Brain Glial Cells Have Unique Circadian Clocks That Break Down in Alzheimer's and Aging

Circadian rhythms—the body's 24-hour biological clock—govern not just sleep-wake cycles but also thousands of genes across virtually every tissue. In the brain, disrupted circadian function is increasingly linked to neurodegenerative diseases, particularly Alzheimer's disease (AD). Yet almost nothing was known about how individual brain cell types maintain their own circadian gene expression programs, or how those programs change with aging or disease pathology.

To address this gap, Sheehan et al. employed two complementary ribosome-tagging strategies: Aldh1l1-RPL10a-eGFP (AstroTRAP) for astrocytes and Cx3cr1-Cre-ERT2;LSL-Rpl22-HA (mgRiboTag) for microglia. Mice were entrained to light-dark cycles, then placed in constant darkness, and sacrificed every two hours over 24 hours. Ribosome-associated mRNA was immunoprecipitated and sequenced, capturing the actively translated transcriptome—the 'translatome'—at each time point. Experiments were performed in wild-type mice, APP/PS1-21 amyloid model mice (collected at 6 months, when plaques are robust), and aged mice (~18 months). Bulk cortex pre-immunoprecipitation samples served as a tissue-level comparison. Two independent experimental cohorts were combined after batch correction, confirming reproducibility.

The data revealed that astrocytes and microglia each harbor highly distinct circadian translatomes that share only partial overlap with bulk cortex rhythms, underscoring that bulk tissue analysis obscures cell-type-specific biology. In healthy mice, glial circadian programs encompassed genes tied to immunometabolism, proteostasis, reactivity, and lysosomal function. Strikingly, nearly half of all known AD genetic risk genes oscillated in a circadian fashion in at least one cell type.

In APP/PS1 amyloid mice, both astrocyte and microglial circadian translatomes were profoundly reprogrammed: many normally rhythmic transcripts lost rhythmicity while new oscillating genes emerged. In bulk cortex, amyloid disrupted circadian regulation of lysosome and autophagy pathways (WT-only rhythmic genes) while driving new rhythmicity in NF-κB and hormone synthesis pathways. Microglia showed particularly dramatic reprogramming, with disease-associated microglial (DAM) gene networks acquiring circadian oscillation in APP mice. Functional assays confirmed that microglial oxidative stress responses and amyloid phagocytosis capacity vary significantly by time of day, with direct implications for how and when these disease-relevant functions operate in vivo.

Aging produced a distinct pattern of circadian reprogramming in both astrocytes and microglia, largely non-overlapping with the amyloid-driven changes, suggesting that aging and disease remodel circadian programs through different mechanisms. The authors also demonstrated that the time of day at which mice are sacrificed substantially influences differential gene expression analysis between WT and APP/PS1 mice, with some genes appearing upregulated or downregulated in the disease model depending solely on collection time. This finding has immediate practical implications for the design and interpretation of transcriptomic studies in AD mouse models.

The entire dataset has been made publicly available as an interactive web resource, providing the field with a high-resolution, cell-type-resolved circadian gene expression atlas for brain health, Alzheimer's disease, and aging.