Whole-body senescent cell clearance alleviates age-related brain inflammation and cognitive impairment in mice

Abstract

Cellular senescence is characterized by an irreversible cell cycle arrest and a pro-inflammatory senescence-associated secretory phenotype (SASP), which is a major contributor to aging and age-related diseases. Clearance of senescent cells has been shown to improve brain function in mouse models of neurodegenerative diseases. However, it is still unknown whether senescent cell clearance alleviates cognitive dysfunction during the aging process. To investigate this, we first conducted single-nuclei and single-cell RNA-seq in the hippocampus from young and aged mice. We observed an age-dependent increase in p16^Ink4a senescent cells, which was more pronounced in microglia and oligodendrocyte progenitor cells and characterized by a SASP. We then aged INK-ATTAC mice, in which p16^Ink4a -positive senescent cells can be genetically eliminated upon treatment with the drug AP20187 and treated them either with AP20187 or with the senolytic cocktail Dasatinib and Quercetin. We observed that both strategies resulted in a decrease in p16^Ink4a exclusively in the microglial population, resulting in reduced microglial activation and reduced expression of SASP factors. Importantly, both approaches significantly improved cognitive function in aged mice. Our data provide proof-of-concept for senolytic interventions' being a potential therapeutic avenue for alleviating age-associated cognitive impairment.

Keywords: SASP; aging; brain; cognition; memory; neurodegeneration; senescence; senolytic; telomeres.

FIGURE 1

Senescent cells accumulate in hippocampus during aging. Hippocampal tissues were isolated from young (4 months) and old (24 months) mice. Single‐cell (sc) or single‐nucleus (sn) suspensions were prepared from hippocampal tissue and used for RNA‐seq. (a) UMAP embeddings of sc‐RNA‐seq (top) or sn‐RNA‐seq (bottom) were investigated in our datasets. Colors represent cell type annotations. Bar graphs on the right show changes in fractions of cell populations isolated from young and old mice. (b) Matching UMAP embeddings with expression of p16^Ink4a for sc‐RNA‐seq (top) or sn‐RNA‐seq (bottom) datasets. Color frames represent cell type annotation as above. % of cells positive for high levels of p16^Ink4a is shown next to each cell population. (c) Heatmap is showing expression of cytokines in various types of brain cells. Bars above the heatmap indicate color‐matched cell types consistent with graphs above. (d) Violin plots show expression of cytokines in cells showing high or no expression of p16^Ink4a. Differential expression was carried out with the Seurat and MAST packages for single‐cell analysis. Multiple testing correction was conducted by computing a false discovery rate (bonferroni correction) using all genes in the dataset. The majority of significantly (fdr <0.05) differentially expressed SASP genes were upregulated in Cdkn2a + cells. Total of 4 mice per condition (young, old) are analyzed, while 2 mice are pooled for each experiment

FIGURE 2

Whole‐body senescent cell clearance alleviates age‐induced spatial memory dysfunction. (a) 4‐month‐old and 25–29‐month‐old mice were split into three groups and assigned to vehicle (v), Dasatinib and Quercetin (D + Q), or AP20187 (AP) treatments. Animals were matched by baseline parameters: body mass, body composition, Rotarod performance, and cognition. After 2 months of drugs or vehicle, mice were tested again (post‐treatment assessment) for changes in body parameters, cognition, and Rotarod performance. Shortly after the post‐treatment assessments, mice were sacrificed. (b) Schematics of the Stone T‐maze are presented. All of the numbers designating dimensions indicate length in centimeters. The parameters of the spatial memory (c), the frequency of errors, and (d) the time to finish the test were assessed in young and old INK‐ATTAC mice using the Stone T‐maze. As the effect of AP, D + Q, or vehicle treatment on memory functional parameters: (e) the frequency of errors and (f) the time to finish the test were assessed in old mice. Post‐treatment results: (g) frequencies of errors and (h) the time to finish the maze were normalized to baseline and analyzed. All data are mean ± SEM with n = 9–11 for old mice, n = 7–8 for young mice. *p < 0.05, **p < 0.01

FIGURE 3

Clearance of senescent cells targets p16^Ink4a‐positive microglia and reduces markers of neuronal senescence in the CA3 region of the hippocampus. (a) Micrographs show immuno‐RNA‐ISH staining against p16^Ink4a mRNA and microglial marker Iba1 on brain sections in hippocampal region of young and old animals (blue = DAPI, green = iba1, red = p16 (in distinct foci), scale bars = 10 μm). Quantification of (b) fraction of p16^Ink4a ‐positive microglia and (c) the total number of microglia in murine hippocampal sections. Quantification of fraction of p16^Ink4a‐positive (d) oligodendrocytes and (e) neurons of CA3 region. (f) Lamin B1 staining Micrographs showing Lamin B1 staining of brain sections in vehicle‐ (v), AP20187‐ (AP), or Dasatinib and Quercetin (D + Q)‐treated animals (blue = DAPI, green = Lamin B1, scale bar 50 μm, white arrows mark Lamin B1‐negative neurons) (g) Quantification of Lamin B1‐negative neurons. (h) Representative images of HMGB1 staining in the hippocampus, HMGB1 stained by NovaRed (Brown) and counterstained with Hematoxylin (Blue) (scale bar is 50 μm). Black arrows heads mark HMGB1‐negative neurons. (i) Quantification of neurons with no nuclear HMGB1 in the pyramidal layer of the CA3 region of the hippocampus. (j) Representative images of p21 RNA‐ISH in the CA3 hippocampus of INK‐ATTAC animals with the indicated treatments. (k) Quantification of p21 mRNA‐positive cells in the pyramidal layer of the CA3 hippocampal region (red = p21 mRNA, blue = DAPI, scale bar 50 μm). White arrows heads mark p21‐positive neurons. All data are mean ± SEM with n = 5–9 for A‐E ad n = 6–10 for F‐K. *p < 0.05, **p < 0.01

FIGURE 4

Treatments targeting senescent cells reduce brain inflammation. (a) Changes in cytokine protein levels [fold change] in the brains of young mice and old mice treated with vehicle (v), AP20187 (AP), or Dasatinib and Quercetin (D + Q). Cytokine concentrations significantly (p < 0.05) change between young and old brains during aging. Only cytokines that showed significant changes between young and old are shown. (b) Graph shows mean number of IL‐1α mRNA foci per cell of CA3 pyramidal layer. (c) Graph shows mean number of MCP‐1 mRNA foci per cell in the CA3 pyramidal layer. (d) Representative images of immunohistochemical staining using antibodies against Iba‐1 protein. (e) Quantification of the average soma size of microglia in the CA3 hippocampal region. (f) Representative images of CD3 immunofluorescence in the area of lateral ventricle (LV) proximal to hippocampus of young vehicle‐treated and old mice treated with vehicle, AP, or D + Q (scale bars 10 μm). (g) Quantification of the mean number of CD3⁺ cells around LV of young and old mice with or without treatment (CD3 is green, DAPI is blue). All data are mean ± SEM n = 4–10. *p < 0.05, **p < 0.01