Neuronal basis, functional role and pathological relevance of grid cell-like representations in the human brain

网格细胞最初发现于啮齿类动物的内嗅皮层（EC），其特征是六边形排列（即六倍旋转对称）的空间放电场。它们被假设通过实现路径整合（PI）机制来支持空间导航。在人类中，可以通过手术前癫痫患者在虚拟空间导航过程中的单单元记录来识别网格细胞。此外，一些功能磁共振成像研究发现了“网格细胞样表征”（GCLRs）的间接证据，这是一种假定的欧共体网格细胞的网络级特征，具有六倍旋转对称性。最近，我们在相同的空间导航任务中对癫痫患者的EC进行了颅内脑电图（iEEG）记录，并根据theta （4-8Hz）振荡的功率观察gclr。然而，到目前为止，网格细胞和gclr之间的直接联系还不能建立起来。其他悬而未决的问题涉及gclr在PI中的确切功能作用及其与理解阿尔茨海默病（AD）的临床相关性，因为AD的神经变性已被提出始于EC。在此，我们提出两项研究来解决这些问题。在第一项研究中，我们将使用脑电图和单单元记录，通过结合微/宏电极同时测量手术前癫痫患者的动作电位和脑电图振荡。患者将在开放的虚拟环境中执行PI任务。这项研究将使我们能够直接将单个网格细胞的活动与gclr联系起来。在第二项研究中，我们将对健康的年轻参与者进行功能磁共振成像记录。一半的参与者携带载脂蛋白E基因（APOE4）的等位基因，这是散发性阿尔茨海默病最重要的遗传风险，而另一半由对照组参与者组成。功能磁共振成像研究由连续两天的两个实验组成。在第一天，参与者将在扫描仪中进行与之前用于识别gclr相同的空间导航任务。实验二由与iEEG/单单元实验相同的路径整合任务组成。我们将量化两个实验中fMRI gclr的稳定性，并将它们与两个实验中的表现和APOE基因型联系起来。总之，这两项研究将更好地了解gclr的神经元机制，它们在PI中的功能作用，以及它们在遗传AD风险携带者中的可能损害。

Grid cells were initially discovered in the entorhinal cortex (EC) of rodents and are characterized by hexagonally arranged (i.e., six-fold rotationally symmetric) spatial firing fields. They have been hypothesized to support spatial navigation by implementing a mechanism for path integration (PI).In humans, grid cells could be identified via single-unit recordings in presurgical epilepsy patients during virtual spatial navigation. In addition, several fMRI studies found indirect evidence for “grid cell-like representations” (GCLRs), a putative network-level signature of grid cells in the EC with six-fold rotational symmetry. More recently, we conducted intracranial EEG (iEEG) recordings in the EC of epilepsy patients during the same spatial navigation task and observed GCLRs based on the power of theta (4-8Hz) oscillations. Until now, however, a direct link between grid cells and GCLRs could not be established. Additional open questions concern the exact functional role of GCLRs for PI and their clinical relevance for understanding Alzheimer’s disease (AD), because neurodegeneration in AD has been proposed to start in EC. Here, we propose two studies to address these issues. In the first study, we will use iEEG and single-unit recordings via combined micro-/macro-electrodes to simultaneously measure action potentials and iEEG oscillations in presurgical epilepsy patients. Patients will perform a PI task in an open-field virtual environment. This study will allow us to directly link the activity of single grid cells to GCLRs. In the second study, we will conduct fMRI recordings in healthy young participants. Half of the participants carry the ɛ4 allele of the Apolipoprotein E gene (APOE4), the most important genetic risk for sporadic AD, while the other half consists of control participants. The fMRI study is composed of two experiments across two consecutive days. On the first day, participants will conduct a same spatial navigation task in the scanner that has been used before to identify GCLRs. Experiment two consists of the same path integration task as in the iEEG/single unit experiment. We will quantify the stability of fMRI GCLRs across the two experiments and relate them to performance in the two experiments and to the APOE genotype. Together, the two studies will provide a better understanding of the neuronal mechanisms of GCLRs, their functional role for PI, and their possible impairment in genetic AD risk carriers.