查看原文
其他

Nature︱去甲肾上腺素在学习过程中的双重作用

Vincent BP 逻辑神经科学 2023-03-10






Dual role for noradrenaline during learning


AuthorVincent Breton-Provencher, Jiesi Feng

Editor-in-ChiefSizhen Wang

EditorBinwei Yang


Figure1: The noradrenergic system in human and mouse.

(Illustration by Vincent Breton-Provencher)


Noradrenaline in the brain is primarily produced by neurons of the locus coeruleus (LC) that are characterized by their highly divergent set of projections to cortical and subcortical areas(Figure 1). In learned behavior, LC-noradrenergic neurons were indirectly linked to distinct components of learned behavior, such as task execution (1) and optimization (2,3). Whether and how activation of noradrenaline-expressing neurons in the locus coeruleus facilitates different components of specific behaviours is unknown. One way LC-noradrenergic activity could mediate multiple functions is suggested by recent evidence of spatial modularity within the LC-noradrenergic neuronal population (4,5). Previous anatomical tracing studies indicated that the distribution of single LC-noradrenergic neurons could be target-specific. However, whether different LC-noradrenergic outputs signal different types of information, and whether behavioural roles for noradrenaline are refined through selective targeting of LC outputs remains unknown.


In a recent article, Breton-Provencher, Drummond, et al. from the lab of Professor Mriganka Sur at the Massachusetts Institute of Technology showed that LC-noradrenergic neurons support different function during learned behavior. Noradrenaline would thus influence both behavioral response to sensory stimuli associated with reward and behavioral optimization following erroneous response to stimuli not associated with reward. To support this role, they recorded from LC-noradrenergic neurons and showed that transient LC activation preceded behavioural execution and followed reinforcement. They also showed that these two components of LC-noradrenergic activity were heterogeneously represented in LC cortical outputs, such that the behavioural response signal was higher in the motor cortex and facilitated task execution, whereas the negative reinforcement signal was widely distributed among cortical regions and improved response sensitivity on the subsequent trial. Modular targeting of LC outputs thus enables diverse functions, whereby some noradrenaline signals are segregated among targets, whereas others are broadly distributed. This work was published in Nature on June 1, 2022.


The screenshot of the work published in Nature

(Adapted from the editor of Logical Neuroscience (逻辑神经科学))


To evaluate the distinct cognitive roles of LC-noradrenaline and measure its underlying activity, Breton-Provencher et al. designed a go/no-go task with graded auditory stimulus evidence and performance. Mice were trained to press a lever to a go tone to obtain a reward and hold still when they hear a no-go tone to avoid a punishment. They recorded the activity in LC-noradrenaline expressing neurons using electrophysiology combined with optogenetics. They demonstrated the temporal signatures of LC- noradrenergic neurons during learned behavior, characterized by a transient activation preceding execution (lever press) and following the reinforcement (Figure 2).


Figure 2: In a sensory-motor task, LC-noradrenergic neurons are transiently activated during the execution (lever press) and following a positive or negative reinforcement.

(Adapted from Breton-Provencher et al., Nature, 2022)


To understand the role of these two components of LC-noradrenergic activity, they used optogenetics inhibition of LC-noradrenaline neurons during the task. They analyzed the effect of inactivating noradrenaline on the execution for different tone intensities. Photoinhibition of LC-noradrenaline activity decreased the lever press probability for low tone intensities, suggesting that LC activity facilitates guessing behavior. They also analyzed the effect of reinforcement on animal’s subsequent behavior and the role of LC in this form of reinforcement learning. They found that following a punishment, the performance of a mouse on the next trial is improved, as seen by an increased lever press to the go-tone. Importantly, they did not observe this behavioral improvement when silencing LC-noradrenergic activity at the time of a punishment, indicating that LC-noradrenergic activity integrated reinforcement signals to increase performance accuracy on the subsequent trial.


Although neuronal activity in cortex has been linked to task execution and response bias the cellular mechanisms producing this activity are unknown. They therefore investigated how the heterogeneous activity at the level of LC neurons maps onto distinct LC-noradrenergic cortical outputs during our task to facilitate behavioural performance. To do so, they used axonal imaging of LC-noradrenergic axons in the motor and prefrontal cortices. They observed enhanced LC-noradrenergic axonal signals at the time of execution in the motor cortex in comparison to the prefrontal cortex (Figure 3). They also found that the LC reinforcement signal is distributed homogeneously in both the motor and prefrontal cortices.


Figure 3: Temporal (top) and spatial (bottom) dynamics of LC-noradrenergic during learned behavior. LC-noradrenergic signals to cortical outputs are targeted modularly to motor cortex during press and distributed focally or broadly following reward or punishment respectively. These distinct spatiotemporal dynamics facilitate task execution (lever movement) and serial response bias.

(Adapted from Breton-Provencher et al., Nature, 2022)


Critical to this finding was the collaboration with Jiesi Feng and Yulong Li at Peking University. They created and optimized a novel genetically encoded GPCR Activation Based fluorescent sensor for NE (GRABNE2m) to further validate the results by comparing axonal calcium activity with the signal of GRABNE. They found that LC-noradrenergic axonal calcium signals reflect the underlying noradrenaline release in the cortex (Figure 4).


Figure 4: Top - Axonal calcium signal averaged for one LC-axon projecting to the motor cortex during false alarm and hit trials. Bottom: Average GRABNE signal during false alarm and hit trials measured in the motor cortex.

(Adapted from Breton-Provencher et al., Nature, 2022)


Conclusion & Discussion, Inspiration Expectation
Using a learned behaviour dependent onlocus coeruleus (LC)-noradrenergic activity, Breton-Provencher et al. demonstrate two concurrently encoded functions for the LC-noradrenergic system: task execution and performance optimization. Pre-execution activity promotes reward-seeking actions by providing the necessary gain modulation in target areas. Performance optimization is due to LC encoding of reinforcement, a role that has also been suggested for acetylcholine and serotonin, and dopamine. Thus, LC-noradrenergic activity would be part of a larger network involving several neuromodulators to facilitate outcome evaluation and implement reinforcement learning.

They also provide evidence that—at the level of LC-noradrenergic outputs—functional modularity exists and supports, at least partially, distinct aspects of learned behaviour. They demonstrate that the LC activity linked to task execution is projected heterogeneously to the cortex such that pre-movement noradrenaline release primarily targets motor regions, facilitating its role in behavioural execution, whereas the reinforcement signal produces broad neuromodulation that is probablyused simultaneously by several regions to bias subsequent behaviour.


Article: Breton-Provencher, V., Drummond, G.T., Feng, J.et al. Spatiotemporal dynamics of noradrenaline during learned behaviour. Nature (2022). https://doi.org/10.1038/s41586-022-04782-2


Vincent Breton-Provencher (left), Gabrielle T. Drummond (middle), Mriganka Sur (right).

(Photo credited from Vincent Breton-Provencher & Mriganka Sur group)


Job opportunity: Exceptionally motivated individual at the PhD or post-doctoral level are invited to apply to Vincent Breton-Provencher lab, email: vincent.breton-provencher@cervo.ulaval.ca , to study the role of noradrenaline and dopamine in learning and attention. Homepage of Vincent Breton-Provencher’s lab: http://www.vbplab.com 


人才招聘
【1】人才招聘︱“ 逻辑神经科学 ”诚聘文章解读/撰写岗位 ( 网络兼职, 在线办公)【2】“ 逻辑神经科学 ”诚聘副主编/编辑/运营岗位(在线办公)【3】“ 逻辑神经科学 ”诚聘编辑/运营岗位 ( 在线办公)
往期文章精选【1】Nat Neurosci︱刘胜/刘奕志/向孟清团队研究证实成年灵长类存在活跃的神经发生现象【2】Mol Psychiatry︱石云/杨建军/柳娜团队揭示谷氨酸受体缺陷引起人类攻击行为的新机制【3】Alzheimers Dement︱申勇团队提出中国阿尔兹海默病的血液诊断新标准【4】Nat Commun︱全舒/薛毅/吴斌团队合作揭示分子伴侣如何不消耗能量而释放底物的机制【5】Nature︱新靶点!冯国平团队揭示靶向丘脑回路拯救帕金森病小鼠运动和情绪缺陷【6】Sci Adv︱刘际团队揭示下丘脑GLP-1信号通过下行迷走神经环路调控血糖代谢的神经机制【7】STAR Protocols︱巫凌刚团队提出新型实时可视化方案观测活细胞内胞吞胞吐的膜动态变化【8】Nat Genet︱最新发现:自闭症表型异质性的遗传相关性【9】Nat Neurosci︱睡眠具有社会性:青少年睡眠与社会互动行为之间的发展联系【10】Frontiers Aging Neurosci︱褪黑素治疗帕金森病睡眠障碍
优质科研培训课程推荐【1】膜片钳与光遗传及钙成像技术研讨会 8月6-7日 腾讯会议【2】单细胞测序与空间转录组学数据分析研讨会(6月18-19日 腾讯在线会议)



References list


(1) Aston-Jones, G. & Cohen, J. D. An integrative theory of locus coeruleus-norepinephrine function: adaptive gain and optimal performance. Annu. Rev. Neurosci. 28, 403–450 (2005).

(2) Bouret, S. & Sara, S. J. Network reset: a simplified overarching theory of locus coeruleus noradrenaline function. Trends Neurosci. 28, 574–582 (2005)

(3) Yu, A. J. & Dayan, P. Uncertainty, neuromodulation, and attention. Neuron 46, 681–692 (2005)

(4) Poe, G. R. et al. Locus coeruleus: a new look at the blue spot. Nat. Rev. Neurosci. 21, 644–659 (2020)

(5) Breton-Provencher, V., Drummond, G. T. & Sur, M. Locus coeruleus norepinephrine in learned behavior: anatomical modularity and spatiotemporal integration in targets. Front. Neural Circuits 15, 638007 (2021).




END


您可能也对以下帖子感兴趣

文章有问题?点此查看未经处理的缓存