TOX-AD: Neurometabolic network reorganization in AD mouse models exposed to Metal/Metalloids

 

Leandro C Freitas-Lima^1*; Kevin P Kotredes²; Juliet A Moncaster^3,4; Scott Persohn¹; Kierra Eldridge¹; Juan Antonio K. Chong Chie¹; Gareth R Howell²; Lee E Goldstein^3,4; Paul R Territo^1,5

¹Indiana University School of Medicine, Indianapolis, IN, USA; ²The Jackson Laboratory, Bar Harbor, ME, USA; ³Boston University Chobanian & Avedisian School of Medicine, Boston, MA, USA; ⁴Boston University Alzheimer’s Disease Research Center, Boston, MA, USA; ⁵Stark Neuroscience Research Institute

 

Background: Alzheimer’s Disease (AD) is a multifactorial disease in which metal/metalloid neurotoxicants are suspected modifiers of AD pathobiology. However, the molecular mechanisms underlying gene-by-environment (GxE) interactions during aging remain unclear, underscoring the importance of investigating the exposome in AD pathobiology. Thus, we hypothesize that arsenic (As), cadmium (Cd), and lead (Pb) exposures alter neural connections and their roles in cognition and disease, thereby potentiating AD pathogenesis in a toxicant-specific, neurodevelopment-sensitive, and age-dependent manner.

Methods: MODEL-AD humanized mouse models (B6J.hAPOE3.hAβ.hMAPT, B6J.hAPOE4.hAβ.hMAPT) were exposed to As, Cd, or Pb via drinking water at human-relevant concentrations (e.g., 20, 50, and 200 ppm, respectively). After exposure, we performed neuroimaging (PET/CT) to assess amyloid-β (Aβ) deposition and neuro-metabolism. Mice were euthanized, perfused, and tissues were harvested. Brains were hemisected for further analysis. Using covariance analysis and consensus clustering, we evaluated functional connectivity changes in the AD stage by calculating whole-brain entropy, degree distribution, nodal connection strength, and clustering coefficients from ¹⁸F-FDG images. Finally, we analyzed ¹⁸F-FDG and ¹⁸F-AV45 separately for positive and negative regional correlations.

Results: After exposure, these neurotoxicants altered animal behavior and accumulated in the brain, leading to aberrant Aβ deposition. Additionally, metabolic connectivity analysis revealed changes in degree distribution, connection strength, and clustering coefficients due to metal/metalloid exposure, suggesting metabolic brain network reorganization, either as compensatory mechanisms or as signs of pathological disruption. Moreover, ¹⁸F-FDG and ¹⁸F-AV45 were significantly correlated in 6/6/4/6 of 107 regions for Control/As/Cd/Pb, respectively, suggesting that Aβ deposition has a small or no significant impact on metabolic brain network reorganization.

Conclusion: MODEL-AD mice exposed to metals/metalloids showed region-specific metabolic alterations in the brain and Aβ plaque deposition. These exposures revealed toxicant-specific neurometabolic alterations, enabling us to quantify and correlate them with similar disease signatures observed in human patients by assessing metabolic alterations in brain connectivity. Thus, investigating GxE interactions highlights the importance of integrating the exposome into the AD field and of analyzing its progression over aging.

Acknowledgments: TOX-AD U01-AG088683 (LG, PT, GH); R56-AG083940 (LG, GH)