Striatal Plasticity in Habit Formation as a Platform to Deconstruct Adaptive Learning

习惯形成中的纹状体可塑性作为解构适应性学习的平台

• 批准号: 10207803
• 负责人: NICOLE CALAKOS
• 金额: $ 99.34万
• 依托单位: DUKE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-09-30 至 2023-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10207803
• 关键词: Acute; Adaptive Behaviors; Animals; Attention; Automobile Driving; Axon; Behavior; Behavioral; Biological Models; Brain; Calcium; Cells; Commuting; Competence; Complement; Complex; Computer Models; Corpus striatum structure; Cre driver; Discipline; Disease; Dopamine; Dorsal; Dystonia; Electrophysiology (science); Elements; Etiology; Event; Excitatory Synapse; Exhibits; Fire - disasters; Foundations; Functional disorder; Generations; Genetic; Gilles de la Tourette syndrome; Goals; Habits; Handwashing; Huntington Disease; Image; Instruction; Interneurons; Knowledge; Learning; Link; Maps; Measurement; Measures; Medial; Mediating; Methods; Modeling; Molecular Genetics; Monitor; Motor; Movement; Mus; Neurons; Obsessive-Compulsive Disorder; Output; Parkinson Disease; Pathway interactions; Pharmaceutical Preparations; Pharmacogenetics; Play; Population; Positioning Attribute; Preparation; Process; Production; Reagent; Reporter; Resources; Role; Series; Signal Transduction; Site; Slice; Specific qualifier value; Substance abuse problem; Synapses; Synaptic Receptors; Synaptic plasticity; Testing; Thalamic structure; Time; Ursidae Family; Weight; Work; Yin; adaptive learning; addiction; autism spectrum disorder; cell type; cohesion; compulsion; density; design; experience; experimental study; habit learning; in vivo; interest; learned behavior; motor skill learning; multidisciplinary; nerve supply; nervous system disorder; network models; neuropsychiatric disorder; neuropsychiatric symptom; novel; novel therapeutic intervention; postsynaptic; predictive modeling; presynaptic; relating to nervous system; response; striosome; success; symptomatology; targeted treatment; therapeutic target; therapy development; tool

ABSTRACT A distinguishing feature of the brain is that its circuitry isn’t computationally static, it adapts to experience. Understanding the circuit mechanisms for adaptive behavior carries two-fold potential benefits - revealing the brain’s learning rules and identifying key behaviorally significant functional “nodes”. These nodes suggest potent sites to target for therapy development and may also be instructive to suggest more basic circuit principles underlying behavior. Using striatal circuitry and habit learning as a model system, we recently uncovered a set of paradigm- challenging findings in a striatum-dependent habit learning task. In particular, we discovered a new circuit-level signature, termed dviLP (direct vs indirect Latency Plasticity), which distinguishes striatal slices prepared from habitual vs goal-directed animals. The features of dviLP shift long-held attention on rate differences between the two principle projection neuron types, those to the direct and indirect pathways, to consider that behaviorally adaptive signals may be generated by plasticity of their relative timing to fire. Moreover, the origin of this plasticity appears to involve striatal fast-spiking interneurons, a highly non-canonical site for the expression of long-lasting plasticity. Beginning with this highly novel foundation, here we propose to generate a robust predictive computational model for striatal-dependent learning mechanisms by joining multiple disciplines and multiple levels of analysis through an iterative process of circuit modeling and experimentation. In Aim 1, we will comprehensively map functional changes in synaptic and cellular activity that define the behavioral transition from goal-directed to habitual in an operant lever press task. We will use a layered suite of molecular genetic tools to assign coordinates that specify inputs, outputs, compartments (striosome/matrix) and regions (medial, dorsal). In Aim 2, we will measure the activity of genetically specified components of the striatum in behaving mice, identifying the dynamic changes that correlate with and cause the shift from goal- directed to habitual behavior. Our team offers multidisciplinary strengths. Dr. Calakos and Yin have expertise in habit behavior, plasticity mechanisms and in vivo circuit dynamics; ideal for spearheading this effort. The success and impact of this effort will be amplified by tightly incorporating Dr. Brunel’s expertise in computationally modeling brain learning mechanisms and Dr. Tadross’s novel pharmacogenetic reagents that are ideally positioned to test causality of synaptic plasticity events, offering the unique opportunity to manipulate a specific synaptic receptor in a genetically defined cell type. Ultimately, we expect that the knowledge gained through this highly collaborative proposal will provide a foundational resource to accelerate understanding of striatal learning rules for adaptive behavior.

摘要 大脑的一个显着特征是，它的电路在计算上不是静态的，它会适应 体验.了解适应行为的电路机制有两个潜在的好处- 揭示大脑的学习规则，并识别关键的行为重要功能“节点”。这些节点 提示了治疗开发靶向有效位点，也可能有益于提示更基本的 行为背后的电路原理 使用纹状体回路和习惯学习作为模型系统，我们最近发现了一组范式- 在纹状体依赖性习惯学习任务中具有挑战性的发现。特别是，我们发现了一个新的电路级 签名，称为dviLP（直接与间接潜伏期可塑性），区分纹状体切片制备， 习惯性动物vs目标导向动物dviLP的特点转移了人们对以下方面的利率差异的长期关注： 两种主要的投射神经元类型，即直接和间接通路，考虑到 行为自适应信号可以通过其相对定时的可塑性来产生。此外，起源 这种可塑性似乎涉及纹状体快速尖峰中间神经元，这是一个高度非典型的网站， 表现出持久的可塑性。从这个非常新颖的基础开始，我们建议在这里生成一个 通过连接多个纹状体依赖性学习机制的鲁棒预测计算模型 通过电路建模和实验的迭代过程，实现多学科和多层次的分析。 在目标1中，我们将全面绘制突触和细胞活动中的功能变化，这些变化定义了 在操作性压杆任务中从目标导向到习惯性的行为转变。我们将使用一套分层的 分子遗传学工具，用于分配指定输入、输出、区室（纹状体/基质）的坐标 和区域（内侧，背侧）。在目标2中，我们将测量基因特异性组分的活性。 行为小鼠的纹状体，识别与目标相关并导致目标- 针对习惯性行为。我们的团队提供多学科的优势。卡拉科斯博士和尹博士擅长 习惯行为，可塑性机制和体内电路动力学;理想的带头这项工作。的 这项工作的成功和影响将通过紧密结合布鲁内尔博士的专业知识来扩大， 计算模拟大脑学习机制和塔德罗斯博士的新型药物遗传学试剂， 理想地定位于测试突触可塑性事件的因果关系，提供了独特的机会， 操纵遗传上确定的细胞类型中的特定突触受体。最终，我们预计， 通过这一高度协作的提案获得的知识将提供基础资源， 理解纹状体学习规则的适应行为。