Coordinated multi‐cellular responses to Alzheimer's disease pathology reveal sex‐specific resilience signatures

BACKGROUND: Alzheimer's disease (AD) triggers multicellular transcriptomic responses that may reflect and moderate disease progression. While most studies observe these likely-concurrent cell-specific changes independently, the current study investigates multi-cellular associations with AD phenotypes. METHOD: We analyzed single-nuclei transcriptomics from the dorsolateral prefrontal cortex (DLPFC) of 427 donors (ROSMAP, Mathys et al.; 2.3M cells). Metacells were generated using k-nearest neighbor clustering, and hierarchical gene clustering identified cell-type-specific gene modules (Figure 1). AD-related gene modules (AGMs) were identified as clusters that showed cross-validated associations with any disease phenotypes. AGMs underwent gene-set enrichment analysis and were entered into a partial least squares (PLS) analysis, deriving multicellular interactions simultaneously predicting multiple AD neuropathological measures and clinical features. RESULT: We identified 28 AGMs across 11 cell subtypes. Cross-validation of PLS regression led to selection of three components, while permutation testing confirmed the model's non-random structure (Figure 2A). Component 1 reflected a combinatorial effect of all significant AD-related gene clusters responding to increased pathology, being strongly associated with amyloid-β (Aβ), tau and TDP-43 pathology, and negatively with APOE E2 carriage (Figure 2B-F). Components 2 (men) and 3 (women) highlighted potential sex-specific resilience responses, positively associated with age and Aβ but negatively with tau pathology (Figure 2B-F). Gene-set enrichment (Figure 3) revealed Component 1 to involve downregulation of defense responses and upregulation of cell junction and adhesion across vulnerable neuronal types, perhaps indicating movement away from defense and toward glia-mediated self-destructive processes. Enrichment further suggested that components 2 and 3 involved oligodendrocyte-mediated synaptic remodeling and neuronal immune/defense responses, with component 2 (men) involving L6B excitatory neurons and component 3 (women) more involving RORB-GABRG excitatory and PVALB-HTR4 inhibitory neurons. Astrocytes and oligodendrocyte precursor cells contributed opposing loadings relative to each other in components 2 and 3. CONCLUSION: Multiple cell types exhibit concurrent shifts, suggesting potential multi-cellular associations with AD pathology. Changes in the first component reflected a generalized response to neurodegenerative pathology, likely representing neuronal death processes of selective neuronal subpopulations. In contrast, two additional multicellular responses emerged, suggesting sex-specific cellular activity moderating resilience to AD pathology. Coordinated cell-type-specific alterations underscore the need to address cross-cellular interactions in AD.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.