19 August 2011. Order can come from chaos. Take α-synuclein, a protein implicated in Parkinson’s disease (PD) and long believed to exist in native form as a monomeric random coil. In the August 14 Nature online, researchers led by Dennis Selkoe at Brigham and Women’s Hospital, Boston, report that the protein actually predominates as a tetramer with an α-helical secondary structure. The tetramer shows little propensity to aggregate, suggesting that it must dissociate before toxic oligomers of α-synuclein assemble in vivo. That could have implications for the study of synuclein and the treatment of PD. Selkoe told ARF that the result were unexpected. “I think the biggest surprise is that synuclein folds into a helical formation, and that is essentially the opposite of what has been taught for years,” he said. Other researchers are not so convinced, however, and the findings seem to be generating some controversy in the field.
The idea that α-synuclein is a monomer with little secondary structure comes from studies of recombinant protein expressed in bacteria or human cell lines (see, e.g., Weinreb et al., 1996; Eliezer et al., 2001). While there were hints that the protein could be larger, including migrations on acrylamide gels and gel filtration chromatography columns slower than usual for a monomeric protein of its size, researchers ascribed those properties to an unfolded protein of large dynamic radius, said Selkoe. “Because many labs repeated the initial findings, people really bought into the idea that the protein was a monomer,” he said.
But when first author Tim Bartels joined Selkoe’s lab to study native forms of α-synuclein, he noticed something curious. On non-denaturing gel electrophoresis, α-synuclein from mouse frontal cortex, human red blood cells, and a variety of cell lines, including M17D dopaminergic neuroblastoma cells, did not behave as a 14.5 kDa α-synuclein monomer, but as a 55-60 kDa protein indicative of a tetramer. To confirm this, Bartels used a cell-permeable crosslinker to covalently stabilize potential oligomers, and then ran the protein on denaturing gels. It ran as a tetramer. Two-dimensional denaturing electrophoresis indicated that this tetramer had the same isoelectric point, that is, the same net charge, as monomers, indicating the tetramer comprises four identical subunits.
The use of native electrophoresis is one of the sticking points raised by researchers in the field who reviewed this work. Hilal Lashuel, who studies α-synuclein at the Swiss Federal Institute of Technology, Lausanne, told ARF that everyone knows that recombinant α-synuclein, whether denatured by boiling or not, runs at around 60 kDa on native gels and size exclusion chromatography. “A good experiment I recommend to those planning to repeat the work is to run the denatured recombinant protein side by side with their cell extracts as a control in these assays,” he told ARF. He also pointed out that while almost all the synuclein Bartels extracted from human red blood cells migrated as a 60 kDa protein on native gels, 90 percent of the protein ran as a monomer on denaturing gels, even after crosslinking, suggesting tetramers represent a minor species. An alternative explanation is that crosslinking is inefficient.
“Clear native gels are certainly not the be all and end all of molecular size determination,” agreed Selkoe, which is why Bartels used a variety of sophisticated biophysical techniques to back up the electrophoresis experiments. Scanning transmission electron microscopy (STEM) and sedimentation analytical ultracentrifugation, an older but highly accurate technique for determining the oligomeric structure of native proteins, also predicted the protein is a tetramer. Lashuel pointed out that the STEM data reflect a sample that is heterogeneous. At least two major species are observed, 30 kDa an 55 kDa, which is inconsistent with the analytical ultracentrifugation results, which show a single ideal species.
“I think these researchers have done a very careful and thorough job,” said Gregory Petsko, from Brandeis University, Waltham, Massachusetts. Petsko, together with Dagmar Ringe and Thomas Pochapsky, also at Brandeis, and Quyen Hoang at the Indiana University School of Medicine, Indianapolis, have evidence that recombinant α-synuclein expressed in bacteria also exists as a tetramer. Their findings will be published in the Proceedings of the National Academy of Sciences USA in the coming weeks.
Petsko told ARF that what is really important about the Selkoe group’s work is the discovery that the tetramer is stable and does not readily form amyloid. Bartels used circular dichroism spectroscopy to study the tetramer from red blood cells and 3D5 neuroblastomas, and found to his surprise that the spectra were characteristic of a protein with α-helical structure. This is in stark contrast to many studies that found the monomer was unstructured. Furthermore, thioflavin T-based assays showed that the tetramer did not form amyloid over a 10-day period, unlike recombinant monomers which began to aggregate around day 4, and continued to form amyloid for the next six days. “In terms of disease, this suggests that the tetramer, which is the predominant species, does not aggregate, and either needs other factors to make it aggregate or, more likely, it needs to disassemble first,” said Selkoe.
That could have important implications for therapeutic approaches. “If the hypothesis that α-synuclein adopts a defined tetrameric structure stands the test of time, which I believe it will, based on the data I’ve seen, then this structural insight represents a new paradigm for developing tetramer stabilizing drugs to prevent α-synuclein aggregation linked to Parkinson’s disease,’’ wrote Jeffery Kelly, from Scripps Research Institute, in an e-mail to ARF. “Numerous biotechnology and pharmaceutical companies will look at this opportunity very carefully,” he added. Kelly has pioneered the same strategy to tackle familial amyloid polyneuropathy, a rare disease caused by misfolding of mutant transthyretin, a protein that migrates from the liver to other tissues. A committee of the European Medicines Agency recently recommended approval of a small molecule arising from that work that stabilizes transthyretin tetramers, preventing them from dissociating into amyloidogenic monomers (see ARF related news story).
Petsko stressed that the synuclein tetramer is not necessarily the physiological form. But he agreed that if it is the physiological form, then stabilizing the tetramer could be a viable therapeutic strategy. “It gives us options that we would not have thought of before,” he said.
In fact, other researchers in the field doubted the physiological relevance, stressing that membrane-bound α-synuclein is the most important. On that point, Bartels found that the tetramer binds lipids with around 10-fold more potency than the monomer, and Selkoe believes it is much more likely that the tetramer is the more physiologically relevant.
Time will tell how important this ordered, tetrameric form of synuclein is. Meanwhile, researchers are concerned that this new order may bring its own form of chaos. Several labs, both academic and private, have tried to reproduce these findings since they were first presented in public, according to some researchers, and so far there are no reports of success, though it could be early days yet. “I am not rejecting the possibility that α-synuclein exists as a structured oligomer. In fact, it would be much more interesting if it did,” said Lashuel, but he added that “it is of critical importance and in the interest of everyone, including Parkinson’s disease patients and their families, to investigate and reproduce this data quickly.”—Tom Fagan.
Bartels T, Choi JG, Selkoe D. a-Synuclein occurs physiologically as a helically folded tetramer that resists aggregation. Nature. 2011 Aug 14. Abstract