Aggregation-dependent epitope sequence and modification fingerprints of anti-Aβ antibodies

A hallmark of Alzheimer's disease (AD), the most common form of dementia, is the progressive accumulation of amyloid-beta (Aβ) peptides across distinct brain regions. Anti-Aβ antibodies (Aβ-Abs) targeting specific Aβ variants are essential tools for AD research, diagnostics, and therapy. The monoclonal antibodies Aducanumab, Lecanemab, and Donanemab have recently been approved as the first disease-modifying treatments for early AD, highlighting the clinical importance of their exact binding profiles. In this study, we systematically characterized the binding and modification requirements of 20 Aβ-Abs, including biosimilars of Aducanumab, Lecanemab, and Donanemab, across monomeric, oligomeric, and aggregated Aβ forms. Array-based analysis of 20,000 modified Aβ peptides defined binding epitopes at single-residue resolution and revealed the impact of sequence variation, including familial AD mutations, as well as diverse post-translational modifications (PTMs). Notably, genetic variants, such as H6R, impaired binding of therapeutic Aβ-Abs like Aducanumab. Donanemab showed strong preference for pyroglutamate-modified AβpE3-17, while Lecanemab and Aducanumab exhibited aggregation- and sequence-context-dependent binding requirements. Comparison of peptide binding profiles with binding of full-length and aggregated Aβ via immunoprecipitation-mass spectrometry, capillary immunoassays, Western blotting, and immunohistochemistry on AD brain tissue revealed distinct aggregation-dependent binding behaviours. The valency- and context-dependence of Aducanumab binding, together with its preference for Ser8-phosphorylated Aβ, supports a dimerization-mediated binding mechanism. For Lecanemab, our data suggest that additional structural contributions beyond the minimal N-terminal epitope are required for binding to aggregated Aβ, which remain to be fully resolved. Together, this work provides the most comprehensive dataset to date on aggregation-dependent sequence and modification selectivity of Aβ-Abs. By integrating mutational, PTM, and aggregation contexts in a unified experimental framework, we establish a resource that enables rational selection of antibodies for research and diagnostic applications and offers mechanistic insights that may inform the design and optimization of future therapeutic antibodies in AD.