Towards quantitative cell type-based mapping of the whole mouse brain

 描述(申请人提供)：小鼠的大脑由7000万个神经元和3000万个神经胶质细胞和其他细胞组成。传统上，神经元是根据其形态、连通性、刺激反应、基因表达和在大脑中的位置进行分类的。虽然我们相当清楚存在于不同大脑位置的主要细胞类型，但我们对全脑细胞类型分布的定量知识很少。此外，基于细胞类型的全脑连接，特别是在单个神经元的投影模式水平上，也在很大程度上仍然没有被映射。这种知识差距使我们无法将积累的基于细胞类型的细胞数据整合到哺乳动物大脑功能的全面电路模型中。在这里，我们建议开发一种高度自动化的方法，用于1)根据整个小鼠大脑中每个解剖区域的完整细胞计数对神经元和神经胶质细胞类型进行定量“图谱”，以及2)对已识别的神经元细胞类型的轴突树枝进行完整的追踪。具体目标是：具体目标1(FOA目标1、4-6)：为分子定义的细胞类型的全脑图谱开发一个通用平台。将开发计算和统计方法，以获得通过序列双光子(STP)断层成像并注册到参考脑图谱的特定细胞类型报告小鼠的整个大脑中准确的细胞数量。然后，这些方法将被用来绘制五种不同类型的GABA能细胞在发育期间、成年男性和女性大脑以及衰老期间的分布。具体目标2(FOA目标6)：通过开发第二代成像平台，提高基于细胞类型的图谱绘制和连通性测绘的吞吐量。将开发一种基于光片荧光显微镜的方法，名为斜面层析成像(OPT)，用于对经Clarity处理的小鼠大脑进行快速高分辨率成像。将建立计算方法，以允许在整个小鼠大脑的细胞类型图谱和投影追踪中使用OPT。具体目标3(FOA目标2-3)：神经信息学基础设施：细胞类型特定数据、工具和分析的综合网络门户。开发的方法，包括用于线上和线下分析的统计工具箱， 生成的数据将通过一个公共网站发布，该网站还将整合有关分子身份、位置、连接性和其他细胞类型特征的相关信息。与公众健康相关：我们将建立一种自动化和定量的方法来研究小鼠大脑中细胞类型的分布和连通性。我们的方法可以很容易地放大并应用于广泛的特定细胞类型的研究，既有正常的大脑功能，也有与人类大脑疾病相关的病理情况。

 DESCRIPTION (provided by applicant): The mouse brain comprises ~70 million neurons and ~30 million glia and other cells. Neurons have been traditionally classified based on their morphology, connectivity, stimulus-response, gene expression, and location in the brain. While we know reasonably well the main cell types that are present at different brain locations, we have little quantitative knowledge about brainwide cell type distribution. In addition, cell type-based brainwide connectivity, especially at the level of projection patterns of single neurons, also remains largely unmapped. This knowledge gap prevents us from incorporating the accumulated cell type-based cellular data into comprehensive circuit models of mammalian brain function. Here we propose to develop a largely automated methodology for 1) quantitative "atlasing" of neuronal and glia cell types on the basis of their complete cell counts per anatomical regions in the whole mouse brain, and 2) complete tracing of axonal arborizations of identified neuronal cell types. The specific Aims are: Specific Aim 1 (FOA goals 1, 4-6): To develop a versatile platform for brainwide atlasing of molecularly defined cell types. Computational and statistical methods will be developed to obtain accurate cell numbers in whole brains of cell type-specific reporter mice imaged by serial two-photon (STP) tomography and registered to reference brain atlases. These methods will then be used to map the distribution of five different GABAergic cell types during development, in the adult male and female brain, and during aging. Specific Aim 2 (FOA goal 6): To enhance the throughput of cell type-based atlasing and connectivity mapping through the development of a second-generation imaging platform. A light-sheet fluorescent microscopy-based method, named Oblique Plane Tomography (OPT), will be developed for fast high- resolution imaging of CLARITY-treated mouse brains. Computational methods will be established to allow the use of OPT in cell type atlasing and projection tracing in the whole mouse brain. Specific Aim 3 (FOA goals 2-3): Neuroinformatics infrastructure: an integrated web portal for cell-type specific data, tools & analytics. The developed methods, including a statistical toolbox for online and offline analytics, and the data generated will be distributed via a public website that will also serve to integrate the relevant information on the molecular identity, location, connectivity and other cell type characteristics. Relevance to public health: We will establish an automated and quantitative approach to the study of cell type distribution and connectivity in the mouse brain. Our methods can be readily scaled up and applied to a broad range of cell type-specific studies of both normal brain functions and pathological conditions related to human brain disorders.