Correlates of motivation and reward

精神兴奋剂能引起食欲和厌恶行为的看似矛盾的混合。为了研究这些行为的神经基础，我们给清醒行为的大鼠全身注射安非他明，同时从外侧链（LHb）进行单单元记录，外侧链是一个先前被证明编码负奖励预测错误的区域。在人群水平上，LHb单位被安非他明注射（1mg/kg和3mg/kg， i.p）以剂量依赖的方式抑制。进一步检查显示X/Y （?? ?）%)的LHb单位被抑制（XX%），而另一半没有反应，或在少数情况下，弱兴奋。这些结果与脑切片中细胞内记录的LHb细胞的结果非常相似。多巴胺（DA, 50uM）对29/49（59.2%）个神经元有抑制作用，这一作用被D2拮抗剂eticlopride （200nM）完全阻断，而喹匹罗对7/13（53.8%）个神经元有抑制作用。其余的神经元要么对DA没有反应，要么轻微兴奋，与体内结果相似。此外，安非他明（n = 5,10 um）导致动作电位放电频率降低66%，这与保持电流变化50pA相关。

Psychostimulant drugs can elicit seemingly paradoxical mixtures of appetitive and aversive behaviors. To investigate the neural bases of these behaviors, we gave awake behaving rats systemic injections of amphetamine while making single-unit recordings from the lateral habenula (LHb), a region previously shown to encode negative reward prediction errors. At a population level, LHb units were inhibited in a dose-dependent manner by amphetamine injections (1mg/kg and 3mg/kg, i.p.). Closer examination showed that X/Y (??%) of LHb units were inhibited (XX%), while the other half showed no response or, in a few cases, weak excitation. These results closely parallel those from intracellularly recorded LHb cells in brain slices. Dopamine (DA, 50uM) application inhibited 29/49 (59.2%) neurons, an effect that was completely blocked by the D2 antagonist eticlopride (200nM), while quinpirole inhibited 7/13 (53.8%) cells. The remaining neurons either did not respond to DA or were slightly excited, similar to the in vivo results. Further, bath applied amphetamine (n = 5, 10uM) resulted in a 66% reduction in action potential firing frequency, which correlated with a 50pA change in holding current. In awake behaving rats, inhibition of LHb units peaked at 15 minutes post-injection, and persisted for nearly an hour. Interestingly, a subset of LHb neurons show compensatory rebound excitations persisting for much longer, in some cases up to 3 hours after injection. These rebound excitations occur at a time when the rewarding effects of amphetamine (as measured by conditioned place preference) are greatly attenuated, possibly reflecting early stages of drug withdrawal. We hypothesize that this delayed excitation reflects a compensatory reaction that contributes to the subjective properties of drug withdrawal. Furthermore, because LHb stimulation inhibits a large percentage of dopamine neurons, presumably via GABAergic neurons in the rostromedial tegmentum (RMTg), our findings suggest the possibility that dopamine neurons and LHb neurons are reciprocally inhibitory, although further study is needed to test this hypothesis.