The Role of Disulfide Bridge in the Folding and Stability of the Recombinant Human Prion Protein
2001; Elsevier BV; Volume: 276; Issue: 4 Linguagem: Inglês
10.1074/jbc.m007862200
ISSN1083-351X
AutoresNilesh Maiti, Witold K. Surewicz,
Tópico(s)Porphyrin Metabolism and Disorders
ResumoIt is believed that the critical step in the pathogenesis of transmissible spongiform encephalopathies is a transition of prion protein (PrP) from an α-helical conformation, PrPC, to a β-sheet-rich form, PrPSc. Native prion protein contains a single disulfide bond linking Cys residues at positions 179 and 214. To elucidate the role of this bridge in the stability and folding of the protein, we studied the reduced form of the recombinant human PrP as well as the variant of PrP in which cysteines were replaced with alanine residues. At neutral pH, the reduced prion protein and the Cys-free mutant were insoluble and formed amorphous aggregates. However, the proteins could be refolded in a monomeric form under the conditions of mildly acidic pH. Spectroscopic experiments indicate that the monomeric Cys-free and reduced PrP have molten globule-like properties, i.e. they are characterized by compromised tertiary interactions, an increased exposure of hydrophobic surfaces, lack of cooperative unfolding transition in urea, and partial loss of native (α-helical) secondary structure. In the presence of sodium chloride, these partially unfolded proteins undergo a transition to a β-sheet-rich structure. However, this transition is invariably associated with protein oligomerization. The present data argue against the notion that reduced prion protein can exist in a stable monomeric form that is rich in β-sheet structure. It is believed that the critical step in the pathogenesis of transmissible spongiform encephalopathies is a transition of prion protein (PrP) from an α-helical conformation, PrPC, to a β-sheet-rich form, PrPSc. Native prion protein contains a single disulfide bond linking Cys residues at positions 179 and 214. To elucidate the role of this bridge in the stability and folding of the protein, we studied the reduced form of the recombinant human PrP as well as the variant of PrP in which cysteines were replaced with alanine residues. At neutral pH, the reduced prion protein and the Cys-free mutant were insoluble and formed amorphous aggregates. However, the proteins could be refolded in a monomeric form under the conditions of mildly acidic pH. Spectroscopic experiments indicate that the monomeric Cys-free and reduced PrP have molten globule-like properties, i.e. they are characterized by compromised tertiary interactions, an increased exposure of hydrophobic surfaces, lack of cooperative unfolding transition in urea, and partial loss of native (α-helical) secondary structure. In the presence of sodium chloride, these partially unfolded proteins undergo a transition to a β-sheet-rich structure. However, this transition is invariably associated with protein oligomerization. The present data argue against the notion that reduced prion protein can exist in a stable monomeric form that is rich in β-sheet structure. prion protein cellular PrP isoform scrapie (proteinase-resistant) PrP isoform human guanidine hydrochloride dithiothreitol 8-anilino-1-naphthalene sulfonic acid Prion diseases, or transmissible spongiform encephalopathies, comprise a group of fatal neurodegenerative disorders that can arise sporadically or can have an infectious or genetic etiology (1Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13383Crossref PubMed Scopus (5052) Google Scholar, 2Weissmann C. FEBS Lett. 1996; 389: 3-11Crossref PubMed Scopus (145) Google Scholar, 3Horiuchi M. Caughey B. Structure. 1999; 7: R231-R240Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar). The best known animal forms of the disease are scrapie in sheep and bovine spongiform encephalopathy in cattle. The human versions include kuru, Creutzfeldt-Jakob disease, fatal familial insomnia, and Gerstmann-Straussler-Scheinker syndrome. These disorders are characterized by vacuolation of neurons, astroglyosis, and cerebral accumulation of an abnormal (scrapie-like) form of prion protein, PrPSc.1 Although the molecular mechanism of transmissible spongiform encephalopathies is controversial (4Chesebro B. Nat. Med. 1997; 3: 491-497Crossref PubMed Scopus (29) Google Scholar), numerous observations point to the central role of PrPSc in the pathogenesis of these disorders (1Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13383Crossref PubMed Scopus (5052) Google Scholar, 2Weissmann C. FEBS Lett. 1996; 389: 3-11Crossref PubMed Scopus (145) Google Scholar, 3Horiuchi M. Caughey B. Structure. 1999; 7: R231-R240Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar). According to the “protein only” hypothesis (1Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13383Crossref PubMed Scopus (5052) Google Scholar, 5Griffith J.S. Nature. 1967; 215: 1043-1044Crossref PubMed Scopus (888) Google Scholar, 6Prusiner S.B. Science. 1982; 216: 136-144Crossref PubMed Scopus (4014) Google Scholar), PrPSc constitutes the sole component of the infectious prion pathogen. PrPSc is derived from a normal (cellular) prion protein, PrPC. Human PrPC is a 209-residue glycoprotein that has two N-glycosylation sites and a disulfide bridge linking Cys residues at positions 179 and 214. PrPC is transported through the secretory pathway and ultimately anchored to the cell surface by a glycosylphosphatidylinositol anchor (1Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13383Crossref PubMed Scopus (5052) Google Scholar, 7Stahl N. Prusiner S.B. FASEB J. 1991; 5: 2799-2807Crossref PubMed Scopus (120) Google Scholar, 8Caughey B. Chesebro B. Trends Cell Biol. 1997; 7: 56-62Abstract Full Text PDF PubMed Scopus (170) Google Scholar). It is localized in cholesterol-rich membrane microdomains called rafts or caveo-like domains (9Vey M. Pilkuhn S. Wille H. Nixon R. DeArmond S.J. Smart E.J. Anderson R.G.W. Taraboulos A. Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14945-14949Crossref PubMed Scopus (485) Google Scholar). The transition between PrPC and PrPSc occurs post-translationally on the cell surface and/or in an endocytic pathway (10Caughey B. Raymond G.J. Race R.E. J. Virol. 1991; 65: 6597-6603Crossref PubMed Google Scholar, 11Borchelt D.R. Taraboulos A. Prusiner S.B. J. Biol. Chem. 1992; 267: 16188-16199Abstract Full Text PDF PubMed Google Scholar). No differences in the covalent structure have been observed between PrPC and PrPSc (12Stahl N. Baldwin M.A. Teplow D.B. Hood L. Gibson B.W. Burlingame A.L. Prusiner S.B. Biochemistry. 1993; 32: 1991-2002Crossref PubMed Scopus (533) Google Scholar). However, the two protein isoforms have profoundly different biochemical and biophysical properties. PrPC is soluble in mild detergents and easily degradable by proteinase K, whereas PrPSc is insoluble in mild detergents and highly resistant to proteinase K digestion (7Stahl N. Prusiner S.B. FASEB J. 1991; 5: 2799-2807Crossref PubMed Scopus (120) Google Scholar, 13Meyer R.K. McKinley M.P. Bowman K.A. Braunfeld M.B. Barry R.A. Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 2310-2314Crossref PubMed Scopus (513) Google Scholar, 14Oesch B. Westaway D. Walchli M. McKinley M.P. Kent S.B. Aebersold R. Barry R.A. Tempst P. Teplow D.B. Hood L.E. Prusiner S.B. Weissmann C. Cell. 1985; 40: 735-746Abstract Full Text PDF PubMed Scopus (1238) Google Scholar). Furthermore, spectroscopic studies have revealed that the two isoforms have markedly different secondary structures; PrPC consists largely of α-helices, whereas PrPSc is rich in β-sheet structure (15Pan K.M. Baldwin M. Nguyen J. Gasset M. Serban A. Groth D. Mehlhorn I. Huang Z. Fletterick R.J. Cohen F.E. Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 10962-10966Crossref PubMed Scopus (2049) Google Scholar, 16Caughey B.W. Dong A. Bhat K.S. Ernst D. Hayes S.F. Caughey W.S. Biochemistry. 1991; 30: 7672-7680Crossref PubMed Scopus (741) Google Scholar, 17Safar J. Roller P.P. Gajdusek D.C. Gibbs Jr., C.J. Protein Sci. 1993; 2: 2206-2216Crossref PubMed Scopus (173) Google Scholar). Because the critical step in the pathogenesis of spongiform encephalopathies appears to be a conformational transition of PrP, there is currently great interest in understanding the biophysical properties of the prion protein. Recent studies have provided a wealth of data on the three-dimensional structure, folding pathway, and thermodynamic stability of the recombinant model of PrPC(18Riek R. Hornemann S. Wider G. Billeter M. Glockshuber R. Wuthrich K. Nature. 1996; 382: 180-182Crossref PubMed Scopus (1114) Google Scholar, 19Zahn R. Liu A. Luhrs T. Riek R. von Schroetter C. Lopez Garcia F. Billeter M. Calzolai L. Wider G. Wuthrich K. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 145-150Crossref PubMed Scopus (935) Google Scholar, 20Liu H. Farr-Jones S. Ulyanov N.B. Llinas M. Marqusee S. Groth D. Cohen F.E. Prusiner S.B. James T.L. Biochemistry. 1999; 38: 5362-5377Crossref PubMed Scopus (197) Google Scholar, 21Donne D.G. Viles J.H. Groth D. Mehlhorn I. James T.L. Cohen F.E. Prusiner S.B. Wright P.E. Dyson H.J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 13452-13457Crossref PubMed Scopus (635) Google Scholar, 22Zhang H. Stockel J. Mehlhorn I. Groth D. Baldwin M.A. Prusiner S.B. James T.L. Cohen F.E. Biochemistry. 1997; 36: 3543-3553Crossref PubMed Scopus (168) Google Scholar, 23Swietnicki W. Petersen R.B. Gambetti P. Surewicz W.K. J. Biol. Chem. 1998; 273: 31048-31052Abstract Full Text Full Text PDF PubMed Scopus (174) Google Scholar, 24Liemann S. Glockshuber R. Biochemistry. 1999; 38: 3258-3267Crossref PubMed Scopus (294) Google Scholar, 25Morillas M. Swietnicki W. Gambetti P. Surewicz W.K. J. Biol. Chem. 1999; 274: 36859-36865Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar, 26Hosszu L.L. Baxter N.J. Jackson G.S. Power A. Clarke A.R. Waltho J.P. Craven C.J. Collinge J. Nat. Struct. Biol. 1999; 6: 740-743Crossref PubMed Scopus (137) Google Scholar, 27Wildegger G. Liemann S. Glockshuber R. Nat. Struct. Biol. 1999; 6: 550-553Crossref PubMed Scopus (129) Google Scholar, 28Swietnicki W. Morillas M. Chen S.G. Gambetti P. Surewicz W.K. Biochemistry. 2000; 39: 424-431Crossref PubMed Scopus (216) Google Scholar). However, the molecular mechanism of conformational transition(s) underlying the conversion of PrPC to PrPSc still remains unknown. In a recent study (29Jackson G.S. Hosszu L.L.P. Power A. Hill A.F. Kenney J. Saibil H. Craven C.J. Waltho J.P. Clarke A.R. Collinge J. Science. 1999; 283: 1935-1937Crossref PubMed Scopus (361) Google Scholar), it was reported that, upon reduction of a single disulfide bridge, the recombinant prion protein could reversibly switch between α-helical conformation and a monomeric form rich in β-sheet structure. It was also postulated that the latter form represents a key monomeric precursor of PrPSc. The notion that a monomeric protein could exist in two profoundly different conformations is highly intriguing and has potentially far reaching implications. To obtain further insight into the molecular basis of conformational transitions in prion protein, we have undertaken detailed studies on the effect of disulfide bridge on the folding and conformational stability of the recombinant human PrP. Our data show that the removal of a disulfide bridge greatly destabilizes the native structure of the protein. In the presence of salt, the reduced or Cys-free protein is highly prone to oligomerization. However, under no experimental conditions could we observe a monomeric β-sheet-rich form of the protein. Both in the presence (28Swietnicki W. Morillas M. Chen S.G. Gambetti P. Surewicz W.K. Biochemistry. 2000; 39: 424-431Crossref PubMed Scopus (216) Google Scholar) and absence of a disulfide bridge, β-structure was formed only upon oligomerization of the recombinant PrP. The plasmids encoding huPrP23–231 and huPrP90–231 with an N-terminal linker containing a His6 tail and a thrombin cleavage site were described previously (25Morillas M. Swietnicki W. Gambetti P. Surewicz W.K. J. Biol. Chem. 1999; 274: 36859-36865Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar). Proteins obtained using these plasmids contain the N-terminal extension Gly-Ser-Asp-Pro (25Morillas M. Swietnicki W. Gambetti P. Surewicz W.K. J. Biol. Chem. 1999; 274: 36859-36865Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar). To remove the last two residues of this extension, the plasmids were amplified by polymerase chain reaction using the primers 5′-CCGCGTGGTTCGAAGAAGCGCCCG and 5′-CGGGCGCTTCTTCGAACCACGC GG for huPrP23–231 or 5′-GCGTGGTTCGGGTCAAGGAG and 5′-CTCCTTGACCCGAACCA CGC for huPrP90–231. The C179A/C214A variant of huPrP23–231 was obtained by site-directed mutagenesis using the primers 5′-C TTT GTG CAC GAC GCC GTC AAT ATC AC and 5′-GT GAT ATT GAC GGC GTC GTG CAC AAA G for Cys179 → Ala replacement and 5′-GTT GAG CAG ATG GCG ATC ACC CAG TAC and 5′-GTA CTG GGT GAT CGC CAT CTG CTC AAC for Cys214 → Ala replacement. All DNA manipulations were carried out according to standard protocols (30Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). DNA sequences of the final constructs were verified by automated sequencing at the Case Western Reserve University Molecular Biology Core Facility. The wild type huPrP23–231 and huPrP90–231 were expressed and purified as described previously (25Morillas M. Swietnicki W. Gambetti P. Surewicz W.K. J. Biol. Chem. 1999; 274: 36859-36865Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar). The C179A/C214A variant was expressed by the same procedure. However, the tendency of the mutant protein to precipitate at neutral pH required different purification strategy. After expression in BL21(DE3) cells, the latter protein was extracted by sonication in a buffer (10 mm Tris, 100 mmK2HPO4, pH 8.0) containing 6 mGdnHCl. The protein was then applied onto a nickel-nitriloacetic acid-agarose column (Amersham Pharmacia Biotech). Weakly bound proteins were removed by washing the column with 10 mm Tris, 100 mm K2HPO4, 50 mmimidazole, pH 8.0. The mutant huPrP23–231 was then eluted in 500 mm imidazole, 6 m GdnHCl, pH 5.8. Protein was then dialyzed against a 10 mm sodium acetate buffer, pH 5.6, and the His tail was cleaved by 15 h of treatment at room temperature with thrombin (10 units/mg of protein). The free peptide and thrombin were removed by selective precipitation of C179A/C214A huPrP23–231 in sodium phosphate buffer, pH 7.0. The pellet was dissolved in 10 mm sodium acetate containing 8m urea, pH 4.0. Finally, the purified protein was refolded by 10-fold dilution in 10 mm sodium acetate, pH 4.0, followed by extensive dialysis against the same buffer. To remove the residual aggregated protein, the dialyzed sample was filtered and ultracentrifuged for 3 h at 100,000 × g. Protein concentration was determined using a molar coefficient at 276 nm of 56,650 m−1cm−1. To reduce the disulfide bond, the wild type huPrP23–231 or huPrP90–231 was unfolded with 8m urea or 6 m GdnHCl in 10 mm Tris, pH 8.0, and treated for 15 h at room temperature with 100 mm DTT. Upon reduction, the protein was refolded by dialysis against 10 mm sodium acetate, 1 mmDTT, pH 4.0. The dialyzed sample was filtered and ultracentrifuged at 100,000 × g for 3 h. Far-UV CD spectra were recorded in a 1-mm quartz cell at a protein concentration of 0.25 mg/ml. Near-UV CD spectra were obtained using a 1-cm cell at a protein concentration of 1 mg/ml. To obtain equilibrium unfolding curves, huPrP23–231, Cys-free huPrP23–231 and reduced huPrP23–231 were diluted (to a final concentration of 0.024 mg/ml) in 10 mm sodium acetate, pH 4.0, containing different concentration of urea. Samples were incubated for 24 h at room temperature, and the ellipticity at 222 nm was measured in a 1-cm cell by averaging the signal over 1 min. The concentration of urea was determined by refractive index measurements. All CD measurements were carried out at room temperature on a Jasco J-810 spectropolarimeter. Protein samples were diluted to a concentration of 0.05 mg/ml in 50 mm sodium acetate buffer, pH 4.0, containing 10 μm ANS. After 1 h of incubation in the dark, fluorescence spectra were measured on an SLM 8100 spectrofluorimeter using the excitation wavelength of 375 nm. The hydrodynamic radius of monomeric forms of PrP was measured by quasi-elastic light scattering on a DynaPro-801 molecular size detector (Protein Solutions Inc.). Data were analyzed with the software provided by the manufacturer using appropriate viscosity and refractive index corrections. Size-exclusion chromatography was performed using Bio-Sil SEC-250 column (Bio-Rad) attached to a RAININ HPLC system. The column was pre-equilibrated with 10 mm sodium acetate, 50 mm NaCl, pH 4.0, and the protein was eluted using the same buffer at the flow rate of 1 ml/min. The elution of the protein was monitored by absorbance at 280 nm. The disulfide bridge in the folded huPrP is buried in a hydrophobic environment (18Riek R. Hornemann S. Wider G. Billeter M. Glockshuber R. Wuthrich K. Nature. 1996; 382: 180-182Crossref PubMed Scopus (1114) Google Scholar, 19Zahn R. Liu A. Luhrs T. Riek R. von Schroetter C. Lopez Garcia F. Billeter M. Calzolai L. Wider G. Wuthrich K. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 145-150Crossref PubMed Scopus (935) Google Scholar, 20Liu H. Farr-Jones S. Ulyanov N.B. Llinas M. Marqusee S. Groth D. Cohen F.E. Prusiner S.B. James T.L. Biochemistry. 1999; 38: 5362-5377Crossref PubMed Scopus (197) Google Scholar, 21Donne D.G. Viles J.H. Groth D. Mehlhorn I. James T.L. Cohen F.E. Prusiner S.B. Wright P.E. Dyson H.J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 13452-13457Crossref PubMed Scopus (635) Google Scholar) and is not accessible to reducing agents such as DTT. However, the bridge could be readily reduced by DTT upon unfolding of the protein in 8 m urea or 6m GdnHCl. At neutral pH, attempts to refold the reduced huPrP23–231 by rapid dilution or dialysis against the denaturant-free buffer (10 mm sodium phosphate, pH 7.0 and 8.0 or Tris HCl, pH 8.0) consistently resulted in a massive precipitation of the protein, leaving no material in a soluble form. Very similar behavior was observed for prion protein variants in which the disulfide bridge was removed by a replacement of Cys residues with alanine (C179A/C214A huPrP23–231). In contrast to the results obtained at neutral pH, dialysis refolding of the disulfide bridge-free huPrP23–231 under acidic conditions (10 mm sodium acetate, pH 4.0) did not lead to visible precipitation of the protein. However, upon complete removal of urea the samples were slightly turbid, indicating that at least part of the protein was self-associated. The aggregated material could be removed by ultracentrifugation for 3 h at 100,000 × g. An analysis of the remaining fraction by quasi-elastic light scattering revealed the presence of a monomeric protein with an apparent Stokes radius of ∼2.8 nm. A comparison of the latter number with the Stokes radius of 2.4 nm for the oxidized wild-type huPrP23–231 suggests that the loss of the disulfide bridge in prion protein is associated with a transition to a less compact conformation. The recovery of the protein in a monomeric form was ∼20% for the reduced huPrP23–231 and less than 10% for the Cys-free mutant. It should be noted that under present experimental conditions, freshly prepared monomeric protein was stable for at least 10 h. However, upon longer incubation at room temperature the protein had a tendency to self-associate. The behavior of the reduced and Cys-free protein contrasts with that of the disulfide bridge-containing huPrP23–231. The latter protein could be unfolded reversibly and refolded both at neutral and acidic pH. The refolded species retained native α-helical conformation and was stable as a monomer for many days or even weeks. The effect of disulfide bridge on the secondary structure of huPrP23–231 was assessed by far-UV circular dichroism spectroscopy. As shown in Fig.1, the spectrum at pH 4 for the wild-type protein with an intact disulfide bridge exhibits a double minimum at 208 and 222 nm (with a mean residue ellipticity at 222 nm, Θ222, of −8000° cm2dmol−1) typical for α-helix-rich proteins (31Woody R.W. Methods Enzymol. 1995; 246: 34-71Crossref PubMed Scopus (691) Google Scholar). This spectrum is consistent with the structure of PrP containing α-helical C-terminal domain and largely unstructured N-terminal region (19Zahn R. Liu A. Luhrs T. Riek R. von Schroetter C. Lopez Garcia F. Billeter M. Calzolai L. Wider G. Wuthrich K. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 145-150Crossref PubMed Scopus (935) Google Scholar, 21Donne D.G. Viles J.H. Groth D. Mehlhorn I. James T.L. Cohen F.E. Prusiner S.B. Wright P.E. Dyson H.J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 13452-13457Crossref PubMed Scopus (635) Google Scholar). The CD spectrum of the monomeric huPrP23–231 with reduced disulfide bridge is profoundly different; it shows a minimum at ∼204 nm, whereas the mean residue ellipticity at 222 nm is reduced to −3000° cm2 dmol−1 (Fig. 1). A very similar spectrum was obtained for the Cys-free variant of huPrP23–231, although in this case Θ222 is somewhat higher (approximately −4000° cm2dmol−1). The spectral data presented above indicate that in the absence of the disulfide bridge, prion protein refolds (as a monomer) into a conformation that is characterized by a major loss of α-helix and increased proportion of an unordered structure. The perturbation of structural integrity upon removal of the disulfide bridge in huPrP is also indicated by near-UV circular dichroism spectroscopy. Near-UV CD spectra provide a very sensitive probe of a global tertiary structure of proteins (31Woody R.W. Methods Enzymol. 1995; 246: 34-71Crossref PubMed Scopus (691) Google Scholar). As shown in Fig.2, upon removal of the disulfide bridge in huPrP23–231, there is a major reduction of the CD signal in the near UV region. This spectral change most likely reflects at least a partial collapse of native tertiary interactions in the protein (31Woody R.W. Methods Enzymol. 1995; 246: 34-71Crossref PubMed Scopus (691) Google Scholar). However, a small part of the signal reduction could also be due to the loss of spectral contribution from the disulfide bond. The conformational properties of the reduced and Cys-free variants of huPrP23–231 were further probed using the hydrophobic dye ANS. This dye, the fluorescence of which greatly increases upon binding to hydrophobic sites, has been used widely to assess the surface hydrophobicity of proteins (32Cardamone M. Puri N.K. Biochem. J. 1992; 282: 589-593Crossref PubMed Scopus (410) Google Scholar). Fig. 3shows that there is very little fluorescence of ANS in the presence of huPrP23–231 with an intact disulfide bridge. However, incubation of the dye in the presence of the reduced PrP or the Cys-free variant results in a dramatic increase in the fluorescence intensity. This indicates binding of the probe to disulfide bridge-free proteins, most likely because of an increased exposure of hydrophobic patches (32Cardamone M. Puri N.K. Biochem. J. 1992; 282: 589-593Crossref PubMed Scopus (410) Google Scholar,33Ptitsyn O.B. Pain R.H. Semisotnov G.V. Zerovnik E. Razgulyaev O.I. FEBS Lett. 1990; 262: 20-24Crossref PubMed Scopus (665) Google Scholar). The effect of the disulfide bridge on the thermodynamic stability of prion protein was studied by equilibrium unfolding in urea. As shown in Fig. 4, the unfolding curve for the oxidized wild-type huPrP23–231 at pH 4 (10 mm sodium acetate) is highly cooperative. Analysis of this curve using a two-state transition model (34Santoro M.M. Bolen D.W. Biochemistry. 1988; 27: 8063-8068Crossref PubMed Scopus (1589) Google Scholar) yields a midpoint unfolding urea concentration of 3.7 m and a free energy difference between the native and unfolded state of ∼12 kJ/m. In contrast to the disulfide bridge-containing protein, under similar experimental conditions the reduced huPrP23–231 and the Cys-free variant unfold over a very broad range of urea concentration, showing no well defined cooperative transition. The reduced (or Cys-free) PrP monomer with the properties described above was observed only at acidic pH in a low ionic strength buffer (10 mmsodium acetate) and in the absence of NaCl. Upon addition of NaCl, there was a rapid change in the CD spectrum to one with a minimum at ∼215 nm (Fig. 5), indicating a conformational transition to a form rich in β-sheet structure (31Woody R.W. Methods Enzymol. 1995; 246: 34-71Crossref PubMed Scopus (691) Google Scholar). Under the present experimental conditions, this conformational transition was very fast, precluding detailed kinetic studies. For example, at a protein concentration of 0.25 mg/ml and a NaCl concentration of 50 mm, the changes in the spectrum were complete within less than 3 min. In contrast to the reduced and Cys-free proteins, the oxidized huPrP23–231 remained α-helical even after 24 h of incubation in the presence of 50 or 100 mm NaCl. The oligomerization state of different forms of huPrP23–231 was assessed by size-exclusion chromatography (Fig.6). Under the conditions corresponding to β-sheet formation (10 mm sodium acetate, 50 mm NaCl), the reduced and Cys-free proteins consistently eluted as high molecular aggregates larger than the 600 kDa fractionation limit of the Bio-Sil SEC-250 column. The elution profile presented above was highly reproducible, and repeated experiments provided no indication for the presence of a monomeric β-sheet-rich species. This is in contrast with the behavior of the oxidized (α-helical) huPrP23–231 that, under the same buffer conditions, eluted at a time corresponding to a monomeric prion protein. Attempts to obtain size exclusion data in the absence of NaCl were unsuccessful because, under these conditions, huPrP23–231 adsorbs to Bio-Sil column and other chromatographic media tested (Superdex, Sephacryl, Superose). However, the monomeric state of the oxidized, reduced, and Cys-free forms of huPrP23–231 in the NaCl-free buffer is clearly indicated by quasi-elastic light scattering experiments as described above. Experiments similar to those described above for huPrP23–231 were also performed with the reduced human prion protein fragment 90–231. In close resemblance to the full-length protein, formation of the β-sheet structure by the reduced huPrP90–231 was invariably associated with rapid oligomerization of the protein, with no indication of the presence of a stable monomeric β-sheet rich form (data not shown for brevity). The central event in the pathogenesis of spongiform encephalopathies is a profound conformational change of the prion protein from an α-helical form, PrPC, to a β-sheet-rich form, PrPSc (1Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13383Crossref PubMed Scopus (5052) Google Scholar, 2Weissmann C. FEBS Lett. 1996; 389: 3-11Crossref PubMed Scopus (145) Google Scholar, 3Horiuchi M. Caughey B. Structure. 1999; 7: R231-R240Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar, 15Pan K.M. Baldwin M. Nguyen J. Gasset M. Serban A. Groth D. Mehlhorn I. Huang Z. Fletterick R.J. Cohen F.E. Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 10962-10966Crossref PubMed Scopus (2049) Google Scholar, 16Caughey B.W. Dong A. Bhat K.S. Ernst D. Hayes S.F. Caughey W.S. Biochemistry. 1991; 30: 7672-7680Crossref PubMed Scopus (741) Google Scholar, 17Safar J. Roller P.P. Gajdusek D.C. Gibbs Jr., C.J. Protein Sci. 1993; 2: 2206-2216Crossref PubMed Scopus (173) Google Scholar). Recent NMR studies have provided high-resolution structural data for the recombinant model of PrPC (18Riek R. Hornemann S. Wider G. Billeter M. Glockshuber R. Wuthrich K. Nature. 1996; 382: 180-182Crossref PubMed Scopus (1114) Google Scholar, 19Zahn R. Liu A. Luhrs T. Riek R. von Schroetter C. Lopez Garcia F. Billeter M. Calzolai L. Wider G. Wuthrich K. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 145-150Crossref PubMed Scopus (935) Google Scholar, 20Liu H. Farr-Jones S. Ulyanov N.B. Llinas M. Marqusee S. Groth D. Cohen F.E. Prusiner S.B. James T.L. Biochemistry. 1999; 38: 5362-5377Crossref PubMed Scopus (197) Google Scholar, 21Donne D.G. Viles J.H. Groth D. Mehlhorn I. James T.L. Cohen F.E. Prusiner S.B. Wright P.E. Dyson H.J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 13452-13457Crossref PubMed Scopus (635) Google Scholar), whereas a detailed structure of the PrPSc isoform is still unknown. It is believed that the propagation of the disease can be described according to the heterodimer (template-assistance type) model (35Prusiner S.B. Science. 1991; 252: 1515-1522Crossref PubMed Scopus (1733) Google Scholar) or the nucleation-dependent polymerization model (36Jarrett J.E. Lansbury Jr., P.T. Cell. 1993; 73: 1055-1105Abstract Full Text PDF PubMed Scopus (1886) Google Scholar). However, the molecular mechanism of the conformational transition and potential folding intermediates underlying the PrPC→PrPSc conversion remain unknown. Within the context of the “protein folding problem,” the fundamental question is whether the β-sheet-rich form of PrP can exist in a stable monomeric state or, as implied by the nucleation-dependent polymerization model, the above form is stable only as a high molecular weight oligomer. Recently, we demonstrated that, under appropriate solvent conditions, the recombinant human PrP undergoes a transition from α-helix to a β-sheet-rich structure (28Swietnicki W. Morillas M. Chen S.G. Gambetti P. Surewicz W.K. Biochemistry. 2000; 39: 424-431Crossref PubMed Scopus (216) Google Scholar). The key requirements for this conversion include acidic pH, mildly denaturing conditions, and the presence of certain ions. 2M. Morillas, D. Vanik, and W. K. Surewicz, unpublished data. The formation of β-sheet structure was associated invariably with oligomerization of prion protein into high molecular weight aggregates. A transition to an oligomeric β-sheet-rich structure was also reported in a related study with a redacted chimeric mouse-hamster PrP consisting of 106 amino acids (37Baskakov I.V. Aagaard C. Mehlhorn I. Wille H. Groth D. Baldwin M.A. Prusiner S.B. Cohen F.E. Biochemistry. 2000; 39: 2792-2804Crossref PubMed Scopus (67) Google Scholar). Importantly, the above two studies provide no evidence for the presence of a monomeric β-sheet-rich conformer of the prion protein. A markedly different behavior was reported for prion protein with a reduced disulfide bond. According to Jackson et al. (29Jackson G.S. Hosszu L.L.P. Power A. Hill A.F. Kenney J. Saibil H. Craven C.J. Waltho J.P. Clarke A.R. Collinge J. Science. 1999; 283: 1935-1937Crossref PubMed Scopus (361) Google Scholar), at neutral pH the reduced huPrP90–231 exists as a native α-helical conformer, whereas at acidic pH it adopts a monomeric form rich in a β-sheet structure. These authors also proposed that the reduction of prion protein in intracellular compartments could play a crucial role in the PrPC → PrPSc conversion. However, this hypothesis is controversial (3Horiuchi M. Caughey B. Structure. 1999; 7: R231-R240Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar), and certain observations argue against the involvement of disulfide bond reduction in the pathogenesis of prion disease (38Turk E. Teplow D.B. Hood L.E. Prusiner S.B. Eur. J. Biochem. 1988; 176: 21-30Crossref PubMed Scopus (266) Google Scholar,39Herrmann L.M. Caughey B. Neuroreport. 1998; 9: 2457-2461Crossref PubMed Scopus (85) Google Scholar). In this study, we have examined the biophysical properties of the disulfide bridge-free recombinant full-length human prion protein. This protein provides a valuable model for probing the role of protein folding intermediates in the PrPC → PrPScconversion. The disulfide bond in huPrP23–231 was reduced by treatment with 100 mm DTT. However, some spectroscopic experiments are not feasible in the presence of concentrated DTT. Dilution (or removal) of the reducing agent could potentially lead to a recreation of the disulfide bond. Therefore, to avoid any ambiguity, in addition to using the reduced protein, the present experiments were also performed with a huPrP23–231 variant in which the two Cys residues were replaced with alanine. Our data show that at neutral pH both the reduced PrP as well as the Cys → Ala variant form insoluble aggregates. The proteins could be recovered in a monomeric form only under the conditions of acidic pH in a NaCl-free buffer. Compared with the PrP with an intact disulfide bond, the monomeric forms of the disulfide-free variants are characterized by a major loss of secondary structure, disruption of native tertiary interactions, a greatly increased affinity for hydrophobic dyes, and a loss of cooperative unfolding transition in urea. Some of these properties are reminiscent of a flexible folding intermediate often referred to as a “molten globule” state (40Ptitsyn O.B. Adv. Protein Chem. 1995; 47: 83-229Crossref PubMed Google Scholar). However, typical molten globules are characterized by a substantially more ordered (native-like) secondary structure than observed for the disulfide bridge-free variants of the prion protein. The finding that removal of the disulfide bridge results in partial unfolding of PrP is not surprising, because disulfide bonds are known to stabilize the folded conformation of many proteins. The contribution of a single disulfide bond to the thermodynamic stability of proteins is usually in the range of several kJ/m(41Fersht A. Structure and Mechanism in Protein Science: A Guide to Enzyme Catalysis and Protein Folding. W. H. Freeman and Company, New York1999Google Scholar), although in certain cases it may be as high as 20 kJ/m (42Bonander N. Leckner J. Guo H. Karlsson B.G. Sjolin L. Eur. J. Biochem. 2000; 267: 4511-4519PubMed Google Scholar). The above described stabilizing effect is believed to be due largely to the reduction of conformational entropy of the unfolded state (41Fersht A. Structure and Mechanism in Protein Science: A Guide to Enzyme Catalysis and Protein Folding. W. H. Freeman and Company, New York1999Google Scholar). Examples of proteins that upon reduction of one or more disulfide bonds adopt molten globule-like or partially unfolded conformation include α-lactalbumin (43Ewbank J.J. Creighton T.E. Biochemistry. 1993; 32: 3694-3707Crossref PubMed Scopus (131) Google Scholar) and papain (44Edwin F. Jagannadham M.V. Biochim. Biophys. Acta. 2000; 1479: 68-82Crossref Scopus (29) Google Scholar). It was reported previously that at acidic pH the reduced recombinant prion protein refolds to a stable monomeric form that is rich in β-sheet structure (29Jackson G.S. Hosszu L.L.P. Power A. Hill A.F. Kenney J. Saibil H. Craven C.J. Waltho J.P. Clarke A.R. Collinge J. Science. 1999; 283: 1935-1937Crossref PubMed Scopus (361) Google Scholar). However, our present results are at odds with this report. We are also not aware of any other protein that would switch from a monomeric α-helical conformation in the oxidized state to a monomeric β-sheet structure in the reduced state. The present study demonstrates that, upon addition of NaCl, the molten-globule-like (disulfide-free) form of huPrP23–231 undergoes a transition to a β-sheet-rich structure. This transition was invariably associated with protein self-association. In contrast to the previous report (29Jackson G.S. Hosszu L.L.P. Power A. Hill A.F. Kenney J. Saibil H. Craven C.J. Waltho J.P. Clarke A.R. Collinge J. Science. 1999; 283: 1935-1937Crossref PubMed Scopus (361) Google Scholar), no monomeric β-sheet-rich conformer could be detected in our experiments with huPrP23–231 and huPrP90–231. The protein self-association reaction and changes in CD spectra occurred very rapidly (within the dead-time of the present experiments), suggesting that the helix → β-sheet transition is an integral part of the oligomerization process. However, the present data do not allow us to exclude the possibility that the self-association step may be preceded by the formation of a transient monomeric β-sheet rich conformer. Such a conformer would be, however, very short-lived and highly prone to aggregation. The finding that the reduced PrP at acidic pH adopts an oligomeric β-sheet structure is in line with previous observations for huPrP90–231 with an intact disulfide bond (28Swietnicki W. Morillas M. Chen S.G. Gambetti P. Surewicz W.K. Biochemistry. 2000; 39: 424-431Crossref PubMed Scopus (216) Google Scholar). However, compared with the latter protein, the reduced form is characterized by a substantially increased propensity to form a polymeric β-sheet-rich form. In contrast to the disulfide bridge-containing PrP, the conversion of the reduced (molten globule-like) protein is triggered by salt alone and does not require denaturing agents such as guanidine HCl or urea. This suggests that destabilization or partial unfolding of the native α-helical conformation is a key factor in the conversion of PrP into a scrapie-like form. The view expressed above is consistent with recent findings that the helix → β-sheet transition of the nonreduced PrP may be modulated by changes in the concentration of urea.2
Referência(s)