TANGO-Inspired Design of Anti-Amyloid Cyclic Peptides

β-Amyloid peptide (Aβ) self-associates into oligomers and fibrils, in a process that is believed to directly lead to neuronal death in Alzheimer’s disease. Compounds that bind to Aβ, and inhibit fibrillogenesis and neurotoxicity, are of interest as an anti-Alzheimer therapeutic strategy. Peptides are particularly attractive for this purpose, because they have advantages over small molecules in their ability to disrupt protein–protein interactions, yet they are amenable to tuning of their properties through chemical means, unlike antibodies. Self-complementation and peptide library screening are two strategies that have been employed in the search for peptides that bind to Aβ. We have taken a different approach, by designing Aβ-binding peptides using transthyretin (TTR) as a template. Previously, we demonstrated that a cyclic peptide, with sequence derived from the known Aβ-binding site on TTR, suppressed Aβ aggregation into fibrils and protected neurons against Aβ toxicity. Here, we searched for cyclic peptides with improved efficacy, by employing the algorithm TANGO, designed originally to identify amyloidogenic sequences in proteins. By using TANGO as a guide to predict the effect of sequence modifications on conformation and aggregation, we synthesized a significantly improved cyclic peptide. We demonstrate that the peptide, in binding to Aβ, redirects Aβ toward protease-sensitive, nonfibrillar aggregates. Cyclic peptides designed using this strategy have attractive solubility, specificity, and stability characteristics.