Abstract BACKGROUND With changing demographics, age-related diseases like neurodegenerative disorders and cancer are becoming increasingly prevalent. We previously demonstrated that over 50% of glioblastoma (GB) patients show Alzheimer’s Disease (AD)-like changes in the tumor-adjacent cortex. However, whether these diseases are biologically linked remains unclear. Here, we investigate cellular senescence, a hallmark of aging, as a shared mechanism. MATERIAL AND METHODS We utilized single cell and single nucleus RNA sequencing datasets from 110 GB patients (GBmap core atlas) and 89 individuals with varying stages of AD (SEA-AD MTG atlas). Following data integration, we assessed senescent cells using three established gene lists, performed differential gene expression and conducted gene set enrichment analysis. Senescence scores were compared across different cell types and patients. Co-expression modules with senescence-associated genes were identified using weighted gene co-expression network analysis (WGCNA). Quantitative neuropathological analysis of published measurements (Iba1, AT8, 6E10) complemented the molecular findings. RESULTS Senescence was found in both GB and AD patient cells including cell types of the microenvironment. Notably, the variability in scores per patient was significantly higher in GB as compared to AD cell types, indicating greater inter-patient heterogeneity. Microglia were specifically enriched for senescence gene sets among glial cells. Subset analysis of microglia through data harmonization identified a proinflammatory GB microglia subtype that shared high similarity with AD microglia and had highest senescence scores. Notably, the increase in senescent microglia was associated with progression to advanced disease stages in both diseases. WGCNA identified 3 consensus modules of senescence-associated genes that were coexpressed with disease-specific genes, like PTEN implicated phosphorylation in the AD module and SPP1 for extracellular matrix remodeling in the GB module. All senescence associated modules showed increased expression in more advanced disease stages. Quantitative analysis of immunohistochemistry measurements in the AD dataset confirmed that samples with a high senescence score showed a significantly higher density of tau pathology and an altered microglia morphology. CONCLUSION Our study identifies cellular senescence as one connection between GB and AD pathology by utilizing atlas-level transcriptomic datasets. Particularly microglia upregulate senescence signatures with disease progression. The co-expression of senescence-associated genes within disease-specific modules underscores the complexity of senescence in this context.
Read full abstract