Involvement of the ghrelin system in the maintenance of oxycodone self-administration: converging evidence from endocrine, pharmacologic and transgenic approaches

Abstract

Ghrelin, an orexigenic hormone, has emerged as a critical biological substrate implicated in drug reward. However, the response of the ghrelin system to opioid-motivated behaviors and the role of ghrelin in oxycodone self-administration remain to be studied. Here, we investigated the reciprocal interactions between the endogenous ghrelin system and oxycodone self-administration behaviors in rats and the role of the ghrelin system in brain stimulation reward (BSR) driven by optogenetic stimulation of midbrain reward circuits in mice. Oxycodone self-administration significantly elevated plasma ghrelin, des-acyl ghrelin and growth hormone and showed no effect on plasma LEAP2, a newly identified endogenous ghrelin receptor (GHS-R1a) antagonist. Oxycodone self-administration produced significant decreases in plasma gastric inhibitory polypeptide and insulin. Acquisition of oxycodone self-administration significantly upregulated GHS-R1a mRNA levels in dopamine neurons in the ventral tegmental area (VTA), a brain region critical in drug reward. Pretreatment with JMV2959, a selective GHS-R1a antagonist, dose-dependently reduced oxycodone self-administration and decreased the breakpoint for oxycodone under a progressive ratio reinforcement in Long-Evans rats. The inhibitory effects of JMV2959 on oxycodone self-administration is selectively mediated by GHS-R1a as JMV2959 showed a similar effect in Wistar wildtype but not in GHS-R knockout rats. JMV2959 pretreatment significantly inhibited BSR driven by selective stimulation of VTA dopamine neurons, but not by stimulation of striatal GABA neurons projecting to the VTA in mice. These findings suggest that elevation of ghrelin signaling by oxycodone or oxycodone-associated stimuli is a causal process by which oxycodone motivates oxycodone drug-taking and targeting the ghrelin system may be a viable treatment approach for opioid use disorders.

Figure 1:

Plasma ghrelin, DAG, LEAP2, growth hormone, GIP and insulin levels as a function of oxycodone self-administration. Ghrelin, DAG and growth hormone levels significantly increased while GIP and insulin levels significantly decreased following oxycodone self-administration. Acquisition of oxycodone self-administration also suppressed basal insulin levels measured immediately before self-administration (0 hr). No differences in LEAP2 levels were found. *P<0.05, **P<0.01 compared to pre-session levels (0 hr), respectively. +P<0.05 compared to basal levels in the saline-exposed group. n=8 in each group.

Figure 2:

Effects of acquisition of oxycodone self-administration on GHS-R1a mRNA expression in the VTA. A-D, representative photomicrographs of VTA slice sections processed by triple-labeled RNAscope in situ hybridization for GHS-R1a, DAT and vGAT and visualized at 60 × magnification. E-F, quantitative analysis of GHS-R1a mRNA signals in DAT and vGAT positive neurons. Arrows in A-D indicate GHS-R1a mRNA puncta and clusters and their colocalization with DAT and vGAT. Open circles in E-F show the individual values in each group. *p<0.05 compared to the saline-exposed group. n=4 in each group.

Figure 3:

Impact of GHS-R1a blockade by JMV2959 on oxycodone self-administration in Long-Evans rats (A, B) and in GHS-R1a KO and wild type (WT) Wistar rats (C, D). JMV2959 dose-dependently decreased oxycodone infusions as compared to both prior session baseline and to the saline-treated group (A). The effects of JMV2959 on oxycodone self-administration were also seen in the WT Wistar rats while they were absent in the KO Wistar rats (C). JMV2959 showed no effects on animals’ responding on the inactive lever (B and D). **P<0.01 compared to the saline group. ++P<0.01 compared to prior session basal levels. n=7 in saline and 2.5 mg/kg groups and n=8 in 1 and 5 mg/kg group of Long-Evans rats; n=8 for both Wistar WT and KO rats.

Figure 4:

Effects of GHS-R1a blockade by JMV2959 on oxycodone self-administration tested under a low oxycodone unit dose (12.5 μg/kg/infusion) and under a PR schedule of reinforcement. JMV2959 dose-dependently decreased oxycodone infusions as compared to both prior session baseline and saline-treated group (A). JMV2959 dose-dependently decreased PR breakpoint measured by either active lever-presses (C) or oxycodone infusions (D). JMV2959 showed no effects on animals’ responding on the inactive lever tested under either FR (B) or PR (data not shown) reinforcement. **P<0.01 compared to saline group. +P<0.05, ++P<0.01 compared to prior session basal levels, respectively. n=7 in each group.

Figure 5:

Impact of GHS-R1a blockade by JMV2959 on BSR maintained by optogenetic self-stimulation of VTA dopamine neurons in DAT-Cre mice and NAS GABA neurons to VTA in vGAT-Cre mice. A, D: Schematic diagrams of the AAV-ChR2-eYFP microinjections into VTA (A) and NAS (D) and intracranial optical fiber implantation into VTA of DAT-Cre (A) and vGAT-Cre (D) mice. B: Immunostaining of brain slice indicating the expression of the injected AAV-ChR2-EYFP in DAT-Cre mice VTA. E, F: Immunostaining of brain slices indicating the expression of the injected AAV-ChR2-EYFP in vGAT-Cre mice NAS and VTA, respectively. Note that VTA injection of AAV-ChR2-EYFP in DAT-Cre mice (B, yellow), but not NAS injection in vGAT-Cre mice (F, red) induced AAV-ChR2-EYFP expression in VTA DA neurons. C: JMV2959 pretreatment (1–5 mg/kg) dose-dependently inhibited active lever-presses maintained by stimulation of VTA DA neurons in DAT-Cre mice as indicated by the rightward shift of response-frequency curve. G: JMV2959 (5–10 mg/kg) showed no effects on BSR maintained by stimulation of NAS-VTA GABA neurons in vGAT-Cre mice. *P<0.05, **P<0.01, as compared to saline pretreatment. n=8 in each treatment.