Scrapie Infectivity Is Independent of Amyloid Staining Properties of the N-Terminally Truncated Prion Protein

Journal of Structural Biology 130, 323–338 (2000) doi:10.1006/jsbi.2000.4242, available online at http://www.idealibrary.com on

Scrapie Infectivity Is Independent of Amyloid Staining Properties of the N-Terminally Truncated Prion Protein Holger Wille,*,† Stanley B. Prusiner,*,†,‡,1 and Fred E. Cohen*,‡,§,㛳 *Institute for Neurodegenerative Diseases and Departments of †Neurology, ‡Biochemistry and Biophysics, §Medicine, and 㛳Cellular and Molecular Pharmacology, University of California, San Francisco, California 94143 Received December 23, 1999, and in revised form February 2, 2000

all caused by an aberrantly folded isoform of the prion protein (PrPSc)2 (Prusiner, 1998). Spectroscopic analysis has shown that the cellular isoform of PrP (PrPC) contains largely ␣-helices and little ␤-sheet, whereas PrPSc, the infectious isoform, is dominated by ␤-sheet structure (Caughey et al., 1991; Gasset et al., 1993; Pan et al., 1993; Prusiner et al., 1983; Safar et al., 1993a). In this article, we explore the impact of changes in the structure of full-length and truncated PrPSc on its infectivity, with special emphasis on the appropriateness of surrogate markers of infectivity such as ␤-sheet formation, Congo red binding, and proteinase K resistance. Discordance between each of these surrogates and infectivity has been observed as a function of the choice of solvent conditions. N-terminal truncation of PrPSc produces PrP 2730, which forms amyloids, as demonstrated by Congo red dye staining and green– gold birefringence when examined under polarized light (Cohen et al., 1982; Prusiner et al., 1983). The ␤-sheet content of PrP 27-30 as measured by FTIR and CD is between 47 and 54% (Caughey et al., 1991; Gasset et al., 1993; Nguyen et al., 1995; Pan et al., 1993; Safar et al., 1993a) and it forms linear polymers that are ultrastructurally and tinctorially indistinguishable

The prion protein undergoes a profound conformational change when the cellular isoform (PrPC) is converted into the disease-causing form (PrPSc). Limited proteolysis of PrPSc produces PrP 27-30, which readily polymerizes into amyloid. To study the relationship between PrP amyloid and infectivity, we employed organic solvents that perturb protein conformation. Hexafluoro-2-propanol (HFIP), which promotes ␣-helix formation, modified the ultrastructure of PrP amyloid and decreased the ␤-sheet content as well as prion infectivity. HFIP reversibly decreased the binding of Congo red dye to the PrP amyloid rods while inactivation of prion infectivity was irreversible. In contrast, 1,1,1-trifluoro-2-propanol (TFIP) did not inactivate prion infectivity but like HFIP, TFIP did alter the morphology of the rods and abolished Congo red binding. Solubilization using various solvents and detergents produced monomeric and dimeric PrP that lacked infectivity. Proteinase K resistance of detergent-treated PrP 27-30 showed no correlation with scrapie infectivity. Our results separate prion infectivity from the amyloid properties of PrP 27-30 and underscore the dependence of prion infectivity on PrPSc conformation. These findings also demonstrate that the specific ␤-sheet-rich structures required for prion infectivity can be differentiated from those required for amyloid formation. © 2000 Academic Press

2 Abbreviations used: AOT, sodium bis(2-ethylhexyl)sulfosuccinate (⫽ dioctylsulfosuccinate, sodium salt; ⫽ aerosol OT); APFO, perfluorooctanoate, ammonium salt; CD, circular dichroism; DLPC, detergent–lipid–protein complex; FTIR, Fourier transform infrared; HFIP, hexafluoro-2-propanol; HFPIP, 1,1,1,3,3,3hexafluoro-2-phenyl-2-propanol; IEP, isoelectric point (of a protein); IR, infrared; NMP, 1-methyl-2-pyrrolidinone; PBSZ, phosphate-buffered saline with zwittergent 3-12; PK, proteinase K; PMSF, phenylmethylsulfonyl fluoride; PrP, prion protein; PrP 27-30, PK-resistant N-terminal fragment of PrPSc (residues ⬃90 to 231); PrPC, cellular isoform of PrP; PrPSc, pathological isoform of PrP; rPrP, recombinant PrP; RT, room temperature; S20,W value, Svedberg sedimentation coefficient normalized to 20°C and the viscosity of water; TFE, 2,2,2-trifluoroethanol; and TFIP, 1,1,1-trifluoro-2-propanol.

Key Words: Congo red dye binding; electron microscopy; HFIP; protein aggregation; protein conformation; PrP 27-30; solubilization.

INTRODUCTION

Scrapie of sheep, bovine spongiform encephalopathy, and Creutzfeldt-Jakob disease of humans are 1

To whom correspondence should be addressed at Institute for Neurodegenerative Diseases, Box 0518, University of California, San Francisco, CA 94143-0518. Fax: ⫹415-476-8386. 323

1047-8477/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

324

WILLE, PRUSINER, AND COHEN

from amyloids (McKinley et al., 1987, 1991b; Prusiner et al., 1983). However, the structure of these polymers is mainly determined by buffer conditions during polymerization (McKinley et al., 1991b). Replication of prions features a profound change in the conformation of PrP. Current hypotheses on the mechanism for nascent PrPSc formation include the template-assistance model (Cohen and Prusiner, 1998) and the seeded nucleation hypothesis (Dealler, 1991; Gajdusek, 1988; Jarrett and Lansbury, 1993). Prion strain-specific information is enciphered in the conformation of PrPSc, arguing that PrPSc must act as a template in the formation of nascent PrPSc (Bessen and Marsh, 1994; Safar et al., 1998; Telling et al., 1996b). The seeded nucleation hypothesis argues for a crystallization-like mechanism whereby seeds consisting of ordered polymers initiate polymerization. Despite the demonstration that PrP amyloid is neither necessary nor sufficient for infectivity (Gabizon et al., 1987; McKinley et al., 1991a; Pan et al., 1993; Prusiner et al., 1990; Wille et al., 1996b), proponents of the seeded nucleation hypothesis persist (Caughey and Chesebro, 1997a; Caughey et al., 1997b). Although large polymeric structures can be excluded from being essential for prion formation, oligomers might serve as templates (Cohen et al., 1998; Eigen, 1996; Harper and Lansbury, 1997; Harrison et al., 1997; Safar, 1996; Wille and Prusiner, 1999). The hydrophobic nature of scrapie prions and their propensity to aggregate impeded attempts to purify and to characterize these pathogens (Millson et al., 1976; Prusiner et al., 1978). Multiple attempts to separate prion infectivity from membranes were unsuccessful (Hunter et al., 1968; Hunter and Millson, 1967; Malone et al., 1978; Marsh et al., 1984; Prusiner et al., 1980; Safar et al., 1991), in accord with the association of PrPSc with membranes. Because of difficulties in solubilization, structural studies of both PrPSc and PrP 27-30 remain elusive. Although solubilization of purified PrP 27-30 employing a combination of detergent and phospholipid to form detergent–lipid–protein complexes (DLPCs) increased prion titers (Gabizon et al., 1987, 1988), the lipids interfered with most analytical procedures. Solubilization of PrP 27-30 with the denaturing detergent SDS, even at low concentrations, resulted in diminished prion infectivity (Prusiner et al., 1983; Riesner et al., 1996). Other reports showed little, if any, solubilization with the infectious prions that migrated with a sedimentation coefficient of ⬃120 S (Akowitz et al., 1990; Kimberlin et al., 1971; Sklaviadis et al., 1992). Both detergents and amphophilic solvents are widely used to solubilize membrane proteins. Deter-

gents can replace the lipids of the membrane without disrupting the lipophilic milieu of the membrane or substantially destabilizing the structure of integral or associated membrane proteins. Short chain alcohols have been shown to solubilize even tightly aggregated proteins by shifting their assembly equilibria toward monomeric and dimeric states (Yang et al., 1993). Fluorinated alcohols exhibit an even stronger effect, partly by increasing their effectiveness as donors in hydrogen-bond formation. Hexafluoro-2-propanol proved to be the strongest solvent in this regard (Yang et al., 1993). On the other hand, it is well documented that addition of alcohols or detergents can induce major shifts in secondary structure (e.g., HFIP and TFE are generally known to favor ␣-helical structures). The effects of various solvents on protein structure can be modified by polyhydric substances such as sucrose, which is often used to stabilize proteins against the action of denaturing reagents (Cioci and Lavecchia, 1994; Lee and Timasheff, 1981). Here we present a synopsis of previously published studies (Wille et al., 1996a,b; Wille and Prusiner, 1999) that employed fluorinated solvents, either in combination with nondenaturing detergents or alone, to alter both the solubility and conformation of PrP 27-30 that polymerized into prion rods. These treatments were also used to dissociate prion infectivity from the amyloid properties of PrP 27-30. Taken together with earlier studies that showed that PrPSc only forms amyloid polymers after it is proteolytically converted into PrP 27-30 (McKinley et al., 1991b), these results argue that amyloid formation is required for neither PrPSc synthesis nor prion propagation. While both scrapie infectivity and amyloid formation require PrPSc with ␤-sheets, the specific ␤-sheet-rich structures required for prion infectivity can be distinguished from those needed for amyloid formation. Whether PrP 27-30 can lose its proteinase K resistance without changes in infectivity is of interest with respect to earlier studies showing that these properties are not causally linked (Hsiao et al., 1994; Kaneko et al., 1995, 1997; Safar et al., 1998; Telling et al., 1996a). Through a process of liquid–liquid extraction via reverse micelles and an organic phase, PrP 27-30 can be solubilized into monomeric and dimeric forms. Although solubilization of PrP 27-30 was not accompanied by any recognizable change in secondary structure as measured by FTIR spectroscopy, it did result in a loss of prion infectivity. While this finding would argue that an oligomer is necessary for PrPSc formation, clearance phenomena that are more efficient with smaller particles than large aggregates might complicate this interpretation. Moreover, ionizing irradiation studies suggest that the

PRION PROTEIN AMYLOID

molecular weight of the smallest infectious particle corresponds to a dimer of PrPSc (Bellinger-Kawahara et al., 1988). MATERIALS AND METHODS Materials and solvents. The solvents and detergents used in this study were purchased from the following sources: Acetonitrile, Baker (Phillipsburg, NJ); AOT, Sigma (St. Louis, MO); APFO, Fluka Chemie (Buchs, Switzerland); HFIP, Aldrich (Milwaukee, WI); HFPIP, Aldrich; NMP, ICN (Aurora, OH); TFIP, Narchem (Chicago, IL); 2-propanol, Sigma. All solvents and detergents were of the highest purity commercially available. Prion protein. PrP 27-30 was prepared as described previously (Prusiner et al., 1983). The last step of preparation included a sucrose gradient centrifugation in a zonal rotor. For the solubilization assays, PrP 27-30 was precipitated out of the sucrose by dilution and an ultracentrifugation step (Beckman rotor SW28, 24 000 rpm, 4°C, 16 h). The protein pellet was resuspended in 50 mM Na Hepes, 0.5 mM NaN3, pH 7.4, at a protein concentration of ⬃1 mg/mL or above and kept frozen until use. Negative staining and electron microscopy. Negative staining was done on carbon-coated 1000-mesh copper grids that were glow discharged prior to staining. Samples of 1 to 5 ␮l were adsorbed for up to 1 min. In the case of solvent-treated PrP 27-30, rods were briefly washed with 0.1 M and 0.01 M NH4 acetate, pH 7.4. All samples were then stained with freshly filtered 2% uranyl acetate or 2% ammonium molybdate. After drying, the samples were viewed in a Jeol JEM 100CX II electron microscope at 80 kV at a standard magnification of 40 000. The magnification was calibrated using negatively stained catalase crystals and ferritin. Bioassays. Bioassays were performed by inoculating 50 ␮l samples intracerebrally into weanling female hamsters (LVG/ LAK) purchased from Charles River Laboratories. Samples were diluted 100-fold into phosphate-buffered saline with 5 mg/mL bovine serum albumin as a carrier. Prion titers were determined by measuring the incubation time intervals from inoculation to the onset of neurological illness (Prusiner et al., 1980). Proteinase K assays. Proteinase K (PK) digestion on solventtreated PrP 27-30 was performed at a final concentration of 8.4 ␮g/mL of PrP 27-30 rods. Samples contained 0.2% (w/v) Sarkosyl, 50% (w/v) sucrose, and various solvents. Organic solvents used to treat PrP 27-30 samples were evaporated either in a stream of dry N2 or by incubating the opened tubes overnight at RT. Water was added after evaporation of the solvents to bring the sample volume up to the original level when necessary. Proteinase K was added to a concentration between 50 and 87.5 ␮g/mL. The reaction mix was incubated at 37°C for 2 to 4 h. The reaction was terminated by adding 5 mM PMSF, and the total sample volume was doubled by adding an equal volume of SDS sample buffer. Prior to SDS–PAGE, all samples were boiled for 5 min at 100°C. The activity of the PK was monitored under various conditions using Chromozym TRY (Boehringer-Mannheim, Mannheim, Germany). Fifty percent sucrose generally decreased the rate at which the proteinase K acted, so that the higher concentration (87.5 ␮g/mL) was used. SDS–PAGE and Western blotting. SDS gel electrophoresis was performed using 12 or 15% polyacrylamide gels. The gels were blotted onto polyvinylidene difluoride membranes. The membranes were saturated with 5% nonfat dry milk in TBST (10 mM Tris–HCl, pH 8.0, 150 mM NaCl, 0.05% (v/v) Tween 20) for 30 to 60 min at RT. The ␣-PrP polyclonal antisera RO73 and N10 were used at 1:4000 dilution in TBST in an overnight incubation at RT (Serban et al., 1990). The primary antibody was detected by using a horseradish peroxidase-coupled secondary antibody and

325

the ECL system (Amersham, Little Chalfont, Buckinghamshire, United Kingdom). Congo red binding assays. The Congo red binding assay was modified from Klunk et al. (1989). Samples contained 50% sucrose and PrP 27-30 rods at a concentration of 1.4 ␮g/mL. After addition of the solvents and 20 min incubation at RT, 10 ␮M Congo red (Chroma Gesellschaft, Stuttgart, Germany) in 100 mM Na phosphate, pH 7.4, 150 mM NaCl was added. Another 2 h of incubation at RT ensured that the complex formation reached completion. Changes in absorption were measured in a 8450A UV-visible spectrophotometer from Hewlett–Packard. Spectra were recorded between 280 and 800 nm. The binding was calculated as the quotient of absorption at 450 nm versus 504 nm. To compensate for small changes in absorption due to the solvent, the equivalent values of blank samples were subtracted from the protein data. Solvent concentrations above 10% for HFIP and TFIP and above 12.8% for NMP altered the Congo red absorption spectra too severely for reliable calculations of binding. Fourier transform infrared spectroscopy. FTIR analysis of PrP 27-30 rods was carried out on samples containing about 10 mg/mL protein, suspended in D2O-containing buffers. PrP 27-30 rods were precipitated with ethanol and centrifuged, and the pellet was resuspended in PBSZ in D2O (10 mM Na phosphate, pD 7.5, 150 mM NaCl, 0.12% Zwittergent 3-12). After an overnight incubation at RT to replace the remaining H2O, the protein was pelleted and resuspended in fresh PBSZ in D2O. An aliquot of this suspension was pelleted and resuspended in 25% sucrose (ultrex grade, Baker) in D2O. The sucrose-containing sample was measured on its own or after addition of HFIP or TFIP. FTIR analysis of PrP 27-30 after solubilization was carried out on samples with up to 1 mg/mL protein in D2O-containing buffers. Solubilization assays were performed as described below except for the use of D2O instead of H2O. FTIR spectra were recorded with a Perkin–Elmer (Norwalk, CT) System 2000 FTIR spectrophotometer with a microscope attachment. The sample was enclosed between 2 AgCl windows (International Crystal Laboratories, Garfield, NJ), creating a path length of 50 ␮m within an airtight sealed chamber. Spectra were recorded in the amide I⬘ region between 1750 and 1550 cm⫺1. Spectral analysis and selfdeconvolution were carried out as previously described (Byler and Susi, 1986; Gasset et al., 1993). Solubilization assay. PrP 27-30 was used mostly in concentrations ranging from 100 to ⬃500 ␮g/mL. The standard buffer was 50 mM Na Hepes, 1 mM PMSF, 0.5 mM NaN3, pH 7.4, unless indicated otherwise. Detergents were used at concentrations of 5 to 10 times the critical micelle concentration (weight per volume) and organic solvents were added at concentrations up to and well above the limits of miscibility with aqueous buffers (volume per volume). Solubilization assays were incubated at 37°C for 4 h (shorter incubation periods proved to be less successful) with frequent agitation in order to keep PrP 27-30 in suspension and the multiple liquid phases mixed. The incubation was followed by an ultracentrifugation at 100 000g for 1 h (Beckman rotor TLA 45, 43 000 rpm, 20°C). After centrifugation, the samples were separated into supernatant and pellet; the supernatant could often be differentiated into an organic and an aqueous phase, with or without particles at the interface. Pellets were resuspended in 50 mM Na Hepes, 0.5 mM NaN3, pH 7.4. Small aliquots of pellets and supernatants were used for SDS gel electrophoresis, bioassays, and negative stain electron microscopy. Sucrose gradient ultracentrifugation. Gradient centrifugation was performed using a Beckman SW 60Ti rotor at 60 000 rpm and 20°C. Centrifugation was set to reach a 兰␻2dt of 1.25ⴱ1012 rad2/s; this value was chosen to sediment particles with a density 0 of ⬃1.3 g/mL and a s 20,W value of 8 S and above. Linear gradients

326

WILLE, PRUSINER, AND COHEN

FIG. 1. Electron microscopy of negatively stained prion rods exposed to organic solvents. (A) Untreated prion rods. (B) Prion rods treated with 10% HFIP, termed “HFIP rods.” (C) Removal of the HFIP by evaporation led to reaggregation into amorphous aggregates. (D) Prion rods treated with 10% TFIP. (E) Prion rods treated with 10% NMP. (F) Prion rods treated with 10% 2-propanol. Minimal changes in rod structure. All panels were solvent treated and stained in the presence of 50% sucrose, negative stain 2% uranyl acetate. Bar, 100 nm. (Data taken from Wille et al. (1996b).)

contained 5–20% sucrose with 10 mM Na Hepes, 50 ␮M NaN3, pH 7.4, with or without 1% TFIP and 0.5% AOT. The gradient volume was 4 mL with 400 ␮l of sample carefully layered onto the top of the sucrose gradient. PrP 27-30 samples were solubilized by 20% TFIP ⫹ 1% AOT. The high density of TFIP made it necessary to dilute the PrP 27-30 sample by a factor of 4 with 10 mM Na Hepes, 50 ␮M NaN3, pH 7.4. This dilution step reduced the PrP 27-30 concentration and the scrapie titer considerably but was necessary to prevent the mixing of the sucrose gradient with the sample prior to centrifugation. After centrifugation, the gradients were harvested from the top into ⬃250-␮l aliquots. The sedimentation coefficients were calculated (Steensgaard et al., 1978). The specific volume for PrP 27-30 was calculated (Durchschlag, 1986), assuming that PrP 27-30 contained 77% protein (⬃16 240 Da; aa 90 –231) with a specific volume of 0.71 cm3/g; 14% lipid (⬃2975 Da in the GPI anchor (Stahl et al., 1992)) with a specific volume of 1 cm3/g; and 9% polysaccharide (⬃1955 Da (Endo et al., 1989)) with a specific volume of 0.61 cm3/g. Taken together these data indicate that PrP 27-30 should have a specific volume of about 0.74 cm3/g (or a density of about 1.35 g/mL). The influence of potentially bound AOT molecules on the density of PrP 27-30 was not taken into account in this estimation. We performed an independent calculation of sedimentation coefficients that employs stan0 dard proteins with known s 20,W values, such as ribonuclease A with 1.6 S and ovalbumin with 3.6 S (Steensgaard et al., 1992), and compares the distances ribonuclease A, ovalbumin, and PrP 27-30 migrated in the sucrose gradient.

RESULTS

Morphology of prion rods exposed to organic solvents. Although prion rods are an artifact of limited proteolysis in the presence of detergent, they do possess the properties of amyloids and thus have a naturally occurring counterpart, i.e., the amyloid fibrils of PrP plaques in the brains of some animals with prion disease. Negatively stained samples of untreated rods showed the typical prion rod structure (Fig. 1A). After addition of 10% HFIP, the rods disassembled into smaller units termed “HFIP rods,” which exhibited a more regular substructure. HFIP rods resemble sheet-like structures with curled up edges, resulting in two light bands visible in the electron micrographs (Fig. 1B). Evaporation of HFIP led to a reaggregation into amorphous structures that showed no regular substructure or similarities to prion rods (Fig. 1C). When diluted 100fold into phosphate-buffered saline, similar to the conditions used for bioassays, the HFIP rods lost their structure and reaggregated into small, amorphous particles (data not shown). Table I shows the concentration dependence of the

PRION PROTEIN AMYLOID

TABLE I Concentration Dependence of the HFIP Effect on Prion Rods Concentration of HFIP (%)a

Prion rodsb

Rods with increased substructurec

HFIP rodsd

0 1 2.5 4 5 6 7 8 10

⫹ ⫹ ⫹ ⫹ ⫺ ⫺ ⫺ ⫺ ⫺

⫺ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺

⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹

Note. (⫹) More than 90% of structures show this appearance; (⫺) fewer than 10% of structures show this appearance. (Table taken from Wille et al. (1996b).) a All HFIP treatments in the presence of 50% sucrose. b Normal PrP 27-30 rods as seen in Fig. 1A. c HFIP rods not included. d HFIP rods as seen in Fig. 1B.

transition between infectious prion rods and HFIP rods. At concentrations up to 4% HFIP, no change in substructure was observed; at 5% some rods could clearly be classified as HFIP rods, while the majority showed intermediate features (data not shown). At 6% HFIP and higher (up to 50%), only HFIP rods could be observed. The addition of TFIP resulted in a similar yet less extensive alteration of prion rod structure compared to HFIP rods (Fig. 1D). The main difference appeared to be the presence of a higher proportion of flat, sheet-like structures. Removal of TFIP led to reaggregation yielding amorphous particles with only a small degree of substructure similar to that seen with HFIP (data not shown). The nonfluorinated solvent 1-methyl-2-pyrrolidinone (NMP) also altered the structure of the rods in a manner different from either HFIP or TFIP (Fig. 1E). Control rods treated with 2-propanol exhibited little change in ultrastructure (Fig. 1F), emphasizing the resistance of prion rods to mild perturbants. HFIP inactivates prion infectivity. Increasing the concentration of HFIP resulted in a progressive reduction in infectivity of the prion rods. Concentrations of up to 2.5% HFIP did not alter infectivity, whereas 5% HFIP reduced the infectivity by more than a factor of 10. At 7% or higher concentrations of HFIP, infectivity of the rods was almost completely abolished (Fig. 2). Unexpectedly, the structurally related solvent TFIP had no adverse effect on infectivity at concentrations up to 50% (Fig. 2). It is notable that TFIP seems to disperse the rods into smaller structures, not unlike HFIP (compare Figs. 1B and 1D), but without detrimental effects on in-

327

fectivity. Neither removal of HFIP by evaporation nor addition of acetonitrile, which promotes ␤-sheet formation, reversed the inactivation of prion infectivity by HFIP (data not shown). Proteinase K resistance and exposure to solvents. Because formation of the prion rods depends on the resistance of PrPSc to limited proteinase K digestion, we measured the resistance of the rods to proteolytic digestion after exposure for 20 min to various solvents. While 10% HFIP abolished the resistance to proteinase K, similar concentrations of TFIP, NMP, and 2-propanol did not (Fig. 3A, lanes 2–5). Because both HFIP and TFIP inhibited the activity of proteinase K, we removed these solvents by evaporation prior to the digestion. However, proteinase K resistance could not be restored by the removal of HFIP. In contrast, NMP and 2-propanol did not inhibit proteinase K activity (data not shown). Concentrations of HFIP up to 2.5% had no effect on the proteinase K resistance of PrP 27-30 in the rods, but at 3%, the amount of PrP 27-30 after proteinase K digestion was substantially reduced. At concentrations of 3.5% or higher, no PrP 27-30 could be detected (Fig. 3B). The faint bands in lanes 7 to 10 are due to the cross-reaction of the antiserum with proteinase K, as shown in lane 12, where reactivity with proteinase K alone is demonstrated. Congo red binding. Since prion rods bind Congo red dye and exhibit green– gold birefringence under polarized light in a manner that is characteristic of amyloid (Prusiner, 1983), we examined the effect of HFIP on Congo red dye binding. In the presence of 50% sucrose, Congo red alone showed an absorption maximum at 504 nm, which shifted to a shorter wavelength of ⬃450 nm when bound to rods (Fig. 4A). The presence of as little as 0.1% HFIP reduced

FIG. 2. Bioassays of prions rods exposed to HFIP or TFIP. Prion rods were exposed to increasing concentrations of solvent for at least 20 min at RT. (Data taken from Wille et al. (1996b).)

328

WILLE, PRUSINER, AND COHEN

The closely related solvent TFIP had essentially the same effect on Congo red binding as HFIP. Congo red dye binding was substantially diminished at 0.8% TFIP and abolished at 10% (v/v) (Fig. 4C). Like HFIP, the inhibition of Congo red binding by TFIP was reversible (data not shown). NMP also inhibited Congo red binding, but the effect was seen only at concentrations above 1.6% NMP (Fig. 4C).

FIG. 3. HFIP abolishes the resistance of prion rods to digestion with proteinase K. SDS–PAGE of samples in 12% polyacrylamide gels; Western blots developed with ␣-PrP rabbit antiserum RO73. (A) Prion rods were incubated with 10% solvent for 20 min and HFIP and TFIP were evaporated prior to digestion with proteinase K, whereas other solvents were not evaporated. (B) Increasing HFIP concentrations ranging from 0 to 5% HFIP. Prion rods were treated with solvents and proteinase K in the presence of 50% sucrose. Molecular weight markers in kilodaltons are shown on the left side. (Figure taken from Wille et al. (1996b).)

the Congo red binding by ⬎50%. Addition of 2.5% HFIP or more disrupted the protein– dye complex, and the Congo red spectrum shifted back to its unbound state, even though prion infectivity remained unaffected at this concentration (Fig. 4B, triangles; Fig. 2). After evaporation of the solvent from HFIPtreated rods, the Congo red binding was restored (Fig. 4B, open squares). Reversing the sequence of addition of HFIP and Congo red gave virtually the same results, arguing that the results were unlikely to be due to interference between HFIP and Congo red (data not shown). PrPC did not bind Congo red under the conditions used here (Fig. 4B, closed square).

Secondary structure conversions induced by HFIP and TFIP. Fourier transform infrared spectroscopy of rods in PBS with 0.12% Zwittergent 3-12 (PBSZ) showed a spectrum indicating a high amount of low-frequency ␤-sheet (between 1641 and 1613 cm⫺1) and a small number of ␣-helical and turn components at 1662–1650 cm⫺1 and 1682–1662.5 cm⫺1, respectively. Transferring the prion rods from PBSZ in D2O to 25% sucrose in D2O shifted the IR spectrum from low-frequency ␤-sheet to one with more ␣-helical and high-frequency ␤-sheet signals (data not shown). This shift from low-frequency to high-frequency ␤-sheet signals can be attributed to a shift from predominantly intermolecular to intramolecular ␤-sheet contacts (Byler and Susi, 1986; Jackson and Mantsch, 1992), during which the percentage of ␤-sheet decreases slightly (Table II). The small increase in ␣-helix content could be explained by the helix-promoting effect of polyols, such as sucrose and glycerol, that has been found in the model system melittin (Bello, 1993). Addition of 10% HFIP increased the amount of ␣-helix and decreased the ␤-sheet content to ⬃17% (Table II), coinciding with a complete loss of infectivity (Fig. 2). The addition of 2.5% HFIP reduced the ␤-sheet while increasing the content of ␣-helix and random coil but did not alter infectivity. Like-

FIG. 4. Binding of Congo red dye to prion rods. (A) Absorption spectra of Congo red bound to prion rods in the presence of 50% sucrose (solid line) and Congo red alone in 50% sucrose (dashed line). (B) Disruption of the Congo red/prion rod complex as a function of HFIP concentration (open triangles) as measured by absorption at 450 and 504 nm. Prion rods exposed to HFIP followed by evaporation of the solvent showed no significant decrease in the binding of Congo red (open squares). Data points for both curves represent the mean of three independent measurements; error bars indicate the standard deviation. PrPC did not bind Congo red (closed square). (C) Prion rods were exposed to HFIP, TFIP, or NMP and absorption was measured at 450 and 504 nm. All measurements in (B) and (C) were done in the presence of 50% sucrose; all data points were subtracted with the according blank control values. (Figure taken from Wille et al. (1996a).)

PRION PROTEIN AMYLOID

TABLE II Secondary Structure Content of PrP 27-30 in Rods after HFIP and TFIP Treatment Secondary structure (%) Protein

␣-Helix

␤-Sheet

Turns

Coil

PrP 27-30 In PBSZ In 25% sucrose ⫹2.5% HFIP ⫹5% HFIP ⫹10% HFIP ⫹2.5% TFIP ⫹5% TFIP ⫹10% TFIP

23 27 41 36 46 34 40 40

48 43 21 25 17 33 27 28

9 13 10 15 16 11 14 12

20 17 28 24 21 22 19 20

Note. All solvent-treated rods were in the presence of 25% sucrose. (Table taken from Wille et al. (1996b).)

wise, addition of TFIP caused the IR spectrum to increase its ␣-helix and random coil signals (Table II) but did not change infectivity levels (Fig. 2). Solubilization assays. The observation that TFIP decreases the aggregate size of the prion rods without interfering with scrapie infectivity made this solvent an ideal candidate for solubilization studies. Of a large number of solvents and detergents tested, only a few yielded any solubilization of PrP 27-30, as judged by ultracentrifugation at 100 000g for 1 h (data not shown). Figure 5 shows that untreated PrP 27-30 (control) was pelleted owing to its aggregated nature. TFIP or the detergent AOT alone led to a limited amount of PrP 27-30 in the supernatants, and only a combination of both resulted in a virtually complete solubilization (Fig. 5). Other combinations that yielded almost complete solubilization involved solvents such as 1,1,1,3,3,3hexafluoro-2-phenyl-2-propanol (HFPIP) and ammonium perfluorooctanoate (APFO) as a detergent (data not shown). We observed that the best results were achieved under conditions that caused a separation into two liquid phases either during the incubation at 37°C or later at the 100 000g ultracentrifugation. Consistently, PrP 27-30 was found primarily in the organic phase and at the interphase. The organic phase was generally at the bottom of the tube because of the high density of the fluorinated alcohols. Solubilization of a protein or hydrophilic molecule into an organic solvent is usually achieved through reverse micelles. Since the detergent AOT is frequently used in studies of reverse micelles, we tested if solubilization occurred via this mechanism here as well. Contrary to solubilization in aqueous systems, the solubilization in reverse micelles through AOT is favored at about 1 pH unit below the isoelectric

329

point (IEP) of the protein and decreases at pHs further away from the IEP (Go¨klen and Hatton, 1987; Wolbert et al., 1989). Subsequently, we found that PrP 27-30 is preferentially solubilized at pH ⬃8.0 (with its IEP near pH 9.0), with a decrease in solubilization at lower pH and at pHs right around and above the IEP of PrP 27-30 (data not shown). Therefore, it appears that PrP 27-30 is solubilized by reverse micelles formed by the detergent AOT. Dynamic light scattering was used to confirm the presence of reverse micelles in the organic phase. A difference in size between empty and protein-filled reverse micelles could be detected (data not shown). Negative staining allowed us to check the aggregation state of solubilized proteins found in the supernatant fractions. The organic solvent TFIP alone reduced the aggregate size, but did not fully solubilize PrP 27-30 (Fig. 6B). AOT alone influenced the structure of PrP 27-30 aggregates even less, barely changing rod shape and ultrastructure (Fig. 6C). Only a combination of both led to a complete disappearance of rod-shaped aggregates (Fig. 6D). Certain combinations of HFPIP, AOT, and APFO were equally successful while the single components alone did not solubilize PrP 27-30 in rods (data not shown). Electron microscopy revealed that solubilization occurred only under conditions in which an ␣-helix-promoting solvent and a detergent were used together. The widely used solubilization criterion of an ultracentrifugation at 100 000g was clearly

FIG. 5. Western blots of PrP 27-30 after solubilization by TFIP, AOT, or a combination thereof; Western blot developed with ␣-PrP rabbit antiserum N10. Samples were centrifuged at 100 000g for 1 h and then separated into pellet and supernatant fractions. The supernatants were split into organic and aqueous phases wherever applicable, with the top phase labeled S1 and the bottom phase labeled S2. The S2 phases of samples treated with 1% AOT showed considerable viscosity, warranting the term detergent pellet more than supernatant 2. The control (buffer only) showed all PrP 27-30 in the pellet. TFIP or AOT alone showed some solubilization, while a combination of both gave the best results. (Data taken from Wille and Prusiner (1999).)

330

WILLE, PRUSINER, AND COHEN

FIG. 6. Negative stain electron microscopy of prion rods before and after solubilization. (A) Highly aggregated prion rods before solubilization. (B) Organic phase (supernatant 2) of PrP 27-30 “solubilized” by 15% TFIP showing amorphous aggregates of various sizes as well as contaminating ferritin complexes (arrowheads). (C) Treatment with 1% AOT changed the prion rod structure but did not disaggregate the polymers. (D) A combination of 15% TFIP ⫹ 1% AOT ⫹ 25% sucrose truly solubilized PrP 27-30, as no protein aggregates could be detected by negative stain electron microscopy (supernatant 2). (A and D) Negative stain 2% ammonium molybdate; (B and C) negative stain 2% uranyl acetate. Bar, 100 nm. (Data taken from Wille and Prusiner (1999).)

insufficient for judging solubilization of PrP 27-30 (Riesner et al., 1996). Prion infectivity and proteinase K resistance of solubilized PrP. Bioassays of samples with PrP 27-30 from solubilization experiments showed that substantial amounts of infectivity could be found in the supernatant fractions after treatment with organic solvents or detergents alone. Solubilization by combinations of TFIP ⫹ AOT ⫹ sucrose led to a titer in the organic phase (S2) that was 10 to 1000 times higher than that of the pellet fraction (Table III). Experiments using the solvent HFPIP either in combination with detergent or alone invariably abolished scrapie infectivity (Table III). Treatment with organic solvents and detergents not only changed the aggregation state of PrP 27-30, but it also modified the resistance of PrP 27-30 to proteinase K digestion (Fig. 7). The control pellet and PrP 27-30 treated by TFIP alone remained resistant to proteinase K digestion. Supernatant fractions containing PrP 27-30 treated with HFPIP, APFO, or a combination of both lost all resistance to proteinase K digestion while PrP 27-30 treated with AOT and combinations of AOT and TFIP showed a substantial reduction in resistance to proteinase K digestion (Fig. 7). Although disaggregation of PrP 27-30 was generally accompanied by a diminution in the resistance

of the protein to digestion by proteinase K, the exceptions to this rule are most important. Even partial disaggregation as seen with HFPIP or APFO (data not shown) led to a reduction in proteinase K resistance. However, treatment with TFIP disrupted PrP 27-30 aggregates without reducing proteinase K resistance, showing that the aggregation state does not always correlate with proteinase K resistance. Bioassays showed that prion rods treated with APFO retained full infectivity while samples treated with HFPIP or a combination of APFO and HFPIP lost infectivity (Table III). The fact that these three conditions abolish proteinase K resistance equally well (Fig. 7) but have different effects on scrapie infectivity suggests that proteinase K resistance might not be required for preservation of prion infectivity. This finding is in accord with other studies on transgenic mice expressing the P101L mutation that generates infectivity spontaneously (Hsiao et al., 1994; Telling et al., 1996a). Interestingly, a proteinase-K-sensitive fraction of PrPSc correlates with the length of the incubation period (Safar et al., 1998). Fourier transform infrared spectroscopy of solubilized PrP 27-30. In contrast to the experiments done on HFIP- and TFIP-treated rods in 25% sucrose (see above), we found that the ␤-sheet content remained almost constant throughout treatments

331

PRION PROTEIN AMYLOID

TABLE III Scrapie Infectivity of PrP 27-30 after Solubilization Log titer ID50/mL (n/n 0 ) Assay conditions

Experiment 1

Experiment 2

Experiment 3

Experiment 4

PrP 27-30, untreated control PrP 27-30 buffer control Pellet Supernatant PrP 27-30 ⫹ 15% TFIP Pellet Supernatant 1 Supernatant 2 PrP 27-30 ⫹ 1% AOT Pellet Supernatant 1 Supernatant 2 PrP 27-30 ⫹ 15% TFIP: Pellet ⫹1% AOT: Supernatant 1 ⫹25% sucrose: Supernatant 2 PrP 27-30 ⫹ 10% HFPIP Pellet Supernatant 1 Supernatant 2 PrP 27-30 ⫹ 1% APFO Pellet Supernatant PrP 27-30 ⫹ 10% HFPIP: Pellet ⫹1% AOT: Supernatant 1 ⫹1% APFO: Supernatant 2

7.5 (4/4)

8.2 (4/4)

8.8 (4/4)

9.3 (4/4)

6.7 (4/4) 1.8 (1/4)

8.3 (4/4) 3.8 (4/4)

9.4 (4/4) 3.2 (3/4)

9.3 (4/4) 3.7 (4/4)

6.4 (4/4) n.d. 3.3 (2/4)

n.d. n.d. n.d.

7.5 (4/4) 4.3 (4/4) 7.8 (4/4)

8.4 (4/4) 2.7 (3/4) 7.1 (4/4)

6.7 (4/4) n.d. 7.8 (4/4)

n.d. n.d. n.d.

9.1 (4/4) 5.7 (4/4) 9.4 (4/4)

9.3 (4/4) 7.2 (4/4) 9.3 (4/4)

3.1 (2/4) n.d. 6.2 (4/4)

5.2 (4/4) 2.5 (4/4) 8.2 (4/4)

6.2 (4/4) 2.6 (4/4) 7.2 (4/4)

6.2 (4/4) 3.3 (4/4) 7.2 (4/4)

⬍1 (0/4) n.d. ⬍1 (0/4)

n.d. n.d. n.d.

1.3 (1/4) ⬍1 (0/4) 3.7 (4/4)

⬍1 (0/4) 1.7 (2/4) 7.4 (3/4)

n.d. n.d.

n.d. n.d.

9.0 (4/4) 8.0 (4/4)

10.0 (4/4) 7.1 (4/4)

n.d. n.d. n.d.

⬍1 (0/4) ⬍1 (0/4) ⬍1 (0/4)

⬍1 (0/4) ⬍1 (0/4) ⬍1 (0/4)

⬍1 (0/4) ⬍1 (0/4) ⬍1 (0/4)

Note. n/n 0 , number of sick animals/number of inoculated animals; n.d., not determined. (Data taken from Wille and Prusiner (1999).)

with either detergents or solvents (Table IV). Only a combination of detergent and solvent with sucrose (15% TFIP ⫹ 1% AOT ⫹ 12.5% sucrose) showed a significant decrease in ␤-sheet structure. This shift from ␤-sheet to ␣-helix was similar to the change observed in the experiments with solvent and sucrose alone (Table II) indicating that sucrose is an essential component for this conformational transition. PrP 27-30 solubilized by HFPIP ⫹ AOT ⫹ APFO could not be analyzed by FTIR spectroscopy because of uninterpretable variations in absorption patterns (data not shown). Sedimentation coefficients of solubilized PrP. We performed sucrose gradient centrifugation to determine the molecular size and aggregation state of solubilized PrP 27-30. The high density of TFIP made it necessary to dilute PrP 27-30 samples with aqueous buffer. This step led, most likely, to a reextraction of the protein into the aqueous phase, since no phase separation was present after dilution. A standard gradient of 5–20% sucrose without addition of either detergent or solvent led to a complete reaggregation during centrifugation (data not shown). The addition of 1% TFIP and 0.5% AOT to

the gradient allowed PrP 27-30 to remain soluble and to migrate according to its size, shape, and 0 density (Fig. 8A). Calculation of the s 20,W values revealed that the majority of solubilized PrP 27-30 migrated as monomers of ⬃2 S, and a small fraction of the protein migrated around 4 S, which corresponds to PrP 27-30 dimers. The pelleted PrP 27-30 had a sedimentation coefficient of ⬎8 S. An indepen0 dent confirmation of the s 20,W values calculated from the first set of sucrose gradients was obtained by use of standard proteins such as ribonuclease A and 0 ovalbumin. The s 20,W values calculated from these gradients (Fig. 8B) matched those that were calculated without the use of calibrated proteins (Fig. 8A); thus, we were able to calculate the sedimentation coefficients by two independent methods. Bioassays of fractions from three independent sucrose gradients revealed that prion infectivity could be found only in the pellet fractions (Fig. 8C). All scrapie prion infectivity that was loaded onto the gradients, as determined by separate bioassays, was recovered in the pellets, showing that only 10% of PrP 27-30 accounted for all infectious units, while the remaining 90% could be separated from infectiv-

332

WILLE, PRUSINER, AND COHEN

FIG. 7. Western blots of proteinase K assays on PrP 27-30 after solubilization. Western blots developed with ␣-PrP rabbit antiserum N10. The supernatants after solubilization by various agents (after evaporation of the organic solvents) were incubated with PK for 2 h at 37°C. Solubilization by 10% HFPIP; 1% APFO; 10% HFPIP ⫹ 1% AOT ⫹ 1% APFO; or 15% TFIP ⫹ 1% AOT ⫹ 12.5% sucrose almost completely abolished PK resistance in PrP 27-30. Treatment by 1% AOT or 20% TFIP ⫹ 1% AOT reduced PK resistance, while 15% TFIP alone had no significant effect. (Figure taken from Wille and Prusiner (1999).)

ity (Fig. 8). Ultracentrifugation without solvent or detergent in the gradient (see above) led to reaggregation and pelleting of PrP 27-30, and all of the infectivity that remained after solubilization was found in the pellet as well. Negative staining did not show any protein aggregates in the peak fractions of monomeric and dimeric PrP. The pellet fraction contained a small number of PrP aggregates showing morphological features of prion rods, indicating the PrP 27-30 had not been solubilized (data not shown). DISCUSSION

The findings reported here with an array of organic solvents and detergents extend earlier studies,

which argued that procedures that diminish ␤-sheet content in PrPSc result in a decrease in scrapie prion infectivity (Gasset et al., 1993; Safar et al., 1993b). In our initial study, we found that denaturing PrP 27-30 isolated by sucrose gradient centrifugation resulted in a diminution of scrapie infectivity and disruption of the rod-shaped polymers that possess the tinctorial properties of amyloid (Prusiner et al., 1983). Earlier studies with other amyloids had shown a correlation between Congo red dye binding and the ␤-sheet content of the amyloid proteins (Glenner et al., 1972). To date, Congo red dye binding is widely used as the defining criterion for amyloid. Proteins of various origins have been found to polymerize via an antiparallel ␤-sheet interaction regardless of their original conformation (Sunde and Blake, 1998). Subsequently, we found that those procedures that diminished prion infectivity also decreased ␤-sheet content of PrP 27-30, as measured by FTIR (Gasset et al., 1993; Wille et al., 1996b). Similar findings were reported using CD (Safar et al., 1993b). Dispersion of PrP 27-30 into detergent–lipid–protein complexes (DLPCs) increased the prion titer by as much as 10-fold (Gabizon et al., 1987, 1988), but the lipids interfered with many biophysical methods of analysis, rendering this procedure impractical. Intermediate concentrations of guanidine hydrochloride (1.5 to 4.5 M) allowed a moderate level of solubilization but suffered from insufficient yields (Safar et al., 1993a). More recently, use of ⬃0.2% SDS and sonication led to disaggregation of PrP 27-30 into small spherical particles with a relatively homogeneous size distribution of around 10 nm and 0 a s 20,W value of ⬃6 S (Riesner et al., 1996), but this procedure failed to produce monomers or dimers of PrP 27-30. In the studies reported here, some ␣-helix-promoting solvents, such as HFIP and HFPIP, decreased

TABLE IV Secondary Structure of PrP 27-30 in Rods after Treatment with Organic Solvents, Detergents, and Combinations Secondary structure (%) Solubilization with

␣-Helix

␤-Sheet

Turns

Coil

PrP 27-30 ⫹buffer only (control): Pellet ⫹1% AOT: Pellet ⫹1% APFO: Pellet ⫹20% TFIP: Pelleta ⫹20% TFIP: Mixture ⫹10% HFPIP: Pelleta ⫹10% HFPIP: Mixture ⫹20% TFIP, 1% AOT: Supernatant 2 ⫹15% TFIP, 1% AOT, 12.5% sucrose: Supernatant 2

24 26 14 22 30 21 38 18 40

41 44 55 54 40 48 32 52 25

9 14 25 7 10 7 10 7 13

26 16 6 17 20 24 20 23 22

Note. (Data taken from Wille and Prusiner (1999).) Actually a mixture of pellet and supernatant 2.

a

PRION PROTEIN AMYLOID

333

ties, aggregational status, and proteinase K resistance, were changed independently, indicating a lack of a causal relationship between these parameters and scrapie infectivity (see Table V).

FIG. 8. Sucrose gradient centrifugation of PrP 27-30 solubilized by 20% TFIP ⫹ 1 % AOT. (A) Western blot of solubilized PrP 27-30 subjected to ultracentrifugation in a 5–20% sucrose gradient. The peak of PrP 27-30 content (fractions 5 to 8) corresponds to mono0 meric protein with a s20,W value of ⬃2 S. A second peak containing less protein (fractions 9 to 11) corresponds to dimeric PrP 27-30, ⬃4 S. A relatively faint band of aggregates with greater than 8 S could be seen in the pellet (fraction 16). (B) A sample of solubilized PrP 27-30 was mixed with small amounts of ribonuclease A and ovalbumin. A Coomassie-stained SDS–PAGE shows the peak of ribonuclease A in fractions 5 and 6 and the peak of ovalbumin in fractions 8 and 9. Sedimentation coefficients for ribonuclease A and ovalbumin from this centrifugation run agree well with published data. The 0 s20,W value of the monomeric PrP 27-30 calculated by comparison to 0 the s20,W values of the ribonuclease A and ovalbumin concurs with the one calculated directly from the sucrose gradient centrifugation. (C) Bioassays of the sucrose gradient fractions showed that scrapie infectivity (dashed line) could only be found in the pellet fractions containing protein aggregates larger than 8 S; neither PrP 27-30 monomers (fractions 5 to 8) nor dimers (fractions 8 to 10) showed significant infectivity. Both curves show the average of three independent experiments. The titer of solubilized PrP 27-30 before the sucrose gradients (diamond on the far left) and that of the pellet fraction (number 16) were sufficiently similar to infer that all infectivity after solubilization resided in the remaining protein aggregates. All other fractions were below or close to the detection limit of 1 infectious unit/mL. (Data taken from Wille and Prusiner (1999).)

scrapie prion infectivity while closely related solvents, like TFIP, remained without effect (Fig. 2). Other parameters, such as amyloid staining proper-

PrP amyloid and prion infectivity. Our attempts to identify structural parameters that correlate prion infectivity with amyloid showed limited success. While the formation of prion rods with the morphology and properties of amyloid has only been found associated with preparations exhibiting scrapie infectivity, such polymers are not obligatory for infectivity (Gabizon et al., 1987; McKinley et al., 1991a; Pan et al., 1993). PrP 27-30 rods exhibited an altered morphology with 5% HFIP, and infectivity diminished substantially; 10% TFIP also altered rod morphology but did not change prion titers (Fig. 1 and Table I). The amyloid properties of prion rods as measured by the binding of Congo red dye to PrP 27-30 were clearly dissociable from scrapie infectivity. For example, 2.5% HFIP abolished Congo red binding to PrP 27-30 rods (Fig. 4B) but did not alter prion titers (Fig. 2). With exposure to 10% HFIP, both Congo red binding and infectivity were abolished. Although Congo red binding could be restored by evaporation of the HFIP, prion infectivity was not. The dissociation between PrP amyloid and scrapie infectivity is of significance because some investigators have argued that PrPSc synthesis proceeds through the addition of PrPC molecules to PrP amyloid (Bessen et al., 1995; Caughey et al., 1995; Dealler, 1991; Gajdusek, 1987, 1994; Jarrett and Lansbury, 1993; Kocisko et al., 1994, 1995). Many lines of evidence argue against this hypothesis (summarized in Table V), including many of the data presented here. First, fractions containing PrPSc purified from the brains of scrapie-infected Syrian hamsters contain only amorphous structures, as judged by electron microscopy after negative staining (McKinley et al., 1991a; Pan et al., 1993). Second, only limited proteolysis of PrPSc in the presence of detergent produced prion rods (McKinley et al., 1991a). Third, dispersion of the rods into DLPCs enhanced prion infectivity ⬃10-fold in many instances (Gabizon, 1987). Fourth, Congo red dye binding did not correlate with scrapie infectivity (Figs. 2 and 4). Fifth, synthetic peptides of PrP can be induced to form amyloid polymers but lack infectivity (Forloni et al., 1993, 1996; Gasset et al., 1992; Inouye and Kirschner, 1997; Pillot et al., 1997). Protease resistance and prion infectivity. Early investigations showed a good correlation between resistance to digestion with proteinase K and scrapie prion infectivity (McKinley et al., 1983). When the ease and speed of the proteinase K diges-

334

WILLE, PRUSINER, AND COHEN

TABLE V Arguments That Certain Properties of PrP 27-30 and PrPSc Do Not Correlate with Scrapie Infectivity, while Others Do Correlate: A Summary Properties of PrP 27-30 and PrPSc that do not correlate with scrapie infectivity PrP polymers

Infectivity in preparations without observable polymers. (McKinley et al., 1991a; Pan et al., 1993) Target size of infectious subunit approx. PrP dimer. (Alper et al., 1966; Bellinger-Kawahara et al., 1988) Dispersion of prion rods into DLPCs increases infectivity. (Gabizon et al., 1987, 1988) Synthetic peptides of PrP and rPrP can polymerize into polymeric forms without acquiring infectivity. (Kaneko et al., 1997)

Amyloid properties HFIP abolishes Congo red dye binding at 2.5%, while infectivity remains unaltered. Evaporation of HFIP recovers Congo red dye binding, while infectivity does not. (This paper, Figs. 2 and 4) HFIP and TFIP both abolish Congo red dye binding, while only HFIP abolishes infectivity. (This paper, Figs. 2 and 4) Infectivity in PrPSc preparations without tinctorial properties of amyloid. (McKinley et al., 1991a; Pan et al., 1993) Only limited proteolysis of PrPSc produces PrP 27-30 amyloid. (McKinley et al., 1991a) Dispersion of prion rods into DLPCs increases infectivity and abolishes amyloid properties. (Gabizon et al., 1987, 1988) Synthetic peptides of PrP form amyloid without acquiring infectivity. (Gasset et al., 1992; Forloni et al., 1993, 1996; Inouye et al., 1997; Pillot et al., 1997)

tion assay is compared to the protracted duration of a bioassay, it is apparent why the proteinase K digestion assay is widely used to approximate infectivity. Intriguingly, proteinase K resistance seems neither necessary nor sufficient for prion infectivity, as shown in the following lines of evidence (summarized in Table V). First, transgenic mice expressing the human GSS mutation P101L transmit disease but fail to produce proteinase-K-resistant PrP (Hsiao et al., 1994; Telling et al., 1996a). Second, various strains of prions show widely differing levels of proteinase K resistance (Safar et al., 1998). Third, certain truncated PrP constructs expressed either in cell culture or in the respective transgenic mice show proteinase K resistance independent of infection by prions (Muramoto et al., 1996; Supattapone et al., 1999). Fourth, synthetic peptides of PrP and recombinantly expressed PrP can acquire proteinase K resistance through a variety of pathways but al-

Proteinase K resistance HFIP-induced loss of PK resistance and loss of infectivity occur at different HFIP concentrations. (This paper, Figs. 2 and 3B)

Properties that correlate with infectivity (␤-sheet content) HFIP-induced reduction of ␤sheet content (below a certain threshold) correlates with loss of infectivity. (This paper, Table II and Fig. 2)

APFO-treated PrP 27-30 rods lose PK resistance but remain infectious. (This paper, Table III and Fig. 7) Transgenic mice expressing the GSS P101L mutation produce infectious PrP that is PKsensitive. (Hsiao et al., 1994; Telling et al., 1996a) Different prion strains show different levels of PK resistance. (Safar et al., 1998)

Reduction in ␤-sheet content of PrPSc decreases infectivity. (Gasset et al., 1993; Safar et al., 1993b) Solubilization of PrP 27-30 by SDS reduces ␤-sheet content and abolishes infectivity. (Riesner et al., 1996)

Truncated PrP species (PrP 106) acquire PK resistance in vivo independently of infectivity. (Muramoto et al., 1996; Supattapone et al., 1999) Synthetic peptides of PrP and rPrP can acquire PK resistance without becoming infectious. (Kaneko et al., 1995, 1997; Hill et al., 1999)

ways lack prion infectivity (Hill et al., 1999; Kaneko et al., 1995, 1997). In the studies presented here, there was only limited correlation between proteinase K resistance and prion infectivity. While 2.5% HFIP altered neither infectivity nor proteinase K resistance, higher concentrations progressively reduced proteinase K resistance around 3% HFIP and infectivity above 5% HFIP (Figs. 2 and 3B). This observation by itself could be attributed to the different sensitivities of Western blot versus bioassay, but more convincing discrepancies can be observed in our solubilization experiments. PrP 27-30 treated by HFPIP, APFO, or a combination of HFPIP ⫹ AOT ⫹ APFO became proteinase K sensitive while only the conditions that include HFPIP resulted in a complete loss of infectivity (Table III and Fig. 7). APFO alone neither reduced prion infectivity nor solubilized PrP 27-30 in prion

PRION PROTEIN AMYLOID

rods, but it completely abolished proteinase K resistance. This comparison argues that proteinase K resistance is not required for scrapie infectivity. Oligomeric PrP and prion infectivity. A successful solubilization of PrP 27-30 into a monomeric or a dimeric form without loss of infectivity would settle the question of the minimal size for an infectious prion. An early study utilizing ionizing radiation and UV light identified a target size of about 150 kDa for the scrapie agent, which at the time was only poorly characterized (Alper et al., 1966). A more recent report improved on the accuracy of the measurements and yielded a target size of about 55 kDa (range 42 to 71 kDa), with the implication that the infectious particles could be composed of PrPSc dimers (Bellinger-Kawahara et al., 1988). So far, all attempts to solubilize and isolate monomeric or dimeric prion protein particles that carry infectivity have been without success, either because the solubilized protein lost infectivity or there was only partial solubilization with large aggregates remaining (see e.g., Akowitz et al., 1990; Malone et al., 1979; Riesner et al., 1996; Sklaviadis et al., 1992, 1989). Our current study generated monomeric and dimeric forms of PrP 27-30 with sedimentation coefficients of 2 S and 4 S, respectively, that lacked infectivity when injected into hamsters. The seeded nucleation hypothesis for prion replication actually uses the absence of solubilized prion infectivity as a key argument that ordered aggregates of PrPSc or PrP 27-30 constitute the infectious particle (Dealler, 1991; Gajdusek, 1988; Jarrett and Lansbury, 1993). The fact that PrPSc preparations show no trace of ordered aggregates within the resolution of negative stain electron microscopy sets narrow limits for the size of such ordered oligomers, should they exist (Caughey and Chesebro, 1997; McKinley et al., 1991a; Pan et al., 1993). Analyses of the kinetics of prion replication conclude that the size of the postulated seed would have to be small (Eigen, 1996) and that differences between the seeded nucleation hypothesis and the template assisted model may be difficult to prove (compare Eigen, 1996; Harper and Lansbury, 1997; Harrison et al., 1997; Safar, 1996). The kinetics of the incubation time associated with prion infection in mice homozygous and hemizygous for the Prnp gene, together with other studies in transgenic animals, argue in favor of the template assistance mechanism (Cohen and Prusiner, 1998). Another parameter that contributes significantly to the issue of the minimal infectious particle is the clearance of molecules and small particles from the brain. Studies on radiolabeled scrapie liposomes or 0.5-␮m polystyrene beads show a 80% clearance

335

within 30 min after intracerebral injection (Mann et al., 1979; Millson et al., 1979). Aggregated infectious prions as found in brain homogenates from diseased laboratory animals or patients were found to be cleared to over 95% between 2 and 10 days postinoculation (Czub et al., 1988; Jendroska et al., 1991; Kimberlin et al., 1986; Manuelidis and Fritch, 1996). It is conceivable that the monomeric and dimeric PrP 27-30 particles produced by our solubilization procedure were cleared from the brains of injected hamsters before an infection could be established. The lack of significant changes in secondary structure after the solubilization, as determined by FTIR spectroscopy (Table IV), would support the notion that these particles may principally still carry infectivity. Experiments are underway to investigate if circumventing rapid clearance by increasing particle size will reestablish prion infectivity in these preparations. Protein oligomers and yeast prions. Growing evidence points to the possibility that the two wellstudied yeast prions [PSI⫹] and [URE3] propagate via the seeded nucleation mechanism. The N-terminal domain of recombinant Sup35p, the protein substrate for the [PSI⫹] phenotype, forms protein filaments that exhibit all the characteristics of amyloid (Glover et al., 1997; King et al., 1997). In fact, fulllength Sup35p from yeast extracts converts into protease-resistant forms when mixed with [PSI⫹] yeast extracts (Paushkin et al., 1997). Support for this assumption might come from transmitting the [PSI⫹] phenotype in vitro through aggregated recombinant Sup35 protein. Similarly, the N-terminal domain of Ure2p, the protein substrate for [URE3], forms amyloid fibers in vitro that may cause the respective prion phenotype by aggregation into polymers (Schlumpberger et al., 2000; Taylor et al., 1999; Thual et al., 1999). The extent to which changes in secondary structure are prerequisites for this polymerization needs to be elucidated for both systems. The distinction between nucleation and template mechanisms depends on the precise nature of the rate-limiting step in replication. To the extent that the conformational change from a monomeric stable non-␤ form into a ␤-rich form that prefers to aggregate is difficult and protein concentration is sufficient, a template mechanism will apply. This appears to be the case for prion replication (Baskakov et al., 2000; Cohen and Prusiner, 1998; Prusiner et al., 1998). By contrast, smaller proteins or protein fragments that exhibit little conformational preference as monomers will form aggregates in a fashion that largely depends upon protein concentration. Peptide models of A␤(1-42) fibril formation follow these guidelines and demonstrate kinetics that fit

336

WILLE, PRUSINER, AND COHEN

the nucleation–polymerization formalism (Jarrett and Lansbury, 1993). We thank Hana Serban and the prep team for the preparation of PrP 27-30, Dr. Marilyn Torchia for help with the bioassays, Guo-Feng Zhang and Stephen D’Angelo for their help with the electron microscopy, Mei-Lie Wong for making available the darkroom facilities, Dr. Martin Schlumpberger for his advice on yeast prions, and Dr. Ronald Wetzel from SmithKline Beecham Pharmaceuticals for his advice on the Congo red binding assay. This work was supported by grants from the National Institutes of Health, the International Human Frontiers of Science Program, and the American Health Assistance Foundation, as well as by a gift from the G. Harold and Leila Y. Mathers Foundation. REFERENCES Akowitz, A., Sklaviadis, T., Manuelidis, E. E., and Manuelidis, L. (1990) Nuclease-resistant polyadenylated RNAs of significant size are detected by PCR in highly purified Creutzfeldt-Jakob disease preparations, Microb. Pathog. 9, 33– 45. Alper, T., Haig, D. A., and Clarke, M. C. (1966) The exceptionally small size of the scrapie agent, Biochem. Biophys. Res. Commun. 22, 278 –284. Baskakov, I. V., Aagaard, C., Mehlhorn, I., Wille, H., Groth, D., Baldwin, M. A., Prusiner, S. B., and Cohen, F. E. (2000) Selfassembly of recombinant prion protein of 106 residues, Biochemistry 39, 2792–2804. Bellinger-Kawahara, C. G., Kempner, E., Groth, D. F., Gabizon, R., and Prusiner, S. B. (1988) Scrapie prion liposomes and rods exhibit target sizes of 55,000 Da, Virology 164, 537–541. Bello, J. (1993) Helix promotion in polypeptides by polyols, Biopolymers 33, 491– 495. Bessen, R. A., Kocisko, D. A., Raymond, G. J., Nandan, S., Lansbury, P. T., and Caughey, B. (1995) Non-genetic propagation of strain-specific properties of scrapie prion protein, Nature 375, 698 –700. Bessen, R. A., and Marsh, R. F. (1994) Distinct PrP properties suggest the molecular basis of strain variation in transmissible mink encephalopathy, J. Virol. 68, 7859 –7868. Byler, D. M., and Susi, H. (1986) Examination of the secondary structure of proteins by deconvolved FTIR spectra, Biopolymers 25, 469 – 487. Caughey, B., and Chesebro, B. (1997) Prion protein and the transmissible spongiform encephalopathies, Trends Cell Biol. 7, 56 – 62. Caughey, B., Kocisko, D. A., Raymond, G. J., and Lansbury, P. T., Jr. (1995) Aggregates of scrapie-associated prion protein induce the cell-free conversion of protease-sensitive prion protein to the protease-resistant state, Chem. Biol. 2, 807– 817. Caughey, B., Raymond, G. J., Kocisko, D. A., and Lansbury, P. T., Jr. (1997) Scrapie infectivity correlates with converting activity, protease resistance, and aggregation of scrapie-associated prion protein in guanidine denaturation studies, J. Virol. 71, 4107– 4110. Caughey, B. W., Dong, A., Bhat, K. S., Ernst, D., Hayes, S. F., and Caughey, W. S. (1991) Secondary structure analysis of the scrapie-associated protein PrP 27-30 in water by infrared spectroscopy, Biochemistry 30, 7672–7680. Cioci, F., and Lavecchia, R. (1994) Effect of polyols and sugars on heat-induced flavin dissociation in glucose oxidase, Biochem. Mol. Biol. Int. 34, 705–712. Cohen, A. S., Shirahama, T., and Skinner, M. (1982) Electron microscopy of amyloid, in Harris, J. R. (Ed.), Electron Micros-

copy of Proteins, Vol. 3, pp. 165–206, Academic Press, New York. Cohen, F. E., and Prusiner, S. B. (1998) Pathologic conformations of prion proteins, Annu. Rev. Biochem. 67, 793– 819. Czub, M., Braig, H. R., and Diringer, H. (1988) Replication of the scrapie agent in hamsters infected intracerebrally confirms the pathogenesis of an amyloid-inducing virosis, J. Gen. Virol. 69, 1753–1756. Dealler, S. (1991) Transmissable spongiform encephalopathy (TSE) agents as crystalline forms of the prion protein (PrP) that multiply by allowing normal metabolic forms of PrP to join the crystal, Med. Hypoth. 36, 131–134. Durchschlag, H. (1986) Specific volumes of biological macromolecules and some other molecules of biological interest, in Hinz, H.-J. (Ed.), Thermodynamic Data for Biochemistry and Biotechnology, pp. 45–128, Springer-Verlag, Berlin. Eigen, M. (1996) Prionics or the kinetic basis of prion diseases, Biophys. Chem. 63, A11–A18. Endo, T., Groth, D., Prusiner, S. B., and Kobata, A. (1989) Diversity of oligosaccharide structures linked to asparagines of the scrapie prion protein, Biochemistry 28, 8380 – 8388. Forloni, G., Angeretti, N., Chiesa, R., Monzani, E., Salmona, M., Bugiani, O., and Tagliavini, F. (1993) Neurotoxicity of a prion protein fragment, Nature 362, 543–546. Forloni, G., Tagliavini, F., Bugiani, O., and Salmona, M. (1996) Amyloid in Alzheimer’s disease and prion-related encephalopathies: Studies with synthetic peptides, Prog. Neurobiol. 49, 287–315. Gabizon, R., McKinley, M. P., and Prusiner, S. B. (1987) Purified prion proteins and scrapie infectivity copartition into liposomes, Proc. Natl. Acad. Sci. USA 84, 4017– 4021. Gabizon, R., McKinley, M. P., and Prusiner, S. B. (1988) Properties of scrapie prion proteins in liposomes and amyloid rods, in Bock, G., and Marsh, J. (Eds.), Novel Infectious Agents and the Central Nervous System, Ciba Foundation Symposium 135, pp. 182–196, Wiley, Chichester, UK. Gajdusek, D. C. (1988) Transmissible and non-transmissible amyloidoses: autocatalytic post-translational conversion of host precursor proteins to ␤-pleated sheet configurations, J. Neuroimmunol. 20, 95–110. Gajdusek, D. C. (1994) Spontaneous generation of infectious nucleating amyloids in the transmissible and nontransmissible cerebral amyloidoses, Mol. Neurobiol. 8, 1–13. Gajdusek, D. C. (1987) The transmissible dementias and other brain disorders caused by unconventional viruses-Relationship of transmissible to nontransmissible amyloidosis of the brain, in Kurstak, E., Lipowski, Z. J., and Morozov, P. V. (Eds.), Viruses, Immunity and Mental Disorders, pp. 3–22, Plenum, New York. Gasset, M., Baldwin, M. A., Fletterick, R. J., and Prusiner, S. B. (1993) Perturbation of the secondary structure of the scrapie prion protein under conditions that alter infectivity, Proc. Natl. Acad. Sci. USA 90, 1–5. Gasset, M., Baldwin, M. A., Lloyd, D., Gabriel, J.-M., Holtzman, D. M., Cohen, F., Fletterick, R., and Prusiner, S. B. (1992) Predicted ␣-helical regions of the prion protein when synthesized as peptides form amyloid, Proc. Natl. Acad. Sci. USA 89, 10940 –10944. Glenner, G. G., Eanes, E. D., and Page, D. L. (1972) The relation of the properties of Congo red-stained amyloid fibrils to the beta-conformation, J. Histochem. Cytochem. 20, 821– 826. Glover, J. R., Kowal, A. S., Schirmer, E. C., Patino, M. M., Liu, J.-J., and Lindquist, S. (1997) Self-seeded fibers formed by

PRION PROTEIN AMYLOID Sup35, the protein determinant of [PSI⫹], a heritable prion-like factor of S. cerevisiae, Cell 89, 811– 819. Go¨klen, K. E., and Hatton, T. A. (1987) Liquid-liquid extraction of low molecular-weight proteins by selective solubilization in reversed micelles, Sep. Sci. Technol. 22, 831– 841. Harper, J. D., and Lansbury, P. T., Jr. (1997) Models of amyloid seeding in Alzheimer’s disease and scrapie: Mechanistic truths and physiological consequences of the time-dependent solubility of amyloid proteins, Annu. Rev. Biochem. 66, 385– 407. Harrison, P. M., Bamborough, P., Daggett, V., Prusiner, S. B., and Cohen, F. E. (1997) The prion folding problem, Curr. Opin. Struct. Biol. 7, 53–59. Hill, A. F., Antoniou, M., and Collinge, J. (1999) Protease-resistant prion protein produced in vitro lacks detectable infectivity, J. Gen. Virol. 80, 11–14. Hsiao, K. K., Groth, D., Scott, M., Yang, S.-L., Serban, H., Rapp, D., Foster, D., Torchia, M., DeArmond, S. J., and Prusiner, S. B. (1994) Serial transmission in rodents of neurodegeneration from transgenic mice expressing mutant prion protein, Proc. Natl. Acad. Sci. USA 91, 9126 –9130. Hunter, G. D., Kimberlin, R. H., and Gibbons, R. A. (1968) Scrapie: A modified membrane hypothesis, J. Theor. Biol. 20, 355–357. Hunter, G. D., and Millson, G. C. (1967) Attempts to release the scrapie agent from tissue debris, J. Comp. Pathol. 77, 301–307. Inouye, H., and Kirschner, D. A. (1997) X-ray diffraction analysis of scrapie prion: Intermediate and folded structures in a peptide containing two putative ␣-helices, J. Mol. Biol. 268, 375– 389. Jackson, M., and Mantsch, H. H. (1992) Halogenated alcohols as solvents for proteins: FTIR spectroscopic studies, Biochim. Biophys. Acta 1118, 139 –143. Jarrett, J. T., and Lansbury, P. T., Jr. (1993) Seeding “onedimensional crystallization” of amyloid: A pathogenic mechanism in Alzheimer’s disease and scrapie? Cell 73, 1055–1058. Jendroska, K., Heinzel, F. P., Torchia, M., Stowring, L., Kretzschmar, H. A., Kon, A., Stern, A., Prusiner, S. B., and DeArmond, S. J. (1991) Proteinase-resistant prion protein accumulation in Syrian hamster brain correlates with regional pathology and scrapie infectivity, Neurology 41, 1482–1490. Kaneko, K., Peretz, D., Pan, K.-M., Blochberger, T., Wille, H., Gabizon, R., Griffith, O. H., Cohen, F. E., Baldwin, M. A., and Prusiner, S. B. (1995) Prion protein (PrP) synthetic peptides induce cellular PrP to acquire properties of the scrapie isoform, Proc. Natl. Acad. Sci. USA 32, 11160 –11164. Kaneko, K., Wille, H., Mehlhorn, I., Zhang, H., Ball, H., Cohen, F. E., Baldwin, M. A., and Prusiner, S. B. (1997) Molecular properties of complexes formed between the prion protein and synthetic peptides, J. Mol. Biol. 270, 574 –586. Kimberlin, R. H., Millson, G. C., and Hunter, G. D. (1971) An experimental examination of the scrapie agent in cell membrane mixtures. III. Studies of the operational size, J. Comp. Pathol. 81, 383–391. Kimberlin, R. H., and Walker, C. A. (1986) Pathogenesis of scrapie (strain 263K) in hamsters infected intracerebrally, intraperitoneally or intraocularly, J. Gen. Virol. 67, 255–263. King, C.-Y., Tittman, P., Gross, H., Gebert, R., Aebi, M., and Wu¨thrich, K. (1997) Prion-inducing domain 2-114 of yeast Sup35 protein transforms in vitro into amyloid-like filaments, Proc. Natl. Acad. Sci. USA 94, 6618 – 6622. Klunk, W. E., Pettegrew, J. W., and Abraham, D. J. (1989) Two simple methods for quantifying low-affinity dye-substrate binding, J. Histochem. Cytochem. 37, 1293–1297.

337

Kocisko, D. A., Come, J. H., Priola, S. A., Chesebro, B., Raymond, G. J., Lansbury, P. T., Jr., and Caughey, B. (1994) Cell-free formation of protease-resistant prion protein, Nature 370, 471– 474. Kocisko, D. A., Priola, S. A., Raymond, G. J., Chesebro, B., Lansbury, P. T., Jr., and Caughey, B. (1995) Species specificity in the cell-free conversion of prion protein to protease-resistant forms: A model for the scrapie species barrier, Proc. Natl. Acad. Sci. USA 92, 3923–3927. Lee, J. C., and Timasheff, S. N. (1981) The stabilization of proteins by sucrose, J. Biol. Chem. 256, 7193–7201. Malone, T. G., Marsh, R. F., Hanson, R. P., and Semancik, J. S. (1978) Membrane-free scrapie activity, J. Virol. 25, 933–935. Malone, T. G., Marsh, R. F., Hanson, R. P., and Semancik, J. S. (1979) Evidence for the low molecular weight nature of the scrapie agent, Nature 278, 575–576. Mann, J. D., Butler, A. B., Johnson, R. N., and Bass, N. H. (1979) Clearance of macromolecular and particulate substances from the cerebrospinal fluid system of the rat, J. Neurosurg. 50, 343–348. Manuelidis, L., and Fritch, W. (1996) Infectivity and host responses in Creutzfeldt-Jakob disease, Virology 216, 46 –59. Marsh, R. F., Dees, C., Castle, B. E., Wade, W. F., and German, T. L. (1984) Purification of the scrapie agent by density gradient centrifugation, J. Gen. Virol. 65, 415– 421. McKinley, M. P., Bolton, D. C., and Prusiner, S. B. (1983) A protease-resistant protein is a structural component of the scrapie prion, Cell 35, 57– 62. McKinley, M. P., Braunfeld, M. B., and Prusiner, S. B. (1987) Scrapie prion ultrastructure, in Prusiner, S. B., and McKinley, M. P. (Eds.), Prions—Novel Infectious Pathogens Causing Scrapie and Creutzfeldt-Jakob Disease, pp. 197–237, Academic Press, Orlando. McKinley, M. P., Meyer, R. K., Kenaga, L., Rahbar, F., Cotter, R., Serban, A., and Prusiner, S. B. (1991a) Scrapie prion rod formation in vitro requires both detergent extraction and limited proteolysis, J. Virol. 65, 1340 –1351. McKinley, M. P., Taraboulos, A., Kenaga, L., Serban, D., Stieber, A., DeArmond, S. J., Prusiner, S. B., and Gonatas, N. (1991b) Ultrastructural localization of scrapie prion proteins in cytoplasmic vesicles of infected cultured cells, Lab. Invest. 65, 622– 630. Millson, G. C., Hunter, G. D., and Kimberlin, R. H. (1976) The physico-chemical nature of the scrapie agent, in Kimberlin, R. H. (Ed.), Slow Virus Diseases of Animals and Man, pp. 243–266, American Elsevier, New York. Millson, G. C., Kimberlin, R. H., Manning, E. J., and Collis, S. C. (1979) Early distribution of radioactive liposomes and scrapie infectivity in mouse tissues following administration by different routes, Vet. Microbiol. 4, 89 –99. Muramoto, T., Scott, M., Cohen, F. E., and Prusiner, S. B. (1996) Recombinant scrapie-like prion protein of 106 amino acids is soluble, Proc. Natl. Acad. Sci. USA 93, 15457–15462. Nguyen, J. T., Inouye, H., Baldwin, M. A., Fletterick, R. J., Cohen, F. E., Prusiner, S. B., and Kirschner, D. A. (1995) X-ray diffraction of scrapie prion rods and PrP peptides, J. Mol. Biol. 252, 412– 422. Pan, K.-M., Baldwin, M., Nguyen, J., Gasset, M., Serban, A., Groth, D., Mehlhorn, I., Huang, Z., Fletterick, R. J., Cohen, F. E., and Prusiner, S. B. (1993) Conversion of ␣-helices into ␤-sheets features in the formation of the scrapie prion proteins, Proc. Natl. Acad. Sci. USA 90, 10962–10966. Paushkin, S. V., Kushnirov, V. V., Smirnov, V. N., and TerAvanesyan, M. D. (1997) In vitro propagation of the prion-like state of yeast Sup35 protein, Science 277, 381–383.

338

WILLE, PRUSINER, AND COHEN

Pillot, T., Lins, L., Goethals, M., Vanloo, B., Baert, J., Vandekerckhove, J., Rosseneu, M., and Brasseur, R. (1997) The 118-135 peptide of the human prion protein forms amyloid fibrils and induces liposome fusion, J. Mol. Biol. 274, 381–393. Prusiner, S. B. (1998) Prions, Proc. Natl. Acad. Sci. USA 95, 13363–13383. Prusiner, S. B., Groth, D. F., Cochran, S. P., Masiarz, F. R., McKinley, M. P., and Martinez, H. M. (1980) Molecular properties, partial purification, and assay by incubation period measurements of the hamster scrapie agent, Biochemistry 21, 4883– 4891. Prusiner, S. B., Hadlow, W. J., Garfin, D. E., Cochran, S. P., Baringer, J. R., Race, R. E., and Eklund, C. M. (1978) Partial purification and evidence for multiple molecular forms of the scrapie agent, Biochemistry 17, 4993– 4997. Prusiner, S. B., McKinley, M. P., Bowman, K. A., Bolton, D. C., Bendheim, P. E., Groth, D. F., and Glenner, G. G. (1983) Scrapie prions aggregate to form amyloid-like birefringent rods, Cell 35, 349 –358. Prusiner, S. B., Scott, M., Foster, D., Pan, K.-M., Groth, D., Mirenda, C., Torchia, M., Yang, S.-L., Serban, D., Carlson, G. A., Hoppe, P. C., Westaway, D., and DeArmond, S. J. (1990) Transgenetic studies implicate interactions between homologous PrP isoforms in scrapie prion replication, Cell 63, 673– 686. Prusiner, S. B., Scott, M. R., DeArmond, S. J., and Cohen, F. E. (1998) Prion protein biology, Cell 93, 337–348. Riesner, D., Kellings, K., Post, K., Wille, H., Serban, H., Groth, D., Baldwin, M. A., and Prusiner, S. B. (1996) Disruption of prion rods generates 10-nm spherical particles having high ␣-helical content and lacking scrapie infectivity, J. Virol. 70, 1714 –1722. Safar, J. (1996) The folding intermediate concept of prion protein formation and conformational links to infectivity, in Prusiner, S. B. (Ed.), Prions Prions Prions, pp. 69 –76, Springer-Verlag, Heidelberg. Safar, J., Ceroni, M., Gajdusek, D. C., and Gibbs, C. J., Jr. (1991) Differences in the membrane interaction of scrapie amyloid precursor proteins in normal and scrapie- or Creutzfeldt-Jakob disease-infected brains, J. Infect. Dis. 163, 488 – 494. Safar, J., Roller, P. P., Gajdusek, D. C., and Gibbs, C. J., Jr. (1993a) Conformational transitions, dissociation, and unfolding of scrapie amyloid (prion) protein, J. Biol. Chem. 268, 20276 – 20284. Safar, J., Roller, P. P., Gajdusek, D. C., and Gibbs, C. J. J. (1993b) Thermal-stability and conformational transitions of scrapie amyloid (prion) protein correlate with infectivity, Protein Sci. 2, 2206 –2216. Safar, J., Wille, H., Itri, V., Groth, D., Serban, H., Torchia, M., Cohen, F. E., and Prusiner, S. B. (1998) Eight prion strains have PrPSc molecules with different conformations, Nat. Med. 4, 1157–1165. Schlumpberger, M., Wille, H., Baldwin, M. A., Butler, D. A., Herskowitz, I., and Prusiner, S. B. (2000) The prion domain of yeast Ure2p induces autocatalytic formation of amyloid fibers by a recombinant fusion protein, Protein Sci. 9, 440 – 451. Serban, D., Taraboulos, A., DeArmond, S. J., and Prusiner, S. B. (1990) Rapid detection of Creutzfeldt-Jakob disease and scrapie prion proteins, Neurology 40, 110 –117. Sklaviadis, T., Dreyer, R., and Manuelidis, L. (1992) Analysis of Creutzfeldt-Jakob disease infectious fractions by gel permeation chromatography and sedimentation field flow fractionation, Virus Res. 26, 241–254.

Sklaviadis, T. K., Manuelidis, L., and Manuelidis, E. E. (1989) Physical properties of the Creutzfeldt-Jakob disease agent, J. Virol. 63, 1212–1222. Stahl, N., Baldwin, M. A., Hecker, R., Pan, K.-M., Burlingame, A. L., and Prusiner, S. B. (1992) Glycosylinositol phospholipid anchors of the scrapie and cellular prion proteins contain sialic acid, Biochemistry 31, 5043–5053. Steensgaard, J., Humphries, S., and Spragg, S. P. (1992) Measurements of sedimentation coefficients, in Rickwood, D. (Ed.), Preparative Centrifugation: A Practical Approach, pp. 187– 232, IRL Press, Oxford. Steensgaard, J., Møller, N. P. H., and Funding, L. (1978) Ratezonal centrifugation: quantitative aspects, in Birnie, G. D., and Rickwood, D. (Eds.), Centrifugal Separations in Molecular and Cell Biology, pp. 115–163, Butterworths, London. Sunde, M., and Blake, C. C. F. (1998) From the globular to the fibrous state: Protein structure and structural conversion in amyloid formation, Q. Rev. Biophys. 31, 1–39. Supattapone, S., Bosque, P., Muramoto, T., Wille, H., Aagaard, C., Peretz, D., Nguyen, H.-O. B., Heinrich, C., Torchia, M., Safar, J., Cohen, F. E., DeArmond, S. J., Prusiner, S. B., and Scott, M. (1999) Prion protein of 106 residues creates an artificial transmission barrier for prion replication in transgenic mice, Cell 96, 869 – 878. Taylor, K. L., Cheng, N., Williams, R. W., Steven, A. C., and Wickner, R. B. (1999) Prion domain initiation of amyloid formation in vitro from native Ure2p, Science 283, 1339 –1343. Telling, G. C., Haga, T., Torchia, M., Tremblay, P., DeArmond, S. J., and Prusiner, S. B. (1996a) Interactions between wildtype and mutant prion proteins modulate neurodegeneration in transgenic mice, Genes Dev. 10, 1736 –1750. Telling, G. C., Parchi, P., DeArmond, S. J., Cortelli, P., Montagna, P., Gabizon, R., Mastrianni, J., Lugaresi, E., Gambetti, P., and Prusiner, S. B. (1996b) Evidence for the conformation of the pathologic isoform of the prion protein enciphering and propagating prion diversity, Science 274, 2079 –2082. Thual, G., Komar, A. A., Bousset, L., Fernandez-Bellot, E., Cullin, C., and Melki, R. (1999) Structural characterization of Saccharomyces cerevisiae prion-like protein Ure2, J. Biol. Chem. 274, 13666 –13674. Wille, H., Baldwin, M. A., Cohen, F. E., DeArmond, S. J., and Prusiner, S. B. (1996a) Prion protein amyloid: separation of scrapie infectivity from PrP polymers, in CIBA Foundation Symposium No. 199: The Nature and Origins of Amyloid Fibrils, pp. 181–201, Wiley, Chichester, England. Wille, H., and Prusiner, S. B. (1999) Ultrastructural studies on scrapie prion protein crystals obtained from reverse micellar solutions, Biophys. J. 76, 1048 –1062. Wille, H., Zhang, G.-F., Baldwin, M. A., Cohen, F. E., and Prusiner, S. B. (1996b) Separation of scrapie prion infectivity from PrP amyloid polymers, J. Mol. Biol. 259, 608 – 621. Wolbert, R. B. G., Hilhorst, R., Voskuilen, G., Nachtegaal, H., Dekker, M., Van’t Riet, K., and Bijsterbosch, B. H. (1989) Protein transfer from an aqueous phase into reversed micelles. The effect of protein size and charge distribution, Eur. J. Biochem. 184, 627– 633. Yang, Y., Barker, S., Chen, M. J., and Mayo, K. H. (1993) Effect of low molecular weight aliphatic alcohols and related compounds on platelet factor 4 subunit association, J. Biol. Chem. 268, 9223–9229.