Human amyloid-β enriched extracts: evaluation of in vitro and in vivo internalization and molecular characterization

Abstract

[Background]: Intracerebral inoculation of extracts from post-mortem human Alzheimer’s disease brains into mice produces a prion-like spreading effect of amyloid-β. The differences observed between these extracts and the synthetic peptide, in terms of amyloid-β internalization and seed and cell-to-cell transmission of cytosolic protein aggregates, suggest that brain extracts contain key contributors that enhance the prion-like effect of amyloid-β. Nevertheless, these potential partners are still unknown due to the complexity of whole brain extracts.

[Methods]: Herein, we established a method based on sequential detergent solubilization of post-mortem samples of human brains affected by Alzheimer’s disease that strongly enrich amyloid-β aggregates by eliminating 92% of the remaining proteins. Internalization of Aβ1–42 from the enriched AD extracts was evaluated in vitro, and internalization of fluorescent-labeled AD extracts was also investigated in vivo. Furthermore, we carried out a molecular characterization of the Aβ-enriched fraction using label-free proteomics, studying the distribution of representative components in the amygdala and the olfactory cortex of additional human AD brain samples by immunohistochemistry.

[Results]: Aβ1–42 from the enriched AD extracts are internalized into endothelial cells in vitro after 48 h. Furthermore, accumulation of fluorescent-labeled Aβ-enriched extracts into mouse microglia was observed in vivo after 4 months of intracerebral inoculation. Label-free proteomics (FDR < 0.01) characterization of the amyloid-β-enriched fraction from different post-mortem samples allowed for the identification of more than 130 proteins, several of which were significantly overrepresented (i.e., ANXA5 and HIST1H2BK; p < 0.05) and underrepresented (i.e., COL6A or FN1; p < 0.05) in the samples with Alzheimer’s disease. We were also able to identify proteins exclusively observed in Alzheimer’s disease (i.e., RNF213) or only detected in samples not affected by the disease (i.e., CNTN1) after the enrichment process. Immunohistochemistry against these proteins in additional tissues revealed their particular distribution in the amygdala and the olfactory cortex in relation to the amyloid-β plaque.

[Conclusions]: Identification and characterization of the unique features of these extracts, in terms of amyloid-β enrichment, identification of the components, in vitro and in vivo cell internalization, and tissue distribution, constitute the best initial tool to further investigate the seeding and transmissibility proposed in the prion-like hypothesis of Alzheimer’s disease.