Accelerated Publication - Enseignements de I

Herein we describe the purification and initial charac- terization of ... Herein, we describe the initial characterization of recombi- nant HET-s ... Downloaded from ...
296KB taille 1 téléchargements 332 vues
THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 277, No. 8, Issue of February 22, pp. 5703–5706, 2002 © 2002 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

Accelerated Publication The HET-s Prion Protein of the Filamentous Fungus Podospora anserina Aggregates in Vitro into Amyloid-like Fibrils* Received for publication, October 23, 2001, and in revised form, November 28, 2001 Published, JBC Papers in Press, December 3, 2001, DOI 10.1074/jbc.M110183200 Suzana Dos Reis‡, Be´ne´dicte Coulary-Salin, Vincent Forge§, Ioan Lascu¶, Joe¨l Be´gueret, and Sven J. Saupe储

The HET-s protein of Podospora anserina is a fungal prion. This protein behaves as an infectious cytoplasmic element that is transmitted horizontally from one strain to another. Under the prion form, the HET-s protein forms aggregates in vivo. The specificity of this prion model compared with the yeast prions resides in the fact that under the prion form HET-s causes a growth inhibition and cell death reaction when co-expressed with the HET-S protein from which it differs by 13 residues. Herein we describe the purification and initial characterization of recombinant HET-s protein expressed in Escherichia coli. The HET-s protein self-associates over time into high molecular weight aggregates. These aggregates greatly accelerate precipitation of the soluble form. HET-s aggregates appear as amyloid-like fibrils using electron microscopy. They bind Congo Red and show birefringence under polarized light. In the aggregated form, a HET-s fragment of ⬃7 kDa is resistant to proteinase K digestion. CD and FTIR analyses indicate that upon transition to the aggregated state, the HET-s protein undergoes a structural rearrangement characterized by an increase in antiparallel ␤-sheet structure content. These results suggest that the [Het-s] prion element propagates in vivo as an infectious amyloid.

Prions are infectious proteins causing mammalian spongiform * This work was supported by a grant from the Program de Recherche sur les ESST et les prions from the CNRS Physique et Chimie du Vivant. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ Recipient of a fellowship from the Ministe`re de l’Enseignement Supe´rieur. 储 To whom correspondence should be addressed: Laboratoire de Ge´ne´tique Mole´culaire des Champignons, Institut de Biochimie et de Ge´ne´tique Cellulaires, UMR 5095 CNRS/Universite´ de Bordeaux 2, 1 rue Camille St Sae¨ns, 33077 Bordeaux cedex. Tel.: 33 5 56 99 90 27; Fax: 33 5 56 99 90 67; E-mail: [email protected]. This paper is available on line at http://www.jbc.org

5703

Downloaded from www.jbc.org at CNRS on October 9, 2006

From les Laboratoires Ge´ne´tique Mole´culaire des Champignons et ¶d’Enzymologie Mole´culaire, Institut de Biochimie et de Ge´ne´tique Cellulaires, UMR 5095 CNRS/Universite´ de Bordeaux 2, 33077 Bordeaux cedex, France et le §Laboratoire de Biophysique Mole´culaire et Cellulaire, UMR 5090, De´partement de Biologie Mole´culaire et Structurale, CEA-Grenoble, FRANCE, 17, rue des Martyrs, 38054 Grenoble cedex 9, France

encephalopathies such as scrapie, mad cow disease, and Creutzfeld-Jakob disease (1). Prions propagate by converting the normal form of the PrP protein into an altered ␤-sheet-rich conformation (2). Prion diseases belong to a larger class of protein deposition diseases that are all characterized by the formation of ␤-sheet-rich fibrillar aggregates termed amyloids (3–5). These include common neurodegenerative diseases such as Alzheimer’s and Parkinson’s disease. However, as yet it is poorly understood why among the various human protein deposition diseases only the prion diseases have a proven infectious form. Prion proteins have also been identified in lower eukaryotes, namely yeast and the filamentous fungus Podospora anserina (6, 7). The URE3 and PSI non-mendelian genetic elements of yeast were shown to correspond to the prion form of the Ure2p and Sup35 proteins, respectively. Ure2p and Sup35p are soluble proteins in normal conformation but aggregate in vivo upon conversion to the prion state (8, 9). In vitro, purified Sup35p and Ure2p proteins undergo self-seeded polymerization into amyloid-like fibrils (10 –13). These elements thus represent valuable model systems to study amyloid formation and prion infectivity. The [Het-s] prion of the filamentous fungus P. anserina controls a cellular recognition process characteristic of filamentous fungi known as heterokaryon incompatibility (7, 14). The het-s locus has two very similar natural polymorphic variants designated as het-s and het-S. The corresponding HET-s and HET-S proteins are 289 amino acids in length and differ by 13 amino acid residues (15). Co-expression of the HET-s and HET-S variants in the same cell triggers a growth alteration and a cell death reaction. het-s and het-S strains are therefore incompatible. Strains expressing the HET-s variant exist as two alternate phenotypes, [Het-s*] and [Het-s]. [Het-s] strains are incompatible with het-S strains, whereas [Het-s*] strains are neutral in incompatibility. A [Het-s*] strain is converted to the reactive [Het-s] phenotype upon contact with a [Het-s] strain. The [Hets] element then spreads as an infectious element from the point of contact between the strains to the entire mycelium at a rate of several mm/hour (16). [Het-s*] strains also spontaneously acquire the reactive [Het-s] phenotype. In practice, with prolonged subculture all [Het-s*] strain ultimately acquire the [Het-s] phenotype. In vivo, a HET-s-GFP fusion protein coalesces into dot-like aggregates upon transition to the infectious [Het-s] state (17). The HET-S variant is devoid of the prion properties and does not aggregate in vivo. We have proposed that [Het-s] propagates as a self-perpetuating aggregated state of the HET-s protein. Compared with the yeast prion models, this system displays the following particular features: (i) the existence of a natural non-prion variant (HET-S), (ii) a cell death reaction associated with co-expression of the [Het-s] prion and HET-S, and (iii) the fact that the prion form corresponds to the biologically reactive form of the HET-s protein. Herein, we describe the initial characterization of recombinant HET-s protein and show that the protein aggregates in vitro into amyloid-like fibrils. These findings may establish [Het-s] as an attractive model system to study self-perpetuating protein aggregation processes and their cellular consequences.

5704

HET-s Aggregates into Amyloid Fibrils EXPERIMENTAL PROCEDURES

RESULTS

Recombinant HET-s Protein Undergoes Self-seeded Polymerization—The HET-s protein was expressed as a C-terminal histidine-tagged construct in E. coli. It was determined that addition of the histidine tag does not affect the biological activity of HET-s in P. anserina. The HET-s protein was purified from inclusion bodies to homogeneity (as judged by SDS-PAGE) under denaturing conditions and then renatured in 150 mM NaCl, 100 mM Tris-HCl, pH 8, and 1 mM dithiothreitol. In size exclusion chromatography experiments, the renatured protein 1

The abbreviation used is: FTIR, Fourier Transform Infrared.

FIG. 1. In vitro aggregation of renatured HET-s protein. Spontaneous aggregation. A, a HET-s solution at two different concentrations (black diamonds, 1.5 mg ml⫺1; open triangles, 1 mg ml⫺1) was incubated at 4 °C, and the amount of soluble protein was measured at various time points. B, inoculation with aggregated protein. 1 ml of renatured HET-s at 0.5 mg ml⫺1 was incubated at 4 °C and inoculated with various amounts of aggregated HET-s protein (black triangles, 5 ␮g; black squares, 50 ␮g; open diamonds, no inoculate).

displayed an apparent molecular mass of ⬃35 kDa in agreement with the calculated monomeric mass (32 kDa) and the apparent molecular mass measured for the native protein in crude P. anserina extracts (17). Renatured HET-s solutions were kept at 4 °C. After a lag phase of about 40 h precipitation started, and after 100 h approximately 50% of the protein had aggregated (Fig. 1A). Aggregation was accelerated by vortexing or brief sonication and also occurred at room temperature, in water, and at various pH values (5 to 9; data not shown). When freshly renatured protein was inoculated with pre-aggregated protein, HET-s aggregation was dramatically accelerated in a dose-dependent manner (Fig. 1B). For instance, using a 1:10 ratio between the aggregated inoculate and the soluble protein, after 1 h more than 80% of the protein had precipitated. Microscopic Characterization of the HET-s Aggregates— Congo Red was used to stain amyloids. Because amyloids are structured aggregates, Congo Red-stained amyloids are not optically amorphous and show a green-orange birefringence under polarized light (19). We incubated HET-s aggregates with Congo Red and collected them by centrifugation. Redstained aggregates were observed by bright field light microscopy, indicating that the aggregates bind the dye (Fig. 2). When observed under cross-polarizers, the aggregates showed greenorange birefringence. However, the birefringence was not observed over the entire aggregate. Using electron microscopy, negatively stained HET-s aggregates appeared as unbranched fibrils of 15–20 nm in width and up to several ␮m in length (Fig. 3). These fibrils have a ropy aspect that may correspond to the lateral bundling of profibrils. The width of HET-s fibrils was comparable with that of fibrils formed from full-length Ure2p or Sup35 (10, 12, 13). Spectroscopic Analysis of Soluble and Aggregated HET-s Protein—The CD spectra of the soluble and aggregate forms of HET-s were recorded (Fig. 4). The spectrum of soluble HET-s displayed two minima at 208 and 222 nm, characteristic of the presence of ␣-helical secondary structures. The deconvolution of the spectrum led to an estimated content of 34% ␣-helical, 16% ␤-sheet, and 50% random coil structure. The CD spectrum of the aggregated form was radically different. It is characterized by a strong reduction in ellipticity, suggesting a decrease in the amount of ␣-helical structure. It displayed a single minimum around 217 nm. Deconvolution of this spectrum yielded the following estimates of secondary structure content: 17% ␣-helix, 32% ␤-sheet, and 50% random coil.

Downloaded from www.jbc.org at CNRS on October 9, 2006

HET-s Expression and Purification—The het-s open-reading frame was amplified by PCR from plasmid DNA using oligos s1 (5⬘-CTCAAACTCATATGTCAGAACCG-3⬘) and s2 (5⬘-ATAAGCTTAGTGATGATGGTGATGGTGATTATCCCAGAACCC-3⬘) and was cloned into the NdeI and HindIII sites of the pET21a vector (Novagen) and introduced into BL21(DE3) pLysS cells. Cells were grown to 0.5 OD in 2⫻ YT medium, and expression was induced by addition of 1 mM isopropyl ␤-D-thiogalactoside. After 4 h, cells were harvested by centrifugation, frozen at ⫺80 °C, and lysed in 150 mM NaCl and 100 mM Tris-HCl, pH 8. The lysate was centrifuged for 20 min at 20,000 ⫻ g. The pellet was washed in the same buffer and resuspended in denaturing buffer (8 M guanidinium HCl, 150 mM NaCl, and 100 mM Tris-HCl, pH 8). The lysate was incubated with Talon Resin (CLONTECH) for 1 h at 20 °C, and the resin was washed with 8 M urea, 150 mM NaCl, and 100 mM Tris-HCl, pH 8. The HET-s protein was eluted from the resin in the same buffer containing 200 mM imidazole. For renaturation 20 mM dithiothreitol were added, and the sample was applied to a HiTrap desalting column (Amersham Biosciences) equilibrated with 150 mM NaCl, 100 mM Tris-HCl, pH 8, and 1 mM dithiothreitol. Spontaneous and Seeded Aggregation Assays—After renaturation, protein samples were incubated at 4 °C. Protein precipitation was analyzed at various time points by centrifuging 50-␮l aliquots for 15 min at 10,000 ⫻ g. Supernatant and pellet fractions were analyzed using SDS-PAGE followed by Coomassie Blue staining, and protein concentration in the supernatant fractions was measured using the Bio-Rad protein assay reagent. For seeding assays, 1 ml of renatured protein solution (at 0.5 mg ml⫺1) was inoculated with various amounts of aggregated protein in a volume of 100 ␮l. The aggregated protein sample was sonicated briefly prior to inoculation. Protein precipitation was analyzed as described above. Microscopy—For Congo Red staining, samples were prepared as described (12) and observed on a Zeiss Axiafold microscope equipped with optimally aligned cross-polarizers. For electron microscopy, 400 mesh copper electron microscopy grids coated with a plastic film (Formvar) were used. A fraction of the protein suspension (at 1 mg ml⫺1) was put onto the grid and sedimented for 10 to 30 min in a moist Petri dish to avoid rapid desiccation. Grids were then rinsed with 15–20 drops of freshly prepared 2% uranyl acetate in water and filtered with 0.22 ␮m of Millipore (18), dried with filter paper, and observed with a Phillips TECNAI 12 Biowin electron microscope at 80 kV. Spectroscopic Methods—CD spectra were recorded at 20 °C using a JASCO JS 810 spectropolarimeter with a quartz cell of 0.1-cm path length. Protein concentration was 0.1 mg ml⫺1. Deconvolution of the spectrum was performed using the K2d algorithm (www.embl-heidelberg.de/%7Eandrade/k2d.html). Fourier Transform Infrared (FTIR)1 spectroscopy spectra were recorded at 20 °C using a JASCO FT/IR-610 spectrometer. Interferograms were acquired at 4 cm⫺1 nominal resolution and are the average of 1000 scans. CaF2 plates and a 100-␮m spacer were used. Protein concentration was 4 mg ml⫺1. For the soluble form the D2O solution was not buffered (pH ⬃6). The sample with fibrils was buffered at pH 8 using Tris-HCl. The software (provided by JASCO) with the spectrometer was used for the spectrum deconvolution (bandwidth, 30 cm⫺1) and for the fitting of the spectra by a sum of lorentzians to estimate the secondary structure contents. Limited Proteolysis—50 ␮g of soluble and aggregated HET-s protein were digested at 37 °C with 1 ␮g of proteinase K in a volume of 50 ␮l. Reactions were stopped by the addition of one volume of SDS-PAGE loading buffer and were heated at 100 °C for 5 min. 10 ␮l of each reaction were analyzed by SDS-PAGE followed by Coomassie Blue staining.

HET-s Aggregates into Amyloid Fibrils

FIG. 2. Birefringence of Congo Red stained HET-s aggregates. HET-s aggregates stained with Congo Red in bright field (left panel) and polarized light (right panel), (scale bar ⫽ 25 ␮m).

FIG. 5. Infrared spectra of the soluble and aggregated forms of HET-s. Raw spectra (left panel) and deconvoluted spectra (right panel) of the soluble (upper) and aggregated form (lower). In the spectrum of the aggregated form, the broad band around 1590 cm⫺1 can be assigned to the COO⫺ group of aspartate and glutamate side-chains. This band is less intense in the spectrum of the soluble form because these groups are partially protonated from different pH and protein conformations. In the spectrum of the soluble form, the COOH groups give a band of ⬃1700 cm⫺1.

form of the protein. Estimations of the secondary structure contents by curve fitting (not shown) give: 30% ␣⫺helix, 20% ␤⫺sheet, and 50% random coil or turn (in the soluble form) and 10% ␣⫺helix, 45% ␤⫺sheet, and 45% random coil or turn (in the aggregated form). These estimated contents are in good agreement with those determined from the CD spectra. Limited Proteolysis of HET-s—To determine whether the transition from the soluble to the aggregated state modifies protease resistance, we submitted freshly renatured protein and aggregated protein to limited proteinase K digestion (Fig. 6). A resistant fragment of ⬃7 kDa was detected for the aggregated form even after 15 min of proteinase K digestion. The corresponding fragment was absent in the digestion profile of the soluble form. In this experiment, a set of fragments of ⬃25 kDa were slightly more resistant in the soluble form compared with the aggregated form. This difference further illustrates the fact that a major structural modification occurs upon transition to the aggregated state. DISCUSSION

FIG. 4. Circular dichroism of soluble and aggregated HET-s. The CD spectrum of soluble HET-s at 0.1 mg ml⫺1 is shown in black, and the CD spectrum of aggregated HET-s at 0.1 mg ml⫺1 is shown in gray.

In the infrared spectrum of the soluble form the amide I⬘ band reached a maximum at 1650 cm⫺1 (Fig. 5). In the spectrum of the aggregated form this maximum was shifted at 1643 cm⫺1, and a shoulder around 1625 cm⫺1 was observed. After deconvolution, three components were observed in the spectrum of the aggregated form. The band at 1645 cm⫺1 can be assigned to random coil structures. The band at 1625 cm⫺1 together with that at 1670 cm⫺1 can be assigned to antiparallel ␤⫺sheets. These two bands are not obvious in the deconvoluted spectrum of the soluble form. This shows that the amount of ␤⫺sheet is much higher within the fibrils than in the soluble

The [Het-s] prion element of the filamentous fungus P. anserina rapidly propagates in vivo and is efficiently transmitted horizontally from one strain to another. We show here that recombinant HET-s protein undergoes self-seeded polymerization in vitro into fibrillar aggregates that have characteristics of amyloids. The aggregates bind Congo Red and show birefringence and contain antiparallel ␤-sheet secondary structures. The behavior of the HET-s protein is thus remarkably similar to that of the two yeast prion proteins and of mammalian PrP. These results allow us to generalize the mechanism for the infectious propagation of prions in micro-organisms. We have shown that in vivo, the transition to the prion state can be detected using the aggregation of a HET-s-GFP fusion protein (17). We suggest, as proposed for the yeast prions, that emergence of the [Het-s] prion corresponds to the transition form of a soluble to an infectious amyloid-like aggregated form. The secondary structure of HET-s was analyzed using CD and FTIR spectroscopy. Despite slight quantitative differences, both spectroscopic methods indicated that soluble HET-s is an ␣/␤ protein. Upon aggregation, the protein undergoes a major structural rearrangement that is characterized by an increase in ␤-sheet content and a decrease in ␣-helical content. In vivo propagation and in vitro aggregation of both yeast prion pro-

Downloaded from www.jbc.org at CNRS on October 9, 2006

FIG. 3. Electron microscopy of HET-s aggregates. Electron micrographs of negatively stained HET-s filaments, (scale bar ⫽ 500 nm and 100 nm in inset).

5705

5706

HET-s Aggregates into Amyloid Fibrils

FIG. 6. Proteinase K digestion of soluble and aggregated HET-s. Soluble (A) and aggregated (B) HET-s proteins were submitted to proteinase K digestion as described under “Experimental Procedures” and analyzed by SDS-PAGE. The size of the molecular mass markers is given in kDa. The arrow indicates the 7-kDa resistant fragment.

Acknowledgments—We thank Eric Dufourc (Centre de Recherche Paul Pascal, CNRS) for use of the polarizing microscope. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

Prusiner, S. B. (1982) Science 216, 136 –144 Prusiner, S. B. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 13363–13383 Dobson, C. M. (1999) Trends Biochem. Sci. 24, 329 –332 Lansbury, P. T., Jr. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 3342–3344 Koo, E. H., Lansbury, P. T., Jr., and Kelly, J. W. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 9989 –9990 Wickner, R. B., Taylor, K. L., Edskes, H. K., Maddelein, M. L., Moriyama, H., and Roberts, B. T. (1999) Microbiol. Mol. Biol. Rev. 63, 844 – 861 Coustou, V., Deleu, C., Saupe, S., and Begueret, J. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 9773–9778 Patino, M. M., Liu, J. J., Glover, J. R., and Lindquist, S. (1996) Science 273, 622– 626 Edskes, H. K., Gray, V. T., and Wickner, R. B. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 1498 –1503 Glover, J. R., Kowal, A. S., Schirmer, E. C., Patino, M. M., Liu, J. J., and Lindquist, S. (1997) Cell 89, 811– 819 King, C. Y., Tittmann, P., Gross, H., Gebert, R., Aebi, M., and Wuthrich, K. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 6618 – 6622 Taylor, K. L., Cheng, N., Williams, R. W., Steven, A. C., and Wickner, R. B. (1999) Science 283, 1339 –1343 Thual, C., Komar, A. A., Bousset, L., Fernandez-Bellot, E., Cullin, C., and Melki, R. (1999) J. Biol. Chem. 274, 13666 –13674 Saupe, S. J. (2000) Microbiol. Mol. Biol. Rev. 64, 489 –502 Turcq, B., Denayrolles, M., and Be´ gueret, J. (1990) Curr. Genet. 17, 297–303 Beisson-Schecroun, J. (1962) Ann. Genet. 4, 3–50 Coustou-Linares, V., Maddelein, M. L., Be´ gueret, J., and Saupe, S. J. (2001) Mol. Microbiol. 42, 1325–1335 Holm Nielsen, E., Nybo, M., and Svehag, S. E. (1999) Methods in Enzymology 309, 491– 495 Westermark, G. T., Johnson, K. H., and Westermark, P. (1999) Methods in Enzymology 309, 3–25 Kheterpal, I., Williams, A., Murphy, C., Bledsoe, B., and Wetzel, R. (2001) Biochemistry 40, 11757–11767

Downloaded from www.jbc.org at CNRS on October 9, 2006

teins involve Q/N-rich regions of the proteins that are thought to be unstructured. The HET-s sequence lacks these Q/N-rich regions. However, analysis of the CD and FTIR spectra suggest that the soluble form of HET-s displays substantial amounts of random coil. It is therefore a possibility that the prion properties of HET-s are also due to the existence of a large unstructured region. Limited proteolysis can identify regions of amyloid peptides that are directly involved in the protective ␤-sheet structure (20). Proteolysis of the aggregated form of HET-s generates a resistant fragment of ⬃7 kDa. This fragment may thus correspond to the core region of the amyloid-like aggregate, the part of the protein that would be directly engaged in a highly hydrogen-bonded antiparallel ␤-sheet structure. Consistent with that hypothesis is the fact that the proteinase K-resistant material remains insoluble (not shown). The results reported herein, suggesting that [Het-s] propagates as an infectious amyloid, lead to an evolutionary enigma. In mammals, amyloid formation represents a cellular catastrophe, a rare failure of the normal folding process that can have dramatic consequences (3). Conversely, [Het-s] emergence is nearly ubiquitous (at least under laboratory conditions). All [Het-s*] strains ultimately acquire the [Het-s] prion phenotype. Moreover, in the [Het-s] system, it is the prion form that is active in the heterokaryon incompatibility phenomenon, whereas the [Het-s*] form is neutral (inactive). Thus, in the

[Het-s] system, the amyloid form would correspond to the biologically reactive form in self/non-self recognition. This raises the intriguing possibility that in this system amyloid formation has been recruited to perform a biological function. Also, the particular feature of the [Het-s] system as compared with the yeast prions is that in the HET-S background, expression of the prion form is associated to a growth inhibition and a cell death reaction. The fact that HET-s forms amyloidlike fibrils may establish this system as the original model to study cellular mechanisms of prion propagation and amyloid toxicity in a tractable micro-organism.