15 June 2009. Wandering from cell to cell seeding aggregation and wreaking neurodegenerative havoc may not be the image that comes to mind when one thinks of the intracellular protein tau, but a study published in the June 7 Nature Cell Biology hints that that might be exactly what goes on in the Alzheimer brain. Researchers report that from a tiny injected seed, tau pathology eventually spreads to anatomically connected areas of the mouse brain. “It suggests that some form of assembled tau is taken up by cells and then promotes formation and release of new assemblies,” said Michel Goedert, MRC Laboratory of Molecular Biology, Cambridge, UK, one of the senior authors on the paper. “The idea has been out there for a while that normal neuronal transport mechanisms may be causing the spread of tau pathology,” suggested Lary Walker of Emory University, Atlanta, Georgia. Walker has studied protein seeding for some time, but was not involved in this study. “The early Braak studies showed that tau pathology progressed from one region of brain to another, starting in the trans-entorhinal cortex. That made sense, based on connectivity of the regions, but this is the first more or less direct evidence that this might be what’s happening,” Walker told ARF (see also comment below).
Goedert said the study, a collaboration with Markus Tolnay’s group at the University of Basel, Switzerland, was, in fact, inspired by Braak staging of Alzheimer disease (AD). The findings add weight to reports that tau pathology can spread from cell to cell in vitro. Marc Diamond, University of California, San Francisco, found that tau aggregates can get into and corrupt normal tau inside cells and that once aggregation occurs in one cell, it can trigger transfer to adjacent cells in culture (see ARF related news story and ARF news story). “I think our data help support their findings and their findings give our work in vivo relevance,” Diamond said in an interview with ARF.
Diamond also suggested the work might have implications that reach far beyond tau pathology and Alzheimer disease (AD). “One of the reasons this paper is so significant is it comes on the heels of observations on Parkinson’s patients who received neural transplants,” said Diamond. Those studies showed that Lewy bodies, which are predominantly made up of aggregates of the intracellular protein α-synuclein, also developed in the grafted tissue (see ARF related news story). “One interpretation of those observations is that some aggregated protein in the neighboring cells can corrupt the protein in the new cells,” said Diamond. If true, that would support the idea that α-synuclein, tau, and possibly other proteins involved in neurodegeneration, such as huntingtin (see ARF related news story), may behave somewhat like prions.
To test the idea that tau pathology can self-propagate, first author Florence Clavaguera and colleagues took brain samples from P301S tau transgenic mice, which develop filamentous tau aggregates, and injected them into the hippocampus or cerebral cortex of another transgenic strain—the ALZ17 mouse. ALZ17 mice overexpress normal human tau but do not develop tau pathology—though they may be more vulnerable to it. Homogenates from six-month-old P301S mice induced the formation of tau filaments when injected into three-month-old ALZ17 mice. Extracts from non-transgenic mice had no effect. Since human tau expressed in the ALZ17 mice has a different N-terminal sequence to that of the P301S tau, the researchers could ensure that the tau filaments formed were indeed due to the endogenous ALZ17 tau, and not to the injected material. Extracts immuno-depleted of tau had no activity.
The induction of tau filaments was dependent on both time and place. The process was slow. Pathology was first apparent at six months and became more robust at 12–15 months. Filaments appeared faster in the hippocampus than in the cortex. The pathology seems to be caused by some form of insoluble tau, since extracts of soluble material were 20-fold less potent. “Currently, all we know is that there is something in the brain extracts that is quite powerful at inducing tau pathology and that it is insoluble,” said Goedert. “We would like to find out what that substance is, whether it is fibrils, some oligomeric form, or what exactly it is that has the activity.” The same question is plaguing researchers trying to find what entity seeds amyloid-β aggregation in vivo (see ARF related news story and Bolmont et al., 2007).
Interestingly, the pathology spread in the injected mice in a manner that is consistent with the Braak staging idea in that it seems to occur between anatomically connected areas of the brain. But it also appeared in regions that are not typically associated with AD tau pathology. “I am surprised by the number of areas where they see positivity,” said Walker. “It’s everything from the medial lemniscus, which is in a sensory pathway in the brain, to the hypothalamus and the zona incerta. A lot of different areas are affected, but they do have connections with areas that are connected to the injection site,” he said. Goedert stressed that he and his collaborators are not suggesting this as a model for tau transmission in the human brain. “All we can say is that regions where pathology appeared over time were all anatomically connected,” he said.
Interestingly, even though the injected ALZ17 mice developed robust tau pathology, it did not lead to neurodegeneration and cell loss that occurs in the P301S mice. “This suggests that the molecular tau species responsible for transmission and neurotoxicity are not identical,” write the authors. They suggest that, much like prions, distinct tau strains are at work in the mice and may also underlie different tauopathies, such as Pick disease, progressive supranuclear palsy, and AD, which are characterized by different tau isoform assemblies. That’s not to say that tau is a prion. “The big difference is the ease with which the diseases are communicated,” said Walker. “There’s no evidence that tauopathy or Aβ can be transmitted in the same way as prion diseases can, so there may be something about prions that makes them better at transmitting structural information that causes disease.” Diamond suggested that one thing that makes prions unique is that they are incredibly tough. “You can eat them and they will get into your brain. Most of these other proteins, tau, and α-synuclein are vulnerable to proteases and readily digested,” he said.
If tau and α-synuclein toxicities do propagate in some prion-like fashion, then might that offer some new clues as to how to treat these diseases? “That might be possible, but the first thing we have to do is understand the mechanism,” said Goedert. Diamond expressed a similar sentiment. “You can think about disease from the standpoint of what happens in one cell, such as tau gets phosphorylated, misfolds, and causes the cell to get sick, but if you are thinking about how the process propagates between cells, that’s an entirely new way of thinking about therapies,” he said. He suggested that the reach of antibody therapy might extend to these intracellular proteins, for example. “If you understand the mechanisms of how aggregates move between cells and corrupt the protein on the inside of cells, that’s a new therapeutic target that could become a silver bullet for all neurodegenerative diseases that are associated with fibrillar proteins, since we have now seen this prion-like property in tau, α-synuclein, and Aβ,” he said.
Walker agreed. “Mechanistically, the fact that these diseases behave so similarly in tissues and cells, suggests there might be some common mechanism that will help us to understand not just Alzheimer disease but maybe 30 to 40 different proteopathies that all involve the accumulation of protein,” he said.—Tom Fagan.
Clavaguera F, Bolmont T, Crowther RA, Abramowski D, Frank S, Probst A, Fraser G, Stalder AK, Beibel M, Staufenbiel M, Jucker M, Goedert M, Tolnay M. Transmission and spreading of tauopathy in transgenic mouse brain. Nature Cell Biology 2008 June 7 advance online publication. Abstract