Border-Associated Macrophage Migrasomes in Alzheimer's Disease: An Emerging Aβ-Senescence-Microglia Axis?

Alzheimer's disease (AD) involves not only the neural parenchyma but also brain-border interfaces in which border-associated macrophages (BAMs) regulate amyloid-β (Aβ) handling, vascular function, and immune surveillance. This review focuses on a pathological axis in which persistent vascular Aβ40 and aging-related stress shift clearance-competent BAMs toward oxidative-stress and senescent-like states. Two experimentally supported but incompletely connected branches are emphasized. In one, Aβ engages macrophage CD36-NOX2 signaling and generates reactive oxygen species that impair neurovascular function. In the other, Aβ40 internalization promotes TSPAN4-dependent migrasome formation and enrichment of CD5L/AIM; vascularly deposited CD5L/AIM lowers endothelial CD59, facilitates C5b-9 formation, and damages the blood-brain barrier. A separate aging study indicates that CD5L/AIM-rich BAM migrasomes can transmit apoptosis resistance and senescence-like dysfunction to microglia. Human evidence is currently strongest for CAA rather than parenchymal AD: a small CAA cohort showed increased circulating CD14-positive migrasomes and monocyte TSPAN4, with an exploratory area under the ROC curve of approximately 0.91 for TSPAN4-positive monocytes versus healthy controls, whereas AD patients selected to lack imaging evidence of CAA did not show increased circulating migrasome counts. Accordingly, the Aβ40-TSPAN4-CD5L/AIM vascular branch should presently be regarded as a CAA-enriched mechanism that may be especially relevant to AD with prominent CAA, not as a universal mechanism of Aβ42-dominant sporadic AD without substantial vascular amyloid. Direct demonstration of BAM-derived migrasomes in human AD brain tissue is still lacking, and neither TSPAN4 nor CD5L/AIM is sufficiently specific to serve as a stand-alone biomarker. We therefore distinguish peer-reviewed human observations, experimental causal evidence, preprint findings, and proposed cross-pathway interactions and outline biomarker validation and pathway-selective therapeutic strategies that preserve beneficial BAM functions.