Generating new prions by targeted mutation or segment duplication

Abstract

Prions are infectious agents composed entirely of protein. Prion activity results from the conversion of soluble proteins into an insoluble, self-templating amyloid form. Nine different amyloid-based prions have been identified in yeast. All but one contain a glutamine/asparagine (Q/N) rich region that is responsible for prion activity. Similar Q/N-rich regions are over-represented in eukaryotic genomes. In humans, aggregation-causing mutations in Q/N-rich proteins have been linked to various degenerative diseases, including ALS. Our lab previously developed a prediction algorithm, PAPA (Prion Aggregation Prediction Algorithm), to predict the aggregation propensity of Q/N-rich proteins, and to predict the effects of mutations on aggregation propensity. Here, we tested the ability of PAPA to design mutations to turn non-prion proteins into prions. We identified four yeast Q/N-rich protein fragments that lacked any detectable aggregation or prion activity. In each case, a small number of designed mutations were sufficient to cause these domains to aggregate, and in two cases, to create bona fide prion activity. We then tested whether simply generating tandem repeats of short, aggregation-prone segments within these domains would likewise be sufficient to create prion activity. We found that such segment duplications consistently increased prion activity in a length-dependent manner. This suggests that duplication of aggregation-prone segments might represent a simple mechanism for evolving new prion domains, potentially explaining why oligopeptide repeats are frequently found in prion proteins.