nuclei). Arrowheads highlight intense FAAH immunoreactivity in the cell bodies of Purkinje neurons (Left).

Figure 2

Pharmacological activity of anandamide in FAAH^+/+ and FAAH^−/− mice. Anandamide elicited dose-dependent pharmacological effects in FAAH^−/− mice (■), but failed to produce any significant effects in FAAH^+/+ mice (○). (A) Hypomotility (locomotor activity), ED₅₀ < 6.25 mg/kg; (B) antinociception (tail-immersion), ED₅₀ [95% confidence limits (CL)] = 13 (5–30) mg/kg; (C) catalepsy, ED₅₀ (95% CL) = 20 (11–35) mg/kg; (D) hypothermia (rectal temperature), ED₅₀ (95% CL) = 11 (6–19) mg/kg. ★, P < 0.05; ★★, P < 0.01; and ★★★, P < 0.001, for FAAH^−/− versus FAAH^+/+ mice receiving the same treatment (planned comparison). *, P < 0.05 and **, P < 0.01 for anandamide-treated versus vehicle-treated FAAH^−/− mice (Dunnett's test). The results are presented as means ± SE. n = 6–8 mice per group.

Figure 3

Time course of the hypothermia (A) and catalepsy (B) in mice treated with either vehicle (FAAH^+/+, open circles; FAAH^−/−, filled circles) or 50 mg/kg anandamide (□, FAAH^+/+; ■ FAAH^−/−). ***, P < 0.001 for anandamide-treated FAAH^−/− mice versus the other three test groups (Scheffé test). The results are presented as means ± SE. n = 6–8 mice per group.

Figure 4

The behavioral effects of anandamide in FAAH^−/− mice are mediated by the CB₁ cannabinoid receptor. The effect of vehicle (open bars) or SR141716A (filled bars) administered 10 min before treatment with anandamide in FAAH^−/− mice. SR141716A (10 mg/kg) completely blocked the hypomotility (A), antinociception (B), catalepsy (C), and hypothermia (D) induced by anandamide (50 mg/kg), as SR141716A-pretreated, anandamide-treated FAAH^−/− mice were indistinguishable in all behavioral assays from FAAH^−/− mice treated with vehicle alone (0 mg/kg anandamide data in Fig. 2 A–D). **, P < 0.01 and ***, P < 0.001 for SR141716A-treated versus vehicle-treated FAAH^−/− mice (planned comparison). The results are presented as means ± SE. n = 6–8 mice per group.

Figure 5

Altered pain responses in FAAH^−/− mice. (A) FAAH^−/− mice (filled columns) exhibited prolonged response latencies in both the tail-immersion (n = 48, 50, and 62 for FAAH^+/+, FAAH^+/−, and FAAH^−/− genotypes, respectively) and hot plate (n = 12–15 mice per group) tests for thermal pain sensation relative to FAAH^+/+ (open columns) and FAAH^+/− mice (hatched bars; for hot plate data with FAAH^+/− mice, see Fig. 6B Left). (B) Duration of licking during the early phase of the formalin test was reduced in FAAH^−/− mice (filled columns) relative to FAAH^+/− (hatched columns) and FAAH^+/+ (open columns) mice; n = 8–10 mice per group. **, P < 0.01 and ***, P < 0.001 for FAAH^−/− versus FAAH^+/− or FAAH^+/+ mice (planned comparisons). The results are presented as means ± SE.

Figure 6

Enhanced endogenous cannabinoid levels and activity in FAAH^−/− mice. (A) FAAH^−/− mice (filled column) possessed greatly increased endogenous brain levels of anandamide relative to FAAH^+/+ mice (open bar); FAAH^+/+, 50 ± 10 pmol/g of tissue; FAAH^−/−, 775 ± 113 pmol/g of tissue; ***, P < 0.001 for FAAH^−/− versus FAAH^+/+ mice (planned comparison); n = 7–8 mice per group. (B) The prolonged response latency of FAAH^−/− mice in the hot plate assay (Left, filled bar) was reversed by treatment with SR141716A (Right, filled bar). Vehicle administration failed to significantly affect the response latencies of FAAH^−/−, FAAH^+/−, and FAAH^+/+ mice (Right, open bars), and SR141716A failed to significantly affect the response latencies of FAAH^+/− and FAAH^+/+ mice (Right, filled bars). n = 8–11 mice per group, assayed 30 min posttreatment. **, P ≤ 0.01 for SR141716A-treated FAAH^−/− mice versus either vehicle-treated FAAH^−/− mice or baseline latencies (planned comparison); ***, P < 0.005, for FAAH^−/− versus FAAH^+/− or FAAH^+/+ mice (planned comparison). The results are presented as means ± SE.