Blood-brain barrier-associated pericytes internalize and clear aggregated amyloid-ß42 by LRP1-dependent apolipoprotein E isoform-specific mechanism.

ABSTRACT

BACKGROUND: Clearance at the blood-brain barrier (BBB) plays an important role in removal of Alzheimer's amyloid-ß (Aß) toxin from brain both in humans and animal models. Apolipoprotein E (apoE), the major genetic risk factor for AD, disrupts Aß clearance at the BBB. The cellular and molecular mechanisms, however, still remain unclear, particularly whether the BBB-associated brain capillary pericytes can contribute to removal of aggregated Aß from brain capillaries, and whether removal of Aß aggregates by pericytes requires apoE, and if so, is Aß clearance on pericytes apoE isoform-specific. METHODS: We performed immunostaining for Aß and pericyte biomarkers on brain capillaries (< 6 µm in diameter) on tissue sections derived from AD patients and age-matched controls, and APPSwe/0 mice and littermate controls. Human Cy3-Aß42 uptake by pericytes was studied on freshly isolated brain slices from control mice, pericyte LRP1-deficient mice (Lrplox/lox;Cspg4-Cre) and littermate controls. Clearance of aggregated Aß42 by mouse pericytes was studied on multi-spot glass slides under different experimental conditions including pharmacologic and/or genetic inhibition of the low density lipoprotein receptor related protein 1 (LRP1), an apoE receptor, and/or silencing mouse endogenous Apoe in the presence and absence of human astrocyte-derived lipidated apoE3 or apoE4. Student's t-test and one-way ANOVA followed by Bonferroni's post-hoc test were used for statistical analysis. RESULTS: First, we found that 35% and 60% of brain capillary pericytes accumulate Aß in AD patients and 8.5-month-old APPSw/0 mice, respectively, compared to negligible uptake in controls. Cy3-Aß42 species were abundantly taken up by pericytes on cultured mouse brain slices via LRP1, as shown by both pharmacologic and genetic inhibition of LRP1 in pericytes. Mouse pericytes vigorously cleared aggregated Cy3-Aß42 from multi-spot glass slides via LRP1, which was inhibited by pharmacologic and/or genetic knockdown of mouse endogenous apoE. Human astrocyte-derived lipidated apoE3, but not apoE4, normalized Aß42 clearance by mouse pericytes with silenced mouse apoE. CONCLUSIONS: Our data suggest that BBB-associated pericytes clear Aß aggregates via an LRP1/apoE isoform-specific mechanism. These data support the role of LRP1/apoE interactions on pericytes as a potential therapeutic target for controlling Aß clearance in AD.