26 May 2012. En route to full-size fibrils, α-synuclein takes shape as two distinct oligomer intermediates, according to a single-molecule analysis published in the May 15 Cell. Early on, monomers of the Parkinson’s disease-linked protein coalesce into small, loosely structured oligomers, which are initially benign oligomers. Over time, these convert to larger, tight-knit versions that are particularly toxic to cells. The period of time between the production of small and large oligomers could be an important therapeutic window, suggest the authors at the University of Cambridge, U.K., laboratories of Christopher Dobson and David Klenerman.
In studying individual oligomers, the authors used a new technique. Many researchers study oligomerization and fibril formation in vitro by analyzing turbidity or dye binding in a solution as a whole. However, the oligomers formed by aggregation-prone proteins like α-synuclein are heterogeneous and often short-lived, and these techniques only offer the “average” result out of a complex population, noted first author Nunilo Cremades in an e-mail to ARF. To zoom in on individual oligomers, she used a single-molecule approach the lab has applied before to amyloidogenic proteins (Orte et al., 2008).
Cremades tagged α-synuclein monomers with either red or green fluorophores, then mixed the populations. Under a microscope, monomers would glow in one or the other color, while oligomers would incorporate both hues. Based on the intensity of the emitted light, Cremades could estimate oligomer size. The dual-color strategy also allowed the team to perform fluorescence resonance energy transfer (FRET) experiments. Exciting only the green fluorophores would activate red ones as well, if the two were in close proximity in the same oligomer. “The great step forward is this ability to characterize the various oligomeric states in a quantitative fashion,” said Vernon Anderson, a former researcher in α-synuclein aggregation. The method could be useful in studying many amyloids, said Anderson, who now works at the National Institute of General Medical Sciences in Bethesda, Maryland. (Anderson was not speaking for the Institute in this instance).
Minutes after mixing the monomers, Cremades observed small oligomers up to approximately 10 monomers in size. After a few hours, her solutions contained structures ranging from two to more than 100 monomers in size. There were two main populations, however. One, deemed type A, emitted a modest FRET signal and ranged from two to 15 monomers in size. Type B emitted higher FRET—indicating the fluorophores were squeezed closer together—and a size range of five to 150. Since the type B oligomers took longer to form, they might have been converted from A types, the authors suggest. The same two populations arose from solutions at physiological α-synuclein concentrations, implying to the authors that they might occur in vivo as well.
To better understand the structures in their solution, the researchers examined how susceptible they are to degradation. While α-synuclein monomers melt away in the presence of proteinase K, fibrils are resistant. What about oligomers? Cremades found that type A oligomers were no more resistant than monomers, suggesting a loose conformation. The type B oligomers, while not as tough as fibrils, exhibited some resistance. This is in line with the FRET results that suggest they bind more tightly, and implies that they contain proteinase K-resistant β-sheets. The result dovetails with previous work showing that β-sheets increase as α-synuclein progresses toward fibrils (Apetri et al., 2006). The researchers have yet to elucidate the precise structures of the oligomers they observed, but the current work sets the stage to do so, Cremades wrote.
What do these oligomers do to cells? Because protein aggregates lead to production of reactive oxygen species (ROS), Cremades examined the oligomers’ effect on the rate of ROS production in rat primary midbrain neurons. Neither monomers nor fibrils affected ROS generation. A solution taken after a day of incubation, with a 4:1 ratio of type A:B structures, boosted ROS output to about 1.5 times normal. A three-day solution, with a 4:3 ratio of A:B, boosted output to nearly 2.5 times normal. From that, the researchers concluded that the larger, proteinase-resistant oligomer was most toxic. Cremades suggested that a positive feedback loop—with more oligomers causing ROS production causing more oligomer formation—could contribute to the development of Parkinson’s.
The study authors propose a model for fibril formation in which monomers initially nucleate type A oligomers. These can either grow into larger A oligomers or convert to the B version. The B oligomers then grow into fibrils. “The implication here … is that the oligomers are both obligate intermediates in the formation of fibrils,” Anderson said. The A-to-B switch would be the crucial event in fibril formation. This conversion, the authors found, had an in-vitro half-time of about 35 hours, compared to the seconds it usually takes proteins to adopt a particular conformation. “The structural conversion is remarkably slow … suggesting there is a significant period of time for the cellular protective machinery to operate and potentially for therapeutic intervention, prior to the onset of cell damage,” the authors wrote.
Cremades also examined how the fluorescently labeled fibrils disaggregated when she put them in a monomer-free solution. Early on, oligomers appearing in the solution were primarily of the B type, suggesting that fibrils break into B-shaped pieces. This further supports her hypothesis that B oligomers are made of β-sheets. Type A oligomers appeared later on, as if B types converted back to A. Cremades and colleagues concluded that α-synuclein fibrils both sequester B oligomers and are a potential source of the toxic species. Scientists designing particularly fibril-busting therapeutics that could spew dangerous B oligomers would do well to take this into account, the authors advised.
“This is a seminal body of work showing that biophysically distinct aggregates can be formed from monomeric species of α-synuclein,” wrote Andrew Dillin of the Salk Institute in La Jolla, California, who was not part of the study team, in an e-mail to ARF. “Whether there will be similar mechanisms at play for amyloid-β, tau, SOD1, and other aggregation-prone proteins will be of utmost importance to the field.” Anderson, for one, predicted that other amyloidogenic proteins would likely follow the two-intermediate pathway. The paper “highlights the importance of looking at single particles … as it relates to these aggregation processes,” Anderson said. He predicts the technique “will be applied to every amyloid that one can imagine.”—Amber Dance.
Cremades N, Cohen SI, Deas E, Abramov AY, Chen AY, Orte A, Sandal M, Clarke RW, Dunne P, Aprile FA, Bertoncini CW, Wood NW, Knowles TP, Dobson CM, Klenerman D. Direct observation of the interconversion of normal and toxic forms of -synuclein. Cell. 2012 May 25;149:1048-59. Abstract