P E R S P E C T I V E Prion diseases are some of the most intriguing infectious disorders affecting the brains of humans and animals. The prevalent hypothesis proposes that the infectious agent is a misfolded protein that propagates in the absence of nucleic acid by transmission of its altered folding to the normal host version of the protein. This article details the evidence for and against the prion hypothesis, including results of recent studies in yeast, in which a prion phenomenon has also been identified. The evidence in favor of the prion model is very strong, but final proof—consisting of the generation of infectious prions in vitro —is still missing. Prion diseases, also called transmissible spongiform encephalopathies (TSEs), are a group of infectious neurodegenerative disorders affect- ing humans and animals 1 . These diseases are characterized by brain vacuolation, astrogliosis, neuronal apoptosis and accumulation of the misfolded, protease-resistant prion protein (PrP res ) in the central nervous system 2 ( Fig. 1 ). Although rare, TSEs have gained notoriety because of the recent appearance of a new variant form of Creutzfeldt–Jakob disease (vCJD) 3 that has been linked with human exposure to the infectious agent from cattle affected by bovine spongiform encephalopathy (BSE) 4 . Interest in TSEs is also spurred by the many unprecedented scientific findings in this area that have directly challenged some of the most established scientific dogmas. Among these, the nature of the infectious agent and its mechanism of propagation constitute one of the most debated findings in biology. Figure 2 outlines the research milestones that have contributed to the current understanding of TSE infectious agents. Emergence of the protein-only hypothesis The transmissibility of TSEs was accidentally demonstrated in 1937, when a population of Scottish sheep was inoculated against a com- mon virus with a formalin extract of brain tissue unknowingly derived from an animal with scrapie. Stunningly, after 2 years, nearly 10% of the flock developed scrapie. Scrapie was subsequently trans- mitted experimentally to sheep 5 and mice 6 . In humans, an infectious route was suspected for the propagation of the TSE kuru among the cannibalistic tribes of New Guinea, and this was demonstrated in 1966 by transmission of kuru to monkeys in the pioneering studies of Carleton Gajdusek 7 . This was followed by observations of transmis- sion to animals of CJD 8 and a familial form of TSE, Gerstmann– Straussler–Scheinker syndrome (GSS) 9 Because of the unusually long incubation period between the time of exposure to the pathogen and the onset of symptoms, the agent was initially thought to be a slow virus 10 . Further research, however, indi- cated that the agent differed substantially from viruses and other con- ventional agents. In 1967, Alper and colleagues reported that the agent responsible for scrapie was extremely resistant to treatments that normally destroy nucleic acids, such as UV and ionizing radia- tion 11 . These results, and the authors’ previous findings that the min- imum molecular weight necessary for infectivity ( ∼ 2 × 10 5 Da) was so small as to exclude viruses and any other known types of infectious agent 12 , led to alternative hypotheses. One was that the infectious agent might be a virino—a small informational molecule (most likely a piece of nucleic acid) encapsulated inside a protein coat 13 . The small size of the nucleic acid and the strong, tight protein coat of a virino could explain the resistance of the nucleic acids to procedures that normally destroy them. Although this hypothesis explains many of the experimental observations, 30 years of research have not consis- tently identified a nucleic acid specifically associated with a TSE infectious particle 2 . In 1967, a visionary article by J.S. Griffith intro- duced, for the first time, the possibility that the material responsible for disease transmission might be a protein that had the surprising ability to replicate in the body 14 . This launched the so-called ‘protein- only’ hypothesis of TSE propagation, subsequently moved forward by Stanley Prusiner’s group, who coined the named ‘prion’ for this new proteinaceous infectious particle 15 Evidence supporting the prion hypothesis A crucial step in understanding the nature of the new infectious agent was the isolation of the protease-resistant prion protein (PrP res ) from the infectious material 16 . PrP res copurified with infec- tivity, and the concentration of the protein was proportional to the infective titer 17 . Highly purified preparations of PrP res , in which no other component was detectable, retained their infectivity. In addi- tion, infectivity was convincingly reduced by agents that destroy protein structure and, more importantly, by antibodies to PrP 17 Indeed, the use of anti-PrP antibodies to reduce infectivity and dis- ease onset has recently been proposed as a therapeutic approach 18,19 . Purification of the protein allowed identification of the gene encoding PrP 20,21 . PrP mRNA proved to be the product of a single host gene, which is present in the brain of uninfected animals and is constitutively expressed by many cell types. It thus became clear that PrP can exist in two alternate forms: the normal cellular The controversial protein-only hypothesis of prion propagation Claudio Soto & Joaquin Castilla Claudio Soto and Joaquin Castilla are in the Department of Neurology, University of Texas Medical Branch, Galveston, Texas, USA. e-mail: clsoto@utmb.edu Published online 1 July 2004; doi:10.1038/nm1069 NEURODEGENERATION JULY 2004 S 6 3 © 2004 Nature Publishing Group http://www.nature.com/naturemedicine P E R S P E C T I V E protein (termed PrP C ) and the pathological isoform (termed PrP res or PrP Sc ). Chemical differences distinguishing these two PrP iso- forms have not been detected 22 , and the conversion seems to involve a conformational change whereby the α -helical content of the nor- mal protein diminishes and the β -sheet content increases 23 . These structural changes are followed by alterations in other biochemical properties, such as protease resistance, solubility and ability to form larger-order aggregates. Genetic studies have shown that most, if not all, familial cases of TSE are linked to mutations in the PrP gene 1,2 . These findings not only provide support for a central role of PrP in disease pathogenesis but also provide strong evidence for the protein-only hypothesis, because the genetic disease can be propagated in an infectious way. Interestingly, a TSE-like disease was produced in mice overexpressing PrP genes with point mutations linked to GSS syndrome 24 . These ani- mals spontaneously developed neurologic dysfunction, spongiform brain degeneration and astrocytic gliosis. Although the initial study indicated that the disease could be transmitted to animals expressing the mutant genes 25 , this transmissibility has been controversial; for instance, another group was unable to reproduce these results by using gene targeting to replace the wild-type mouse PrP gene with the mutant gene 26 A particularly strong piece of evidence in favor of the prion hypothesis came from the group of Charles Weissmann 27 , who showed that mice lacking the PrP gene were resistant to scrapie infec- tion, neither developing signs of scrapie nor allowing propagation of the disease. Another important finding was the successful propaga- tion of infectivity in neuroblastoma cells 28,29 . These cells could be chronically infected with brain homogenate containing PrP res and infectivity was maintained over several months. A further milestone in support of the prion hypothesis was the finding that the pathological protein catalyzed the cell-free conver- sion of PrP C into PrP res . The original system developed by Caughey and co-workers 30 , using purified PrP C mixed with stoichiometric amounts of purified PrP res under nonphysiological conditions, produced a low yield of PrP res formation. Nonetheless, the PrP res –induced transformation of the normal protein was important evidence for the prion hypothesis. More recently, we have reported a new in vitro conversion system to transform large quantities of PrP C using minute amounts of PrP res (ref. 31). This system, called PMCA (protein misfolding cyclic amplification), confirms a crucial facet of the prion hypothesis: that prion replication is a cyclical process, with newly produced PrP res triggering further misfolding to maintain prion propagation 31 Criticisms and missing evidence Although the prion hypothesis neatly explains many observed fea- tures of TSE, it has one particular weakness that has long been used as evidence against it 32 . Scrapie and other TSEs occur in multiple ‘strains’ characterized by different incubation periods, clinical fea- tures and neuropathology 2 . In infectious diseases, different strains generally arise from mutations or polymorphisms in the genetic makeup of the infectious agent. To reconcile the strain phenomenon with an infectious agent composed exclusively of protein, it has been proposed that PrP res from different prion strains has slightly different conformation or aggregation states that can replicate faithfully at the expense of the host PrP C (ref. 2). These different states of PrP res could have distinct abilities to catalyze PrP conversion and could selectively target different brain regions, producing the diversity of clinical symptoms and neuropathological alterations characteristic of prion strains. Support for this concept has come from various studies show- ing that PrP res isolated from different strains have distinct secondary S 6 4 JULY 2004 NEURODEGENERATION Figure 1 Histopathology of scrapie-infected mouse brain, showing the typical spongiform degeneration and deposition of PrP res aggregates. Evidence of the transmissible nature of TSE in sheep inoculated with a vaccine prepared from formalin-treated sheep brain Scrapie experimentally transmitted to mice Experimental transmission of scrapie Kuru transmitted to chimpanzees First enunciation of the protein-only hypothesis A hereditary human disease (GSS) is transmitted to animals Protease-resistant and hydrophobic protein discovered in scrapie- infected hamster brain Propagation of infectivity in cells Infectivity neutralized by anti-PrP antibodies Gene encoding PrP cloned Prion concept enunciated Scrapie agent shown to be highly resistant to DNA destruction Demonstration of small size of scrapie agent Purification and characterization of prion protein Production of transgenic animals expressing mutant PrP that develop clinical and pathological signs of TSE Cell-free conversion of PrP C into PrP res First mutation in PrP gene associated with familial TSE PrP knockout mice found to be resistant to scrapie Expansion of the prion concept to yeast proteins 1937 1939 1961 1966 1967 1980 1981 1982 1984 1985 1988 1989 1990 1993 1994 Conformational differences between PrP C and PrP res reported Figure 2 Timeline outlining the most relevant milestones related to TSE infectivity and the nature of the infective agent. © 2004 Nature Publishing Group http://www.nature.com/naturemedicine P E R S P E C T I V E structures 33,34 and can faithfully transfer their properties to PrP C in vitro 35 . However, so far it has not been shown whether such differ- ences in PrP res conformation are the cause, or simply another mani- festation, of the prion strain phenomenon. The presence and quantity of PrP res usually correlates with infectiv- ity 2,17 . However, there have been reports of infectivity being propa- gated in the absence of detectable PrP res (ref. 36) and other reports of samples with abundant PrP res but little or no infectivity 37 . To explain these results, prion proponents argue that although protease resist- ance is a typical biochemical feature of the misfolded infectious pro- tein, it does not necessarily equate with infectivity and indeed only part of the infectious protein is protease resistant 33 Although no nucleic acid consistently associated with infectious preparations of PrP res has been identified 2 , several groups have reported the presence of small quantities of nucleic acids in infectious samples 32,38 . Moreover, PrP res interacts with high affinity with nucleic acids, especially RNA 39,40 , and it has recently been shown that RNA may help to catalyze the conversion of PrP C into PrP res in vitro 41 Another argument often used against the prion hypothesis is the lack of infectious origins for other neurodegenerative and systemic disorders associated with protein misfolding and aggregation, such as Alzheimer’s disease, Parkinson’s disease and peripheral amyloidosis. In many of these diseases, interaction between normal and misfolded proteins leads to the formation of new pathological proteins by a process known as seeding-nucleation polymerization, which is markedly similar to the PrP conversion mechanism 42 . The apparent lack of infectivity of these other misfolded proteins suggests that the protein interactions and induced misfolding events common to these diseases might not explain the unique transmissibility of TSE. It is not clear, however, whether some of these diseases may be transmissible under certain conditions; indeed, some intriguing recent data provide preliminary support for prion-like transmission of other dis- eases 43,44 . If transmissibility can be convincingly demonstrated for other protein misfolding diseases, this will provide strong evidence in favor of the prion hypothesis What is arguably the most important piece of supporting evidence for the protein-only hypothesis has not been obtained: the generation of infectivity in the test tube. If the infectious agent is misfolded PrP res and its replication is promoted by interaction with PrP C , then the whole process should be possible to reproduce in vitro This experiment is considered by many scientists in the field as the ultimate proof for the protein-only hypothesis. One proposed strategy for generating de novo infectivity is to pro- duce PrP mutations that cause inherited TSEs. Several mutant PrP res - like molecules have been generated, some of which can acquire various biochemical properties of PrP res , but so far none has been shown to be infectious 45,46 . Another approach has been to induce misfolding of recombinant protein or short PrP synthetic peptides into β -sheet rich structures that have some of the biochemical and biological properties of PrP res (ref. 47). The hope in these experi- ments was that if even a very small percentage of the protein altered in vitro adopted the ‘infectious folding’, infectivity would result. However, this approach has systematically failed to generate infec- tious protein. A strategy considered more promising is the generation of infectivity by in vitro conversion of PrP C , because in such experi- ments the conversion process is triggered and catalyzed by brain- derived PrP res . In the cell-free system developed by Caughey and co-workers 30 , the low conversion yield made it difficult to distinguish potentially newly generated infectivity from the vast amount of infec- tivity used to begin the conversion. However, Collinge and co-workers took advantage of the species barrier phenomenon to test the infec- tivity of newly generated PrP res under conditions in which the PrP res from the inoculum would not be infectious 48 . Their results indicated that PrP res generated by their cell-free system was not infectious. The protein-misfolding cyclic amplification assay (PMCA) provides a new opportunity for evaluating the infectious properties of PrP res generated in vitro , because after amplification, >99% of protease- resistant protein is composed of newly produced PrP res . The high yield after conversion is essential to distinguish newly generated infectivity from that used to initiate the reaction. The PMCA technology was ini- tially difficult to reproduce, because many variables had to be opti- mized, but it has recently been replicated by several groups 41,49 Yeast prions In 1994 Reed Wickner 50 expanded the prion concept to explain the unusual nonmendelian transmission of two yeast genetic elements termed [URE3] and [PSI + ], which he proposed were prion forms of the proteins Ure2 (which is involved in modulating nitrogen metabo- lism) and Sup35 (which is essential in translation termination), respectively. Later, several other proteins showing prion-like behavior were identified in yeast and other fungi 51 . A yeast prion is generally defined as an infectious protein that behaves as a nonmendelian genetic element, transmitting biological information in the absence of nucleic acid. Diverse genetic, biochemical and structural evidence has been provided in support of the prion yeast phenomenon 51 . This evi- dence includes findings demonstrating that (i) the nonmendelian genetic element can be transmitted by cytoplasmic transference in the absence of nucleic acids; (ii) the prion phenotype can be reversed by protein denaturation and can arise again spontaneously (without introduction of new DNA) at a low frequency; (iii) the chromosomal gene encoding the normal form is required to propagate the prion form; (iv) overproduction of the normal protein increases sponta- neous occurrence of the prion form; (v) like PrP, the yeast prions can exist in two states—a normal, soluble, protease-sensitive state and an insoluble, protease-resistant state with a high tendency to form β -sheet-rich fibrillar aggregates; (vi) the conversion process can be reproduced in vitro using highly purified prion forms and can be sequentially repeated after serial dilutions, mimicking the continuous propagation of prions; and (vii) the prion-forming domain of Sup35p is modular and transferable; indeed, artificial prions have been generated by fusing a mammalian receptor to the yeast prion domain. NEURODEGENERATION JULY 2004 S 6 5 Mutant PrP produced in cells exhibits some biochemical properties of PrP res but lacks infectivity In vivo transmission of yeast Sup35 prion generated in vitro High-efficiency in vitro conversion of PrP C into PrP res by PMCA In vivo propagation of different yeast prion strains generated from recombinant protein in vitro 1996 2000 2001 2004 © 2004 Nature Publishing Group http://www.nature.com/naturemedicine P E R S P E C T I V E Although yeast prions were discovered fairly recently, the rapid progress in this area has already had important implications for the validation of the prion hypothesis—in particular, in enhancing the understanding of the strain phenomenon and in attempts to generate in vitro infectious protein. The first breakthrough came from studies by Sparrer and co-workers 52 showing that introducing in vitro –converted, purified Sup35 prion domain (N-terminal residues 1–254) into cells by liposome-based transformation resulted in the appearance of [PSI + ] prion in 1–2% of the transformed cells. However, the low efficiency of this procedure left open the possibility that infected protein was pro- duced de novo as a result of a high local con- centration of Sup35p. Another step forward came from studies by Maddelein and col- leagues 53 using the [Het-s] prions from the fungus Podospora anserine The authors showed that inserting fibrils made in vitro from renatured recombinant HET-s into the mycelia of P. anserine efficiently produces [Het-s] prions. The most compelling evidence yet for in vitro generation of prions, however, was recently published by King and Diaz-Avalos 54 and Tanaka and colleagues 55 . In the first study, bacterially produced N-terminal fragments of Sup35p labeled with green fluorescent pro- tein were transformed into amyloid fibrils ( Fig. 3 ) through incuba- tion with yeast-derived infectious aggregates. These fibrils were able to propagate the prion phenotype to yeast cells 54 De novo generation of infectivity was demonstrated by incubation and sonication of sam- ples containing infectious particles with recombinant Sup35 frag- ments, followed by serial dilutions into more of the recombinant protein. After three rounds of incubation, sonication and dilution, the sample contained a 160-fold dilution of the original infectious mate- rial. The newly converted material retained infectivity, whereas no activity was observed in the original seeding material at the same dilution 54 . Moreover, the in vitro –converted protein propagated the characteristics of the different strains used to begin the conversion reaction, implying that the heritable information defining distinct strains resides exclusively in the folding patterns of the protein. The same conclusion was obtained independently by Tanaka and co-workers, who were able to generate diverse stable, propagating conformations of a recombinant N-terminal fragment of Sup35p by exposing the proteins to different temperatures 55 . Infection of yeast with these different conformers led to the generation of distinct [PSI + ] strains in vivo , indicating that differences in the conformation of the infectious protein determine prion strain variation. Conclusions and perspectives The nature of the infectious agent associated with TSE has been one of the most heated debates in the biological sciences. The evidence in favor of the heretical protein-only hypothesis is extensive and is transforming the prion concept into a new dogma with vast implications for diverse areas of biology. Prions, defined as infectious proteins with the ability to transmit biological information through the propagation of alterna- tive protein folding, represent an entirely new mechanism for expand- ing phenotypic diversity without changes in the genome 2,51 . Skeptics argue, however, that the prion hypothesis is still not definitively proven 32,38 . It is widely agreed that the final proof of the protein-only hypothesis will require the engineering in vitro of a synthetic infectious protein capable of propagating a prion in vivo . These experiments have now been completed successfully using the yeast prion Sup35 pro- tein 54,55 , coming tantalizingly close to be the long-awaited definitive proof of the prion hypothesis. But do these studies demonstrate that prions are the infectious agents associated with TSEs? No—they demonstrate that the biological principle underlying prions is correct and that the nature of the strain phenomenon indeed resides in the propagation of alternative protein folding. However, the challenge of transmitting disease for an infectious agent composed only of a protein is greater in a multicellular organism than in yeast. The protein must resist biological clearance mechanisms, get to the right place in the brain, be propagated from cell to cell and induce specific cerebral dam- age that is different depending on the exact folding of the infectious agent. Until successful generation of infectivity by in vitro production or amplification of PrP res can be demonstrated, it remains possible that misfolded PrP is not the only component of the prion infectious agent. ACKNOWLEDGMENTS We thanks K. Maundrell (Serono), K. Abid and P. Saa (University of Texas Medical Branch) for critical reading of the manuscript and for helpful discussions. COMPETING INTERESTS STATEMENT The authors declare that they have no competing financial interests. HOW TO CITE THIS ARTICLE Please cite this article as supplement to volume 10 of Nature Medicine , pages S63–S67. Published online at http://www.nature.com/focus/neurodegen/ 1. Collinge, J. Prion diseases of humans and animals: their causes and molecular basis. Annu. Rev. Neurosci. 24 , 519–550 (2001). 2. Prusiner, S.B. Prions. Proc. Natl. Acad. Sci. USA 95 , 13363–13383 (1998). 3. Will, R.G. et al. A new variant of Creutzfeldt-Jakob disease in the UK. Lancet 347 , 921–925 (1996). 4. Collinge, J. Variant Creutzfeldt-Jakob disease. Lancet 354 , 317–323 (1999). 5. Cullie, J. & Chelle, P.L. Experimental transmission of trembling to the goat. C. R. Seances Acad. Sci. 208 , 1058–1160 (1939). 6. Chandler, R.L. Encephalopathy in mice produced by inoculation with scrapie brain material. Lancet 1 , 1378–1379 (1961). 7. Gajdusek, D.C., Gibbs, C.J. & Alpers, M. Experimental transmission of a Kuru-like syndrome to chimpanzees. Nature 209 , 794–796 (1966). 8. Gibbs, C.J. Jr et al. Creutzfeldt-Jakob disease (spongiform encephalopathy): trans- mission to the chimpanzee. Science 161 , 388–389 (1968). S 6 6 JULY 2004 NEURODEGENERATION Figure 3 In vitro formation of amyloid fibrils by recombinant fragment of yeast Sup35 labeled with GFP seeded by Sup35 aggregates. The micrographs show yeast-derived Sup35 particles (left), and growth by incorporating the recombinant GFP-labeled protein after 48 h of incubation (right). Reprinted from ref. 54. Chi-yen King; reprinted from Nature © 2004 Nature Publishing Group http://www.nature.com/naturemedicine P E R S P E C T I V E 9. Masters, C.L., Gajdusek, D.C. & Gibbs, C.J., Jr. Creutzfeldt-Jakob disease virus iso- lations from the Gerstmann-Straussler syndrome with an analysis of the various forms of amyloid plaque deposition in the virus-induced spongiform encephalopathies. Brain 104 , 559–588 (1981). 10. Cho, H.J. Is the scrapie agent a virus? Nature 262 , 411–412 (1976). 11. Alper, T., Cramp, W.A., Haig, D.A. & Clarke, M.C. Does the agent of scrapie replicate without nucleic acid? Nature 214 , 764–766 (1967). 12. Alper, T., Haig, D.A. & Clarke, M.C. The exceptionally small size of the scrapie agent. Biochem. Biophys. Res. Commun. 22 , 278–284 (1966). 13. Kimberlin, R.H. Scrapie agent: prions or virinos? Nature 297 , 107–108 (1982). 14. Griffith, J.S. Self-replication and scrapie. Nature 215 , 1043–1044 (1967). 15. Prusiner, S.B. Novel proteinaceous infectious particles cause scrapie. Science 216 , 136–144 (1982). 16. Bolton, D.C., McKinley, M.P. & Prusiner, S.B. Identification of a protein that purifies with the scrapie prion. Science 218 , 1309–1311 (1982). 17. Gabizon, R., McKinley, M.P., Groth, D. & Prusiner, S.B. Immunoaffinity purification and neutralization of scrapie prion infectivity. Proc. Natl. Acad. Sci. USA 85 , 6617–6621 (1988). 18. Enari, M., Flechsig, E. & Weissmann, C. Scrapie prion protein accumulation by scrapie-infected neuroblastoma cells abrogated by exposure to a prion protein anti- body. Proc. Natl. Acad. Sci. USA 98 , 9295–9299 (2001). 19. Peretz, D. et al. Antibodies inhibit prion propagation and clear cell cultures of prion infectivity. Nature 412 , 739–743 (2001). 20. Oesch, B. et al. A cellular gene encodes scrapie PrP 27-30 protein. Cell 40 , 735–746 (1985). 21. Chesebro, B. et al. Identification of scrapie prion protein-specific mRNA in scrapie- infected and uninfected brain. Nature 315 , 331–333 (1985). 22. Stahl, N. et al. Structural studies of the scrapie prion protein using mass spectrom- etry and amino acid sequencing. Biochemistry 32 , 1991–2002 (1993). 23. Pan, K.M. et al. Conversion of α -helices into β -sheets features in the formation of scrapie prion poteins. Proc. Natl. Acad. Sci. USA 90 , 10962–10966 (1993). 24. Hsiao, K.K. et al. Spontaneous neurodegeneration in transgenic mice with mutant prion protein. Science 250 , 1587–1590 (1990). 25. Hsiao, K.K. et al. Serial transmission in rodents of neurodegeneration from trans- genic mice expressing mutant prion protein. Proc. Natl. Acad. Sci. USA 91 , 9126–9130 (1994). 26. Barron, R.M. & Manson, J.C. A gene-targeted mouse model of P102L Gerstmann- Straussler-Scheinker syndrome. Clin. Lab. Med. 23 , 161–173 (2003). 27. Bueler, H. et al. Mice devoid of PrP are resistant to Scrapie. Cell 73 , 1339–1347 (1993). 28. Rubenstein, R., Carp, R.I. & Callahan, S.M. In vitro replication of scrapie agent in a neuronal model: infection of PC12 cells. J. Gen. Virol. 65 , 2191–2198 (1984). 29. Race, R.E., Fadness, L.H. & Chesebro, B. Characterization of scrapie infection in mouse neuroblastoma cells. J. Gen. Virol. 68 , 1391–1399 (1987). 30. Kocisko, D.A. et al. Cell-free formation of protease-resistant prion protein. Nature 370 , 471–474 (1994). 31. Saborio, G.P., Permanne, B. & Soto, C. Sensitive detection of pathological prion pro- tein by cyclic amplification of protein misfolding. Nature 411 , 810–813 (2001). 32. Chesebro, B. BSE and prions: uncertainties about the agent. Science 279 , 42–43 (1998). 33. Safar, J. et al. Eight prion strains have PrP(Sc) molecules with different conforma- tions. Nat. Med. 4 , 1157–1165 (1998). 34. Caughey, B., Raymond, G.J. & Bessen, R.A. Strain-dependent differences in β - sheet conformations of abnormal prion protein. J. Biol. Chem. 273 , 32230–32235 (1998). 35. Bessen, R.A. et al. Non-genetic propagation of strain-specific properties of scrapie prion protein. Nature 375 , 698–700 (1995). 36. Lasmezas, C.I. et al. Transmission of the BSE agent to mice in the absence of detectable abnormal prion protein. Science 275 , 402–405 (1997). 37. Hill, A.F. et al. Species-barrier-independent prion replication in apparently resist- ant species. Proc. Natl. Acad. Sci. USA 97 , 10248–10253 (2000). 38. Narang, H. A critical review of the nature of the spongiform encephalopathy agent: protein theory versus virus theory. Exp. Biol. Med. 227 , 4–19 (2002). 39. Weiss, S. et al. RNA aptamers specifically interact with the prion protein PrP. J. Virol. 71 , 8790–8797 (1997). 40. Derrington, E. et al. PrPC has nucleic acid chaperoning properties similar to the nucleocapsid protein of HIV-1. C. R. Acad. Sci. III 325 , 17–23 (2002). 41. Deleault, N.R., Lucassen, R.W. & Supattapone, S. RNA molecules stimulate prion protein conversion. Nature 425 , 717–720 (2003). 42. Soto, C. Unfolding the role of protein misfolding in neurodegenerative diseases. Nat. Rev. Neurosci. 4 , 49–60 (2003). 43. Kane, M.D. et al. Evidence for seeding of β -amyloid by intracerebral infusion of Alzheimer brain extracts in β -amyloid precursor protein-transgenic mice. J. Neurosci. 20 , 3606–3611 (2000). 44. Lundmark, K. et al. Transmissibility of systemic amyloidosis by a prion-like mech- anism. Proc. Natl. Acad. Sci. USA 99 , 6979–6984 (2002). 45. Lehmann, S. & Harris, D.A. Two mutant prion proteins expressed in cultured cells acquire biochemical properties reminiscent of the scrapie isoform. Proc. Natl. Acad. Sci. USA 93 , 5610–5614 (1996). 46. Chiesa, R., Piccardo, P., Ghetti, B. & Harris, D.A. Neurological illness in trans- genic mice expressing a prion protein with an insertional mutation. Neuron 21 , 1339–1351 (1998). 47. Jackson, G.S. et al. Reversible conversion of monomeric human prion protein between native and fibrilogenic conformations. Science 283 , 1935–1937 (1999). 48. Hill, A.F., Antoniou, M. & Collinge, J. Protease-resistant prion protein produced in vitro lacks detectable infectivity. J. Gen. Virol. 80 , 11–14 (1999). 49. Barret, A. et al. Evaluation of quinacrine treatment for prion diseases. J. Virol. 77 , 8462–8469 (2003). 50. Wickner, R.B. [URE3] as an altered URE2 protein: evidence for a prion analog in Saccharomyces cerevisiae. Science 264 , 566–569 (1994). 51. Uptain, S.M. & Lindquist, S. Prions as protein-based genetic elements. Annu. Rev. Microbiol. 56 , 703–741 (2002). 52. Sparrer, H.E., Santoso, A., Szoka, F.C., Jr & Weissman, J.S. Evidence for the prion hypothesis: induction of the yeast [PSI+] factor by in vitro –converted Sup35 pro- tein. Science 289 , 595–599 (2000). 53. Maddelein, M.L., Dos, R.S., Duvezin-Caubet, S., Coulary-Salin, B. & Saupe, S.J. Amyloid aggregates of the HET-s prion protein are infectious. Proc. Natl. Acad. Sci. USA 99 , 7402–7407 (2002). 54. King, C.Y. & Diaz-Avalos, R. Protein-only transmission of three yeast prion strains. Nature 428 , 319–323 (2004). 55. Tanaka, M., Chien, P., Naber, N., Cooke, R. & Weissman, J.S. Conformational vari- ations in an infectious protein determine prion strain differences. Nature 428 , 323–328 (2004). NEURODEGENERATION JULY 2004 S 6 7 © 2004 Nature Publishing Group http://www.nature.com/naturemedicine