Computationally engineered cyclic peptides reduce prion levels in vitro

Prion diseases are neurodegenerative disorders associated with the structural conversion of the cellular prion protein (PrPc) into its misfolded infectious isoform (PrPSc). Despite substantial efforts, no disease-modifying therapy or cure is currently available. Here, we present an integrated computational-experimental pipeline for the rational design of cyclic peptides targeting PrPc to inhibit its pathogenic conversion. Starting from crystal structures of antibody-bound mouse PrPc, we develop a rational design strategy combined with iterative molecular dynamics simulations and sequence optimization to generate peptides with enhanced binding and structural impact. Three candidates were selected for experimental validation. Our results show that [Formula] (49YGPDPSDSYT58, antibody numbering) that binds stably to the 2-3 interface most effectively reduced PrPSc levels in GT1-7 cells, essentially by inducing allosteric re-arrangements that reinforce the intramolecular helical bundle. [Formula] (89GQSNTKPYT97) and [Formula] (89RQSNTWPYT97) binding the {beta}1-1/3 junction exerted more modest effects due to the potential competition of the flexible tail to bind at this site. These results establish a mechanistic link between peptide-induced stabilization of PrPc and inhibition of prion propagation and provide a generalizable framework for designing conformational stabilizers of aggregation-prone proteins.