Differential presence of cerebral microbleeds in an amyloid mouse model compared with a CADASIL mouse model.

INTRODUCTION: Cerebral microbleeds are commonly observed on susceptibility-based magnetic resonance imaging (MRI) in Alzheimer's disease (AD) and are often interpreted as markers of small-vessel disease. However, how microbleed occurrence differs between amyloid-associated vascular pathology and non-amyloid vascular conditions remains incompletely understood. In this study, we performed a comparative analysis between an amyloidosis mouse model and a non-amyloid arteriopathy cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) model to assess whether microbleed occurrence differs across these distinct vascular contexts. METHODS: Multimodal in vivo MRI at 11.7T, including gradient echo (GRE), spin echo, and diffusion-weighted imaging, was performed in 5xFAD mice and a CADASIL mouse model across 9 to 20 months of age. Ex vivo high-resolution GRE imaging and Prussian blue staining were used to validate microbleeds. Quantitative analyses focused on microbleed counts, parenchymal and cerebrospinal fluid volumes, and regional apparent diffusion coefficient (ADC). In addition, blood-brain barrier (BBB) integrity was assessed in the CADASIL mice. RESULTS: Both in vivo and ex vivo GRE MRI consistently revealed hippocampal microbleeds in 5xFAD mice, whereas no microbleeds were detected in CADASIL mice at any examined age. Microbleeds in 5xFAD mice occurred in the absence of brain atrophy or ventricular enlargement. ADC elevation was observed selectively in the midbrain of 5xFAD mice but not in other regions or in the CADASIL cohort. BBB permeability remained normal in CADASIL mice, indicating preserved vascular barrier integrity despite vascular smooth muscle cell loss. DISCUSSION: Cerebral microbleeds emerged selectively in the amyloid model and were absent in a non-amyloid arteriopathy driven by vascular smooth muscle cell degeneration, suggesting that microbleed occurrence may depend on the underlying pathology. These findings show consistency with the possibility that cerebral microbleeds may reflect disease-associated vascular conditions, refining their interpretation as translational MRI biomarkers in AD.