From spatiotemporal restriction of imprecise synaptic partner choice to a robust representation of navigational information in the fly brain

密切相关的神经元亚型相邻的成人大脑往往表现出高度特异性，不同的连接。果蝇视觉系统中的模式特异性感光神经元及其突触伙伴代表了这种现象的一个实例：当眼睛的主要部分和潜在的视神经球处理运动和颜色特异性视觉输入时，背缘区域（DRA）处理天窗偏振，最终导致在中枢大脑中呈现导航信息，该信息在各种刺激条件下都很强大。我们最近在RobustCircuit pi的研究团队发现，颜色敏感的R7神经元在接触后很容易与不正确的伙伴形成突触连接，通过在时间和空间上分离的突触发生相互作用，确保了发育中的野生型大脑中的突触特异性。因此，在P3中，我们直接比较了DRA光感受器R7和R8对处理天窗信号的特定下游神经元的选择性靶向，以及它们对颜色敏感的对应体。基于我们对DRA区域神经元的初步数据，我们假设DRA神经元处理的信息的鲁棒表征是基于它们通过在空间和时间上对突触发生相遇的差异调节来混杂形成突触的能力。P3旨在通过三种互补的方法来验证这一假设：(i)通过多光子离体成像对突触前和突触后细胞类型的丝状动力学进行定量研究；（ii）通过基于em的连接组学量化神经元形态和突触分布的可变性；（iii）利用光生理学（钙成像）对单细胞和整体水平上产生稳健细胞反应的可变性和不精确性进行量化。(i)和（iii）都将包括野生型大脑，以及有针对性的操作来破坏茎突动力学和神经元相遇。然后，我们将把这些研究扩展到连接视叶和中央大脑的神经元，从而在那里获得天窗信息的健壮的地形表示。当这些研究结束时，我们将建立一个完整的4维描述机制，用于在完整的大脑中基于随机动力学的导航信息的稳健表示的发展。

Closely related neuronal subtypes neighboring each other in the adult brain often exhibit highly specific, divergent connectivity. Modality-specific photoreceptor neurons and their synaptic partners in the Drosophila visual system represent an instance of this phenomenon: While the main portion of the eye and underlying optic neuropils process motion- and color-specific visual input, the dorsal rim area (DRA) processes skylight polarization, ultimately leading to a representation of navigational information in the central brain that is robust over a variety of stimulus conditions. P3 is motivated by our recent team effort of RobustCircuit PIs which discovered that color-sensitive R7 neurons readily form synaptic connections with incorrect partners upon contact, ensuring synaptic specificity in the developing wild type brain by separated synaptogenic interactions in time and space. In P3, we therefore directly compare the selective targeting of DRA photoreceptors R7 and R8 to their specific downstream neurons which process skylight cues, to their color-sensitive counterparts. Based on our preliminary data on neurons in the DRA region, we hypothesize that the robust representation of information processed by DRA neurons are based on their ability to form synapses promiscuously through the differential regulation of synaptogenic encounters in space and time. P3 is devised to test this hypothesis using three complementary approaches: (i) the quantitative investigation of filopodial dynamics of both pre- and postsynaptic cell types via multi-photon ex vivo imaging; (ii) the quantification of variability of neuronal morphology and synaptic distribution by EM-based connectomics; (iii) the quantification of variability and imprecision for creating robust cellular responses, both on a single cell-, as well as on a bulk level, using optophysiology (calcium imaging). Both (i) and (iii) will include wild type brains, as well as targeted manipulations to disrupt filipodial dynamics and neuronal encounters. We will then expand these studies towards neurons that connect the optic lobes with the central brain leading to a robust topographic representation of skylight information there. When these studies are concluded, we will have established a complete 4-dimensional description of the mechanisms for the development of a robust representation of navigational information based on stochastic dynamics in an intact brain.