Brain region-specific neural activation by low-dose opioid promotes social behavior

Abstract

The opioid system plays crucial roles in modulating social behaviors in both humans and animals. However, the pharmacological profiles of opioids regarding social behavior and their therapeutic potential remain unclear. Multiple pharmacological, behavioral, and immunohistological c-Fos mapping approaches were used to characterize the effects of μ-opioid receptor agonists on social behavior and investigate the mechanisms in naive mice and autism spectrum disorder-like (ASD-like) mouse models, such as prenatally valproic acid-treated mice and Fmr1-KO mice. Here, we report that low-dose morphine, a μ-opioid receptor agonist, promoted social behavior by selectively activating neurons in prosocial brain regions, including the nucleus accumbens, but not those in the dorsomedial periaqueductal gray (dmPAG), which are only activated by analgesic high-dose morphine. Critically, intra-dmPAG morphine injection counteracted the prosocial effect of low-dose morphine, suggesting that dmPAG neural activation suppresses social behavior. Moreover, buprenorphine, a μ-opioid receptor partial agonist with less abuse liability and a well-established safety profile, ameliorated social behavior deficits in two mouse models recapitulating ASD symptoms by selectively activating prosocial brain regions without dmPAG neural activation. Our findings highlight the therapeutic potential of brain region-specific neural activation induced by low-dose opioids for social behavior deficits in ASD.

Keywords: Behavior; Neurodevelopment; Neuroscience; Pharmacology; Therapeutics.

Figure 1. Morphine increased social behaviors in mice only at low doses.

(A–D) Effects of systemic administration of morphine on social behaviors in the single-chamber social interaction test (SIT). Single-housed C57BL/6J mice were s.c. injected with morphine (0.03 or 5 mg/kg) or vehicle at 12–13 weeks of age. The time spent in the interaction zone (A and C) and the avoidance zone (B and D) during a 3-minute test period with an unfamiliar mouse kept in a cage (target session) is shown. n = 21–22 (A and B) and 11 animals (C and D). (E) Effects of systemic administration of morphine on social interaction deficits in mice prenatally exposed to valproic acid (VPA) in the reciprocal SIT. VPA (500 mg/kg) or saline was intraperitoneally injected into pregnant CD-1 mice at embryonic day 12.5. Male offspring at 8 weeks of age were s.c. injected with morphine (0.03, 0.1, 0.3, 1, 3 mg/kg) or vehicle, and the duration of sniffing was measured during a 20-minute test period. n = 5–13 animals. (F) Effects of morphine on thermal nociception in prenatally VPA-treated mice. The withdrawal latency at 49°C was measured with the hot-plate test at 1 hour after administration of morphine (0.1, 1 mg/kg, s.c.). n = 12 animals. Data are shown as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, by the parametric tests (A, D, and E) or nonparametric ones (B, C, and F) compared with vehicle-treated mice. ^##P < 0.01 compared with mice prenatally exposed to saline (control). Mor, morphine.

Figure 2. Brain area–specific activation by systemic morphine administration in mice.

Morphine (0.03 or 5 mg/kg) or vehicle was s.c. administered to naive C57BL/6J mice 1.5 hours before sampling the brain. (A) Representative images of c-Fos–immunostained brain sections of the NAc, mPFC, VTA, and dorsal PAG with the brain atlas map (31). Scale bar: 100 μm. (B–G) Quantification of c-Fos–positive cells in each brain region. Results from the NAc (B, 17–18 sections/group from 4 mice/group), mPFC (C, 23–28 sections/group from 4 mice/group), VTA (D, 18–26 sections/group from 4 mice/group), dmPAG (E, 32–43 sections/group from 4 mice/group), dlPAG (F, 29–42 sections/group from 4 mice/group), and lPAG (G, 32–42 sections/group from 4 mice/group) are shown. Data are shown as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.01, ****P < 0.0001, by the parametric tests (C, F, and G) or nonparametric ones (B, D, and E) compared with vehicle-treated mice. Mor, morphine; NAc, nucleus accumbens; mPFC, medial prefrontal cortex; VTA, ventral tegmental area; dmPAG, dorsomedial periaqueductal gray; dlPAG, dorsolateral periaqueductal gray; lPAG, lateral periaqueductal gray.

Figure 3. Social behavior induced by morphine was antagonized by topical administration of morphine into the dorsal PAG.

(A) Timeline of the experiment. (B) Schematic of cannula implantation into the dorsomedial part of the PAG, and a representative image of mouse brain sections obtained after the experiment. Scale bar: 500 μm. (C and D) Effects of intra-PAG administration of morphine on social behaviors in the single-chamber social interaction test (SIT). Single-housed C57BL/6J mice were s.c. injected with morphine (0.03 mg/kg) or vehicle 1 hour before the test. Immediately before the test, morphine (5 μg) or vehicle was infused through the cannula (see Supplemental Methods). The time spent in the interaction zone (C) and the avoidance zone (D) during a 3-minute test period with an unfamiliar mouse kept in a cage (target session) is shown. n = 7–8 animals. *P<0.05, **P < 0.01, ****P < 0.0001, by parametric (C) and nonparametric (D) tests compared with the group treated with systemic morphine 0.03 mg/kg plus vehicle intra-PAG infusion. PAG, periaqueductal gray; Mor, morphine.

Figure 4. Low-dose buprenorphine increased social behaviors in VPA-treated and Fmr1-KO mice.

(A) Effects of systemic buprenorphine administration on social behavior deficits in mice prenatally exposed to valproic acid (VPA) in the reciprocal social interaction test. Buprenorphine (1, 3, 10, or 30 μg/kg) or vehicle was s.c. injected into male offspring at 8 weeks of age. One hour after administration, the sniffing duration was measured during a 20-minute test period. (B) Effects of buprenorphine on thermal nociception in VPA-treated mice. The withdrawal latency at 49°C was measured with a hot-plate test 1 hour after buprenorphine administration (3 or 30 μg/kg, s.c.). (C and D) Effects of buprenorphine (3 or 30 μg/kg, s.c.) on social interaction deficits in VPA-treated mice were examined at 3 (C) or 12 hours (D) after administration. (E) The effects of buprenorphine (3 μg/kg, s.c.) on social interaction deficits were antagonized by naloxone (1 mg/kg, s.c.), an MOR antagonist, in VPA-treated mice. (F) Effects of systemic buprenorphine administration (3 μg/kg, s.c.) on disrupted social preference in Fmr1-KO mice. The time spent in the zone with or without an unfamiliar mouse was measured during a 10-minute test period. n = 11–18 animals. Data are shown as the mean ± SEM. ^##P < 0.01, by parametric test (A) or nonparametric one (F) compared with vehicle-treated control mice prenatally exposed to saline (A) or vehicle-treated wild-type (C57BL/6J) mice (F). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, by parametric tests (A–C) or nonparametric ones (D–F) compared with vehicle-treated mice prenatally exposed to VPA (A–E) or vehicle-treated Fmr1-KO mice (F). Bup, buprenorphine; MOR, μ-opioid receptor; Fmr1, fragile X mental retardation 1.

Figure 5. Brain area–specific activation by systemic buprenorphine administration in VPA-treated mice.

Male offspring born to mothers injected with valproic acid (500 mg/kg, intraperitoneal) were subjected to immunohistochemical analysis at 8 weeks of age. Buprenorphine (3 or 30 μg/kg) or vehicle was s.c. administered 1.5 hours before sampling the brain. (A) Representative images of c-Fos–immunostained brain sections containing the NAc, mPFC, VTA, and dorsal PAG. Scale bar: 100 μm. (B–G) Quantification of c-Fos–positive cells in each brain region. The results in the NAc (B, 21–24 sections/group from 4 mice/group), mPFC (C, 8–22 sections/group from 3–7 mice/group), VTA (D, 12–13 sections/group from 2 mice/group), dmPAG (E, 29–40 sections/group from 6 mice/group), dlPAG (F, 29–40 sections/group from 6 mice/group), and lPAG (G, 29–40 sections/group from 6 mice/group) are shown. Data are shown as the mean ± SEM. *P < 0.05, ***P < 0.001, by parametric (B and D) or nonparametric (C and E–G) tests compared with vehicle-treated mice. Bup, buprenorphine; NAc, nucleus accumbens, mPFC, medial prefrontal cortex; VTA, ventral tegmental area; dmPAG, dorsomedial periaqueductal gray; dlPAG, dorsolateral periaqueductal gray; lPAG, lateral periaqueductal gray.