An efficient biochemical method for characterizing and classifying potentially amyloidogenic and therapeutic peptides

Abstract

Amyloidosis are proteinopathies characterized by systemic or organ-specific deposition of proteins in the form of amyloid fibers. Nearly forty proteins have been identified to play a role in these pathologies and the structures of the associated fibers are beginning to be determined by Cryo-EM. However, the molecular events underlying the process, such as fiber nucleation and elongation, are poorly understood, which impairs developing efficient therapies. In most cases, only a few dozen amino acids of the pathological protein are found in the final structure of the fibers, while amyloid peptides comprising 5 to 10 amino acids are involved in fibers nucleation process. The identification and biochemical characterization of these peptides is therefore of major scientific and clinical importance. In silico approaches are limited due to the peptides small size and long-distance intra- and intermolecular interactions that occur during nucleation. To address this problem, we developed a novel biochemical method for characterizing and classifying batches of related peptides. Initial work to optimize our approach is based on the reference peptide PHF6 ({beta}1) from Microtubule Associated Protein Tau (MAPT) as compared to 22 related peptides. We classified these peptides into groups displaying different biochemical properties, and thereby identified new amyloid peptides and peptides with therapeutic potential. We underline that our method is applicable to any family of peptides and could be scaled up for high-throughput analyses.