504b Folding and Aggregation of Polyglutamine Peptides and Proteins

Regina M. Murphy, Robert Walters, Matthew Tobelmann, and Joseph Bernacki. Chemical and Biological Engineering, University of Wisconsin, 1415 Engineering Drive, Madison, WI 53706

Huntington's is the most prevalent example of the “CAG diseases”. Each of these neurodegenerative diseases is associated with a unique protein and unique clinical symptoms, but they share two common features: the disease-associated protein has an abnormally expanded polyglutamine (polyQ) domain, and the protein aggregates into nuclear and/or cytoplasmic inclusions. A causal link between aggregation and toxicity is supported by many studies, but the mechanism of toxicity is still unknown. We are interested in developing a quantitative description of the kinetics of aggregation of peptides and proteins with expanded polyQ domains. We believe this information is relevant in determining the size and nature of the polyQ aggregates that are responsible for toxicity, and in developing new therapeutic approaches for inhibiting toxicity. Our first objective is to examine the mechanism of aggregation of synthetic polyQ peptides, and the role of the length of the polyQ domain in moderating aggregation. The dominant hypothesis in the literature is that aggregation proceeds via formation of a monomeric nucleus that is thermodynamically unfavorable. We show, through statistical model discrimination methods, that the published data do not support this conclusion. Further, we present experimental data indicating that aggregation proceeds through formation of soluble oligomers that lack defined secondary structure. Through FRET analysis, we show that shorter, aggregation-resistant polyQ peptides are expanded in physiological buffer whereas longer, aggregation-prone peptides are collapsed, and we propose a mechanism by which the extent of collapse is related to aggregation. Our second objective is to examine the influence of polyQ domains on protein misfolding and aggregation. We report on a strategy for generating “length libraries” of proteins with variable-length polyQ inserts. We have produced recombinant proteins containing polyQ inserts and have characterized how folding, misfolding, and aggregation of these proteins is correlated with polyQ length.