Multiplexed Sensing and Control of Neuromodulators and Peptides in the Awake Brain

摘要 神经调节剂和其他化学信号水平的失衡导致了一系列神经疾病。 然而，以前描述这些影响的研究通常一次只检查一个分子，并且通常提供 对大脑或脑脊液(CSF)中的信号水平的静态描述，它沐浴着所有的神经元。在……里面 现实中，数十种信号在不同状态下呈现出动态变化，例如安静的醒着、社交或非社交 觉醒，在疾病中会发生改变。对神经活动模式的跟踪和操纵一直是 对神经科学的最新进展至关重要。我们缺乏类似的工具来估计和控制动态模式 神经调节信号，这可能会彻底改变对大脑状态的研究，并有效地恢复健康 横跨神经学和精神障碍的州。此外，我们不理解任何人是如何被赋予 神经精神科药物动态影响内源性神经调节剂和多肽水平 脑脊液或脑，从而阻碍合理设计最佳给药策略，以最大限度地提高疗效和 将副作用降至最低。这些盲点是由于技术限制：虽然细胞成像和光遗传学 在跟踪和操纵脑细胞方面实现了越来越高的精确度，但我们缺乏 准确地(I)实时记录或(Ii)同时控制多个神经调节信号。我们是 克服第一个挑战，通过开发新的方法来对一组 基于绿色荧光蛋白的疾病相关神经调制信号光学传感器(目标1)： 加压素、催产素、生长抑素、多巴胺、去甲肾上腺素、5-羟色胺、乙酰胆碱、组胺、褪黑素、 促肾上腺皮质激素释放因子、血管活性肠肽和腺苷。简而言之，培养的细胞组 将表示单个传感器组合在施加到梯度前面的3D水凝胶传感器阵列中 折射率(GRIN)透镜，它被插入清醒的头部固定的小鼠的脑脊液或脑组织中 慢性插管。然后使用传感器阵列的3D双光子成像来估计信号浓度 通过后自组织机器人将相同的传感器阵列浸入不同浓度的 体外神经调节剂。一旦我们建立了这种追踪神经调节成分的方法 小时或天数和跨行为状态(目标1)，我们将使用闭环交付方法来控制动态 清醒小鼠大脑中多达12个神经调节信号的模式，并评估哪些模式驱动 行为偏好或回避(目标2)。这些实验受益于使用荧光寿命和 以及荧光强度测量，允许对整个流体成分进行定量评估 延长的时间(几小时到几天)，漂白的影响最小。总而言之，这些工具提供了一种新颖的、 研究和控制大脑中多种神经调节剂的整体框架。敏感的， 实时、多路传输的小容量信号读出补充了微透析，并使 应用于基础和临床神经科学研究的大多数领域的闭环控制。

Summary Imbalanced levels of neuromodulators and other chemical signals contribute to a host of neurological disorders. Yet, previous studies describing these effects often examine only one molecule at a time, and typically provide a static description of signal levels in the brain or in the cerebrospinal fluid (CSF) that bathes all neurons. In reality, dozens of signals exhibit dynamic changes across states such as quiet waking and social or non-social arousal, which are altered in disease. The tracking and manipulation of patterns of neural activity has been critical to recent neuroscience progress. We lack analogous tools for estimation and control of dynamic patterns of neuromodulatory signals, which could revolutionize the study of brain states and effectively restore healthy states across neurologic and psychiatric disorders. Moreover, we do not understand how any given neuropsychiatric drug dynamically influences the levels of endogenous neuromodulators and peptides in the CSF or brain, thus impeding the rational design of optimal drug delivery strategies to maximize efficacy and minimize side effects. These blind spots are due to technical limitations: while cellular imaging and optogenetics have enabled ever-increasing precision in tracking and manipulation of brain cells, we lack the ability to accurately (i) record or (ii) control multiple neuromodulatory signals simultaneously in real time. We are overcoming the first challenge by developing novel methods for multiplexed, quantitative imaging of a panel of green fluorescent protein-based optical sensors of disease-relevant neuromodulatory signals (Aim 1): vasopressin, oxytocin, somatostatin, dopamine, norepinephrine, serotonin, acetylcholine, histamine, melatonin, corticotropin-releasing factor, vasoactive intestinal peptide, and adenosine. Briefly, sets of cultured cells expressing individual sensors are combined in a 3D hydrogel sensor array applied to the front of a gradient refractive index (GRIN) lens, which is inserted into the CSF or brain tissue of an awake, head-fixed mouse via a chronic cannula. Estimates of signal concentration using 3D two-photon imaging of the sensor array are then calibrated via post-hoc robotic dipping of the same sensor array into varying concentrations of each neuromodulator ex vivo. Once we have established this approach to track neuromodulatory composition across hours or days and across behavioral states (Aim 1), we will use closed-loop delivery methods to control dynamic patterns of up to a dozen neuromodulatory signals in the brain in awake mice and evaluate which patterns drive behavioral preference or avoidance (Aim 2).These experiments benefit from the use of fluorescence lifetime and well as fluorescence intensity measurements, allowing quantitative assessment of fluid composition across extended periods of time (hours to days) with minimal effects of bleaching. Together, these tools offer a novel, holistic framework for the study and control of multiple neuromodulators in the brain. The sensitive, real-time, multiplexed readout of signals in small volumes complements microdialysis and enables closed-loop control with applications to most domains of basic and clinical neuroscience research.