(631a) A Cyclic Peptide Mimic of the Beta-Amyloid Binding Domain on Transthyretin

Self-association of β-amyloid (Aβ) into soluble oligomers and fibrillar aggregates is associated with Alzheimer’s disease (AD) pathology, motivating the search for compounds that selectively bind to and inhibit Aβ oligomerization and/or neurotoxicity. Numerous small molecules that can alter Aβ production and/or aggregation have been extensively explored as therapeutic agents against AD, although to date none have been clinically effective. An alternative approach is to exploit the use of peptides and peptidomimetics that bind to Aβ, since peptides have potential advantages over small molecules in terms of better target affinity and specificity. In general, two divergent strategies have been employed in the search for peptides that bind to Aβ: (i) using self-complementation and (ii) screening random peptide-based libraries.

Alternatively, one could design peptides that bind to Aβ based on knowledge of complementary binding proteins. Transthyretin (TTR), a plasma transport protein that is also abundant in cerebrospinal fluid, has been shown to bind to Aβ, inhibit aggregation, and reduce its toxicity. Previously, we identified strand G of TTR as a specific Aβ binding domain, and developed TTR-derived peptides as Aβ inhibitors. G16, a 16-mer peptide with a sequence matching that of strand G of TTR, was shown to be protective against Aβ toxicity. Although both TTR and G16 bound to Aβ and inhibited toxicity, they differed significantly in their effect on Aβ aggregation. Further, G16 was less effective than TTR at protecting cells from Aβ toxicity.

In an effort to bridge the gap in efficacy between native TTR and G16, we took the concept one step further by mimicking not only the sequence of the binding epitope, but also its structure. The sequence of G16 spans two β-strands (strand G and H) connected by a short loop. However, G16 lacks the β-strand/β-hairpin/β-strand structure of the binding domain in TTR. To mimic the alignment of each residue in binding epitope, we designed a cyclic peptide where a hairpin turn is enforced via a dipeptide D-Pro-L-Pro. Two cyclic peptides with 18 and 22 amino acids were synthesized using an orthogonally protected glutamic acid derivative and solid-phase synthesis. Synthesis was confirmed by mass spectrometry and circular dichroism. We examined the ability to bind to Aβ by photo-induced cross-linking and proteolytic fragmentation. We observed that the cyclic peptide with 22 amino acids binds to Aβ while the shorter cyclic peptide does not. We next evaluated the effect of cyclic peptide on Aβ aggregation kinetics by dynamic light scattering. The cyclic peptide increases both the average hydrodynamic size and the rate of growth of Aβ aggregates, and the effect was stronger compared to that of G16. Finally, we observed that the cyclic peptide was able to protect neurons against Aβ toxicity at relatively low concentration. Therefore, a beta hairpin mimetic of Aβ binding epitope on TTR can interact with Aβ specifically and retain the protective role of TTR.