Factors Driving Amyloid Beta Fibril Recognition by Cell Surface Receptors: A Computational Study

Alzheimer's disease (AD) has been studied extensively and is characterized by plaques deposited throughout the brain. Plaques are made of beta-amyloid (<i>A</i>β) peptides which have undergone fibrillogenesis to form insoluble <i>A</i>β fibrils (<i>fA</i>β) that are neurotoxic. Receptor for Advanced Glycation End end products (RAGE), toll-like receptors (TLRs) 2 and 4, and co-receptor CD14 recognize negatively charged binding regions on <i>fA</i>β to activate microglia and release proinflammatory cytokines. In this study, we used two experimentally resolved <i>fA</i>β structures (type I and II) isolated from AD brain tissue to elucidate binding patterns of <i>fA</i>β with RAGE, TLR2, TLR4, and CD14 and investigated whether binding was affected by fibril structure or system pH. Receptors TLR2 and RAGE formed tight complexes with both type I and II fibrils, while TLR4 showed selectivity for type I. CD14 binding was less tight and selective for type II. Binding was pH dependent for CD14, TLR4, and RAGE but not TLR2. We explored the effects of familial mutations on fibril structure to determine whether mutants of type I or II structures are feasible. Finally, we investigated whether mutations affected binding interactions of <i>fA</i>β with proteins. The Arctic (Glu22Gly), Dutch (Glu22Gln), and Iowa (Asp23Asn) mutations showed similar effects on binding affinity. Italian (Glu22Lys) mutations abrogated binding, whereas type I and II fibrils with Flemish (Ala21Gly) mutations were not shown to be feasible. Results highlight the adaptability of immune receptors in recognizing damaging molecules, with fibril structure and pH being the main recognition determinants predicated on disease progression. In silico mutations showed that aggregates similar to type I and II structures were plausible for some familial mutations.