Summary Microglia-mediated synaptic pruning is highly regulated during developmental critical periods of synaptic refinement, but its activation in vulnerable brain regions in disease models suggests that disease and development share common regulators and mechanisms of pruning. Synaptic refinement is an activity- dependent process where weak synapses are preferentially pruned. We hypothesize that neural activity is a key upstream activator and regulator of microglia-mediated pruning in development and disease. In support of this hypothesis, we and others demonstrated that microglia phagocytose synaptic elements and are capable of sensing and responding to activity-related signals as they preferentially engulf less active synapses. However, how microglia determine which synapses to engulf and which to avoid, and the identity of the upstream neuronal signals that detect neural activity and transmit this information to microglia are not known. In the immune system, phagocytosis is carefully governed by both phagocytic, “eat me” and anti- phagocytic, “don’t eat me” molecules. In the brain, we identified neuronal CD47, a “don’t eat me” signal protects synapses from inappropriate removal. We further found that exposed phosphatidylserine (PS), an “eat me signal” drives microglial recognition of synapses for engulfment. Furthermore, we have identified the tyrosine kinases, Pyk2 and JAK2 as neuronal signals that are activated at inactive synapses and necessary for the elimination of these inputs. Finally, we found that PS exposure is elevated early in the hippocampus in an Alzheimer's disease (AD) mouse model. Based on these and other data, we propose to test the hypothesis that activity-dependent signals within neurons, such as Pyk2 and JAK2, dynamically regulate "eat me" and "don't eat me" signals on synapses to drive proper microglial phagocytosis of inactive synapses. We further hypothesize that aberrant activation of such signals lead to abnormal synapse loss in neurodegenerative diseases, such as Alzheimer’s Disease (AD). Specifically we aim to test the following: Aim 1) Investigate mechanisms of activity-regulated PS exposure and function at developing synapses; Aim 2: Determine activity-dependent neuronal signaling that regulates "eat me" and "don't eat me" signals for microglial engulfment; and Aim 3: Investigate the mechanisms of early synaptic dysfunction and microglial synaptic targeting associated with AD. We will use interdisciplinary in vivo and in vitro approaches with molecular/cell biological, histological, mouse genetic, imaging, and electrophysiological techniques with various newly developed systems to address our aims. Our studies could also provide new targets for therapeutic intervention, as restoring the balance of protective and elimination signals could protect against synapse loss in early stages of AD and other diseases.