Using a nuclear magnetic resonance technique, University of Illinois at Chicago chemists have obtained the first molecular-level images of precursors of bundled fibrils that form the brain plaques seen in Alzheimer's disease.

Untangling the molecular structure of these pre-fibril forms, which may be the key neurotoxins in Alzheimer's, may help identify targets for new drugs to combat many neurodegenerative diseases.

Microscopic bundled fibrils made of proteins called amyloid-beta are presumed to be the toxic culprits in the senile plaques found in the brain with Alzheimer's. But there is increasing evidence that even smaller assemblies of amyloid-beta found prior to formation of pre-fibrils are the real nerve-killers. Scientists have been frustrated that electron microscope images of these nanometer-scale spherical assemblies have failed to reveal any distinct molecular structure.

Yoshitaka Ishii, assistant professor of chemistry at UIC, and graduate student Sandra Chimon have now determined this structure using a methodology developed with high-resolution solid-state nuclear magnetic resonance, or SSNMR. Details were reported in a Communication article last month in JACS, the Journal of the American Chemical Society.

"This is the first case showing that these intermediate species, the smaller assemblies, have a well-defined structure," said Ishii, who conducted a two-year search to map the structure of the pre-fibril assemblies, then spent another year confirming his findings.

Ishii's technique uses what is called time-resolution SSNMR to view nanoscale spectral images of this chemical formation.

Thioflavin, a dye commonly used to stain amyloid senile plaques, is applied to detect formation of the intermediate assemblies in florescence. The intermediate sample is then frozen to capture quickly changing spectral images of the molecules before they can self-assemble into fibril-forming sheets.

The resulting SSNMR "snapshots" provide a structural diagram for finding molecular binding targets that may stop proteins from misfolding, which may stop Alzheimer's disease from developing.

"We're interested in how the molecules assemble in this shape, and eventually into fibrils," Ishii said. "We wanted to find out what kind of structure each amino acid takes in a certain site of a protein at the atomic level. It gives us an idea of how these molecules interact with each other to make this structure."

Ishii said the SSNMR technique may be used to study small chemical subunits involved in diseases such as Parkinson's and prion diseases like mad cow or Creutzfeld-Jacob, to name just some of the 20 or so neurodegenerative diseases characterized by misfolding proteins.

"You want to design molecules that will interact and prevent this," said Ishii. "That's been difficult. Now we have a new clue to learn how."

Ishii's research work was supported by the Alzheimer's Association and the National Science Foundation.