The development of feature selectivity across the visual system

背景：哺乳动物的视觉系统是以一种分层的方式组织的，随着你从视网膜转移到更高的大脑区域，神经元反应的复杂性增加。这种复杂性的增加可以通过不同脑区神经元的特征选择反应来衡量，例如视网膜的中心-环绕反应，视觉皮质的定向和方向选择反应，以及颞叶皮质的视觉对象选择反应。然而，关于复杂的特征选择性是如何产生的，无论是在大脑区域内还是跨大脑区域，人们都知之甚少。为了更好地了解特征选择性是如何产生的，可以通过直接比较单个突触后神经元的特征选择性和其突触前输入的特征选择性，在单细胞功能连接的水平上研究神经元回路。目标1：比较初级视觉皮质中单个第四层起始神经元的特征选择性和它们在视觉皮质中突触前输入的特征选择性。为了做到这一点，我们将用修改后的狂犬病病毒从视皮层第四层进行单细胞启动的跨突触电路追踪。我们将向小鼠提供视觉刺激，并使用体内双光子钙成像技术测量突触连接神经元中的光反应。*目的2：比较丘脑外侧膝状体(LGN)内单个第4层起始神经元与其突触前神经元的特征选择性。采用与目标1相同的实验范式，我们将使用活体双光子钙成像技术监测单个皮质第四层起始细胞及其突触前LGN输入的神经元活动。*目标3：比较为视觉皮质第四层单个神经元提供双突触输入的视网膜神经节细胞的特征选择性。我们将实施一种新颖的串联电路跟踪策略，涉及使用单细胞启动的狂犬病病毒跟踪来自多个脑区的单个细胞，以驱动狂犬病从单个LGN神经元-即突触前到第四层皮质起始神经元-跳到突触前视网膜神经节细胞。然后，我们将在视觉刺激过程中对体外培养的视网膜神经节细胞进行双光子钙成像，以检测这些神经元的特征选择性。*结果：这些实验将提供强有力的直接证据，证明早期视觉系统如何出现明显的特征选择性反应，特别是方向和方向选择性反应。这将是第一个通过关注三个不同大脑区域的突触连接神经元来直接监测单个特征如何在多个大脑区域产生选择性的项目。在未来，类似的方法可以用来理解更复杂的特征选择性反应，例如视觉对象选择性。

Background: The mammalian visual system is organized in a hierarchical manner, with the complexity of neuronal responses increasing as you move from the retina to higher brain regions. This increase in complexity can be measured via the feature selective responses of neurons in different brain areas, such as center-surround responses in the retina, orientation and direction selective responses in visual cortex, and visual object selective responses in temporal cortex. However, little is known about how complex feature selectivity arises, both within and across brain regions. To gain a better understanding of how feature selectivity is generated, neuronal circuits can be studied at the level of single-cell functional connectivity, by directly comparing the feature selectivity of a single postsynaptic neuron to the feature selectivity of its presynaptic inputs.******Objective 1: Compare the feature selectivity of individual layer 4 ‘starter' neurons in primary visual cortex to the feature selectivity of their presynaptic inputs in visual cortex. To do this, we will perform single-cell initiated transsynaptic circuit tracing from layer 4 of visual cortex with modified rabies viruses. We will present mice with visual stimuli and measure the light responses in the synaptically connected neurons using in vivo 2-photon calcium imaging. ******Objective 2: Compare the feature selectivity of individual layer 4 starter neurons to their presynaptic neurons in the lateral geniculate nucleus (LGN) of the thalamus. Using the same experimental paradigm as outlined in Objective 1, we will monitor the neuronal activity of individual cortical layer 4 starter cells and their presynaptic LGN inputs using in vivo 2-photon calcium imaging.******Objective 3: Compare the feature selectivity of the retinal ganglion cells that provide di-synaptic input to individual neurons in layer 4 of visual cortex. We will implement a novel, serial circuit tracing strategy, involving the use of single-cell-initiated rabies virus tracing from individual cells in multiple brain regions, to drive rabies to jump from a single LGN neuron – that is presynaptic to the layer 4 cortical starter neuron – to presynaptic retinal ganglion cells. We will then perform 2-photon calcium imaging from retinal ganglion cells in vitro during visual stimulation to examine the feature selectivity of these neurons.******Outcomes: These experiments will provide strong direct evidence regarding how distinct feature selective responses, in particular orientation and direction selective responses, arise in the early visual system. This will be the first project to directly monitor how a single feature selectivity arises across multiple brain regions by focusing on synaptically connected neurons in three distinct brain regions. In the future, similar approaches could be used to understand more complex feature selective responses, such as visual object selectivity.