Back

Cell composition, transcriptomic, and functional pathway changes in the hippocampus in Alzheimer's disease and overlap with lead (Pb) exposure signatures

Wang, H.; Matei, E.; Dou, J.; Morgan, R. K.; Colacino, J.; Bakulski, K. M.

2026-07-22 toxicology
10.64898/2026.07.21.26358590 medRxiv
Show abstract

Background: Lead (Pb) is associated with Alzheimer's disease (AD); however, the relationships between Pb and AD hippocampal transcription remains unclear. We evaluated overlap between Pb-response signatures and cell-type-independent AD transcriptomic signatures. Method: Three toxicology studies (two neuronal cell lines, one mouse hippocampus) provided Pb-response genes. Five human postmortem hippocampal AD case-control transcriptional datasets (n=90 AD, n=106 normal cognition) were cell type deconvoluted and tested with beta regression. Differential gene expression, adjusted for age, sex, and estimated cell-types, were meta-analyzed. Overlapping Pb and AD genes and biological pathways were identified (padj<0.05). Results: Consistent Pb response was observed at 25 genes (INPP5F, KIF20B, KIFC1) and 47 pathways (ensheathment of neurons, glial cell differentiation, regulation of nervous system processes). Relative to controls, AD samples had fewer neurons (-2.46%), greater microglia (0.42%), astrocytes (0.31%), oligodendrocytes (0.46%), and endothelial cells (0.95%), and 1,455 differentially expressed genes, which were enriched for cellular energy production and metabolism pathways. Six genes (EHD3, LAP3, NRXN3, PPP1R16B, RPL29, THRA) and four pathways (synaptic vesicle maturation, vesicle docking) overlapped between Pb and AD. Conclusion: We identified overlapping Pb and AD transcriptomic signatures and pathways, providing molecular context for epidemiologic associations.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.