Classifying Cortical Neurons by Correlating Transcriptome with Function

 描述（由申请人提供）：将神经元分类为不同类型是神经科学中的一项基本奋进。神经元分类允许人们深入了解神经系统的构建模块，对于神经系统功能的机械理解至关重要，并且是研究人员之间明确沟通的先决条件。对于哺乳动物大脑皮层中的神经元，还没有一个明确的分类方案。基于形态学特征的神经元分类方法使我们对神经系统的认识有了巨大的进步，但由于许多形态学特征难以参数化，因此在皮层神经元中的分类往往是模糊的。其他基于免疫组织化学或电生理学的分类方法很有帮助，但单独使用无法捕捉皮层神经元的丰富多样性。有证据表明，不同的神经元类型表达不同的基因。因此，原则上，基因表达模式可用于生成明确和客观的分类方案。此外，基于基因表达的分类将允许人们使用分子方法选择性地标记和干扰给定的神经元类型，用于基础研究和临床目的。然而，仅基于它们的基因表达模式对神经元进行分类是先验的，关于它们的功能、位置或整合到皮层网络中是没有信息的。的 这个提议的目标是根据那些最能预测神经元功能和位置的基因对皮层神经元进行分类。因此，为了找到这些基因，我们需要将哺乳动物大脑皮层中神经元的转录谱与其功能和位置相关联。我们建议研究初级视觉皮层，因为它是皮层感觉区，神经元的功能已被描述得最详细。我们将对小鼠的初级视觉皮层进行钙成像，以确定视觉皮层神经元对视觉刺激的反应。我们将用光激活的GFP标记成像的神经元。我们将从标记的神经元中收集RNA。我们将进行下一代RNAseq，以揭示体内成像的每个神经元的转录谱。我们将把神经元的转录谱与其对视觉刺激的特异性反应联系起来。最后，我们将使用聚类算法和主成分分析，根据与功能最相关的基因将神经元分类为不同类型。其结果将是一种基于遗传的分类方法，提供有关细胞类型的功能信息。

 DESCRIPTION (provided by applicant): The classification of neurons into distinct types is a fundamental endeavor in neuroscience. Neuronal classification allows one to gain insight into the building blocks of the nervous system, is essential for a mechanistic understanding of the function of the nervous system and is a prerequisite for unambiguous communication between investigators. No single unequivocal categorization scheme exists yet for neurons in the mammalian cerebral cortex. The classification based on morphological characteristics has led to tremendous advances in our understanding of the nervous system, yet is often ambiguous in cortical neurons because many morphological properties are difficult to parameterize. Other classifications based on immunohistochemistry or electrophysiology have been helpful but, alone, fail to capture the rich diversity of cortical neurons. Evidence indicates that distinct neuron types express different genes. Thus, in principle, the gene expression pattern could be used to generate an unambiguous and objective classification scheme. Furthermore, a classification based on gene expression would allow one, using molecular approaches, to selectively tag and perturb a given neuron type both for basic research and for clinical purposes. However, classifying neurons exclusively based on their gene expression pattern is, a priori, uninformative with regard to their function, location or integration into the cortical network. The goal of this proposal is to classify cortical neurons based on those genes that best predict neuronal function and location. Thus, to find those genes we need to correlate the transcriptional profile of a neuron in the mammalian cerebral cortex to its function and location. We propose to investigate the primary visual cortex because it is the cortical sensory area where the function of neurons has been described in greatest detail. We will perform calcium imaging of the primary visual cortex of mice to determine the tuning of visual cortical neurons in response to visual stimuli. We will tag the imaged neurons with photoactivatable GFP. We will harvest the RNA from the labeled neurons. We will perform next generation RNAseq to reveal the transcriptional profile of each individual neuron imaged in vivo. We will correlate the transcriptional profile of a neuron with its specific response to visual stimuli. Finally, we will se clustering algorithms and principal component analysis to classify neurons in different types based on those genes that best correlate with function. The result will be a genetically based classification method that provides functional information about cell types.