(575b) β Sheet Not Prevalent in Amyloid Nucleus

Authors: 
Heldt, C. L., Michigan Technological University
Sorci, M., Rensselaer Polytechnic Institute
Belfort, G., Rensselaer Polytechnic Institute
Kurouski, D., University at Albany, SUNY
Lednev, I. K., University at Albany, SUNY
Grafeld, E., Rensselaer Polytechnic Institute


Amyloid diseases, including Alzheimer's disease and Parkinson's disease, are characterized by aggregation of normally functioning proteins into ordered, β-sheet rich, fibrils. The transformation to β-sheet rich structures is not well understood, but is currently hypothesized to be a nucleation dependent reaction due to the presence of a lag time prior to spontaneous fibril elongation. We have separated and characterized the nucleus of the amyloid insulin through reverse-engineering of the system and were surprised to find no detectable β-sheet formation in the nucleus. The nuclei of insulin fibrils were isolated by dissolving the fibrils at high pH. After disaggregation, two protein populations remain, a population that seeds the fibril reaction and increases the rate of fibril formation and a population that does not increase the rate of fibril formation. The population that does not increase fibril formation after fibril disaggregation has similar secondary structure to native insulin (by circular dichroism (CD) and deep UV Raman Resonance (DUVRR) spectroscopy). The population that seeds the reaction is less than 1% of the total protein and is likely to contain the nucleus. We recovered the protein in this sample using diafiltration with a 100 kDa membrane and compared its secondary structure with native insulin. The α-helix content of this population was less than that for native insulin (measured by CD). However, β-sheet content did not increase as compared with native insulin (measured by DUVRR spectroscopy). We hypothesize that the nucleus is not dissolved at high pH and remains in the solution. All of the insulin, however, that joined the nucleus to form fibrils had a different conformation and could be removed from the fibril and return to native dissolved insulin. This is the first time that a stable nucleus of an amyloid reaction has been separated and characterized. This nucleus is not β-sheet rich, but has a more disordered structure, and it also has a distinct structure from the insulin that elongates the fibril.