The Transcellular Spread of Cytosolic Amyloids, Prions, and Prionoids

Neuron

Perspective

The Transcellular Spread of Cytosolic Amyloids, Prions, and Prionoids

Adriano Aguzzi1,* and Lawrence Rajendran2,* 1Institute of Neuropathology, University Hospital of Zu¨ rich, Schmelzbergstrasse 12, CH-8091 Zu¨ rich, Switzerland 2Systems and Cell Biology of Neurodegeneration, Psychiatry Research, University of Zurich, CH-8008 Zu¨ rich, Switzerland *Correspondence: adriano.aguzzi@usz.ch (A.A.), rajendran@bli.uzh.ch (L.R.) DOI 10.1016/j.neuron.2009.12.016

Recent reports indicate that a growing number of intracellular proteins are not only prone to pathological aggregation but can also be released and ‘‘infect’’ neighboring cells. Therefore, many complex diseases may obey a simple model of propagation where the penetration of seeds into hosts determines spatial spread and disease progression. We term these proteins prionoids, as they appear to infect their neighbors just like prions—but how can bulky protein aggregates be released from cells and how do they access other cells? The widespread existence of such prionoids raises unexpected issues that question our understanding of basic cell biology.

Imagine that you are a neuroscientist vacationing on Mars. One day you encounter a colony of Martians that, as it happens, look similar to water bottles. The Martians are highly distressed and seek your advice, as their community is plagued by an enigmatic transmissible disease. Intrigued, you agree to help. It turns out that the bodies of your exobiotic friends consist of bottles filled with a supersaturated salt solution. At some point crystals have started forming in one individual, and then crystallization has somehow been transferred to other community members. Lacking molecular insight, you would initially conclude that the Martians are affected by an infectious agent. Through ingenuity and technology, you may then discover that the infectious agent is exceedingly simple and homogeneous, that it lacks informational nucleic acids, and that it is generated both by ordered aggregation of an intrinsic precursor and by appositional growth of extrinsically added seeds. Your discovery will earn you the Intergalactic Nobel Prize, yet two crucial questions remain unanswered: how do the crystals transfer between individuals, and what can be done to prevent this from happening?

Middle-aged readers may feel reminded of the plot for Andromeda Strain, a stunningly prescient novel published in 1969 by the late Michael Crichton. But the sci-fi scenario described above is also the blueprint of Prusiner’s hypothesis of prion propagation. Over time, we have learned that prions consist of PrPSc, higher-order aggregates of a physiological protein termed PrPC. Accordingly, prions propagate through elongation and breakage of PrPSc aggregates (Aguzzi and Polymenidou, 2004)—not unlike the crystals vexing our extraterrestrial friends.

There is mounting evidence (Clavaguera et al., 2009; Frost et al., 2009; Ren et al., 2009; Desplats et al., 2009; Luk et al., 2009) suggesting that the events sketched above, far from being confined to science-fiction and prion diseases (whose incidence in humans is just z1/106/year), may underlie highly prevalent human diseases of the brain and many other organs. The unifying characteristics of all these diseases is the aggregation of proteins into highly ordered stacks, henceforth termed ‘‘amyloids’’ irrespective of their size. Since PrPSc undoubtedly fulfills the latter definition of amyloid, one is led to wonder whether the prion principle may be much more pervasive than previously appreciated and whether many more diseases of unknown cause may eventually turn out to rely on prion-like propagation (Table 1, upper panel). Even more intriguingly, a number of proteins appear to exert normal functions when arranged in highly ordered stacks that are similar to amyloids and to prionoids (Table 1, lower panel).

Prions and Prionoids

There is one crucial difference between bona fide prion diseases and all other amyloids and prion-like phenomena hitherto described in uni- and pluricellular organisms (Table 1). Prions are infectious agents, transmissible between individuals, and tractable with microbiological techniques—including, e.g., titer determinations. Even if certain amyloids of yeast and mammals appear to infect neighboring molecules and sometimes neighboring cells, they do not propagate within communities, and none of them were found to cause macroepidemics such as Kuru and bovine spongiform encephalopathy. We have therefore termed these self-aggregating proteins ‘‘prionoids’’ (Aguzzi, 2009), since the lack of microbiological transmissibility precludes their classification as true prions.

Some prionoids may soon qualify for an upgrade to prion status. At least in select settings, amyloid A (AA) amyloidosis may exist as a truly infectious disease based on a self-propagating protein. AA amyloid consists of orderly aggregated fragments of SAA protein, whose deposition can damage many organs of the body. Somewhat bizarrely, AA aggregation is also present in the liver of force-fed geese, hence contributing to the pathophysiology of foie gras (Solomon et al., 2007). AA seeds can induce amyloidosis upon transfer of white blood cells (Sponarova et al., 2008). Furthermore, AA seeds are excreted with the feces, and AA amyloidosis is endemic in populations of cheetah (Zhang et al., 2008). It is therefore tantalizing to suspect that amyloid may entertain the complete life cycle of an infectious agent, including transmission by the orofecal and hematogenous route—similarly to enteroviruses and, perhaps, scrapie prions. While there may be many other good reasons to avoid foie gras, including, e.g., animal welfare concerns, gourmets may not need to panic: under experimental conditions, AA amyloidosis is only transmitted to AgNO3-pretreated mice that display elevated levels of the SAA precursor protein.

Alzheimer’s disease (AD) has long been suspected to be a transmissible disease, but these suspicions have never materialized in epidemiological studies. On the other hand, Mathias Jucker and Lary Walker observed that injection of the Ab peptide from human AD brains induced robust and convincing aggregation of Ab in transgenic mice overexpressing the Ab precursor protein, APP (Kane et al., 2000; Meyer-Luehmann et al., 2006). Jucker’s finding raises an epistemologically significant question: if aggregation depends on the introduction of seeds and on the availability of the monomeric precursor, and if amyloid represents the primordial state of all proteins (Chiti and Dobson, 2006), wouldn’t all proteins—under appropriate conditions— give rise to prionoids in the presence of sufficient precursor?

The issues sketched above go well beyond AD and prions. There are many other diseases—not necessarily involving the nervous system—whose pathogenesis involves ordered aggregation of proteins, but for which there is no evidence of transmission between individuals. The best-studied of these are the systemic amyloidoses, which come about through the nucleation of some aggregation-prone proteins such as transthyretin and immunoglobulin light chains. Yet ordered protein aggregation is by no means confined to the ‘‘classical’’ amyloidoses and extends to a number of conditions, some of which have been rather unexpected.

Type II diabetes is yet another disease whose pathogenesis may involve ordered protein aggregation. Evidence to support this idea was discovered over a century ago (Opie, 1901) but was largely forgotten until recently. It is now evident that aggregation of islet amyloid polypeptide (IAPP) is an exceedingly frequent feature of type II diabetes. IAPP amyloids damage the insulin-producing b cells within pancreatic islets and may crucially contribute to the pathogenesis of diabetes (Hull et al., 2004). It is unknown, however, whether IAPP deposition simply accrues linearly with IAPP production or whether it spreads prion-like from one pancreatic islet to the next.

A body of recent work supports the idea that many aggregation proteinopathies are, in one way or another, transmissible. A recent report showed that a-synuclein is released from neurons and is then taken up by the neighboring cells, thereby aiding in a progressive spread of the protein (Desplats et al., 2009; Lee et al., 2005). When exogenously added to cultured cells, fluorescently labeled, recombinant a-synuclein was internalized from the extracellular milieu into the cytosol. Furthermore, injection of GFP-labeled mouse cortical neuronal stem cells into the hippocampus of a-synuclein-transgenic mice led to the efficient uptake of the host a-synuclein into the grafted cells after just 4 weeks. These findings are reminiscent of the observation that healthy fetal tissue, grafted into the brains of Parkinson’s disease patients, acquired intracellular Lewy bodies. The latter phenomenon is somewhat anecdotal and has been disputed (Mendez et al., 2008), yet it would be entirely compatible with the hypothesis that a-synuclein aggregates are prionoids (Li et al., 2008). A similar study conclusively demonstrated that exogenous a-synuclein fibrils induced the formation of Lewy body-like intracellular inclusions in vitro (Luk et al., 2009). This study also showed that the conversion of the host cell a-synuclein was accompanied by dramatic changes, including hyperphosphorylation and ubiquitination of a-synuclein aggregates—thus recapitulating some key features of the human pathology.

In experiments conceptually analogous to those discussed above, polyglutamine-containing protein aggregates similar to those present in Huntington’s disease and in spinocerebellar ataxias exhibited prion-like propagation (Ren et al., 2009). There, aggregation of huntingtin progressed from the extracellular space to the cytosol and eventually to the nucleus. What is more, similar phenomena occurred upon exposure of cells to Sup35 aggregates, which consist of a yeast protein for which there are no known mammalian paralogs. This suggests that the prionoid properties are intrinsic to amyloids and are not tied to the origin or function of their monomeric precursor protein.

In another work, Tolnay and colleagues report a similar phenomenon in a mouse model of ‘‘tauopathy,’’ a neurodegenerative disease due to intraneuronal aggregation of the microtubule- associated tau protein (Clavaguera et al., 2009). Aggregation- prone mutant tau, when extracted from the brain of transgenic mice, induced tauopathy in mice overexpressing wild-type tau. Assuming that tau pathology wasn’t elicited by some indirect pathway (tau-overexpressing mice develop tangles when exposed to Ab aggregates [Go¨ tz et al., 2001]), these transgenic mice appear to behave like the Martian bottles, since tauopathy was not induced in mice expressing normal levels of tau. In yet another study, the microtubule binding part of the full-length tau was found to attack and penetrate cells when added exogenously, and this again induced host tau misfolding (Frost et al., 2009). This study also showed that aggregated intracellular Tau spontaneously transferred between two cocultured cell populations (Frost et al., 2009). In the case of both tau and polyglutamines, the protein aggregates appear to gain access to the cytosol and to cause further aggregation of their host counterparts—presumably by nucleation.

The unifying characteristics of all these diseases is the aggregation of proteins into highly ordered stacks, termed amyloids irrespective of their size; the growth of these structures also exhibits generic features (Knowles et al., 2009) shared with a wide class of self-assembly phenomena characterized by elongation and fragmentation, such as the formation of analogous aggregates in micro-organisms and in vitro. Two conclusions can be drawn from the recent studies: (1) an unexpected number of amyloidogenic proteins can be released from affected cells in the form of extracellular amyloid seeds, and (2) even more surprisingly, these seeds can then re-enter other cells and nucleate the aggregation of their intracellular counterparts—in the cytosol or even in the nucleus. The biological and practical implications are far-reaching. On the one hand, cell therapies of aggregation diseases may be more difficult than anticipated, as the transplanted cells may undergo infection. A possible remedy could consist in the removal of the genes encoding the precursor of the offending proteins from the cells utilized for therapy—e.g., using the zinc-finger nuclease strategy (Hockemeyer et al., 2009). On the other hand, a novel paradigm of amyloid pathogenesis is emerging from these data, whereby each prionoid behaves as a self-assembling and self-replicating nanomachine.

Conversely, these findings raise a number of enigmas for which we are lacking any satisfactory answer. Whereas PrPC and the Ab are luminally exposed, a-synuclein and tau are cytoplasmic— and huntingtin is even nuclear. Aggregates of both Ab and PrPSc, as well as their monomeric precursors, are found in the extracellular space; it is hence intuitive that the nucleation process can propagate spatially across large distances. Instead, the propagation of cytoplasmic prionoids challenges our basic cell-biological understanding, since it posits that protein aggregates are released into the extracellular space and can subsequently reenter—and wreak havoc—in the cytosol of other cells. The release of cytosolic amyloids is supported by the amelioration of Lewy body pathology in a-synuclein transgenic mice immunized with human a-synuclein (Masliah et al., 2005). Similarly, anti-tau oligomer immunotherapy reduced brain pathology (Asuni et al., 2007), and immunization with mutant SOD1 led to clearance of SOD1 and delayed the onset of the disease in mice (Urushitani et al., 2007). All of these results indicate that cytosolic amyloids are somehow accessible to extracellular antibodies. This raises the question of how these proteins are released into the extracellular space (‘‘cytosol to lumen’’) and how they subsequently re-enter cellular cytosol (‘‘lumen to cytosol’’). Both events require trespassing lipid bilayer barriers—by no means a trivial feat for proteins, let alone highmolecular- weight aggregates.

snip...

Conclusion

The wave of these recent reports on the prion-like behavior of disparate pathogenic proteins raises many more questions than it answers. Here we have highlighted a number of open issues related to mechanisms of cell-to-cell spread of prionoids. The resolution of such issues may constitute the first step toward the development of rational strategies aimed at blocking transcellular propagation. There is justified hope that the latter may decelerate the progression of pathology and, consequently, help toward fighting the devastating outcome of aggregation proteinopathies.

http://www.cell.com/neuron/abstract/S0896-6273(09)01006-X



Sunday, July 18, 2010

Alzheimer's Assocition International Conference on Alzheimer's Disease (updated diagnostic criteria) 2010 July 10 - 15 Honolulu, Hawaii

http://betaamyloidcjd.blogspot.com/2010/07/alzheimers-assocition-international.html



Saturday, April 24, 2010

New connection between Alzheimer’s and prionic illnesses discovered

http://betaamyloidcjd.blogspot.com/2010/04/new-connection-between-alzheimers-and.html



Sunday, June 7, 2009

ALZHEIMER'S DISEASE IS TRANSMISSIBLE

http://betaamyloidcjd.blogspot.com/2009/06/alzheimers-disease-is-transmissible.html



Wednesday, April 14, 2010

Food Combination and Alzheimer Disease Risk A Protective Diet

http://betaamyloidcjd.blogspot.com/2010/04/food-combination-and-alzheimer-disease.html



Alzheimer's and CJD


http://betaamyloidcjd.blogspot.com/




Terry S. Singeltary Sr. P.O. Box 42 Bacliff, Texas USA 77518