Sometimes good proteins go bad. They change their shapes – from a soluble alpha helical conformation to an insoluble beta sheet one – stick to other proteins, and form fibrils that grow into clumps. Along the way cells die. When proteins slip into this so-called “amyloid” state they frequently cause disease. Devastating examples of pathogenic amyloids include tau and amyloid-β (Alzheimer’s disease), alpha-synuclein (Parkinson’s disease), huntingtin (Huntington’s disease), superoxide dismutase (amyotrophic lateral sclerosis) and PrPSc (the protein behind the prion diseases-e.g.bovine spongiform encephalopathy (BSE), scrapie, kuru, Creutzfeldt Jakob disease (CJD), fatal familial insomnia, and Gertsmann-Straussler-Scheinker disease).
Most scientists now accept the once heretical notion that amyloid diseases propagate without the mediation of viruses, bacteria or fungi – that’s without DNA or RNA (although microbes may trigger the start of a disease cascade). Amyloids spread, rather, via a process akin to crystallization; a misfolded protein acts as a seed or template that converts normal proteins to adopt the rogue conformation, which in turn impose this shape on other protein molecules in a chain reaction that propagates from cell to cell. In Alzheimer’s disease, Parkinson’s disease, and amyotrophic lateral sclerosis, the disease spreads only within the animal. In prion diseases like scrapie, the misfolded protein is contagious and can jump from individual to individual and even cross species barriers – in the case of the recent Mad Cow disaster from sheep (scrapie) to cow (BSE) to human (vCJD).
It gets weirder. In the last decade, researchers have convincingly argued that amyloids not only spread like microbes but also come in multiple “strains”. Under evolutionary pressure, prion proteins with identical sequences but slightly different conformations display different phenotypic behaviors, from longer/shorter disease incubation times, to increased/decreased likelihoods of “cross seeding” – i.e. triggering the spread of a different amyloid protein. Researchers have identified some 20 strains of scrapie, for example, a disease that’s been around since the early 1700s. MITs Susan Lindquist has recently shown that varieties of wild yeast contain prions with multiple conformational strains that confer positive as well as negative traits. Such findings, which raise the specter of Lamarckian-style “inheritance without genes,” are controversial. But I wouldn’t bet against them.
Do mainstream neurodegenerative diseases like AD and PD also have conformational strains? If they do it may help explain both the enormous clinical variability seen in life and the pathological heterogeneity found post mortem. Parkinson’s disease is a case in point. Clinicians have long observed significant variation among patients in onset age, clinical subtype, rate of progression, and prevalence of non-motor symptoms like dementia. Pathologists, likewise, report that deceased PD and AD patients’ brains rarely manifest just the pure hallmarks of a disease – e.g. amyloid plaques and tau tangles for Alzheimer’s disease and Lewy bodies and Lewy neurites for Parkinson’s disease. Neuropathologist Tom Beach of the Arizona Study of Aging and Neurodegenerative Diseases, for example, says he sees “a huge amount of heterogeneity post mortem. Most cases of Dementia with Lewy bodies have Alzheimer's disease as well. A third of Parkinson’s disease cases have developing Alzheimer’s disease as well. We see many cases of people dying with both Parkinson’s disease and Progressive supranuclear palsy.”
According to UPenn’s neurobiological power couple, Virginia Lee and John Trojanowski, some of this heterogeneity may be due to different amyloid conformational strains.
In a clever study published in Cell, Lee, Trojanowski, Guo and colleagues demonstrate that alpha-synuclein can manifest different amyloid strains that lead, via cross seeding, to different pathologies. Using a variety of approaches – experiments in vitro, experiments in vivo (with transgenic mice), and postmortem analysis of brain tissue derived from people with Parkinson’s disease dementia (PDD), the team demonstrates that that one strain (A) of alpha-synuclein fibrils, which cannot initially cause tau to form fibrils, “evolved” into another strain (B) – with the same amino acid sequence but a different conformation; one that could induce tau to fibrillize.
What’s true of alpha-synuclein is likely true of other rogue proteins like Amyloid-β, tau, and huntingtin. So the implications for neurodenerative diseases may be significant.
The paper has already created a lot of buzz. The University of Tübingen’s Mathias Jucker calls it “a milestone paper for the field…. very impressive, very interesting, and very exciting.” What’s even more exciting for patients is the idea that there are a number of feasible ways to reduce the amount of alpha-synuclein, and thereby potentially halt PD’s progression. Teams are working on small molecule drugs and antibodies. But cancer drugs may also have potential. In the May 10, 2013 edition of Human Molecular Genetics (on line), Georgetown’s Charbel Moussa et al demonstrate that very low doses of a FDA approved leukemia drug called nilotnib activates neurons’ garbage disposal machinery, enhancing the clearance of alpha synuclein and tau in a mouse model. In high doses, the drug causes neurons to self-destruct. In micro doses, it increases lysosomal storage and degradation of the rogue protein. A phase II trial in humans is being planned.