Amyloid fibril formation caught in action
April 19, 2012 § Leave a comment
Neurodegenerative diseases like Alzheimer’s and Parkinson’s and metabolic diseases like Type 2 diabetes are characterized by clumps of proteins. These clumps are made of amyloid fibrils. A paper appears in the latest issue of Science that describes, for the first time, the view of the intermediate folding state of globular proteins that are on their way to becoming amyloid fibrils. As David Eliezer of Weill Cornell Medical College explains in a related Perspectives article, the work provides “a view, at an unprecedented level of detail, into the structural changes that convert a normally soluble protein into an aggregation-prone precursor.”
The work was carried out by a team led by Lewis Kay at the University of Toronto in Canada and Michele Vendruscolo at Cambridge University in the U.K. The researchers used a combination of nuclear magnetic resonance and computational methods to study a 60-amino-acid peptide chain that contained a SH3 domain. SH3 domains are involved in signal transduction pathways and are evolutionarily conserved in prokaryotes and eukaryotes.
Kay and colleagues showed that, in an intermediate state, the last four amino acids of the SH3 domain became rearranged and didn’t cap the protein’s end. When these researchers created a version of the SH3 domain missing the last four amino acids, the peptide spontaneously assembled into amyloid fibrils. The work demonstrates that those last four amino acids play a protective role.
The reason why researchers haven’t been able to take a good look at intermediate folding states to date , explains Kay, is that intermediates are rare beings. “In the case of the SH3 domain investigated in our work, the lifetime of the intermediate is approximately one millisecond and at any given time about two percent of the molecules are populating it,” he says. “As a result, structural information from conventional biophysical techniques predominantly report the remaining 98 percent of the molecules that are in the stable native conformation.”
Kay says detailed characterization of intermediates in protein folding or other biological processes demand techniques that simultaneously offer sufficient sensitivity to detect lowly populated states; sufficient time resolution; and the near-atomic resolution in space. A number of NMR laboratories, including the Kay group, have been developing methods that provide a detailed glimpse into the nature of these ‘invisible’ states.
Paul Fraser at the University of Toronto, an expert in molecular mechanisms of neurodegenerative diseases who wasn’t affiliated with the work, says the transition from a globular protein to an amyloid fibril “involves partial unfolding of the native state, but this has always been something of a black box for those of us working in the field, especially when it comes to a precise molecular structure of the unstable intermediates in this pathway.”
Fraser explains that the work by Kay and colleauges now opens up the question whether other globular proteins known to form amyloid fibrils in humans undergo similar structural transitions. “The SH3 domain, as indicated in the paper, is not associated with any clinical phenotype or amyloid-related disease, at least as far as we know,” he says. “But it does provide an excellent model of the amyloid misfolding pathway. Now that [Kay and colleagues] have solved the structure of this model system, I suspect other researchers in the field will be applying similar NMR approaches to determine if other disease-causing amyloidogenic proteins undergo a comparable structural rearrangement.”
Eliezer also points out a different study that appeared last month in Science by a group led by David Eisenberg at the University of California, Los Angeles. The study provided a complementary view of the formation of protein clumps. Eliezer notes that two papers “suggest a potentially general pathway for amyloid formation, in which protective native interactions are removed”.