• Top page
  • Timetable
  • Per session
  • Per presentation
  • How to
  • Meeting Planner



Control of neural circuit function by the endocannabinoid 2-arachidonoylglycerol

開催日 2014/9/12
時間 15:00 - 17:00
会場 Room F(302)
Chairperson(s) 狩野 方伸 / Masanobu Kano (東京大学大学院医学系研究科 神経生理学分野 / Department of Neurophysiology, Graduate School of Medicine, The University of Tokyo, Japan)
渡辺 雅彦 / Masahiko Watanabe (北海道大学大学院医学研究科 解剖学講座 解剖発生学分野 / Department of Anatomy, Hokkaido University Graduate School of Medicine, Japan)

The role of endocannabinoid signaling in the striatum for reinforcement learning

  • S2-F-2-6
  • 上田 康雅 / Yasumasa Ueda:1 岡崎 哲 / Satoru R Okazaki:4 山中 航 / Ko Yamanaka:2 鮫島 和行 / Kazuyuki Samejima:2 榎本 一紀 / Kazuki Enomoto:2 狩野 方伸 / Masanobu Kano:3 木村 實 / Minoru Kimura:2 
  • 1:関西医科大学 / Kansai Medical University, Japan 2:玉川大学 脳科学研究所 / Tamagawa Univ. Brain Sci. Inst, Japan 3:東京大学・医学系研究科・生理学講座・神経生理学 / Dept. Neurophysiol., Univ. Tokyo, Japan 4:京都府立医科大学 / Kyoto Pref. Univ. Med, Japan 

Many studies suggested that basal ganglia are involved in, motor control, reinforcement learning, and reward-based decision making. The striatum, the first input channel of the basal ganglia, receive input from many brain areas. It has also been shown that presynaptic terminals of the striatal projection neurons have cannabinoid CB1 receptors which modulate transmitter release. However, little is known regarding whether or how the striatal CB1 receptors are involved in motor control, learning and decision making.
To address this issue, we trained two monkeys (Macaca fuscata) to perform reward-based adaptive choice task and pharmacologically blocked CB1 receptors in the striatum when two monkeys performing the task. Handle turns (left or right) were rewarded either with a large (0.2 ml) or a small amount (0.07 ml) of juice with pre-set probabilities of 10, 50 and 90%. The reward probabilities for both options were fixed for a block of 20-25 trials. At the beginning of each trial block, monkeys were given only a cue that only indicated the start of a new block without any information about reward probabilities associated with each option (left or right). Monkeys had to find the higher value option by trial-and-error manner to repeatedly receive a large reward during the new block. Before a block started, we pressure injected either a CB1 receptor antagonist (Rimonabant, 1.0-100 μ g/ μ l, 1.6 μ l) dissolved in dimethyl sulfoxide (DMSO) or DMSO only (1.6 μ l) as a control into the putamen by pressure injection.
We found that monkeys tended to choose a low value option after CB1 receptors antagonist injection when compared to vehicle-injection, even if they had repeatedly received a small reward repeatedly after choosing the same option.
These results suggest that endocannabinoid-CB1 signaling in the striatum is involved in reinforcement learning probably through regulating update of action values.

Copyright © Neuroscience2014. All Right Reserved.