Unraveling shared immunotoxicological mechanisms linking exposure to PPCPs with Alzheimer's disease and atherosclerosis: integrative multi-omics, network toxicology, and experimental validation.
TL;DR
To explore how exposure to PPCPs may contribute to AD-AS comorbidity through shared immune-toxicology mechanisms, and to identify natural compounds as candidate interventions against prioritized disease-relevant targets, integrated AD and AS bulk transcriptomic datasets were integrated.
Abstract
Background
Pharmaceuticals and personal care products (PPCPs) are pseudo-persistent contaminants with potential exposure risks. Atherosclerosis (AS) and Alzheimer's disease (AD) share neurovascular and myeloid-immune remodeling, but whether chemical-target networks associated with exposure to PPCPs converge on shared mechanisms linking AD and AS remains unclear.
Purpose
To explore how exposure to PPCPs may contribute to AD-AS comorbidity through shared immune-toxicology mechanisms, and to identify natural compounds as candidate interventions against prioritized disease-relevant targets.
Methods
We integrated AD and AS bulk transcriptomic cohorts with single-cell datasets to define shared disease targets and locate them within relevant immune-cell states. Chemical-gene interactions associated with exposure to PPCPs were then overlaid onto this AD-AS target network, and candidate natural compounds were prioritized by reverse network pharmacology, structure-based virtual screening, molecular docking, molecular dynamics and genetic evidence analyses. The prioritized targets and intervention hypotheses were further examined in APP/PS1 AD mice, ApoE-knockout high-fat-diet AS mice, RAW264.7 macrophage-like cells and HMC3 microglial cells.
Results
The integrated analyses identified TREM2, SPP1, TGFBI and MRGPRF as four AD-AS signature genes, with diagnostic discrimination in independent AD (AUC 0.783-0.996) and AS (AUC 0.892-0.924) cohorts. Among them, TREM2 and SPP1 emerged as the key signature genes, were highly expressed in disease-associated microglial and macrophage states, and formed an SPP1-linked myeloid communication axis with neighboring disease niches. Metabolic scoring and in silico TREM2 perturbation pointed to lipid-metabolic remodeling in AD microglia and macrophage metabolic, adhesive and inflammatory-migration programs in AS lesions. Bisphenol A, bisphenol F, estradiol and acetaminophen were identified as PPCPs related to signature-gene regulation. Salvianolic acid B, β-sitosterol, quercetin and tanshinone IIA were identified as potential phytochemical modulators of TREM2, SPP1, TGFBI and MRGPRF, respectively, with favorable structural-stability evidence. MR analyses supported genetic causal associations between the signature genes and AD/AS outcomes, while TREM2-centered SMR and GTEx v10 eQTL colocalization for TREM2, SPP1 and TGFBI provided locus-level support. In vivo, TREM2, OPN/SPP1, and TGFBI were upregulated in both AD and AS tissues, whereas MRGPRF showed disease-context-dependent regulation, with increased expression in AD tissues but decreased expression in AS tissues. In vitro, BPA decreased TREM2 and increased OPN/SPP1, accompanied by enhanced lipid deposition, CD86 activation, and inflammatory transcription, whereas salvianolic acid B increased TREM2, suppressed OPN/SPP1, and attenuated lipid deposition and inflammatory responses.
Conclusions
TREM2, SPP1, TGFBI, and MRGPRF constitute a recurrent but directionally heterogeneous molecular signature across AD and AS, with MRGPRF showing disease-context-dependent regulation. Within this signature, TREM2 and SPP1 showed the strongest mechanistic convergence in disease-associated myeloid states. BPA disrupted the TREM2/OPN-SPP1 axis in microglia and macrophages, whereas salvianolic acid B partially counter-regulated these changes and attenuated associated inflammatory and lipid-metabolic disturbances.