This RO1 grant proposal is in response to FOA: PAR-19-070, Notice NOT-AG-19-033 “Selective Cell and Network Vulnerability in Aging and Alzheimer's Disease”. Memory loss in Alzheimer's disease (AD) reflects a progressive failure of neural network function. The overall goal of this proposal is to explore a novel paradigm for the role of neural networks and the REST transcription factor in brain aging and AD. We have shown that REST is activated in the aging human brain but not in AD, where it fails beginning at the stage of mild cognitive impairment. Furthermore, we have recently demonstrated that regulation of neural network activity by REST is involved in a conserved mechanism of longevity and cognitive preservation. Our preliminary studies indicate that REST loss-of- function gives rise to cognitive decline, and markedly augments amyloid and tau pathology in several AD mouse models. It remains to be determined, however, whether REST selectively affects critical neural circuits or acts more globally. We have generated brain conditional floxed REST knockout mice, and now propose to generate REST knockout mice that selectively target entorhinal cortical neurons, CA1 or CA3 hippocampal neurons, or inhibitory GABAergic interneurons. These novel conditional REST knockout lines, together with two established AD transgenic lines, will be used to interrogate the role of REST in the entorhinal (EC)-hippocampal circuit that plays an essential role in memory formation and is central to the onset of AD. Specifically, selective effects of targeted REST deletion on neural network excitation, synaptic plasticity, memory, amyloid deposition and the propagation of tau pathology will be assessed. Our working hypothesis is that REST will play a central role in the regulation of EC-hippocampal network homeostasis, and that a breakdown of network function is an early component of AD. To complement these in vivo studies, we have established a cerebral organoid model of sporadic AD and APOE4 to interrogate neural network function. Cerebral organoids derived from sporadic AD and APOE4 gene-edited iPS cells will be examined for altered neural network activity and tau propagation, and the role of REST. Non-invasive electrical stimulation will be explored in cerebral organoids as an intervention for ameliorating network dysfunction. Finally, we will employ a powerful new platform for single cell RNA-seq and ATAC-seq in the brain. This approach will be applied to mouse models and the human brain to discover gene networks that protect against age- and AD-related memory loss. This multidisciplinary approach will involve collaborating investigators with expertise ranging from genomic biology to electrophysiology and neuropathology, in a coordinated effort to advance the understanding of selective neural network dysfunction in AD.