Activity-dependent modeling of developing dendrites in a retinal neuron

格兰特：视网膜神经元树突发育的活动依赖模型**神经元通常以其树枝的形状为特征。了解它们在体内发育过程中如何获得形态是神经生物学的一个关键目标。位置和精确的分支模式不仅控制它们通过电路布线接收的突触输入的类型，而且还关键地决定信号如何被处理和转换为棘波活动。视网膜中编码向上运动的定向选择性神经节细胞(DSGCs)提供了形式和功能之间这种关系的显著例子，因为它们的整个树突树指向它们编码的方向(特伦霍尔姆等人，2011年)。这个项目的总体目标是了解不对称树枝是如何在发育过程中建立的。*在中枢神经系统的不同部分，神经元以不同的方式建立它们的不对称树枝。或者，它们可以通过定向向突触前伙伴生长而变得不对称。或者，神经元可能首先建立无定形的树枝状分支，树突分支指向随机方向，然后成熟为高度定向的树枝。要区分这些可能性是具有挑战性的，因为给定脑区的神经元由各种亚型组成，当树突尚未获得其特征形态时，这些亚型在未成熟状态下尤其难以区分。为此，我们最近发现了一种转基因小鼠，其中GFP在不对称的DSGCs中选择性表达，使我们能够跟踪这些神经元的早期发育。使用传统的解剖学方法，我们将首先描述树突状生长与突触前伙伴(双极和星状爆发式无长突细胞)的关系，以确定DSGCs获得不对称树突状突起的步骤。*虽然已知树突结构的正确发育和建立是由遗传程序(Jan和Jan，2003)指定的，但它可以在很大程度上通过钙信号介导的活性依赖过程(McAllister等人，1997，Nedivi等人，1998，Rajan和Cline，1998，Wu等人，1999，Polleux等人，2000，Sin等人，2002，Grueber等人，2003，Haas等人，2006，Janushi-Nakao等人，2007，Jan和Jan，2010)进行修改。在发育中的视网膜，自发活动(独立于光)是由NMDA和乙酰胆碱受体介导的，已被假设为精炼树突状分支。在第二个目标中，我们将使用Cre-lox策略来特异性干扰向上编码的DSGCs的活动。我们将研究通过NMDA和乙酰胆碱受体介导的钙信号的作用，使用基因策略敲除NMDA受体或沉默胆碱能神经元以抑制乙酰胆碱的释放。相反，我们将使用DSGCs的光遗传刺激(使用通道视紫红质的选择性表达)来提高活性水平。这样的操作将使我们能够评估活动对DSGCs在其自然环境中树突树发育的影响。*最近对透明斑马鱼和蝌蚪中的单个神经元进行的活体时间推移成像显示，树突状分支的运动量和周转量非常大。在这笔赠款的第三个目标中，我们将描绘不同发育阶段的树突的动态行为，并研究依赖活动的过程(在目标2中建立)如何影响哺乳动物神经系统中树突的动态重新建模。

Grant: Activity-dependent modeling of developing dendrites of a retinal neuron**Neurons are often characterized by the shape of their dendritic arbours. Understanding how they acquire their morphology during development in vivo is a key goal in neurobiology. The location and precise branching patterns not only controls the type of synaptic input they receive by virtue of the circuit wiring, but also critically determines how the signals are processed and translated into spike activity. Directionally selective ganglion cells (DSGCs)that code upward motion in the retina, offer a striking example of this relationship between form and function, as their entire dendritic tree points in the direction they code (Trenholm, et al., 2011). The overall aim of this project is to understand how asymmetric dendritic arbors are established during development.*In different parts of the CNS, neurons establish their asymmetric dendritic arbors in distinct ways. Either, they can become asymmetric by directed outgrowth towards their presynaptic partners. Alternatively, neurons may first establish amorphous arborisations with dendritic branches pointing in random directions and subsequently mature into a highly oriented arbor. To distinguish these possibilities is challenging because neurons in a given brain area consist of a variety of subtypes which are especially hard to distinguish in the immature state when their dendrites have not acquired their characteristic morphologies. To this end, we recently identified a transgenic mouse line in which GFP is expressed selectively in asymmetric DSGCs allowing us to follow these neurons through early development. Using a conventional anatomical approach we will first characterize the development of dendritic growth in relation with presynaptic partners (bipolar and starburst amacrine cells), to determine the steps by which DSGCs acquire their asymmetric dendritic arbors. *While the proper development and establishment of dendritic structure is known to specified by genetic programs (Jan and Jan, 2003), it can modified to a large extent by activity-dependent processes (McAllister et al.,1997, Nedivi et al., 1998, Rajan and Cline, 1998, Wu et al., 1999, Polleux et al., 2000, Sin et al., 2002,Grueber et al., 2003, Haas et al., 2006, Jinushi-Nakao et al., 2007, Jan and Jan, 2010)mediated by Ca signaling. In the developing retina, spontaneous activity (independent of light) is mediated by NMDA and acetylcholine receptors, has been postulated to refine dendritic arboursizations. In the second aim, we will use Cre-lox strategies to specifically perturb activity in upward coding DSGCs. We will examine the role of Ca signals mediated through NMDA and acetylcholine receptors using genetic strategies to either knock out NMDA receptors or silence cholinergic neurons to inhibit release of acetlycholine. Conversely, we will use optogenetic stimulation of DSGCs (using selective expression of channelrhodopsin) to increase activity levels. Such manipulations will allow us to assess the impact of activity on the development of the dendritic trees of DSGCs in their native environment.*Recent in vivo time lapse imaging of individual neurons in transparent zebra fish and tadpoles have revealed an exceptional amount of motility and turnover of dendritic branches. In the third aim of this grant we will image the dynamic behavior of DSGC dendrites during different stages of development and examine how an activity-dependent process (established in Aim 2) affects dynamic re-modeling of dendrites in the mammalian nervous system.