Membrane Environment Alters the Conformational Structure of the Recombinant Human Prion Protein
1999; Elsevier BV; Volume: 274; Issue: 52 Linguagem: Inglês
10.1074/jbc.274.52.36859
ISSN1083-351X
AutoresManuel Morillas, Wieslaw Swietnicki, Pierluigi Gambetti, Witold K. Surewicz,
Tópico(s)Neurological diseases and metabolism
ResumoThe prion protein (PrP) in a living cell is associated with cellular membranes. However, all previous biophysical studies with the recombinant prion protein have been performed in an aqueous solution. To determine the effect of a membrane environment on the conformational structure of PrP, we studied the interaction of the recombinant human prion protein with model lipid membranes. The protein was found to bind to acidic lipid-containing membrane vesicles. This interaction is pH-dependent and becomes particularly strong under acidic conditions. Spectroscopic data show that membrane binding of PrP results in a significant ordering of the N-terminal part of the molecule. The folded C-terminal domain, on the other hand, becomes destabilized upon binding to the membrane surface, especially at low pH. Overall, these results show that the conformational structure and stability of the recombinant human PrP in a membrane environment are substantially different from those of the free protein in solution. These observations have important implications for understanding the mechanism of the conversion between the normal (PrPC) and pathogenic (PrPSc) forms of prion protein. The prion protein (PrP) in a living cell is associated with cellular membranes. However, all previous biophysical studies with the recombinant prion protein have been performed in an aqueous solution. To determine the effect of a membrane environment on the conformational structure of PrP, we studied the interaction of the recombinant human prion protein with model lipid membranes. The protein was found to bind to acidic lipid-containing membrane vesicles. This interaction is pH-dependent and becomes particularly strong under acidic conditions. Spectroscopic data show that membrane binding of PrP results in a significant ordering of the N-terminal part of the molecule. The folded C-terminal domain, on the other hand, becomes destabilized upon binding to the membrane surface, especially at low pH. Overall, these results show that the conformational structure and stability of the recombinant human PrP in a membrane environment are substantially different from those of the free protein in solution. These observations have important implications for understanding the mechanism of the conversion between the normal (PrPC) and pathogenic (PrPSc) forms of prion protein. prion protein human prion protein 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoserine, lyso-PC, lysophosphatidylcholine lysophosphatidylserine glycosylphosphatidylinositol Prion diseases, also known as spongiform encephalopathies, are fatal neurodegenerative diseases that can be sporadic, genetically determined, or acquired by infection (1Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13387Crossref PubMed Scopus (5151) Google Scholar, 2Weissmann C. FEBS Lett. 1996; 389: 3-11Crossref PubMed Scopus (145) Google Scholar, 3Gajdusek D.C. Science. 1977; 197: 943-960Crossref PubMed Scopus (723) Google Scholar). One of the hallmarks of prion diseases is cerebral accumulation of an abnormal form of prion protein (PrP),1PrPSc, which is derived from the normal cell surface glycoprotein, PrPC. According to the "protein only" hypothesis, PrPSc is the sole agent responsible for the pathogenesis of spongiform encephalopathies (1Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13387Crossref PubMed Scopus (5151) Google Scholar, 4Prusiner S.B. Science. 1982; 216: 136-144Crossref PubMed Scopus (4103) Google Scholar). The conversion of PrPC into PrPSc occurs as a post-translational process without any detectable covalent modifications to the protein molecule (5Stahl 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 (536) Google Scholar). The chemical identity of PrPC and PrPSc contrasts with their markedly different physical properties. Thus, PrPC is monomeric and readily degradable by proteinase K, whereas PrPSc forms highly insoluble aggregates and shows a remarkable resistance to proteolytic digestion (6Meyer 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 (517) Google Scholar, 7Oesch B. Westaway D. Walchli M. McKinley M.P. Kent S.B.H. 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 (1253) Google Scholar). These characteristics most likely reflect different conformations of the two protein isoforms. Indeed, optical spectroscopic data indicate that PrPC isolated from normal brain is highly α-helical, whereas PrPSc contains a large proportion of β-sheet structure (8Pan 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 (2076) Google Scholar, 9Caughey B.W. Dong A. Bhat K.S. Ernst D. Hayes S.F. Caughey W.S. Biochemistry. 1991; 30: 7672-7680Crossref PubMed Scopus (742) Google Scholar, 10Safar J. Roller P.P. Gajdusek D.C. Gibbs C.J. Protein Sci. 1993; 2: 2206-2216Crossref PubMed Scopus (174) Google Scholar). It is believed that the conversion of PrPC into PrPSc constitutes the key molecular event in the pathogenesis of prion diseases. However, the molecular mechanism of the conformational transition between the normal and pathogenic forms of prion protein remains poorly understood. Structural and biophysical studies of prion protein have been hampered by difficulties encountered in obtaining highly purified material from the brain. Only recently have expression systems been developed for large scale production of the recombinant prion protein inEscherichia coli (11Hornemann S. Glockshuber R. J. Mol. Biol. 1996; 262: 614-619Crossref Scopus (74) Google Scholar, 12Zhang 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, 13Swietnicki W. Petersen R.B. Gambetti P. Surewicz W.K. J. Biol. Chem. 1997; 272: 27517-27520Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 14Horneman S. Korth C. Oesch B. Riek R. Wider G. Wuthrich K. Glockshuber R. FEBS Lett. 1997; 413: 277-281Crossref PubMed Scopus (167) Google Scholar, 15Zahn R. von Schroetter C. Wuthrich K. FEBS Lett. 1997; 417: 400-404Crossref PubMed Scopus (248) Google Scholar). Following this development, the three-dimensional structure of the recombinant PrP in solution was determined by NMR spectroscopy (16Riek R. Hornemann S. Wider G. Billeter M. Glockshuber R. Wüthrich K. Nature. 1996; 382: 180-182Crossref PubMed Scopus (1128) Google Scholar, 17James T.L. Liu H. Ulyanov N.B. Farr-Jones S. Zhang H. Donne D.G. Kaneko K. Groth D. Mehlhorn I. Prusiner S.B. Cohen F.E. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 10086-10091Crossref PubMed Scopus (429) Google Scholar, 18Donne 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 (637) Google Scholar, 19Riek R. Hornemann S. Wider G. Glockshuber R. Wuthrich K. FEBS Lett. 1997; 413: 282-288Crossref PubMed Scopus (664) Google Scholar). The availability of the recombinant protein also allowed biophysical studies on the thermodynamic stability and folding pathway of the cellular prion protein (11Hornemann S. Glockshuber R. J. Mol. Biol. 1996; 262: 614-619Crossref Scopus (74) Google Scholar, 12Zhang 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, 13Swietnicki W. Petersen R.B. Gambetti P. Surewicz W.K. J. Biol. Chem. 1997; 272: 27517-27520Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 20Swietnicki W. Petersen R.B. Gambetti P. Surewicz W.K. J. Biol. Chem. 1998; 47: 31048-31052Abstract Full Text Full Text PDF Scopus (176) Google Scholar, 21Hornemann S. Glockshuber R. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 6010-6014Crossref PubMed Scopus (240) Google Scholar, 22Liemann S. Glockshuber R. Biochemistry. 1999; 38: 3258-3267Crossref PubMed Scopus (295) Google Scholar, 23Jackson 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 (365) Google Scholar). However, a critical limitation of all these studies is that they have been performed in an aqueous solution, whereas PrPC in a living cell is associated with cellular membranes (24Stahl N. Borchelt D.R. Hsiao K. Prusiner S.B. Cell. 1987; 51: 229-240Abstract Full Text PDF PubMed Scopus (907) Google Scholar). To bridge the gap between the cellular studies with the authentic PrPC and the experiments with the recombinant PrP in an aqueous solution, we have characterized the biophysical properties of the recombinant prion protein associated with model lipid membranes. We provide evidence that the membrane environment strongly affects, in a pH-dependent manner, the conformational structure and stability of the protein. These findings have important implications for understanding the conversion of PrPC to PrPSc. cDNA coding for huPrP-(23–231), huPrP-(23–145), and huPrP-(90–231) was amplified from a plasmid pVZ21 (25Petersen R.B. Parchi P. Richardson S.L. Ulrig C.B. Gambetti P. J. Biol. Chem. 1996; 271: 12661-12668Abstract Full Text Full Text PDF PubMed Scopus (136) Google Scholar) by polymerase chain reaction. The primers used were 5′-CAT GGT GGTGGA TCC GGG TCA AGG AGG and 5′-GAG GAT CGA GCT GAG AAT TCC TCT CCA CCT G for huPrP-(90–231), and 5′-GTG ACC TGG GGG ATC CGA AGA AGC GCC CG and 5′-GAG GAT CGA GCT GAG AAT TCC TCT CCA CCT G for huPrP-(23–231) and huPrP-(23–145). After digestion with BamHI and EcoRI enzymes (sites indicated in bold), the restricted fragments and a double-stranded linker (5′P-CT AGC CTG GTT CCG CGT GGT TCG-3′/5′-P-G ATC CGA ACC ACG CGG AAC CAG G-3′) were ligated into pRSETB vector (Invitrogen) that was predigested with EcoRI andNheI enzymes. The final constructs coded for appropriate huPrP fragments fused to the N-terminal linker containing His6 tail and a thrombin cleavage site. A Gly-Ser-Asp-Pro extension at the N terminus remained after cleavage of the linker. DNA sequences of all constructs were verified by automated DNA sequencing. Protein was expressed and purified essentially as described previously (15Zahn R. von Schroetter C. Wuthrich K. FEBS Lett. 1997; 417: 400-404Crossref PubMed Scopus (248) Google Scholar). The modifications to the above protocol included different thrombin cleavage conditions of the His tail (1-h treatment at room temperature with 10 units of thrombin/mg protein in 10 mm phosphate, pH 6.4) and subsequent ion exchange chromatography on a CM-Sepharose column using a linear 0–500 mm gradient of NaCl in 10 mm phosphate, pH 6.4. The purity of the final product was better than 98% as judged by SDS-polyacrylamide gel electrophoresis. The identity of each protein was further confirmed by mass spectrometry. Protein concentration was determined spectrometrically using the molar extinction coefficient ε280 of 56,795, 21,495, and 43,600 m−1 cm−1 for huPrP-(23–231), huPrP-(90–231), and huPrP-(23–145), respectively. To obtain small unilamellar vesicles, lipid mixture of a desired composition was dissolved in chloroform and dried under a stream of nitrogen. The resulting film of dried lipid was hydrated in an appropriate buffer, vortexed, and finally sonicated on ice for approximately 15 min using a probe-type sonifier. Lysophospholipid micelles were prepared by dissolving the mixture of lysophospholipids in chloroform/methanol/water (4:4:2). After drying under nitrogen, the lipid film was dispersed by vortexing in a desired buffer. Binding of huPrP variants to lipid vesicles was studied at room temperature by fluorescence spectroscopy. For this purpose, small aliquots of concentrated vesicle suspension were successively added to 1.8 μm protein solution in a desired buffer. After each addition of lipid, the solution was thoroughly mixed and left to equilibrate for 5 min. Fluorescence emission spectra were measured on an SLM 8100 spectrofluorometer using a 4-mm cell and an excitation wavelength of 280 nm (huPrP-(23–231) and huPrP-(23–145)) or 295 nm (huPrP-(90–231)). Spectra were corrected for light scattering effect by subtracting lipid blanks in the same buffer. The fluorescence titration curves were analyzed as described previously (26Surewicz W.K. Epand R.M. Biochemistry. 1984; 23: 6072-6077Crossref PubMed Scopus (96) Google Scholar, 27Butko P. Huang F. Pusztai-Carey M. Surewicz W.K. Biochemistry. 1997; 36: 12862-12868Crossref PubMed Scopus (49) Google Scholar), providing an apparent binding constantK d /n, where K d is the dissociation constant of the lipid-protein complex, and n is the number of binding sites per lipid. The ratioK d /n represents the reciprocal of the classic first association constant for the protein-lipid interaction and provides a measure of the overall affinity of a protein to the membrane. Tryptophan fluorescence quenching experiments were performed by titrating the solution of free and membrane-associated proteins with freshly prepared solution of acrylamide. Fluorescence intensities at each acrylamide concentration were measured at the wavelength corresponding to the emission maximum of each protein and corrected for dilution, blanks, and the inner filter effect. The excitation wavelength was 295 nm. Quenching curves were analyzed in terms of the effective Stern-Volmer constantK SV(eff); this effective parameter represents a weighted average of the quenching constants of individual Trp residues and may contain contributions from both dynamic and static quenching processes. K SV(eff values were calculated as described previously (28Etfink M.R. Ghiron C.A. Anal. Biochem. 1981; 114: 199-227Crossref PubMed Scopus (1625) Google Scholar) from the inverse slope of theF O/ΔF versus1/[Q] plots, where F O andF are the fluorescence intensities in the absence and presence of the quencher, respectively, ΔF =F O − F, and [Q] is the molar concentration of acrylamide. The effect of the lipid environment on the secondary structure of huPrP variants was studied by following changes in far UV CD spectra. The measurements were performed at 25 °C using a JASCO-600 spectropolarimeter and a 1-mm path length cell. Typically, six spectra were averaged and smoothed to improve signal to noise ratio. Thermal stability of the free and lipid-associated proteins was studied by following changes in ellipticity at 222 nm as a function of temperature. The measurements were performed using a JASCO-600 spectropolarimeter equipped with a 1-mm water-jacketed cylindrical cell and a computer-controlled water bath. If not mentioned otherwise, the following buffers were used throughout this study: pH 7, 17 mm potassium phosphate; pH 6, 40 mm potassium phosphate; pH 4 and 5, 50 mm sodium acetate. All buffers contained 130 mmNaCl and had a constant ionic strength of 180 mm. Tryptophan residues provide a convenient spectroscopic probe that allows the measurements of protein-membrane binding by fluorescence spectroscopy (26Surewicz W.K. Epand R.M. Biochemistry. 1984; 23: 6072-6077Crossref PubMed Scopus (96) Google Scholar, 27Butko P. Huang F. Pusztai-Carey M. Surewicz W.K. Biochemistry. 1997; 36: 12862-12868Crossref PubMed Scopus (49) Google Scholar). The full-length huPrP-(23–231) contains seven tryptophans that are located within the N-terminal portion of the protein (residues 23–99). The fluorescence emission spectrum of huPrP-(23–231) in an aqueous solution (pH range between 4 and 7) has a maximum at 348 nm and is indicative of a highly polar environment of the Trp residues. This is consistent with the NMR structural data showing that the N-terminal region is highly flexible and largely unordered (18Donne 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 (637) Google Scholar). Incubation of the protein with membrane vesicles consisting solely of zwitterionic phosphatidylcholine (POPC) did not result in any measurable spectral changes, suggesting the lack of protein-membrane interaction. However, the fluorescence spectra changed markedly when the protein was titrated with the vesicles containing acidic phosphatidylserine (POPS). As shown in Fig. 1, upon addition of the vesicles containing the latter phospholipid the emission maximum of huPrP-(23–231) is shifted to a shorter wavelength, and there is an increase in the fluorescence intensity. The observed blue shift reflects the increase in the hydrophobicity of the Trp environment and is indicative of protein binding to the membrane. The representative titration curves obtained at different pH values by measuring changes in the fluorescence intensity or the wavelength of the emission maximum at increasing concentrations of POPC/POPS (2:1) vesicles are shown in Fig. 1. These curves indicate that the interaction between huPrP-(23–231) and phosphatidylserine-containing vesicles is pH-dependent, becoming considerably stronger under acidic conditions. The apparent binding parameters derived from data of Fig. 1 A are shown in Table I.Table IProtein-membrane binding constantspHK d /nhuPrP-(23–231)huPrP-(23–145)huPrP-(90–231)μm7.0605 ± 10592 ± 10No interaction6.0419 ± 2179 ± 8NDaND, not determined.5.0184 ± 4575 ± 1155 ± 44.0109 ± 1642 ± 6NDa ND, not determined. Open table in a new tab Binding of huPrP-(23–231) to lipid vesicles was further verified by filtration on a 100-kDa cut-off Microcon membrane (Amicon). Free protein as well as the protein preincubated with POPC vesicles easily passed through the membrane, and all of it was recovered in the filtrate. However, when huPrP-(23–231) was preincubated with the vesicles containing 33 mol % of acidic POPS (at a saturating lipid concentration as assessed from the titration data), essentially all protein, together with the vesicles, was retained on the filter. These data provide independent evidence for the binding of prion protein to acidic lipid-containing vesicles. Furthermore, it confirms the validity of the fluorescence-based approach used in this study. To gain insight into the mechanism of huPrP-(23–231)-membrane interaction, the fluorescence titration experiments were performed using membrane vesicles containing different proportions of acidic phosphatidylserine. As shown in Fig. 2, up to approximately 50 mol % POPS, the binding affinity is strongly dependent on the density of negatively charged lipid in the bilayer. Such behavior suggests that the protein contains multiple binding sites that can combine with acidic lipids (29Mosior M. McLaughlin S. Biophys. J. 1991; 60: 149-159Abstract Full Text PDF PubMed Scopus (110) Google Scholar). To identify the lipid-interacting region(s) in PrP, binding experiments similar to those described above were performed for prion protein fragments huPrP-(23–145) and huPrP-(90–231). As shown in Fig. 3 A, the N-terminal part 23–145 binds to acidic lipid-containing membranes both at acidic and neutral pH. The apparent membrane affinity of this fragment shows relatively little dependence on pH. In contrast to huPrP-(23–145), the fluorescence spectrum of huPrP-(90–231) at neutral pH remained essentially unchanged in the presence of the vesicles, suggesting the lack of binding to the membrane. Association of the latter protein with POPS-containing vesicles could be detected only at acidic pH. Under such conditions, the emission maximum of Trp-99 in huPrP-(90–231) was shifted to lower wavelength (Fig. 3 B), with little change in the fluorescence intensity. The apparent lipid binding affinities for PrP fragments appear to be higher than those for huPrP-(23–231), suggesting that some of the potential lipid-binding sites may be masked in the full-length protein. It should also be noted that in the presence of huPrP-(23–231) or its fragments there was a time-dependent increase in light scattering by the vesicles, especially at low pH. Although this effect was not explored further in the present study, it suggests that protein binding may induce aggregation and/or fusion of phospholipid vesicles. To characterize further the interaction between the recombinant prion protein variants and membrane lipids, we have used the technique of fluorescence quenching with a polar quencher acrylamide. This approach allows one to assess the exposure of Trp residues in proteins (28Etfink M.R. Ghiron C.A. Anal. Biochem. 1981; 114: 199-227Crossref PubMed Scopus (1625) Google Scholar). Fig. 4 shows representative Stern-Volmer plots for acrylamide quenching of Trp fluorescence of various huPrP variants (huPrP-(23–231), huPrP-(23–145), and huPrP-(90–231)) at pH 5 in the absence and presence of membrane vesicles at a saturating lipid concentration. The effective quenching constants derived from these curves are shown in Table II. TheK SV(eff) values for the free proteins in buffer are in the range of 10–12 m−1 and indicate a complete exposure of Trp residues to water (28Etfink M.R. Ghiron C.A. Anal. Biochem. 1981; 114: 199-227Crossref PubMed Scopus (1625) Google Scholar). However, the accessibility of all three prion protein variants to acrylamide quenching was markedly reduced in the presence of POPS-containing vesicles, clearly indicating protein binding to the membrane. The extent of the protection from acrylamide quenching as well as the magnitude of the lipid-induced blue shift of λmax(approximately 7 nm) suggest that the protein interacts largely with the membrane surface, with possible penetration into the lipid head group region but not the hydrophobic interior of the bilayer. A similar effect was observed at neutral pH, with the exception that under these conditions very little protection from the quencher was found for huPrP-(90–231) (Table II). These data support the notion that huPrP-(23–231) and huPrP-(23–145) interact with membrane vesicles both at acidic and neutral pH (although with different affinities), whereas the C-terminal region 90–231 shows appreciable affinity for lipids only at acidic pH.Table IIFluorescence quenching constants for prion protein variantsSystemK SV (eff)huPrP-(23–231)huPrP-(23–145)huPrP-(90–231)m −1Protein alone, pH 7.012 ± 0.611.9 ± 0.412.1 ± 0.5Protein + vesicles, pH 7.04.5 ± 0.54.3 ± 0.410.6 ± 0.6Protein alone, pH 5.011 ± 0.410.3 ± 0.511.6 ± 0.6Protein + vesicles, pH 5.04.1 ± 0.44.0 ± 0.53.9 ± 0.5 Open table in a new tab The secondary structure of proteins in a membrane-like environment was studied by CD spectroscopy. To avoid potential distortion of circular dichroism spectra due to light scattering on membrane vesicles, single chain lysophospholipids were used in the CD experiments. Since lysophospholipids form relatively small micellar structures, they produce much less light scattering than phospholipid vesicles. In a series of control fluorescence experiments, it has been verified that the binding characteristics of PrP variants to lysophospholipid micelles are very similar to those described for the corresponding phospholipid membranes (data not shown for brevity). In accord with the previous data (14Horneman S. Korth C. Oesch B. Riek R. Wider G. Wuthrich K. Glockshuber R. FEBS Lett. 1997; 413: 277-281Crossref PubMed Scopus (167) Google Scholar), the CD spectrum of huPrP-(23–231) in buffer shows a double minimum at approximately 222 and 208 nm (Fig. 5); this spectrum reflects a largely α-helical structure of the protein. Consistent with the binding data, lyso-PC alone did not have any measurable effect on the protein CD spectrum. However, in the presence of acidic lyso-PS (or a mixture of lyso-PS with lyso-PC), there was an increase in the negative ellipticity above 200 nm and a small increase in the positive ellipticity at lower wavelengths. The above spectral changes (Fig. 5 A) suggest that the lipid environment produces some rearrangement in the backbone conformation of the protein, resulting in an increased proportion of ordered secondary structure. To identify the domain of the protein involved in this conformational transition, CD spectra in the presence and absence of lyso-PS-containing micelles were measured for huPrP-(23–145) and huPrP-(90–231). The spectrum of huPrP-(23–145) in buffer at pH 7 or 5 has a minimum at approximately 197 nm and is characteristic of an essentially random conformation. The lack of ordered secondary structure in the N-terminal region is in accord with the NMR data for hamster and murine PrP (18Donne 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 (637) Google Scholar, 19Riek R. Hornemann S. Wider G. Glockshuber R. Wuthrich K. FEBS Lett. 1997; 413: 282-288Crossref PubMed Scopus (664) Google Scholar). Upon addition of lyso-PS-containing micelles, there is a marked increase in the negative ellipticity in the entire region between 200 and 240 nm, and the minimum of the CD spectrum shifts to approximately 205 nm (Fig. 5 B). This indicates that in a lipid environment at least part of the 23–145 region adopts a more ordered conformation. Although the present data are insufficient to characterize fully the molecular nature of this conformational transition, the increase in negative ellipticity between 200 and 240 nm is suggestive that a fragment of huPrP-(23–145) may form α-helix or a β-sheet structure. In contrast to huPrP-(23–231) and huPrP-(23–145), the CD spectrum of the C-terminal domain huPrP-(90–231) remains unchanged upon incubation with lyso-PS-containing micelles (Fig. 5 C). Altogether, these data indicate that the lipid-induced increase in the CD signal of the full-length PrP reflects a conformational change (partial ordering) within the N-terminal region of the protein. The effect of lipid binding on the stability of huPrP-(23–231) was studied by following changes in the ellipticity at 222 nm as a function of temperature (19Riek R. Hornemann S. Wider G. Glockshuber R. Wuthrich K. FEBS Lett. 1997; 413: 282-288Crossref PubMed Scopus (664) Google Scholar). In the absence of lipid the protein undergoes a highly cooperative thermal transition, with a midpoint unfolding temperature, T m, of 71.8 and 71.2 °C at pH 7 and 5, respectively (Fig. 6). This transition reflects thermal denaturation of the folded C-terminal domain of prion protein. The unfolding curves remained essentially unchanged in the presence of lyso-PC micelles. However, binding of huPrP-(23–231) to lyso-PS-containing micelles resulted in a structural destabilization of the protein. Although at neutral pH the above effect was relatively small (reduction of T m to 68.8 °C), lipid-induced protein destabilization becomes much more dramatic at pH 5. In the latter case, the apparent midpoint unfolding temperature of huPrP-(23–231) was reduced upon lipid binding to approximately 54 °C, and there was a drastic reduction in the cooperativity of the thermal unfolding (Fig. 6 B). Under the present experimental conditions, the thermal denaturation of the free protein was essentially reversible (recovery of the original CD signal after cooling better than 80%). However, considerably smaller reversibility was found for the protein in a lipid environment. Therefore, no attempt was made to derive thermodynamic parameters from the unfolding curves of Fig. 6. A wealth of data indicate that the key molecular event in the pathogenesis of prion disorders is a conformational change in PrPC, resulting in the conversion of a portion of the protein molecule from an α-helical conformation to a β-sheet-rich, proteinase K-resistant oligomeric structure (1Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13387Crossref PubMed Scopus (5151) Google Scholar, 2Weissmann C. FEBS Lett. 1996; 389: 3-11Crossref PubMed Scopus (145) Google Scholar, 8Pan 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 (2076) Google Scholar, 9Caughey B.W. Dong A. Bhat K.S. Ernst D. Hayes S.F. Caughey W.S. Biochemistry. 1991; 30: 7672-7680Crossref PubMed Scopus (742) Google Scholar, 10Safar J. Roller P.P. Gajdusek D.C. Gibbs C.J. Protein Sci. 1993; 2: 2206-2216Crossref PubMed Scopus (174) Google Scholar). A critical step toward dissecting the complex mechanism of this conversion reaction is to understand the structural and biophysical properties of PrPC and its folding intermediates. The recombinant prion protein has been recently used as a structural model of PrPC, providing fundamentally important information about the three-dimensional structure of the protein in an aqueous solution (16Riek R. Hornemann S. Wider G. Billeter M. Glockshuber R. Wüthrich K. Nature. 1996; 382: 180-182Crossref PubMed Scopus (1128) Google Scholar, 17James T.L. Liu H. Ulyanov N.B. Farr-Jones S. Zhang H. Donne D.G. Kaneko K. Groth D. Mehlhorn I. Prusiner S.B. Cohen F.E. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 10086-10091Crossref PubMed Scopus (429) Google Scholar, 18Donne 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 (637) Google Scholar, 19Riek R. Hornemann S. Wider G. Glockshuber R. Wuthrich K. FEBS Lett. 1997; 413: 282-288Crossref PubMed Scopus (664) Google Scholar), its thermodynamic stability, as well as the folding pathway (11Hornemann
Referência(s)