PROJECT ABSTRACT Astrocytes are a group of glial cells with multiple roles in the brain. They extend fine processes that localize near synapses and blood vessels and thus locally regulate synapse formation, elimination and function, as well as blood-brain-barrier permeability. This local regulation more broadly impacts on the function of larger neural circuits within the brain. Astrocytes from different brain regions show distinct morphological and gene expression profiles, which suggest regional specificity of function. Links between neuronal degeneration, distinctive of brain maladies such as Parkinson’s disease (PD) and Alzheimer’s disease (AD), and astrocytic malfunction are suspected but not well characterized mechanistically. Mitochondria in astrocytes have recently emerged as specialized subcellular organelles with robust spontaneous Ca2+ fluxes capable of responding to neurotransmitter agonists, a distinct functional proteomic profile, and spatial segregation into somata versus processes with regard to Ca2+ fluxes and morphology. These unique characteristics lead us to hypothesize that abnormalities in astrocytic mitochondria play a major role in accelerating neurodegenerative processes in the brain. However, the lack of tools to functionally ablate mitochondria selectively in astrocytes from specified brain regions in vivo is a major barrier towards defining how astrocytic mitochondrial dysfunction contributes to neurodegeneration. Specifically, developing such a tool would enable a mechanistic understanding of the role of astrocytic mitochondria in PD and AD. Here, we propose to develop and characterize an adeno-associated virus (AAV)-based tool to functionally ablate astrocytic mitochondria in two brain regions relevant to PD and AD, viz. the striatum and posterior cingulate cortex, respectively. In addition, we will examine the cell-autonomous and non-cell autonomous effects of functionally ablating astrocytic mitochondria on astrocytic and neuronal gene expression profiles with the future goal of applying the knowledge gained to in vivo models of PD and AD.