(276b) Folding, Misfolding, And Aggregation Of Polyglutamine Peptides And Proteins

Murphy, R. M. - Presenter, University of Wisconsin
Lee, C. C. - Presenter, University of Wisconsin
Walters, R. H. - Presenter, University of Wisconsin
Tobelmann, M. - Presenter, University of Wisconsin

There are at least 9 neurodegenerative diseases associated with proteins that contain an unusually expanded polyglutamine domain, of which the best known is Huntington's disease. In all of these disease, the mutant protein aggregates into neuronal inclusions; it is generally believed that protein aggregation is the underlying cause of the observed neuronal degeneration. Importantly, the length of the expanded polyglutamine domain markedly affects the age of onset of disease and the severity of symptoms. In vitro, longer domains are correlated with faster aggregation kinetics. Synthetic peptides have been used as model systems for detailed examination of the mechanism of polyglutamine-mediated aggregation. The published literature suggests that aggregation follows a nucleation-elongation mechanism characterized by a significant lag time during which the peptide is monomeric, and that the nucleus is a monomer that attains a thermodynamically unfavorable conformation. We re-examined this hypothesis by measuring the aggregation kinetics of a polyglutamine peptide, K2Q23K2, using sedimentation, static and dynamic light scattering, electron microscopy, and size exclusion chromatography. Our data show that during the putative ?lag time' there is in fact substantial assembly of the peptide into soluble elongated aggregates. These aggregates lack defined secondary structure. Over time, the aggregates grow by lateral alignment and entanglement mechanisms; eventually, through growth, dehydration, and/or conformation change, insoluble aggregates begin to form. Once these insoluble aggregates appear, monomer loss from the solution accelerates, possibly by a templated assembly mechanism. Our data challenge the robustness of the ?monomeric nucleus' model of polyglutamine aggregation. We have expanded these studies to examine folding, stability and aggregation in proteins containing a polyglutamine insert. As a model protein we chose apomyoglobin, a predominantly helical protein that will aggregate into beta-sheet fibrils at high temperature. We developed a method for rapid generation of ?length libraries? of mutant apomyoglobin wherein varying lengths of polyglutamine inserts at a single insertion site are generated in a single transformation into recA+ cells. Using this method, we produced plasmids with inserts of 16, 18, 20, 24, 28, 34, 35, 38, 46, 52, and 102 glutamines. Importantly, the lengths obtained ranged from well below to well above the putative pathological length. We selected 6 of these mutants and produced and purified quantities of soluble folded protein. We have begun detailed biophysical characterization of these mutants. Preliminary circular dichroism and tryptophan fluorescence data indicate that there is a loss of helix and an increase in both beta-sheet and random coil content with increasing glutamine length, but that these structural changes are limited to the region near the insert. To date, significant aggregation has been detected only in the Q46 mutant. Further characterization of the correlation between polyQ length, protein folding, stability, and aggregation kinetics is underway.