Our research centers on the principles of protein structures and their associations with the goal of relating structure and function. We are interested in protein folding and misfolding. Amyloid formation is an intriguing problem with vast practical implications. We study both model peptide systems and proteins. Among the protein aggregation diseases, Huntington and other diseases associated with extended glutamine repeats that lead to fibril formation are most clearly linked to molecular and genetic factors. T he Huntington phenotype directly correlates with the number of CAG tracts in the gene coding for the Huntingtin protein. Healthy individuals have polyglutamine (polyQ) tracts in their huntingtin protein that are between 11 and 35 residues long. Higher numbers lead to symptoms. Further, there is a linear correlation between the number of excessive Gln's, the pathology onset age and the severity of symptoms, becoming a severe juvenile disease when the tracts are larger than 80. Thus, the single gene and number of repeats dependence, make it an attractive system for conformational studies. Huntingtin aggregation shares most of the features displayed by other amyloidogenic proteins. In vitro incubation of fresh protein samples leads to a lag phase characterized by dynamic conformational changes with a marked increase in beta-sheet content. The fiber has a cross-beta structure with hydrogen bonds parallel to the super helical axis. This axis represents the direction of fiber growth. Inclusion body formation is toxic, leading to neuronal death. There is not much detailed information on the ultra structure of the polyQ -based fibers. EM of fibers matured in concentrated solutions of homo oligomers showed a reasonable resemblance to the more 'classical' amyloid-like fibers. PolyQ aggregates are less organized and although in some cases they bind Congo red, the majority of the fibers are morphologically branched and short. This feature generally indicates incomplete fibril formation. In terms of molecular organization, to date there is not much detail. It has only been shown that the organization that leads to fiber formation is reached via beta-sheets, similar to other amyloid fibril aggregates. There has been considerable effort to elucidate the molecular conformation that allows polyQ rich peptide chains to form stable insoluble fibers. The first molecular model dates to 1994. Following solid state IR studies, Perutz suggested a flat long beta-sheet with anti-parallel strands. Later, some variations were suggested, such as parallel strands or beta-turn rich organization. These studies involved modeling a rigid conformation and minimizing the energy using classical force fields. Others have suggested a helical organization, however, there is not much experimental information supporting a secondary structure that may radically differ from a beta sheet based arrangement. In an attempt to solve the polyQ organization puzzle, Perutz and his colleagues suggested an innovative model, in-between the beta-helix conformational motif and nanotubes observed in synthetic polymers. Based on the X-ray diffraction pattern obtained from D2Q15K2 peptides, Perutz et al. inferred a helical template that presented tubular shape with 20 residues per turn, with parallel beta-sheet hydrogen bonding interactions between successive turns of the helix. Further, the topological distribution of the residues allowed each amino acid to form amide - amide hydrogen bonds with residues set at positions i+20 and i-20. We have modeled peptides rich in glutamine, through series of molecular dynamics simulations. Starting from a rigid nanotube-like conformation, we have obtained a new conformational template that shares structural features of a tubular helix and of a beta-helix conformational organization. Our new model can be described as a super-helical arrangement of flat beta-sheet segments linked by planar turns or bends.
Showing the most recent 10 out of 42 publications
