Menu
Background: Anti-amyloid monoclonal antibody (mAb) treatment has been associated with vasogenic edema and microhemorrhages, known as amyloid-related imaging abnormalities (ARIA), in some Alzheimer’s disease (AD) patients, especially in ApoE4 carriers. Here, we compared recombinant 3D6-L, a murine version of the anti-amyloid mAb bapineuzumab, which induced ARIA in AD patients, and an isotype control IgG2a mAb to investigate potential mechanisms, including complement activation, involved in these vascular side effects following passive immunization.
Methods: Plaque-rich 16.5-mo-old APP/PS1dE9; hApoE4 mice were treated weekly (i.p.) for 13 weeks with ~15 mg/kg (500 μg) 3D6-L or IgG2a (n=5/group). Amyloid-beta (Aβ) and ApoE levels were measured in the brain homogenates. Complement C1q and C3 levels in the brain and serum were determined using ELISA, and microhemorrhages in the brain were quantified using Prussian-blue hemosiderin staining. Additionally, 20-mo-old APP/PS1dE9;hApoE4 mice underwent a shorter 7-week passive immunization (500 μg/week 3D6-L or IgG2a, n= 3/group) followed by RNAseq to evaluate differentially expressed genes (DEGs) in the brain and FACS sorted CD31+ vascular endothelial cells. Pathological analyses to investigate downstream complement signaling and immune cell infiltration are ongoing.
Results: In the 13-week study, 3D6-L significantly reduced guanidium-HCL-soluble Aβx-42 and Aβx-40; ApoE levels showed a trend for reduction. Soluble C3 levels in the brain rose significantly, correlating with increased microhemorrhages in leptomeninges and large blood vessels in the cortex and cerebellum. The peripheral serum C3 levels in the 3D6-L treated mice were significantly lower compared to IgG2a treated mice, while C1q levels had a non-significant trend of reduction in the 3D6-L treated mice. In the shorter immunization study, brain transcriptome analysis of bulk RNA revealed significant upregulation of many genes related to complement activation (C1qa, C1qb, Cq1qc, C4b, C3), glial activation (GFAP), IgG receptor activity (Fcgr1, Fcgr2, Fcgr3), lysozyme (Lyz2) and Galectin-3 (LGALS3). Endothelial cell transcriptome revealed increased expression of genes involved in inflammation and lipid metabolism. Some of these include IL1r1, Cxcl12, Lcn2, Abca1, CH25H, vWf, and Cfh. In both studies, 3D6-L induced microhemorrhages and were associated with an increase in cerebral amyloid angiopathy (CAA)-associated complement activation and endothelial inflammation. Based on our RNAseq data, we are investigating potential markers, including iC3b, C5b-9, MMP-9, and vWf, involved in immunotherapy-associated microhemorrhages.
Conclusion: Passive immunization against Aβ using 3D6-L amplified CAA-associated complement activation and altered expression of genes related to extracellular matrix degradation and vascular inflammation. The observed microhemorrhages appear to result from complement-mediated vascular inflammation induced by the binding of 3D6-L to vascular amyloid. These findings suggest potential biomarkers or targets for therapeutic interventions in the context of anti-Aβ immunization.
