Multimodal MRI Reveals Cerebral and Vascular Amyloid‐Driven Myeloarchitectural Disorganization in a Mouse Model of Alzheimer's Disease

Alzheimer's disease (AD) is a multifactorial neurodegenerative disorder involving a complex interaction of cerebral and vascular amyloid-beta (Aβ) accumulation, myelin disruption, lipid alterations, and cerebrovascular dysfunction. The early detection and differentiation of these interconnected pathologies remain challenging. To understand the effect of cerebral and vascular Aβ on regional myeloarchitecture and lipid composition, we developed a novel multimodal neuroimaging approach integrating quantitative MRI (qMRI), chemical exchange saturation transfer (CEST) MRI, and immunohistochemistry (IHC). The framework was applied to a 10-month-old mouse model exhibiting both cerebral and vascular amyloid pathologies. High-resolution in vivo MRI was performed using a 9.4 Tesla scanner. The results suggest region-specific vulnerability to Aβ pathology with significant regional increases in apparent transverse relaxation rate (R2*, p < 0.05; Hedges' g = 1.22) and quantitative susceptibility mapping (χ, p < 0.01; Hedges' g = 1.99) within the hippocampus of ARTE10 mice compared to wild-type littermates. Multislice CEST-based Z-spectra were used with multipool Lorentzian fitting and quantitative T1 longitudinal relaxation maps to obtain nuclear Overhauser enhancement (NOE) weighted apparent exchange-dependent relaxation (AREX) maps. NOE _{(-3.5 ppm)}-sensitive CEST imaging contrast showed region-specific changes in the hippocampus (p < 0.01; Hedges' g = -1.81), corpus callosum (p < 0.01; Hedges' g = -2.39), and thalamus (p < 0.01; Hedges' g = -2.64) of ARTE10 animals relative to WT littermates. Hippocampal Aβ burden, iron load, and myelin density were quantified using immunohistochemistry, suggesting strong Aβ plaque presence and elevated iron load in the hippocampi of ARTE10 mice compared to WT mice. Collectively, our results demonstrate the utility of this multimodal MRI framework in identifying sensitive and specific biomarkers of amyloid-driven myeloarchitectural and molecular changes. The proposed framework offers a valuable tool for enhancing early detection, understanding different pathophysiological pathways, and facilitating therapeutic monitoring and targeted intervention strategies.