1 Scrapie is a highly infectious disease of sheep that is caused by neurotoxic prion protein aggregation.
PHOTO: WAYNE HUTCHINSON/MINDEN PICTURES
Paradigm shifts are drivers of scientific progress, yet the shifters of the paradigms often experience scorn rather than immediate applause. That was the fate of Stanley Prusiner’s 1982 paper claiming—to the initial amusement of his colleagues—that scrapie, a degenerative disease that affects the central nervous system of sheep, is caused by “proteinaceous infectious particles,” which he called prions (1). Prusiner’s intuition, which earned him the 1997 Nobel Prize, is influencing our approach to an ever-expanding variety of seemingly unrelated diseases and physiological processes, and its implications reverberate to the present day.
The two decades preceding Prusiner’s paper had witnessed the immense success of molecular biology, including the cracking of the genetic code; the elucidation of DNA replication, transcription, and translation; and the cloning of genes. These discoveries prompted Francis Crick to conceptualize the “central dogma”: Information flows unidirectionally from DNA to proteins. But although religious dogmas may be eternal, the shelf life of scientific dogmas is inevitably limited. Prusiner postulated that prions carry on their replicative cycle without the participation of nucleic acids. This hypothesis, reminiscent of John Griffith’s 1967 suggestion of the existence of self-replicating proteins (2), had the potential to explain the prodigious resistance of the scrapie agent to DNA-damaging radiation.
Daniel Carleton Gajdusek, who won a Nobel Prize for showing that Kuru was a human disease transmitted by cannibalism in Papua New Guinea, proposed in 1959 that the neurodegenerative disorders Kuru, scrapie, and Creutzfeldt-Jakob disease (CJD) are caused by “slow viruses.” Indeed, prions behave similarly to neurotropic viruses in many surprising ways, including the colonization of extraneural organs followed by neuroinvasion of the brain through peripheral nerves (3). Yet, Prusiner purified the agent and found it to be smaller than a virus: No informational nucleic acid would fit into it. Over time, the group of prion diseases grew to include other human (fatal familial insomnia) and animal (bovine spongiform encephalopathy, also called mad cow disease, and chronic wasting disease) disorders, but no causative virus has been identified and their prion etiology is now well accepted.
But prions did not contradict Crick’s central dogma after all. Charles Weissmann, refusing to believe that a protein could exist without its respective gene, discovered in hamsters the gene encoding the cellular prion protein (PrPC), whose misfolding yields tightly packed aggregates called PrPSc. It is generally believed that prion replication occurs when coalesced PrPSc is broken down into smaller species. Those species then accrue further PrPSc, in a process akin to the growth of crystals, and eventually break again, perpetuating their replicative cycle. Infectious prion seeds then move to neighboring cells and wreak havoc in the central nervous system by inducing vacuolation (“spongiosis”) within neurons (see the figure).
Does this mean that PrPSc is the prion? Weissmann’s discovery in 1993 that prion protein (Prnp)–ablated mice are resistant to scrapie (4) was designed to disprove the protein-only hypothesis but failed to do so. However, Prnp deletion in mice also fell short of proving the prion hypothesis. If PrPC were the receptor of an imaginary “scrapie virus,” its ablation may also render mice resistant to scrapie. More direct evidence for Prusiner’s ideas emerged in 2001 from Claudio Soto’s landmark experiment: Repeated cycles of PrPSc fragmentation, when followed by addition of PrPC and aggregate regeneration, can multiply prions ad libitum (5). These findings strengthen the hypothesis that the transfer of structural information can occur horizontally between proteins.
More recently, the prion concept has been applied, sometimes overenthusiastically, to virtually all diseases characterized by progressive deposition of aggregated proteins in the central nervous system, whether infectious or not—and even to physiological processes such as memory formation (6). α-Synuclein aggregates can self-propagate in the brains of Parkinson’s disease patients (7), in cultured cells, and in mice (8). This implies that a-synuclein is a de facto prion and that its handling demands high biosafety standards. Similar arguments were made for tau and amyloid-β (Aβ) aggregates, the major hallmarks of Alzheimer’s disease (9). However, prions caused many epidemics, whereas infectiousness has not been conclusively demonstrated for other protein aggregates—and specifically not through oral transmission. Protein aggregates that were not shown to be serially transmissible across multiple generations of hosts are better regarded as “prionoids,” even if they share molecular mechanisms of amplification with bona fide prions in vitro.
The timeline shows key prion-related discoveries. In 1982, Prusiner suggested that the prion protein (PrP) is the infectious cause of spongiform encephalopathies, including Kuru, scrapie, and Creutzfeldt-Jakob disease (CJD). These insights have had implications for many neurodegenerative diseases involving prionoids, but many questions still remain unanswered.
GRAPHIC: MELISSA THOMAS BAUM/SCIENCE; (IMAGES, LEFT TO RIGHT) K. FRONTZEK; U. S. HERMANN ET AL., SCI. TRANSL. MED. 7, 299RA123 (2015); G. SPAGNOLLI, ADAPTED FROM (12)
As predicted by Prusiner in the closing lines of his paper, the “prion revolution” boosted research in the field of neurodegeneration by providing an intellectual framework that might explain many aspects of Alzheimer’s disease, Parkinson’s disease, and many other neurodegenerative diseases featuring protein aggregates. Although cellular PrPC is now known to be crucial for the maintenance of peripheral myelin (10), our understanding of prions has essentially stagnated for more than a decade and may now be lagging behind that of prionoids. What is really known about prions, after almost 40 years since Prusiner’s discovery?
One crucial obstacle to advancing prion research is the lack of high-resolution structures of PrPSc owing to its insolubility, its noncrystalline aggregational state, and the persistent difficulties in preparing highpurity infectious material de novo from recombinant protein (11). This raises the possibility that infectious aggregates may constitute a sparsely populated conformational variant within such preparations. If so, most material aggregated in vitro may be noninfectious and may not be informative of the structure of the prion or of its replicative mechanism.
Of all the models that have been proposed so far, the most plausible suggests that the prion consists of fibrils arranged as fourrung b-solenoids (12) stacked either head-totail or head-to-head. Cryo–electron microscopy of purified glycosylphosphatidylinositol (GPI)–anchorless prion fibrils (13) supports this model, thus providing the first highmagnification images of infectious prions, albeit the resolution does not suffice to determine the precise arrangement of the monomers within the fibrils. These structures are quite different from those of tau, α-synuclein, and Aβ and also differ from recombinant PrP fibrils—all of which are arranged in long fibers with no cavity. Hence, PrPSc has distinctive structural characteristics, but it is unknown whether and how these peculiarities relate to their frightening infectivity.
The link between the generation of PrP aggregates and their neurotoxicity is also unclear. A large body of evidence (14) indicates that PrPC is necessary for toxicity, perhaps because extracellular PrPSc oligomers dock to PrPC on the surface of diverse cell types. Another aspect specific to prion infections pertains to the peculiar morphology of the damage that it wreaks on the brain. Of all aggregation-prone proteins, prions are the only ones that cause extensive intraneuronal vacuolation (spongiosis), the severity of which increases during disease progression. This phenomenon is as much intriguing as it is mysterious. To date, almost nothing is known about the cellular and molecular pathologies underlying vacuole formation; yet its ubiquity in all known prion diseases suggests that vacuolation is a prime driver of toxicity—and therefore also a target for therapeutic interventions.
High-resolution three-dimensional structures of prions are also required to solve the long-standing question of prion strains, which share the same PrP sequence and yet cause distinct diseases (e.g., “hyper” and “drowsy” phenotypes in minks), the traits of which are maintained over successive rounds of infection. Viral strains are defined by specific polymorphisms in their respective genomes, and the existence of strains in prion diseases was long thought to be incontrovertible evidence for the involvement of nucleic acids. However, after four decades of failed attempts to isolate any scrapie-specific genomes, strains are now thought to be caused by different PrPSc conformations that can be distinguished with conformer-sensitive fluorescent polythiophenes.
Embarrassingly for the prion field, no definitive structural evidence for these presumptions has come forward, and the “strainness” of bona fide infectious prions is still diagnosed using imperfect surrogate biomarkers such as differential resistance to disaggregation and proteolysis. By contrast, conformational heterogeneity was reported to correlate with distinct clinical phenotypes in some prionoid pathologies, although the stability of different conformations in serial transmission experiments is not yet fully established.
But how stable are prion strains across generations? RNA viruses achieve maximal fitness by creating quasispecies, clouds of variants in precarious equilibrium between adaptive mutagenesis and error catastrophe. Notably, prions can also engender quasispecies whose monoclonal constituents can be isolated from cultured cells by applying various kinds of selective pressure (15). The structural mechanisms underlying this phenomenon are unknown and may involve conformational selection of distinct PrPSc species. The conformational selection model predicates the coexistence of multiple conformers within a single infected organism, some of which may replicate more efficiently in their host under certain environmental circumstances. The incubation time of prion infections can vary immensely between different strains, and the delay in the onset of the pathology might reflect the time needed for such selection to occur
PrPSc conformer heterogeneity may also underlie the barriers that control interspecies prion transmission, the strength of which is variable and depends both on host factors and on prion strains. Although prion propagation from cows to humans results in variant CJD, sheep prions appear to be largely innocuous to humans. This species barrier relies both on the structural diversity of the PrPSc contained in the inoculum and the PrPC of the host, which cannot always interact with the misfolded conformer efficiently.
The ideas promulgated by Prusiner have undergone a marked metamorphosis. Templated nucleation of protein aggregates is now known to underlie not only diseases but also many physiological processes, some of which bear little resemblance to the original set of diseases that attracted Prusiner’s attention. Notably, the structural predictions of the prion model were verified for several prionoids but not for prions. As such, many of the questions raised by Prusiner in 1982—prion structure, mechanism of replication, and drivers of toxicity—are still open. Based on historical evidence, addressing these questions in the prion arena may, once again, provide answers that will also apply to more prevalent neurodegenerative diseases.
Acknowledgments: The authors thank K. Frontzek for the micrograph of a CJD brain and G. Spagnolli for the PrPSc molecular model. The Aguzzi lab is supported by the Nomis Foundation and the Swiss National Science Foundation.