Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Nov 10;112(45):14084-9.
doi: 10.1073/pnas.1509795112. Epub 2015 Oct 26.

Endocannabinoid Signaling Mediates Oxytocin-Driven Social Reward

Affiliations
Free PMC article

Endocannabinoid Signaling Mediates Oxytocin-Driven Social Reward

Don Wei et al. Proc Natl Acad Sci U S A. .
Free PMC article

Abstract

Marijuana exerts profound effects on human social behavior, but the neural substrates underlying such effects are unknown. Here we report that social contact increases, whereas isolation decreases, the mobilization of the endogenous marijuana-like neurotransmitter, anandamide, in the mouse nucleus accumbens (NAc), a brain structure that regulates motivated behavior. Pharmacological and genetic experiments show that anandamide mobilization and consequent activation of CB1 cannabinoid receptors are necessary and sufficient to express the rewarding properties of social interactions, assessed using a socially conditioned place preference test. We further show that oxytocin, a neuropeptide that reinforces parental and social bonding, drives anandamide mobilization in the NAc. Pharmacological blockade of oxytocin receptors stops this response, whereas chemogenetic, site-selective activation of oxytocin neurons in the paraventricular nucleus of the hypothalamus stimulates it. Genetic or pharmacological interruption of anandamide degradation offsets the effects of oxytocin receptor blockade on both social place preference and cFos expression in the NAc. The results indicate that anandamide-mediated signaling at CB1 receptors, driven by oxytocin, controls social reward. Deficits in this signaling mechanism may contribute to social impairment in autism spectrum disorders and might offer an avenue to treat these conditions.

Keywords: anandamide; endocannabinoid; oxytocin; reward; social behavior.

Conflict of interest statement

Conflict of interest statement: D.P. is an inventor on patents protecting a compound (URB597) used in the present study.

Figures

Fig. 1.
Fig. 1.
Anandamide signaling at CB1 receptors mediates social reward. (A) Schematics of the experimental protocol. (BD) Levels of anandamide (AEA; B), 2-arachidonoyl-sn-glycerol (2-AG; C), and oleoylethanolamide (OEA; D) in nucleus accumbens (NAc), ventral-to-mid hippocampus (vHC), amygdala, dorsal striatum (dStr), and ventral midbrain (vMB) of isolated (Iso) or resocialized (Soc) C57Bl6 mice. (E) Schematics of the social conditioned place preference (sCPP) test. (F and G) sCPP in wild-type (wt) or faah−/− mice treated with vehicle (VEH) or CB1 inverse agonist AM251 (2 mg-kg−1, intraperitoneal). (H) sCPP in mice treated with FAAH inhibitor URB597 (intraperitoneal). Data are shown as means ± SEM; *P < 0.05, compared using Student’s unpaired t test, n = 4–5 (BD), n = 12–14 (F), two-way ANOVA with Tukey’s post hoc test, n = 12–14 (G), or one-way ANOVA with Bonferroni’s post hoc test, n = 10 (H).
Fig. S1.
Fig. S1.
Pictures of punches of tissue used for LC/MS-based lipid analysis. Each region starts with a picture of the coronal section before the punch was taken, and then progresses through what the punch looks like as sections were cut toward the posterior of the brain. Arrows indicate the nucleus accumbens (NAc; A), dorsal striatum (dStr; B), S2 cortex (C), dorsal hippocampus (dHC; D) and ventral-to-mid hippocampus (vHC; E). Arrowheads indicate the amygdala (C) and the piriform cortex (D). (F) Punches of the ventral midbrain, which likely includes the ventral tegmental area and the substantia nigra.
Fig. S2.
Fig. S2.
Survey of anandamide levels during socialization. Levels of anandamide (AEA) in isolated (Iso) or resocialized (Soc) C57Bl6 mice in the dorsal hippocampus (dHC), S2 cortex, and piriform (Pir) cortex. The experimental protocol followed the schematic in Fig. 1A. Data are shown as means ± SEM; n = 4–6.
Fig. S3.
Fig. S3.
Survey of 2-arachidonoyl-sn-glycerol levels during socialization. Levels of 2-arachidonoyl-sn-glycerol (2-AG) in isolated (Iso) or resocialized (Soc) C57Bl6 mice in the ventral-to-mid hippocampus (vHC), amygdala, dorsal striatum (dStr), dorsal hippocampus (dHC), S2 cortex, and piriform (Pir) cortex. The experimental protocol followed the schematic in Fig. 1A. Data are shown as means ± SEM; n = 5–6.
Fig. S4.
Fig. S4.
High-fat food or cocaine CPP is not overtly altered in faah−/− mice. (AC) Schematic of the experimental paradigm (A), high-fat chamber times (B), and high-fat chamber preference (C). (DF) Schematic of the experimental paradigm (D), cocaine chamber times (E), and cocaine chamber preference (F). Data are shown as means ± SEM; n = 8–10.
Fig. S5.
Fig. S5.
Anandamide enhancement does not change social approach performance. (A) Schematic of the experimental paradigm. (B) Chamber and sniffing times of wild-type (wt) or faah−/− mice. (C) Chamber and sniffing times of C57Bl6J mice treated with vehicle (VEH) or FAAH inhibitor URB597 (1 mg-kg−1, intraperitoneal). Data are shown as means ± SEM; n = 8–10.
Fig. S6.
Fig. S6.
Anandamide enhancement by the FAAH inhibitor URB597 administered during isolation in an inverted sCPP does not alter development of preference to the social environment. (A) Schematic of the experimental paradigm, in which isolation conditioning precedes socialization and is paired with URB597, in contrast to the normal sCPP procedure according to Fig. 1E. (B and C) Social chamber time and social preference of C57Bl6J mice treated with vehicle (VEH) or FAAH inhibitor URB597 (1 mg-kg−1, intraperitoneal) during isolate conditioning. Data are shown as means ± SEM; n = 10–12.
Fig. 2.
Fig. 2.
Oxytocin transmission enhances anandamide mobilization in the nucleus accumbens (NAc). (A) Anandamide (AEA) levels in the NAc of isolated (Iso) or resocialized (Soc) mice treated with vehicle (VEH) or oxytocin receptor antagonist L-368,899 (5 mg-kg−1, intraperitoneal). (B) AEA levels in the NAc of isolated mice treated with the oxytocin agonist WAY-267,464 (5 nmol, intracerebroventricular), with or without the oxytocin receptor antagonist L-368,899 (0.5 nmol, intracerebroventricular). (C and D) Levels of AEA (C) and 2-AG (D) in the NAc of isolated mice with virally directed, oxytocin promoter-restricted expression of DREADD hM3Dq receptors in the paraventricular nucleus of the hypothalamus and were treated with VEH or clozapine-N-oxide (CNO, 5 mg-kg−1, intraperitoneal). Data are shown as means ± SEM; *P < 0.05, **P < 0.01, compared using two-way ANOVA with Tukey’s post hoc test, n = 4–5 (A), one-way ANOVA with Bonferroni’s post hoc test, n = 4–5 (B), or Student’s unpaired t test, n = 4–5 (C and D).
Fig. S7.
Fig. S7.
Effects of oxytocin on anandamide mobilization in the ventral hippocampus (vHC). (A) Levels of anandamide (AEA) in the vHC of isolated mice treated with the oxytocin agonist WAY-267,464 (5 nmol, intracerebroventricular), with or without the oxytocin receptor antagonist L-368,899 (0.5 nmol, intracerebroventricular). (B) Levels of AEA in the vHC of isolated mice with virally directed, oxytocin promoter-restricted expression of DREADD hM3Dq receptors in the paraventricular nucleus of the hypothalamus and were treated with VEH or clozapine-N-oxide (CNO, 5 mg-kg−1, intraperitoneal). (C) Levels of AEA in the NAc of isolated (Iso) or resocialized (Soc) mice treated with vehicle (VEH) or oxytocin receptor antagonist L-368,899 (5 mg-kg−1, intraperitoneal). Data are shown as means ± SEM; *P < 0.05, compared using One-way ANOVA with Bonferroni’s post hoc test, n = 4–5 (A), or Student’s unpaired t test, n = 4–5 (B-C).
Fig. 3.
Fig. 3.
Anandamide mediates oxytocin-dependent social reward. (A and B) sCPP of wild-type (wt) or faah−/− mice treated with vehicle (VEH) or oxytocin receptor antagonist L-368,899 (5 mg-kg−1, intraperitoneal). (C and D) Representative images and normalized quantification of cFos immunolabeling after isolation or resocialization, following the protocol in Fig. 1A. (EH) cFos levels in socialized faah+/+ mice and faah−/− mice treated with or without the oxytocin receptor antagonist L-368,899 (5 mg-kg−1, intraperitoneal). Scale bar represents 0.1 mm. Dotted line delineates core (inside) and shell (outside). Data are shown as means ± SEM; *P < 0.05, compared using Student’s unpaired t test, n = 10–14 (A), n = 3–4 (C and D), n = 4–5 (EH), or two-way ANOVA with Tukey’s post hoc test, n = 10–14 (B).
Fig. S8.
Fig. S8.
Effects of oxytocin and anandamide on cFos expression in the ventral hippocampus. (A) Normalized quantification of cFos immunolabeling after isolation or resocialization, following the protocol in Fig. 1A. (B) cFos levels in socialized faah+/+ mice and faah−/− mice treated with or without the oxytocin receptor antagonist L-368,899 (5 mg-kg−1, intraperitoneal). Data are shown as means ± SEM; *P < 0.05, compared using Student’s unpaired t test, n = 3 (A), n = 4–5 (B).

Similar articles

See all similar articles

Cited by 41 articles

See all "Cited by" articles

Publication types

MeSH terms

Substances

Feedback