Single-cell multi-omic and multi-region atlas of the Aging Mouse Brain

Aging is the primary risk factor for many neurodegenerative diseases, primarily attributed to progressive changes in cellular epigenomes. Despite this known association, the complexity of the mammalian brain complicates the identification of brain regions and cell types that are most susceptible to aging effects. While alterations in gene expression have been observed in aging neural cells, comprehension of the epigenetic regulatory mechanisms underlying the transcriptomic changes remains largely elusive. To unveil these complexities in a comprehensive manner, we utilized single-nucleus methylome sequencing (snmC-seq3) and multi-omic sequencing (snm3C-seq) to produce extensive datasets comprising 132,551 methylomes and 72,666 chromatin conformation-methylome joint profiles from eight distinct brain regions of C57BL/6J mice aged 2, 9, and 18 months. These regions include the anterior hippocampus (AHC), posterior hippocampus (PHC), frontal cortex (FC), amygdala (AMY), nucleus accumbens (NACB), PAG/PCG, entorhinal cortex (ENT), and caudate putamen (CP). We utilized this dataset to investigate age-related changes in DNA methylomes and transcription factor motifs potentially affected by methylation alterations. We also examined methylation changes in transposable elements, observing locus-specific retrotransposon activation. At the chromatin conformation level, we analyzed aging-related changes in compartmentalization, topologically associating domains (TADs), and chromatin loops. Furthermore, we explored regional heterogeneity during aging within identical cell types and validated our findings using the MERFISH dataset. C57BL/6J (males and females) mice of different ages were purchased from Jackson Laboratories and maintained in the Salk animal barrier facility on 12-hour dark-light cycles with food ad-libitum for up to 4 weeks (Housing condition: Temperature: 21-23 °C, relative humidity: 61-63%). Brains were extracted, sliced, and dissected in an ice-cold dissection buffer as previously described (Liu et. al. 2021). Females were synchronized for their estrous cycle by exposing them to male bedding for three days, and collected during diestrus. For snmC-seq3 and snm3C-seq samples, brains were sliced coronally at 600 μm intervals from the frontal pole across the whole brain, yielding 18 slices, and dissected brain regions were obtained according to the Allen Brain Reference Atlas Common Coordinate Framework version 3 (CCFv3) as previously described (Liu et. al. 2023). For nuclei isolation, each region was dissected from 2-7 animals and pooled to make 2-3 biological replicates. No statistical methods were used to predetermine sample size. We empirically determined to use two to three biological experiments for all single-cell epigenomic experiments to achieve minimum reproducibility for this large-scale project. Blinding and randomization is not performed during handling the tissue samples. Additionally, all dissected regions were digitally registered into CCFv3 using ITK-SNAP60 (v4.0.0) at a 25 μm resolution.