All-optical interrogation of the hippocampal neural code underlying episodic memory

我们对特定事件形成记忆的能力，被称为情景记忆，是我们身份和与世界互动的核心。海马体是参与存储和提取情景记忆的关键大脑结构。海马体的损伤，或在神经退行性疾病如阿尔茨海默氏症期间海马体回路的中断，可以破坏现有记忆的回忆并阻止新的情景记忆的形成。海马体中位置细胞的发现为情景记忆的形成提供了一种可能的细胞机制。然而，目前还不知道位置细胞放电是否与记忆形成有因果关系。此外，海马体用来存储和检索跨越空间和时间的记忆的神经代码的性质是未知的。我们将通过利用一种新的策略对完整大脑中的神经回路进行“全光学”询问来解决这些基本问题。这种方法使用光来同时读出神经元的活动，同时在行为小鼠中进行具有细胞分辨率的靶向光遗传学刺激。这使我们能够识别表现出某些类型活动的神经元，并选择性地刺激它们，以测试这种活动的因果功能作用，我们可以通过观察动物的行为表现来评估。通过改变回路中激活神经元的数量、时间和模式，我们可以识别支持情景记忆的神经编码，我们的实验将探讨海马神经元在情景记忆的空间和时间两个方面的形成中的作用。我们将使用两种不同的行为任务来识别和操纵相关的神经代码。对于空间记忆，小鼠将在虚拟现实环境中执行空间导航任务，沿着轨道运行，并学习在它们获得奖励的区域停下来并舔。对于时间记忆，我们将使用嗅觉任务，动物必须在延迟期间记住气味的身份，然后评估其与第二种气味的关系。然后，我们将使用全光学询问来测试记忆的形成和提取，通过识别和人工增加相关活动模式的可靠性，以提高这两项任务的学习率。此外，我们将通过人工驱动这些模式并观察相关的行为改变来测试位置细胞和序列在记忆提取中的作用。在这个初步功能测试的基础上，我们将系统地改变我们的刺激参数，包括细胞数量、同步水平和顺序激活模式。这将提供海马神经元的特定活动模式和行为之间的第一个因果关系，以及揭示情景记忆背后的神经代码的基本属性。通过确定使记忆成为可能的神经活动模式，我们将大大加深我们对健康和患病大脑的理解。我们对疾病状态如何影响健康神经活动模式的表现的知识仍然有限，部分原因是我们还不知道健康大脑中神经代码的性质。一旦了解了健康大脑中支持记忆的活动模式，就可以设计新的治疗方法，以便在患病的大脑发生故障时保留或恢复它们。我们的发现对于指导痴呆和神经退行性疾病的研究尤其重要，这些疾病由于海马系统中细胞功能和编码的破坏而导致认知缺陷。

Our ability to form memories of specific events, known as episodic memory, is central to our identity and our interactions with the world. A key brain structure involved in storing and retrieving episodic memories is the hippocampus. Lesions of the hippocampus, or disruptions of the hippocampal circuit during neurodegenerative diseases such as Alzheimer's, can disrupt recall of existing memories and prevent formation of new episodic memories. The discovery of place cells in the hippocampus, which are active in specific regions of the environment, provided a possible cellular mechanism supporting the formation of episodic memories. However, it is not yet known whether place cell firing is causally linked to memory formation. Furthermore, the nature of the neural code that the hippocampus utilises to store and retrieve memories across space and time is unknown. We will address these fundamental questions by harnessing a novel strategy for 'all-optical' interrogation of neural circuits in the intact brain. This approach uses light to simultaneously read out the activity of neurons while performing targeted optogenetic stimulation with cellular resolution in behaving mice. This allows us to identify neurons which exhibit certain types of activity and to stimulate them selectively in order to test the causal functional role of this activity, which we can assess by observing the behavioural performance of the animal. By changing the number, timing and pattern of activated neurons in the circuit we can identify the neural code which supports episodic memory.Our experiments will probe the role of hippocampal neurons in the formation of both spatial and temporal aspects of episodic memory. We will use two different behavioural tasks in order to identify and manipulate the relevant neural codes. For spatial memory, mice will perform a spatial navigation task in a virtual reality environment, running down a track and learning to stop and lick in an area where they receive a reward. For temporal memory we will use an olfactory task where animals must remember the identity of an odor across a delay and then assess its relationship to a second odor. We will then use all-optical interrogation to test both memory formation and retrieval by identifying and artificially increasing the reliability of the relevant activity patterns in order to increase learning rates on both tasks. Additionally, we will test the role of place cells and sequences in memory retrieval by artificially driving these patterns and observing relevant behavioural alterations. Building upon this initial test of function we will then systematically vary our stimulation parameters, including the number of cells, level of synchrony and pattern of sequential activation. This will provide the first causal links between specific activity patterns of hippocampal neurons and behaviour, as well as revealing the fundamental properties of the neural code underlying episodic memory.By determining the neural activity patterns enabling memory we will substantially further our understanding of both the healthy and diseased brain. Our knowledge of how disease states impact the manifestation of healthy neural activity patterns is still limited, in part because we do not yet know the nature of the neural code in the healthy brain. Once the activity patterns supporting memory in the healthy brain are understood, new treatments can be designed to preserve or restore them when the diseased brain malfunctions. Our findings will be especially important for guiding research in dementias and neurodegenerative diseases, which cause cognitive deficits due to disruptions of cellular function and coding in the hippocampal system.