NR AKHC

AU Saborio,G.P.; Permanne,B.; Soto,C.

TI Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding

QU Nature 2001 Jun 14; 411(6839): 810-3

PT evaluation studies; journal article

AB Prions are the infectious agents responsible for transmissible spongiform encephalopathies. The principal component of prions is the glycoprotein PrPsc, which is a conformationally modified isoform of a normal cell-surface protein called PrPc (ref. 1). During the time between infection and the appearance of the clinical symptoms, minute amounts of PrPsc replicate by conversion of host PrPc, generating large amounts of PrPsc aggregates in the brains of diseased individuals. We aimed to reproduce this event in vitro. Here we report a procedure involving cyclic amplification of protein misfolding that allows a rapid conversion of large excess PrPc into a protease-resistant, PrPsc-like form in the presence of minute quantities of PrPsc template. In this procedure, conceptually analogous to polymerase chain reaction cycling, aggregates formed when PrPsc is incubated with PrPc are disrupted by sonication to generate multiple smaller units for the continued formation of new PrPsc. After cyclic amplification more than 97% of the protease-resistant PrP present in the sample corresponds to newly converted protein. The method could be applied to diagnose the presence of currently undetectable prion infectious agent in tissues and biological fluids, and may provide a unique opportunity to determine whether PrPsc replication results in the generation of infectivity in vitro.

VT Prions are the infectious agents responsible for transmissible spongiform encephalopathies. The principal component of prions is the glycoprotein PrPsc, which is a conformationally modified isoform of a normal cell-surface protein called PrPc (ref. [1]). During the time between infection and the appearance of the clinical symptoms, minute amounts of PrPsc replicate by conversion of host PrPc, generating large amounts of PrPsc aggregates in the brains of diseased individuals. We aimed to reproduce this event in vitro. Here we report a procedure involving cyclic amplification of protein misfolding that allows a rapid conversion of large excess PrPc into a protease-resistant, PrPsc-like form in the presence of minute quantities of PrPsc template. In this procedure, conceptually analogous to polymerase chain reaction cycling, aggregates formed when PrPsc is incubated with PrPc are disrupted by sonication to generate multiple smaller units for the continued formation of new PrPsc. After cyclic amplification more than 97% of the protease-resistant PrP present in the sample corresponds to newly converted protein. The method could be applied to diagnose the presence of currently undetectable prion infectious agent in tissues and biological fluids, and may provide a unique opportunity to determine whether PrPsc replication results in the generation of infectivity in vitro.
To evaluate the cyclic amplification procedure (Fig. 1), we diluted brain homogenate from scrapie-affected hamsters until the signal of PrPsc was barely detectable by immunoblot after digestion with proteinase K (Fig. 2a, lane 1). The same dilution of scrapie brain homogenate was incubated with brain homogenate from healthy hamsters as a source of PrPc. After incubation, an increase in the signal of a protease-resistant PrPsc-like protein (PrPres) with relative molecular mass 27,000-30,000 (Mr 27-30K) could be detected (Fig. 2a, lane 2). Using the same conditions, but subjecting the mixture to five cycles of incubation-sonication, the amount of PrPres dramatically increased (Fig. 2a, lane 3). This process of in vitro replication of protein conformation is called protein-misfolding cyclic amplification (PMCA). To estimate the rate of amplification, we performed densitometric analysis of immunoblots from five independent experiments done under the same conditions. The average amplification rate was 57.9 +/- 19.9, indicating that after five PMCA cycles the newly converted protein corresponds to 97.4-98.7% of the total amount of PrPres. This proportion can be further increased by subjecting the sample to a larger number of amplification cycles (see below). To rule out artefacts of protein blotting, rat brain homogenate was added to the control sample of scrapie hamster brain homogenates to achieve equal protein concentrations (Fig. 2b). Only the hamster-derived protein was measured as the antibody used to reveal PrPres in the immunoblot does not react with rat prion protein. The conversion was dependent on the presence of PrPsc, as no PrPres formation was observed when the normal hamster brain homogenate was incubated alone, under the same conditions, with or without sonication (Fig. 2c, lanes 2 and 3). In addition, the increase in PrPres signal was not due to higher reactivity of initial PrPsc as a result of the treatment, because incubation-sonication of the material in the absence of a source of PrPc resulted in no change in the signal (Fig. 2c, lanes 5 and 7). We noted that in the control experiment in which brain homogenates containing PrPc and PrPsc were incubated, but not subjected to PMCA (Fig. 2a, lane 2), PrPres formation was greater than that reported previously using purified proteins [2,3] or cell lysates [4]. The higher efficiency of conversion might be due to the presence of additional factors present in the brain homogenate that catalyse the conversion [4,5]. The amplification system could be useful to isolate such additional factors by providing an assay to monitor their activity.
----------------------------------------------------------------
Figure 1 Diagrammatic representation of the PMCA procedure.
Amplification is based on multiple cycles of PrPsc incubation in the presence of excess PrPc followed by sonication. During the incubation periods, the size of oligomeric PrPsc is increased by incorporation of PrPc into the growing aggregate, while during sonication the aggregates are disrupted, producing an expanded population of converting units.
----------------------------------------------------------------
Figure 2 Amplification of PrPsc by sonication cycles.
Scrapie hamster brain homogenate was diluted 500-fold into 10% healthy rat brain homogenate (lane 1 in a and b, control experiment) or into 10% healthy hamster brain homogenate (lanes 2 and 3 in a and b). Samples were incubated during 5 h without (lanes 1 and 2 in a and b) or with (lane 3 in a and b) five PMCA cycles. All samples were treated with proteinase K (PK). Samples were immunoblotted with the 3F4 anti-PrP antibody (a) or loaded onto a gel and stained for total protein with Coomassie blue (b). c, Control experiments with brain homogenate of healthy hamsters containing PrPc (lanes 1-3) or brain homogenate of scrapie-affected hamsters containing PrPsc (lanes 4-7), were diluted 100-fold in healthy hamster brain homogenate (H), healthy rat brain homogenate (R) or PBS buffer (B), as indicated. Samples in lanes 1, 2, 4 and 6 were incubated for 5 h at 37 °; samples in lanes 3, 5 and 7 were subjected to five PMCA cycles.
----------------------------------------------------------------
To evaluate the minimum PrPsc concentration needed to trigger amplification and enhanced detection of abnormal protein, we serially diluted brain homogenate from hamsters with scrapie directly into brain homogenate from healthy hamsters and assayed for PrPres signal with or without incubation-sonication cycles. Without PMCA, the PrPres signal diminished progressively until it was not longer detectable at ~ 640-fold dilution (Fig. 3a, c). By contrast, when parallel samples were subjected to 10 PMCA cycles, PrPres was readily detected at this dilution (Fig. 3b, c). Indeed, PrPres was detected even after >10,000-fold dilution under these conditions. On the basis of our estimation of PrPsc detection by immunoblot of known amounts of recombinant hamster PrP, the minimum amount of PrPsc detectable under these conditions is ~6-12 pg or 0.2-0.4 x 10^-15 mol. The amount of newly converted PrPres is ~250 pg or 8.3 x 10^-15 mol. This detection limit was calculated using low-sensitivity immunoblot assays and therefore it might be further decreased, to reach a similar or even better sensitivity than the biological assay of infectivity, by using detection systems with higher sensitivity and/or by performing a larger number of amplification cycles (see below).
----------------------------------------------------------------
Figure 3 Sensitivity of the PMCA system.
a, Scrapie hamster brain homogenate diluted serially into 5% rat brain homogenate (control). b, Two-fold serial dilutions of the scrapie brain homogenate into a 5% healthy hamster brain homogenate and subjected to 10 PMCA cycles. All samples in a and b were treated with proteinase K. The dilutions of scrapie brain homogenate in lanes 1-10 of a and b were 20, 40, 80, 160, 320, 640, 1.280, 2.560, 5.120 and 10.240, respectively. c, Densitometric quantification of five independent experiments to evaluate the sensitivity threshold of the cyclic amplification under these conditions. To normalize differences between experiments, the band intensity at each dilution is expressed as a percentage of the PrPres band intensity obtained at a 100-fold dilution of the scrapie brain homogenate without any treatment.
----------------------------------------------------------------
To determine the effect of different numbers of PMCA cycles on PrPres formation, we diluted scrapie brain homogenate 1,000-fold and incubated it with an excess of healthy hamster brain homogenate. Samples were treated with 0, 5, 10, 20 or 40 cycles and the resultant PrPres signal determined by immunoblot. As anticipated, the levels of PrPres increased with the number of cycles (Fig. 4a). Control samples containing identical mixtures were incubated for the same time, but without sonication. No increase in PrPres signal strength was detected for these samples (Fig. 4b). The observed increase in PrPres per cycle fitted an exponential curve (r^2 = 0.973, Fig. 4c), limited by an underestimation of the amplified material at higher cycle numbers. Samples subjected to more than 20 cycles formed strong aggregates that could not be converted to monomeric protein during electrophoresis and hence were excluded from the quantification.
----------------------------------------------------------------
Figure 4 Relationship between the extent of the conversion and the number of amplification cycles. A 1,000-fold dilution of hamster scrapie brain homogenate was made into 5% healthy hamster brain homogenate. Aliquots were subjected to 0, 5, 10, 20 or 40 PMCA cycles (a) or incubated for the same times, but without sonication (b). All samples were treated with proteinase K and PrPres signal was evaluated by immunoblot. To appreciate more clearly the differences in the band intensity, one tenth of the samples were loaded in the gel. c, Densitometric quantification of the blots in a (PMCA, solid line) and b (control, dotted line); the inset shows the same data in a logarithmic scale.
----------------------------------------------------------------
Propagation of transmissible spongiform encephalopathies (TSE) is believed to depend on the replication of PrPsc at the expense of the normal protein [1,6]. Although the molecular details of the replication process are not completely known, it involves changes in conformation that are stabilized upon protein oligomerization [6-8]. It is assumed that this process occurs in vivo, taking months or even years after infection of the host for PrPsc replication to progress sufficiently to trigger appearance of the disease. PrP replication has been performed in vitro in a cell-free system, by mixing purified PrPc with an equimolar concentration of partially denatured PrPsc (ref. 2). This system has been used to study the molecular mechanism of PrP conversion [3], the sequence specificity of PrPsc formation [9], and to identify and evaluate inhibitors of PrP transformation [10,11]. However, the efficiency of the cell-free conversion is low, as the amount of newly converted protein is much less than the initial concentration of PrPsc needed to trigger the conversion reaction. This problem has precluded the study of the structural and infectious properties of the newly converted PrPres [6]. Our aim was to develop a system that mimics the replication of prions in the body during infection, by starting with undetectable concentrations of PrPsc, mixing it with a large excess of PrPc, and finishing the reaction with most of the molecules in the altered conformation. Using the PMCA procedure we estimate that the initial amount of PrPsc corresponds to less than 3% of the total concentration of PrPres produced at the end of the conversion. The amplification procedure requires the presence of several factors: exogenous PrPsc acting as a template for the conversion; PrPc that serves as a substrate; and unknown factors present in brain homogenate that catalyse the reaction (G.P.S. and C.S., unpublished observations). The strategy to speed up the conversion, and thus reproduce in a few hours in vitro a process that takes months or years for completion in vivo, was to perform cycles of incubation-sonication. It has been proposed that the infective unit of PrPsc is a ß-sheet-rich oligomer that converts the normal protein by integrating it into the growing aggregate, where it acquires the properties associated with the abnormal protein [12,13]. After incubation of the two forms of PrP, the oligomeric species increase in size by recruiting and transforming PrPc molecules. In the PMCA system, PrPsc is incubated with an excess of non-pathogenic conformer to enlarge the oligomeric converting units, followed by a sonication step to break down the aggregates into smaller units, each of which is capable of initiating further rounds of growth (Fig. 1). Our results demonstrate for the first time that the folding and biochemical properties of a protein can be transferred cyclically to other protein molecules, resulting in amplification of protein conformation in a manner conceptually analogous to DNA amplification by polymerase chain reaction (PCR).
A highly debated issue in the field of TSE is the nature of the infectious agent [14-16]. Strong evidence supports the 'protein-only' hypothesis of TSE propagation [1,6]. However, the conclusive proof of de novo production of prions is still lacking [6,15]. Our results show for the first time a high-efficiency replication of PrP conformation in vitro, mimicking the process of PrPsc replication thought to occur in vivo during the disease. Therefore, our findings reinforce one of the aspects of the prion hypothesis, that minute amounts of PrPsc have the ability to replicate its biochemical properties by converting large amounts of PrPc. Cyclic amplification of PrPsc in the presence of PrPc provides a unique opportunity to generate infectivity in vitro. The protein generated with the PMCA procedure shares several of the biochemical properties typical of PrPsc extracted from the brains of animals with scrapie, including protease-resistance, detergent insolubility (data not shown) and the ability to further convert PrPc in vitro. Because of the high yield of conversion, the infectious properties of the newly generated PrPsc-like protein can be tested and distinguished from the small amount of PrPsc used to begin the conversion reaction. We are currently determining the infectious properties of the PrPsc-like protein generated by PMCA.
The definitive diagnosis of TSE is based on the demonstration of PrPsc in brain tissue of the affected host [17]. Indeed, PrPsc is the only validated surrogate marker for the disease and the only known component of the TSE infectious agent [1]. The lack of a prion-specific nucleic acid prevents the use of highly sensitive PCR-based diagnostic tests. Given the recent widespread distribution of cases of bovine spongiform encephalopathy (BSE) in Europe and the strong arguments linking it to variant Creutzfeldt-Jakob disease (CJD) in humans [18,19], development of a sensitive diagnostic test that can reliably identify the disease in animals and people during the pre-symptomatic period is a top priority [20]. Currently available methods to detect PrPsc are limited by the amount of the abnormal protein present in the tissues [21]. This limitation results in late-stage diagnosis, in most cases from brain tissue obtained at autopsy [20,21]. However, as shown by infectivity studies, the infectious agent is present pre-symptomatically in many tissues in addition to the brain [21,22]. Thus, early detection of PrPsc from non-brain sources should be possible with highly sensitive methods [21]. Our findings constitute a strategy to detect low quantities of PrPsc, by means of amplifying undetectable amounts of the protein to a detectable level. The PMCA procedure can be combined with any of the existing detection systems to reach a further reduction of detection threshold. Preliminary results indicate that the PMCA system can also be applied to human brain samples obtained from sporadic CJD (G.P.S. and C.S., unpublished observations). Therefore, the PMCA method opens a new possibility for TSE diagnosis that could be applied to systemic tissues or fluids during the pre-symptomatic phase of the disease.
Methods
Preparation of brain homogenates
Brains from healthy rats (F.344) and Syrian golden hamsters healthy or infected with the adapted scrapie strain 263 K were obtained after decapitation and immediately frozen in dry ice and kept at -80 ° until used. Brains were homogenized in PBS buffer containing protease inhibitors (Complete cocktail from Boehringer Mannheim) at a 1 x final concentration. Detergents (0.5% Triton X-100, 0.05% SDS, final concentrations) were added and samples clarified with low-speed centrifugation (1,000g) for 1 min, using an Eppendorf 5415 centrifuge.
Cyclic amplification
Serial dilutions of the scrapie brain homogenate were made directly in the healthy brain homogenate. We incubated 60 µl of these dilutions at 37 ° with agitation. Every hour one cycle of sonication (five pulses of 1 s each) was done using a microsonicator (Bandelin Electronic, Sonopuls) with the probe immersed in the sample and the power setting fixed at 40%. These cycles were repeated 5-40 times.
Detection of PrPres
The samples were digested with proteinase K (100 µg/ml for 60 min at 37 °) and the reaction was stopped with 50 mM phenyl-methyl sulphonyl fluoride. Samples were separated by SDS-PAGE and electroblotted into nitrocellulose membrane in 3-(cyclohexylamino)-1-propane sulphonic acid or Tris-glycine transfer buffer with 10% methanol during 45 min at 400 mA. For immunoblotting, the membranes were blocked with 5% non-fat milk and incubated for 2 h with the monoclonal antibody 3F4 (ref. 23) (1:50,000). Four washes of 5 min each were performed with PBS and 0.3% Tween20 before incubation with secondary anti-mouse antibody labelled with horseradish peroxidase (1:5,000) for 1 h. After washing, the reactivity in the membrane was developed with an ECL Chemilumicense Kit (Amersham) according to the manufacturer's instructions. Densitometric analyses of western blots were performed with the program SigmaGel v1.0 (Jandel Scientific). The concentration of PrPsc was estimated by densitometric analysis and comparison with pure recombinant hamster PrP, the concentration of which was determined by amino-acid analysis.
Received 13 February 2001; accepted 10 May 2001
References
1. Prusiner, S. B. Prions. Proc. Natl Acad. Sci. USA 95, 13363-13383 (1998).
2. Kocisko, D. A. et al. Cell-free formation of protease-resistant prion protein. Nature 370, 471-474 (1994).
3. Horiuchi, M. & Caughey, B. Prion protein interconversions and the transmissible spongiform encephalopathies. Structure Fold Des. 7, R231-R240 (1999).
4. Saborio, G. P. et al. Cell-lysate conversion of prion protein into its protease-resistant isoform suggests the participation of a cellular chaperone. Biochem. Biophys. Res. Commun. 258, 470-475 (1999).
5. Telling, G. C. et al. Prion propagation in mice expressing human and chimeric PrP transgenes implicates the interaction of cellular PrP with another protein. Cell 83, 79-90 (1995).
6. Aguzzi, A. & Weissmann, C. Prion research: the next frontier. Nature 389, 795-798 (1997).
7. Cohen, F. E. & Prusiner, S. B. Pathologic conformations of prion proteins. Ann. Rev. Biochem. 67, 793-819 (1998).
8. Caughey, B., Kocisko, D. A., Raymond, G. J. & Lansbury, P. T.Jr Aggregates of scrapie-associated prion protein induce the cell-free conversion of protease-sensitive prion protein to the protease-resistance state. Chem. Biol. 2, 807-817 (1995).
9. Kocisko, D. A. et al. 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 (1995).
10. Chabry, J., Caughey, B. & Chesebro, B. Specific inhibition of in vitro formation of protease-resistant prion protein by synthetic peptides. J. Biol. Chem. 273, 13203-13207 (1998).
11. Caughey, W. S., Raymond, L. D., Horiuchi, M. & Caughey, B. Inhibition of protease-resistant prion protein formation by porphyrins and phthalocyanines. Proc. Natl Acad. Sci. USA 95, 12117-12122 (1998).
12. Harper, J. D. & Lansbury, P. T.Jr Models of amyloid seeding in Alzheimer's disease and scrapie: mechanistic truths and physiological consequences of the time-dependent solubility of amyloid proteins. Ann. Rev. Biochem. 66, 385-407 (1997).
13. Brown, P., Goldfarb, L. G. & Gajdusek, D. C. The new biology of spongiform encephalopathy: infectious amyloidoses with a genetic twist. Lancet 337, 1019-1022 (1991).
14. Chesebro, B. BSE and prions: uncertainties about the agent. Science 279, 42-43 (1998).
15. Mestel, R. Putting prions to the test. Science 273, 184-189 (1996).
16. Dormont, D. Agents that cause transmissible subacute spongiform encephalopathies. Biomed. Pharmacother. 53, 3-8 (1999).
17. Johnson, R. T. & Gibbs, C. J. Creutzfeldt-Jakob disease and related transmissible spongiform encephalopathies. New Engl. J. Med. 339, 1994-2004 (1998).
18. Collinge, J. Variant Creutzfeldt-Jakob disease. Lancet 354, 317-323 (1999).
19. Bruce, M. E. et al. Transmissions to mice indicate that new variant CJD is caused by the BSE agent. Nature 389, 498-501 (1997).
20. Schiermeier, Q. Testing times for BSE. Nature 409, 658-659 (2001).
21. Brown, P., Cervenakova, L. & Diringer, H. Blood infectivity and the prospects for a diagnostic screening test in Creutzfeldt-Jakob disease. J. Lab. Clin. Invest. 137, 5-13 (2001).
22. Glatzel, M. & Aguzzi, A. Peripheral pathogenesis of prion diseases. J. Gen. Virol. 81, 2813-2821 (2000).
23. Kascsak, R. J. et al. Mouse polyclonal and monoclonal antibody to scrapie-associated fibril proteins. J. Virol. 61, 3688-3693 (1987).
Acknowledgements. We thank S. Peano, A. Conz, L. Anderes and M.-J. Frossard for technical assistance and R. J. Kascsak for providing 3F4 anti-PrP antibody. We are grateful to M. Pocchiari, S. Fumero, T. Wells, K. Maundrell and J. DeLamarter for reading the manuscript and providing comments. We also thank C. Herbert for help in the preparation of the figures.

IN Als McKinley et al. 1986 scrapie-assoziierte Fibrillen aus infizierten Hamsterhirnen mit Ultraschall in kleine Fragmente zerlegten, da bemerkten sie keinen Einfluß dieser Prozedur auf die Infektiosität [AIFH]. Dies ist auch nicht verwunderlich, denn mann kann durch intensive Ultraschallbehandlung mit den Fibrilen auch die Infektiosität auflösen [AKAC]. Serio et al. stellten aber fest, dass sich die Konversionsrate von Hefeprionprotein drastisch steigern ließ, wenn man die Prionen zuvor durch Ultraschallbehandlung in viele kleine Stücke teilt [AKSD]. Aus diesem Ansatz entwickelten nun die Autoren eine Methode, die im Prinzip der PCR ähnelt, weil sich Ultraschallbehandlung und Amyloidbildung abwechseln. Inzwischen erreichen die Autoren nach 5 Zyklen 97,4-98,7% aus PrPc neu umgewandeltes proteaseresistentes Prionprotein PrPres. Mit 10 Amplifikationszyklen konnten sie eine PrPres-Ausgangsmenge von 6-12 pg nachweisen, indem rund 250 pg PrPres neu umgewandelt wurden. Die PrPres-Menge nimmt mit der Zahl der Zyklen exponentiell zu.
Sollten Nukleinsäuren oder Ribosomen durch die Ultraschallbehandlungen praktisch zerstört werden, dann konnten sich in diesem zellfreien System keine Viren vermehren. In dem Fall hätte diese Arbeit gemäß meiner Vorhersage von 1996 [ANDC] bereits im Jahr 2001 die Prionhypothese bewiesen, ohne das dies jemandem aufgefallen wäre.

MH Animal; Brain Chemistry; Chemistry, Analytical/methods; Hamsters; Phosphoprotein Phosphatase/chemistry; PrPsc Proteins/*analysis/chemistry; Prion Diseases/diagnosis/etiology; *Protein Folding; Rats; Scrapie/metabolism; Templates, Genetic

AD Serono Pharmaceutical Research Institute, CH1228 Geneva, Switzerland, Correspondence and requests for materials should be addressed to C.S. (e-mail: claudio.soto@serono.com) or to G.P.S. (e-mail: gabriela.saborio@serono.com).

SP englisch

PO England

EA pdf-Datei und HTML-Version

ZF kritische Zusammenfassung von Roland Heynkes

Autorenindex - authors index
Startseite - home page