Metabolic determinants of reactive astrogliosis and cognitive heterogeneity in aging

ABSTRACT The greatest risk factor for Alzheimer’s disease and related dementias (ADRD) is aging, but the mechanisms that link aging to ADRD disease processes are largely unknown. Of particular interest, recent single-cell transcriptomic studies of aged hippocampal astrocytes, and multi-transcriptomic analysis of astrocytes from human AD samples

and a mouse model of AD, show commonalities in the decline of mitochondrial oxidative phosphorylation (OXPHOS) and mitophagy related pathways. Although cognitive function can decline with age, many human and animal model subjects retain function. Current behavioral/cognitive testing paradigms are inefficient at discerning

cognitive trajectories in aging. We have developed a unique automated home-cage testing paradigm for spatial memory that can reliably stratify cognitive performance in aged mice into cognitively ‘intact’ and ‘impaired’ subgroups. Molecular analyses show significant declines in OXPHOS genes, mitochondrial function and

concomitant increases in markers of astrogliosis specifically in ‘impaired’ mice. Whether impairments in astrocyte mitochondrial function are a driver of reactive astrogliosis associated with cognitive impairment in aging and AD is unresolved and is a critical question since it guides future therapeutic interventions. Our central hypothesize

is that deficits in astrocyte mitochondrial function and mitochondrial proteostasis drive reactive astrogliosis that contribute to cognitive impairment in aging and AD, and that therapies that improve astrocyte mitochondrial function will delay cognitive decline. The following aims are proposed: Aim 1.

Identify astrocyte subtypes with dysfunctional mitochondria in hippocampus of cognitively impaired aged mice. We will use high resolution respirometry to measure mitochondrial function in isolated hippocampal mitochondria from aged cognitively stratified mice. We will perform scRNAseq on hippocampal astrocytes to

identify subtypes that are specific to cognitive impairment and allow us to differentiate between healthy aging and cognitive impairment. Aim 2. Define the role of mitochondrial proteostasis in reactive astrogliosis of aged mice. We will assess mitochondrial protein turnover rates as well as turnover of targeted mitochondrial

proteins using D2O labeling of proteins in astrocytes from young, aged-intact and -impaired mice and determine the contribution of mitophagy to astrocyte-mediated cognitive decline in aging. We expect these experiments will link decreased mitochondrial protein turnover rate and mitophagy to the genesis of reactive astrogliosis and

cognitive impairment. Aim 3. Determine whether interventions that improve astrocyte mitochondrial function alleviate cognitive impairment in aged and APP/PS1 mice. We will use AAV5-mediated overexpression of SOD2 specifically in astrocytes delivered via stereotactic injections into the hippocampus of

cognitively stratified aged, and APP/PSI mice. We expect that the intervention will improve mitochondrial proteostasis, reduce reactive astrogliosis, and improve cognition. These experiments are the first to address the molecular determinants of cognitive heterogeneity in aging.