Optogenetic dissection of cellular and circuit mechanisms of network dysfunction and amyloid deposition in mouse models of Alzheimer's disease in vivo

摘要 临床试验失败的一个原因是转基因(Tg)的翻译限制。 过表达模型，其中生理和内源性调节的致病作用不是 已实现。此外，标准的行为测试可能缺乏识别健壮和可重复的行为的敏感性 新开发的敲入(KI)车型存在缺陷。为了解决或减轻这些限制，我们建议研究 新一代无转基因过表达的Ki小鼠AD模型的建立 用于行为表型和慢性无线脑电/肌电记录的机器学习方法。 具体地说，我们建议研究在小鼠控制下表达人源化抗体的敲入(KI)小鼠 带有或不带有FAD突变的APP基因座，包括AppAβ/Aβ、AppNL-F/NL-F和AppNL-G-F/NL-G-F小鼠 最新的慢性活体电生理记录和机器学习方法 表型鉴定。父母的资助主要集中在AD的J20TG模型上，因为App-Ki小鼠没有表现出 标准行为测试中的显著认知缺陷，包括莫里斯水迷宫测试，10因为它 尚不清楚App-Ki是否存在J20和其他TG-APP鼠标型号中描述的强健网络异常 AD患者，包括伽玛振荡改变、β-伽马偶联、癫痫样棘波和癫痫发作。 15.拟议的实验和基因分型扩大了我们父母赠款的重点，并将开发工具和 父母拨款将直接使用的程序，以改善行为和大脑网络特征 AD的TG模型和KI模型。我们提出了以下目标：目标1.发展和应用机器学习 识别晚发性APPA-β/A、早发性APPNL-F/NL-F和APPNL-G-F/NL-β行为改变的方法 G-F AD小鼠。目的2.通过慢性无线脑电/肌电记录确定电生理表型 晚发性AppAβ/Aβ和早发性AppNL-F/NL-F和AppNL-G-F/NL-G-F AD小鼠的衰老和疾病进展。 目的3.确定晚发性行为改变与神经网络功能障碍之间的关系 AppAβ/Aβ和早发性AppNL-F/NL-F和AppNL-G-F/NL-G-F AD小鼠。 此外，这项多样性补充补助金将显著增强候选人的研究潜力，并 进一步发展她追求研究事业的能力。应聘者将获得宝贵的相关经验 资助期间的电生理数据分析和小鼠行为分析。这些新奇的活体 方法和技术技能将帮助应聘者解决AD发病机制的未探索问题， 从而创造了一条通向独立的技术和概念道路。拟议的补编将提供 支持编制《公约》初步数据所需的概念和技术基础 候选人未来的F32拨款申请。

SUMMARY A contributing factor to the failure of clinical trials has been the translational limitations of transgenic (TG) overexpression models in which physiological and endogenously-regulated pathogenic interactions are not achieved. In addition, standard behavioral tests may lack sensitivity to identify robust and reproducible behavioral deficits in newly develop knock-in (KI) models. To address or mitigate these limitations, we propose to study the next generation of newly developed KI mouse models of AD without transgene overexpression using the latest machine learning approaches for behavioral phenotyping and chronic wireless EEG/EMG recordings. Specifically, we propose to study knock-in (KI) mice that express humanized Ab under the control of the mouse App locus with or without FAD mutations, including AppAβ/Aβ, AppNL-F/NL-F, and AppNL-G-F/NL-G-F mice using state-of- the-art chronic in vivo electrophysiological recordings and machine learning approaches for behavioral phenotyping. The parent grant heavily focuses on the J20 TG model of AD, because App-KI mice show no prominent cognitive deficits in standard behavioral tests, including the Morris water maze test,10 and because it is unknown if App-KI have robust network abnormalities described in J20 and other TG-APP mouse models and AD patients, including altered gamma oscillations, theta-gamma coupling, epileptiform spikes and seizures.1,3,9,11- 15. The proposed experiments and genotypes expand the focus of our parent grant and will develop tools and procedures that will be directly used by the parent grant to improve behavioral and brain network characterization of TG and KI models of AD. We propose the following aims: Aim 1. Develop and apply machine learning approaches to identify behavioral alterations in late-onset AppAβ/Aβ and early-onset AppNL-F/NL-F and AppNL-G-F/NL- G-F AD mice. Aim 2. Determine electrophysiological phenotypes by chronic wireless EEG/EMG recordings during aging and disease progression in late-onset AppAβ/Aβ and early-onset AppNL-F/NL-F and AppNL-G-F/NL-G-F AD mice. Aim 3. Determine relationships between behavioral alterations and neuronal network dysfunction in late-onset AppAβ/Aβ and early-onset AppNL-F/NL-F and AppNL-G-F/NL-G-F AD mice. In addition, this diversity supplement grant will significantly enhance the research potential of the candidate and further her ability to pursue a research career. The candidate will gain valuable relevant experience with electrophysiological data analysis and mouse behavioral assays during the funding period. These novel in vivo approaches and technical skills will help the candidate to address unexplored questions of AD pathogenesis, thereby creating a technical and conceptual path towards independence. The proposed supplement will provide the conceptual and technological foundation needed to support the production of preliminary data for the candidate’s future F32 grant application.