Acidic pH and Detergents Enhance in Vitro Conversion of Human Brain PrPC to a PrPSc-like Form
2002; Elsevier BV; Volume: 277; Issue: 46 Linguagem: Inglês
10.1074/jbc.m203611200
ISSN1083-351X
AutoresWen-Quan Zou, Neil R. Cashman,
Tópico(s)RNA regulation and disease
ResumoIn the presence of a low concentration of denaturants or detergents, acidic pH triggers a conformational transition of α-helices into β-sheets in recombinant prion protein (PrP), likely mimicking some aspects of the transformation of host-encoded normal cellular PrP (PrPC) into its pathogenic isoform (PrPSc). Here we observed the effects of acidic pH and guanidine hydrochloride (GdnHCl) on the physicochemical and structural properties of PrPC derived from normal human brain and determined the ability of the acid/GdnHCl-treated PrP to form a proteinase K (PK)-resistant species in the absence and presence of PrPSc template. After treatment with 1.5 mGdnHCl at pH 3.5, PrPC from normal brain homogenates was converted into a detergent-insoluble form similar to PrPSc. Unlike PrPSc, however, the treated brain PrPC was protease-sensitive and retained epitope accessibility to monoclonal antibodies 3F4 and 6H4. Brain PrPC treated with acidic pH/GdnHCl acquired partial PK resistance upon further treatment with low concentrations of sodium dodecyl sulfate (SDS). Formation of this PrPSc-like isoform was greatly enhanced by incubation with trace quantities of PrPSc from Creutzfeldt-Jakob disease brain. Acid/GdnHCl-treated brain PrP may constitute a "recruitable intermediate" in PrPSc formation. Further structural rearrangement seems essential for this species to acquire PK resistance, which can be promoted by the presence of a PrPSc template. In the presence of a low concentration of denaturants or detergents, acidic pH triggers a conformational transition of α-helices into β-sheets in recombinant prion protein (PrP), likely mimicking some aspects of the transformation of host-encoded normal cellular PrP (PrPC) into its pathogenic isoform (PrPSc). Here we observed the effects of acidic pH and guanidine hydrochloride (GdnHCl) on the physicochemical and structural properties of PrPC derived from normal human brain and determined the ability of the acid/GdnHCl-treated PrP to form a proteinase K (PK)-resistant species in the absence and presence of PrPSc template. After treatment with 1.5 mGdnHCl at pH 3.5, PrPC from normal brain homogenates was converted into a detergent-insoluble form similar to PrPSc. Unlike PrPSc, however, the treated brain PrPC was protease-sensitive and retained epitope accessibility to monoclonal antibodies 3F4 and 6H4. Brain PrPC treated with acidic pH/GdnHCl acquired partial PK resistance upon further treatment with low concentrations of sodium dodecyl sulfate (SDS). Formation of this PrPSc-like isoform was greatly enhanced by incubation with trace quantities of PrPSc from Creutzfeldt-Jakob disease brain. Acid/GdnHCl-treated brain PrP may constitute a "recruitable intermediate" in PrPSc formation. Further structural rearrangement seems essential for this species to acquire PK resistance, which can be promoted by the presence of a PrPSc template. An insoluble, β-sheet-rich isoform of the prion protein (herein generically designated PrPSc) 1The abbreviations used are: PrPSc, pathogenic prion protein; PrPC, cellular prion protein; PK, proteinase K; rPrP, recombinant prion protein; hurPrP, recombinant human PrP; GdnHCl, guanidine hydrochloride; CJD, Creutzfeldt-Jakob disease; PMCA, protein misfolding cyclic amplification; PBS, phosphate-buffered saline; BSA, bovine serum albumin is the only known component of the infectious prion particle associated with the prion diseases, which are a group of transmissible, fatal neurodegenerative diseases in humans and animals. PrPSc is derived from its normal cellular isoform (PrPC), which is rich in α-helical structure, by a posttranslational process involving a conformational transition (1Prusiner S.B. Scott M.R. DeArmond S.J. Cohen F.E. Cell. 1998; 93: 337-348Abstract Full Text Full Text PDF PubMed Scopus (831) Google Scholar). While the primary structure of PrPC is identical to that of PrPSc, secondary and tertiary structural changes are responsible for the distinct physicochemical properties of the two isoforms. PrPC exists as a detergent-soluble monomer and is readily degraded by protease K (PK), whereas the infectious isoform PrPSc forms detergent-insoluble aggregates and shows high resistance to PK digestion and to phosphatidylinositol-specific phospholipase C-mediated release (2Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13383Crossref PubMed Scopus (5168) Google Scholar). According to the "protein only" theory of prion infectivity, PrPC is converted to PrPSc by a template-directed process catalyzed by PrPSc. This disease-related in vivo transition has been modeled in vitro, in which PrPC can be converted to a protease-resistant form by contact with PrPSc (3Kocisko D.A. Come J.H. Priola S.A. Chesebro B. Raymond G.J. Lansbury P.T. Caughey B. Nature. 1994; 370: 471-474Crossref PubMed Scopus (792) Google Scholar, 4Saborio G.P. Soto C. Kascsak R.J. Levy E. Kascsak R. Harris D.A. Frangione B. Biochem. Biophys. Res. Commun. 1999; 258: 470-475Crossref PubMed Scopus (61) Google Scholar). Recently, it has been reported that this conversion can be promoted by protein misfolding cyclic amplification (PMCA), a process analogous to the polymerase chain reaction for nucleic acids (5Saborio G.P. Permanne B. Soto C. Nature. 2001; 411: 810-813Crossref PubMed Scopus (1022) Google Scholar). Studies using recombinant PrP (rPrP) have indicated that the structural transition of α-helix to β-sheet and concomitant self-association can be triggered by acidic pH combined with detergents in vitro (6Swietnicki W. Petersen R. Gambetti P. Surewicz W.K. J. Biol. Chem. 1997; 272: 27517-27520Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 7Jackson G.S. Hill A.F. Joseph C. Hosszu L. Power A. Waltho J.P. Clarke A.R. Collinge J. Biochim. Biophys. Acta. 1999; 1431: 1-13Crossref PubMed Scopus (113) Google Scholar, 8Hornemann S. Glockshuber R. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 6010-6014Crossref PubMed Scopus (240) Google Scholar). We have also demonstrated that acid-induced conformational transition and aggregation may be associated with protonation of the acidic amino acids aspartate (Asp) and glutamate (Glu) using a peptide (195–213) corresponding to the C-terminal region of PrP (9Zou W.Q. Yang D.S. Fraser P.E. Cashman N.R. Chakrabartty A. Eur. J. Biochem. 2001; 268: 4885-4891Crossref PubMed Scopus (7) Google Scholar). These data suggest that PrP in acidic organelles may participate in specific folding pathways, forming molecular species structurally and physicochemically distinct from PrPC. Although studies using rPrP have provided important information about PrP structural transition, three-dimensional structure, thermodynamic stability, and folding pathways, the absence of co-factor molecules and posttranslational modifications, such as N-linked glycans and the C-terminal glycosylphosphatidylinositol anchor, may confound attempts to model faithfully in vivo PrP conversion events. In the present study, we have observed the effect of acidic pH and guanidine hydrochloride (GdnHCl) on the physicochemical and structural properties of cellular PrP derived from normal human brain. Importantly, we find that acid/GdnHCl-treated human brain PrP is a superior substrate for in vitro conversion than untreated PrP. Phenylmethylsulfonyl fluoride and proteinase K were purchased from Sigma. Magnetic beads (Dynabeads M-280 Tosylactivated) were from Dynal Co. (Dynal AS, Oslo, Norway). Mouse monoclonal antibody 6H4 from Prionics Co. (Zürich, Switzerland) recognizes the sequence DYEDRYYRE in the prion protein (human PrP residues 144–152). Mouse monoclonal antibody 3F4 from Signet Laboratories, Inc. (Dedham, MA) recognizes an epitope of human PrP residues 109–112 including residues MKHV. Horseradish peroxidase-conjugated sheep anti-mouse antibody was purchased fromAmersham Biosciences. Necropsied human brain tissue was collected within 24 h of death. The normal human brain was obtained from an individual determined by histology to be free of neurological disorders, and a prion-infected brain was from an individual with Creutzfeldt-Jakob disease (CJD) confirmed by histological examination and Western blot analysis to show the presence of PrPSc. Both samples were homozygous for Met/Met at codon 129. Mouse brains were from wild-type and PrP−/− mice. All tissues were frozen immediately after collection and stored at −80 °C. 10% (w/v) brain homogenates were prepared in lysis buffer (100 mm NaCl, 10 mm EDTA, 0.5% Nonidet P 40 (Nonidet P-40), 0.5% sodium deoxycholate, 10 mm Tris-HCl, pH 7.5). Samples were mixed with an equal volume of 2× electrophoresis loading buffer (6% sodium dodecyl sulfate (SDS), 5% 2-mercaptoethanol, 4 mm EDTA, 20% glycerol, 125 mm Tris-HCl, pH 6.8) and boiled for 10 min. Proteins were separated by 12% SDS-PAGE and electrotransferred onto polyvinylidene difluoride membranes at 125 mA for 2 h. The membranes were blocked with 5% nonfat milk in TBST (10 mm Tris-HCl, pH 7.6, 150 mm NaCl, 0.05% Tween 20) overnight at 4 °C or 1 h at 37 °C prior to incubation with antibodies. Membrane-bound proteins were probed with 6H4 antibody at 1:5,000 or with 3F4 antibody at 1:50,000. Washing with TBST, the blot was incubated with horseradish peroxidase-conjugated sheep anti-mouse antibody at 1:3,000. After washing with TBST, the proteins were visualized by enhanced chemiluminescence + plus (ECL + Plus, Amersham Biosciences). 100 μl of 10% brain homogenate was mixed with an equal volume of 3.0m GdnHCl (final concentration 1.5 m) in PBS at pH 7.4 or pH 3.5 adjusted with 1 n HCl, followed by incubation at room temperature with shaking. After 5 h, samples were mixed with 5 volumes of prechilled methanol and incubated at −20 °C for 2 h to precipitate the proteins. The samples were subjected to centrifugation at 16,000 × g for 20 min at 4 °C to remove the acidic buffer and GdnHCl, and pellets were resuspended in 100 μl of lysis buffer or in 100 μl of 0.05% SDS, 0.5% Triton X-100 in PBS, pH 7.4, according to the experimental design. The samples treated at pH 7.4 were designated as mock-treated samples. 100 μl of the treated or mock-treated sample in lysis buffer was centrifuged at 100,000 × g (Beckman, TL-100 Ultracentrifuge) at 4 °C for 1 h. Supernatants (containing detergent-soluble PrP) were transferred to clean tubes and pellets (containing detergent-insoluble PrP) were resuspended in an equal volume of lysis buffer. The distribution of detergent-soluble PrP or detergent-insoluble PrP was determined by immunblotting. To determine the PK-resistance of the treated PrP, 20 μl of sample was incubated with PK at 100 μg/ml for 1 h at 37 °C, and the digestion reaction was terminated by addition of phenylmethylsulfonyl fluoride to 2 mm final concentration. The sample was mixed with equal volumes of loading buffer, boiled for 10 min, and subjected to SDS-PAGE and immunoblotting. Anti-PrP monoclonal antibodies (6H4 and 3F4) at 30 μg/ml were coupled to magnetic Dynabeads in PBS at 37 °C for 20 h and washed twice with washing buffer (0.1% BSA/PBS). The antibody-conjugated beads were incubated with 0.1% BSA, 0.2 m Tris-HCl, pH 8.5 at 37 °C for 4 h to block nonspecific binding sites and then washed twice with 0.1% BSA/PBS. The antibody-conjugated beads were stored in PBS at 4 °C. For immunoprecipitation of PrP, 50 μl of antibody-conjugated beads was incubated with 945 μl of lysis buffer in the presence of 5 μl of 10% (w/v) brain homogenate (mock-treated, acidic pH/GdnHCl-treated, or CJD brain) at room temperature for 3 h. The immune complex-containing beads were washed three times with washing buffer (2% Nonidet P-40, 2% Tween 20, PBS, pH 7.4). After the last wash, all liquids were removed and 30 μl of loading buffer was added (without reducing agents such as dithiothreitol and β-mercaptoethanol to prevent antibody fragments from eluting off the beads). The samples were heated at 95 °C for 5 min and then centrifuged at 3,000 rpm for 3 min. The supernatants were subjected to SDS-PAGE and immunoblotting. To perform in vitro conversion of PrP, proteins precipitated by prechilled methanol were resuspended in equal volumes of 0.05% SDS, 0.5% Triton X-100 in PBS, pH 7.4, rather than in lysis buffer. This buffer has been used in PMCA of PrPSc (5Saborio G.P. Permanne B. Soto C. Nature. 2001; 411: 810-813Crossref PubMed Scopus (1022) Google Scholar). Conversion in vitro was performed in an 80-μl volume of the appropriate test substrate material (79.4 μl of sample and 0.6 μl of 10% CJD brain homogenate for the template-directed experiments), and the sample was incubated in a thermomixer at 37 °C for 12 h with shaking. After PK digestion at 37 °C for 1 h and boiling in loading buffer, the samples were subjected to SDS-PAGE and immunoblotting. Consistent with the observations of other groups (6Swietnicki W. Petersen R. Gambetti P. Surewicz W.K. J. Biol. Chem. 1997; 272: 27517-27520Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 7Jackson G.S. Hill A.F. Joseph C. Hosszu L. Power A. Waltho J.P. Clarke A.R. Collinge J. Biochim. Biophys. Acta. 1999; 1431: 1-13Crossref PubMed Scopus (113) Google Scholar, 8Hornemann S. Glockshuber R. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 6010-6014Crossref PubMed Scopus (240) Google Scholar), we have found that recombinant mouse PrP-(23–231) undergoes an α-helix to β-sheet conformational transition upon treatment with 1.5 m GdnHCl in PBS at pH 3.5, as indicated by circular dichroism spectroscopy. 2W. Q. Zou and N. R. Cashman, unpublished data. We examined the effect of these conditions on the physicochemical properties of cellular PrP derived from normal human brain tissue. After removal of acidic buffer and detergent from normal human brain homogenates, the detergent-soluble (S) and insoluble fractions (P) of the samples in lysis buffer were separated by ultracentrifugation. The distribution of PrP in the supernatants and in the pellets was determined by immunoblotting. As shown inpanel a of Fig. 1, while PrPC in the mock-treated samples was predominantly found in the detergent-soluble fraction, PrP in the treated brain homogenates was recovered mostly in the detergent-insoluble fraction. Therefore, acidic pH and GdnHCl cannot only induce a conformational transition of the recombinant protein, but can also induce a physical conversion from detergent-soluble PrPC into detergent-insoluble PrPSc-like species in native, properly posttranslationally modified PrP from brain tissue. Interestingly, treated protein retained the property of insolubility even when returned to physiological pH for at least 7 days (Fig. 1 a and not shown). To further characterize the effect of pH and GdnHCl on the solubility of PrP in non-denaturing detergents, we performed titrations of these two treatments on human brain homogenates. Panel b of Fig. 1demonstrates the pH-dependent insolubility of PrP in the presence of low concentrations of GdnHCl. At pH less than or equal to 3.5, PrP from human brain became insoluble, while at pH equal to or greater than 4.5 PrP was soluble. This pH range corresponds to the pK a of the side chains of the Asp and Glu, suggesting that the pH-dependent change in solubility of PrP could be associated with the protonation of these acidic residues. To determine the effect of GdnHCl on the solubility of PrP at low pH, brain homogenates were incubated with various concentrations of GdnHCl at pH 3.5. As shown in panel c of Fig. 1, when the concentration of GdnHCl was increased to 2.5 m or higher, most of the brain PrP became soluble. However, PrP still remained insoluble at the concentration of GdnHCl less than and equal to 1.5m, a concentration of GdnHCl at which most other brain proteins were soluble on Coomassie Blue gel analysis (data not shown). Therefore, acidic pH-treated PrP may possess a unique structure at 1.5m GdnHCl, the physicochemical properties of which may not be shared by many other brain proteins. Partial protease resistance is a hallmark of the PrPSc isoform, presumably resulting from structural conversion of the protein. To determine whether brain PrP treated with acid/GdnHCl possesses this property, samples were incubated with PK at various concentrations at 37 °C for 1 h. As shown in Fig.2, both mock-treated and acidic pH/GdnHCl-treated PrP display an intrinsic PK resistance at low concentrations of PK (equal to or less than 1 μg/ml PK), which is consistent with the observations of Buschmann et al. (10Buschmann A. Kuczius T. Bodemer W. Groschup M.H. Biochem. Biophys. Res. Commun. 1998; 253: 693-702Crossref PubMed Scopus (27) Google Scholar); however, there was no difference in PK sensitivity between the two, suggesting that despite acquiring PrPSc-like detergent insolubility, the new species of PrP is not identical to PrPSc. Monoclonal antibodies to diverse epitopes of PrP have been used to probe conformational rearrangement in PrP structural isoforms (11Peretz D. Williamson R.A. Matsunaga Y. Serban H. Pinilla C. Bastidas R.B. Rozenshteyn R. James T.L. Houghten R.A. Cohen F.E. Prusiner S.B. Burton D.R. J. Mol. Biol. 1997; 273: 614-622Crossref PubMed Scopus (319) Google Scholar, 12Li R. Liu T. Wong B.S. Pan T. Morillas M. Swietnicki W. O'Rourke K. Gambetti P. Surewicz W.K. Sy M.S. J. Mol. Biol. 2000; 301: 567-573Crossref PubMed Scopus (106) Google Scholar, 13Kanyo Z.F. Pan K.M. Williamson R.A. Burton D.R. Prusiner S.B. Fletterick R.J. Cohen F.E. J. Mol. Biol. 1999; 293: 855-863Crossref PubMed Scopus (51) Google Scholar, 14Yokoyama T. Kimura K.M. Ushiki Y. Yamada S. Morooka A. Nakashiba T. Sassa T. Itohara S. J. Biol. Chem. 2001; 276: 11265-11271Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar, 15Leclerc E. Peretz D. Ball H. Sakurai H. Legname G. Serban A. Prusiner S.B. Burton D.R. Williamson R.A. EMBO J. 2001; 20: 1547-1554Crossref PubMed Scopus (57) Google Scholar, 16Cashman, N. R., Paramithiotis, E., Pinard, M., Lawton, T., LaBossiere, S., Leathers, V., Zou, W. Q., Estey, L., Kondejewski, L., Haghighat, A., Spatz, S. J., Tonelli, Q., Ledebur, H. C., and Chakrabartty, A. (2001) 31st Annual Meeting of Neuroscience, November 10–15, San Diego, CA, Abstract 657.2 (www.sfn.org).Google Scholar). We used immunoprecipitations with 3F4 antibody (against residues 109–112) or 6H4 antibody (against residues 144–152) to identify differences in epitope accessibility between mock-treated and acid/GdnHCl-treated brain proteins. In lysis buffer containing low concentrations of non-denaturing detergents (0.5% Nonidet P-40 and 0.5% sodium deoxycholate), 6H4 and 3F4 antibodies precipitated PrP equally well from both mock-treated and acid/GdnHCl-treated brain homogenates (Fig. 3 a), suggesting that these two epitopes are not obscured in the structural changes induced by the treatment conditions. In contrast, the amount of PrP precipitated from CJD brain homogenates by the two antibodies was much less than that from normal brain homogenates (Fig. 3 b). Since 6H4 recognizes only native PrPC and not native PrPSc under these immunoprecipitation conditions (17Korth C. Streit P. Oesch B. Methods Enzymol. 1999; 309: 106-122Crossref PubMed Scopus (39) Google Scholar), and 3F4 is likewise poorly accessible in native PrPSc (11Peretz D. Williamson R.A. Matsunaga Y. Serban H. Pinilla C. Bastidas R.B. Rozenshteyn R. James T.L. Houghten R.A. Cohen F.E. Prusiner S.B. Burton D.R. J. Mol. Biol. 1997; 273: 614-622Crossref PubMed Scopus (319) Google Scholar), the small amounts of PrP detected in the CJD brain homogenates could simply be residual PrPC. These data confirm that residues 109–112 and 144–152 are cryptic in PrPSc molecules (11Peretz D. Williamson R.A. Matsunaga Y. Serban H. Pinilla C. Bastidas R.B. Rozenshteyn R. James T.L. Houghten R.A. Cohen F.E. Prusiner S.B. Burton D.R. J. Mol. Biol. 1997; 273: 614-622Crossref PubMed Scopus (319) Google Scholar, 17Korth C. Streit P. Oesch B. Methods Enzymol. 1999; 309: 106-122Crossref PubMed Scopus (39) Google Scholar) and suggest that the content of PrPC may be decreased in the prion disease-affected brain (14Yokoyama T. Kimura K.M. Ushiki Y. Yamada S. Morooka A. Nakashiba T. Sassa T. Itohara S. J. Biol. Chem. 2001; 276: 11265-11271Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar). Moreover, acidic pH/GdnHCl-treated human brain PrP does not share the epitope obscuration properties of native PrPSc. Recently, SDS has been found to induce structural conversion of α-helices to β-sheets and aggregation of hamster rPrP (residues 23–231) (18Xiong L.W. Raymond L.D. Hayes S.F. Raymond G.J. Caughey B. J. Neurochem. 2001; 79: 669-678Crossref PubMed Scopus (58) Google Scholar, 19Jansen K. Schafer O. Birkmann E. Post K. Serban H. Prusiner S.B. Riesner D. Biol. Chem. 2001; 382: 683-691Crossref PubMed Scopus (71) Google Scholar). Also, a low concentration of SDS is present in the PMCA protocol pioneered by Saborio et al.(5Saborio G.P. Permanne B. Soto C. Nature. 2001; 411: 810-813Crossref PubMed Scopus (1022) Google Scholar). To determine whether acid/GdnHCl-treated protein can undergo further structural rearrangement upon treatment with SDS, treated and mock-treated brain samples were incubated in 0.05% SDS, 0.5% Triton X-100 in PBS, pH 7.4 at 37 °C for 12 h with shaking. As shown in Fig. 4, small amounts of a PK-resistant fragment were found in the acid/GdnHCl-treated PrP sample (Fig. 4 a, lane 4), while none was observed in mock-treated sample (Fig. 4 a, lane 3), suggesting that SDS and/or Triton X-100 may induce conformational change in treated brain PrP. Although SDS has been shown to induce aggregation and structural transition of recombinant hamster PrP, these preparations do not acquire PK resistant at pH 6.5 (18Xiong L.W. Raymond L.D. Hayes S.F. Raymond G.J. Caughey B. J. Neurochem. 2001; 79: 669-678Crossref PubMed Scopus (58) Google Scholar), which is consistent with our data generated with mock-treated PrPCfrom brain tissue. Remarkably, the formation of PK-resistant PrP from the treated PrP was greatly enhanced by incubation with trace quantities of PrPSc from CJD brain homogenate (Fig. 4 a,lane 2). Converted PrP displayed a protease-induced gel mobility shift similar to that displayed by CJD brain (Fig.4 b). Enhanced conversion of treated human brain PrP in the presence of a PrPSc template was confirmed in four independent experiments. Little or no PK-resistant PrP was derived from similar experiments with mock-treated brain PrP (the very faint PK-resistant fragments seen in lane 1 of Fig.4 a are protease-resistant fragments of the exogenous PrPSc template). Similarly, no amplification of PrPSc was observed when human template was incubated with brain homogenates from treated wild-type or PrP−/− mice (Fig. 4 c), although an input template of human PrPSc persisted in the wild-type brain incubations and did not persist in PrP−/− brain or PBS incubations. These mouse control experiments establish that the incubation conditions did not merely enhance the recovery and detection of the template PrPSc, but are consistent with authentic amplification of protease-resistant PrP in the test system. Moreover, a series of experiments with the mouse PrP-reactive 6H4 antibody revealed the lack of conversion of mouse PrP by human template (not shown), apparently recapitulating species barrier phenomena observed in prion protein conversion in vivo (1Prusiner S.B. Scott M.R. DeArmond S.J. Cohen F.E. Cell. 1998; 93: 337-348Abstract Full Text Full Text PDF PubMed Scopus (831) Google Scholar, 2Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13363-13383Crossref PubMed Scopus (5168) Google Scholar) and in vitro (3Kocisko D.A. Come J.H. Priola S.A. Chesebro B. Raymond G.J. Lansbury P.T. Caughey B. Nature. 1994; 370: 471-474Crossref PubMed Scopus (792) Google Scholar, 18Xiong L.W. Raymond L.D. Hayes S.F. Raymond G.J. Caughey B. J. Neurochem. 2001; 79: 669-678Crossref PubMed Scopus (58) Google Scholar). Furthermore, the persistence of template PrPSc in wild-type mouse brain, but not PrP −/− brain or PBS (Fig.4 c), is reminiscent of the recent finding that persistencein vivo of hamster scrapie agent in mouse brain is dependent upon expression of endogenous PrP (20Race R. Chesebro B. Nature. 1998; 392: 770Crossref PubMed Scopus (106) Google Scholar, 21Race R. Raines A. Raymond G.J. Caughey B. Chesebro B. J. Virol. 2001; 75: 10106-10112Crossref PubMed Scopus (140) Google Scholar). Notably, the acid-guanidine facilitated conversion reaction system detailed herein contained only 0.6 μl of 10% CJD brain homogenate in 79.4 μl of 10% normal brain homogenate, and each sample inlanes 1 and 2 of Fig. 4 a contained only 65 nl of CJD brain homogenate, representing a 3–10-fold template-driven signal enhancement with treated brain by gel densitometry (not shown). According to the method of Saborio and colleagues (5Saborio G.P. Permanne B. Soto C. Nature. 2001; 411: 810-813Crossref PubMed Scopus (1022) Google Scholar) ∼70–90% of the PK-resistant PrP in our human brain system was derived from substrate in treated brain homogenate, compared with 97% in their hamster brain PMCA system (5Saborio G.P. Permanne B. Soto C. Nature. 2001; 411: 810-813Crossref PubMed Scopus (1022) Google Scholar). In this study, we have demonstrated that acidic pH and GdnHCl induces a physical transition of cellular PrP derived from normal human brain homogenates. Treated PrP is detergent-insoluble, similar to PrPSc, but displays PrPC-like protease sensitivity and epitope accessibility. A small proportion of acidic pH/GdnHCl-treated human brain PrP, but not mock-treated PrP, acquires PK resistance upon further treatment with a low concentration of SDS. Trace quantities of PrPSc template greatly enhance conversion of acidic pH/GdnHCl/SDS-treated human brain PrP to a PrPSc-like PK-resistant species. Our data suggest that conversion of PrPC to PrPSc may progress through two discrete stages, which might be recapitulated in vitro. Low pH and denaturants may induce the first stage of structural rearrangement, in which treated PrP becomes more "recruitable" than native PrPC. The second stage is dependent upon further rearrangement that is driven by a PrPSc template, which is fostered in vitro by low concentrations of SDS, a denaturing anionic detergent. Low pH and denaturants can induce β-sheet conformational change and self-association in recombinant PrP from multiple species (6Swietnicki W. Petersen R. Gambetti P. Surewicz W.K. J. Biol. Chem. 1997; 272: 27517-27520Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 7Jackson G.S. Hill A.F. Joseph C. Hosszu L. Power A. Waltho J.P. Clarke A.R. Collinge J. Biochim. Biophys. Acta. 1999; 1431: 1-13Crossref PubMed Scopus (113) Google Scholar, 8Hornemann S. Glockshuber R. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 6010-6014Crossref PubMed Scopus (240) Google Scholar, 16Cashman, N. R., Paramithiotis, E., Pinard, M., Lawton, T., LaBossiere, S., Leathers, V., Zou, W. Q., Estey, L., Kondejewski, L., Haghighat, A., Spatz, S. J., Tonelli, Q., Ledebur, H. C., and Chakrabartty, A. (2001) 31st Annual Meeting of Neuroscience, November 10–15, San Diego, CA, Abstract 657.2 (www.sfn.org).Google Scholar), and in a prion peptide (PrP 195–213; Ref. 9Zou W.Q. Yang D.S. Fraser P.E. Cashman N.R. Chakrabartty A. Eur. J. Biochem. 2001; 268: 4885-4891Crossref PubMed Scopus (7) Google Scholar). The narrow pH range (pH 3.5–4.5) in which these transitions occur is consistent with protonation of acidic amino acids (9Zou W.Q. Yang D.S. Fraser P.E. Cashman N.R. Chakrabartty A. Eur. J. Biochem. 2001; 268: 4885-4891Crossref PubMed Scopus (7) Google Scholar). Molecular dynamics simulations of Syrian hamster rPrP reveals the importance of protonation of acidic amino acids in acid-induced structural changes of PrP species (22Alonso D.O. DeArmond S.J. Cohen F.E. Daggett V. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 2985-2989Crossref PubMed Scopus (158) Google Scholar). At neutral pH, Asp-178 forms a charge-stabilized hydrogen bond with Tyr-128 in molecular dynamics simulations; however, this interaction between Asp and Tyr is broken when Asp is protonated at low pH (22Alonso D.O. DeArmond S.J. Cohen F.E. Daggett V. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 2985-2989Crossref PubMed Scopus (158) Google Scholar). Furthermore, recombinant human PrP (hurPrP) (90–231) showed a pH-dependent exposure of hydrophobic patches on the surface of the molecule, as evidenced by an increase in fluorescence intensity of bound 1,1′-bi(4-anilino)naphthalene-5,5′-disulfonic acid at pH lower than 5.5 (6Swietnicki W. Petersen R. Gambetti P. Surewicz W.K. J. Biol. Chem. 1997; 272: 27517-27520Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar). Thus, acquisition of detergent insolubility of human brain PrP resulting from treatment with low pH and GdnHCl could be concomitant with conformational changes and associated aggregation, driven by hydrophobic interactions. Human brain PrP treated at acidic pH/GdnHCl acquires the insolubility of PrPSc, but does not display the protease resistance or epitope obscuration possessed by this isoform. Several lines of evidence indicate that β-structure formation in a region spanning residues ∼90–120 contributes to the acquisition of PK resistance of PrPSc (11Peretz D. Williamson R.A. Matsunaga Y. Serban H. Pinilla C. Bastidas R.B. Rozenshteyn R. James T.L. Houghten R.A. Cohen F.E. Prusiner S.B. Burton D.R. J. Mol. Biol. 1997; 273: 614-622Crossref PubMed Scopus (319) Google Scholar, 23Zhang H. Kaneko K. Nguyen J.T. Livshits T.L. Baldwin M.A. Cohen F.E. James T.L. Prusiner S.B. J. Mol. Biol. 1995; 250: 514-526Crossref PubMed Scopus (194) Google Scholar, 24Cohen F.E. Prusiner S.B. Annu. Rev. Biochem. 1998; 67: 793-819Crossref PubMed Scopus (473) Google Scholar). Using anti-PrP antibody-mediated structural mapping, Peretz et al. (11Peretz D. Williamson R.A. Matsunaga Y. Serban H. Pinilla C. Bastidas R.B. Rozenshteyn R. James T.L. Houghten R.A. Cohen F.E. Prusiner S.B. Burton D.R. J. Mol. Biol. 1997; 273: 614-622Crossref PubMed Scopus (319) Google Scholar) reported epitopes in this region (residues ∼90 to 120) were accessible in PrPCbut largely cryptic in PrP 27–30. The obscuration in PrPScof the 3F4 epitope (residues 109–112) is consistent with this hypothesis, although the decreased accessibility to the 6H4 epitope (residues 144–152) suggests that the final rearranged region extends C-terminal to residues 90 to 120. If PrPSc formation is dependent upon structural changes in codons ∼90–120, then structural changes in other regions of the molecule are responsible for the decreased solubility and enhanced recruitability of acidic pH/GdnHCl-treated human brain PrP. Precedents for structural rearrangements occurring outside residues 90–120 include: an acidic pH-induced α-β transition in a C-terminal domain of hamster rPrP-(121–231) (8Hornemann S. Glockshuber R. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 6010-6014Crossref PubMed Scopus (240) Google Scholar) and the acquisition of a novel binding site for Cu(II) in mouse rPrP-(121–231) treated at pH range 3–5 (25Cereghetti G.M. Schweiger A. Glockshuber R. Van Doorslaer S. Biophys. J. 2001; 81: 516-525Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar), both of which are outside of the 90–120 region. Acidified, but not non-acidified, human brain PrP acquires some protease resistance characteristic of PrPSc in the presence of low concentration SDS and Triton X-100. The anionic detergent SDS may impact the hydrophobicity of treated rPrP molecules (18Xiong L.W. Raymond L.D. Hayes S.F. Raymond G.J. Caughey B. J. Neurochem. 2001; 79: 669-678Crossref PubMed Scopus (58) Google Scholar) and induce intermolecular interactions and further structural rearrangement. The PK-resistant conversion of acidified, but not non-acidified, human brain PrP was greatly enhanced by the presence trace amounts of native PrPSc. These data suggest that acidic pH/GdnHCl-treated PrP constitutes an acceptable substrate for the PrPC to PrPSc recruitment process. Moreover, the acquisition of PrPSc-like properties by acidic pH/GdnHCl brain PrP is template-directed, similar to the PrP conversion occurring in prion disease. PrPC is synthesized in the rough endoplasmic reticulum and transits through the Golgi apparatus to the cell surface, where it is attached to the outer leaflet by a glycosylphosphatidylinositol anchor in lipid rafts or caveolae (26Stahl N. Borchelt D.R. Hsiao K. Prusiner S.B. Cell. 1987; 51: 229-240Abstract Full Text PDF PubMed Scopus (913) Google Scholar, 27Caughey B. Race R.E. Ernst D. Buchmeier M.J. Chesebro B. J. Virol. 1989; 63: 175-181Crossref PubMed Google Scholar). PrPC is internalized into endocytic compartments from which most of the molecules are recycled intact to the cell surface (28Harris D.A. Clin. Microbiol. Rev. 1999; 12: 429-444Crossref PubMed Google Scholar, 29Shyng S.L. Huber M.T. Harris D.A. J. Biol. Chem. 1993; 268: 15922-15928Abstract Full Text PDF PubMed Google Scholar). A small proportion of the endocytosed molecules are proteolytically cleaved, followed by externalization of the cleaved products (29Shyng S.L. Huber M.T. Harris D.A. J. Biol. Chem. 1993; 268: 15922-15928Abstract Full Text PDF PubMed Google Scholar, 30Taraboulos A. Scott M. Semenov A. Avrahami D. Laszlo L. Prusiner S.B. Avraham D. J. Cell Biol. 1995; 129: 121-132Crossref PubMed Scopus (518) Google Scholar). Ultimately, PrPC is degraded in the endosomal-lysosomal pathway (29Shyng S.L. Huber M.T. Harris D.A. J. Biol. Chem. 1993; 268: 15922-15928Abstract Full Text PDF PubMed Google Scholar, 31Mayer R.J. Tipler C. Laszlo L. Arnold J. Lowe J. Landon M. Biomed. Pharmacother. 1994; 48: 282-286Crossref PubMed Scopus (9) Google Scholar), a series of progressively acidic intracellular organelles. Our finding that acidified PrP is an enhanced substrate for PrP conversion implicates an acidic environment for conversion in vivo. Indeed, acidic denaturation participates in amyloid fibril formation of transthyretin (32Lai Z. Colon W. Kelly J.W. Biochemistry. 1996; 35: 6470-6482Crossref PubMed Scopus (511) Google Scholar), immunoglobulin light chain (33Hurle M.R. Helms L.R., Li, L. Chan W. Wetzel R. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 5446-5450Crossref PubMed Scopus (329) Google Scholar), and lysozyme (34Pepys M.B. Hawkins P.N. Booth D.R. Vigushin D.M. Tennent G.A. Soutar K. Totty N. Nguyen O. Blake C.C. Terry C.J. Feest T.G. Zalin A.N. Hsuan J.J. Nature. 1993; 362: 553-557Crossref PubMed Scopus (567) Google Scholar). Metabolic labeling studies indicate that PrPSc formation takes place on the cell surface or a proximate postsurface compartment (35Caughey B. Raymond G.J. J. Biol. Chem. 1991; 266: 18217-18223Abstract Full Text PDF PubMed Google Scholar, 36Caughey B. Raymond G.J. Ernst D. Race R.E. J. Virol. 1991; 65: 6597-6603Crossref PubMed Google Scholar, 37Taraboulos A. Raeber A.J. Borchelt D.R. Serban D. Prusiner S.B. Mol. Biol. Cell. 1992; 3: 851-863Crossref PubMed Scopus (240) Google Scholar, 38Borchelt D.R. Taraboulos A. Prusiner S.B. J. Biol. Chem. 1992; 267: 16188-16199Abstract Full Text PDF PubMed Google Scholar, 39McKinley M.P. Taraboulos A. Kenaga L. Serban D. Stieber A. DeArmond S.J. Prusiner S.B. Gonatas N. Lab. Invest. 1991; 65: 622-630PubMed Google Scholar, 40McKinley M.P. Meyer R.K. Kenaga L. Rahbar F. Cotter R. Serban A. Prusiner S.B. J. Virol. 1991; 65: 1340-1351Crossref PubMed Google Scholar) and that PrPScsubsequently accumulates in the endosomal-lysosomal system (30Taraboulos A. Scott M. Semenov A. Avrahami D. Laszlo L. Prusiner S.B. Avraham D. J. Cell Biol. 1995; 129: 121-132Crossref PubMed Scopus (518) Google Scholar, 31Mayer R.J. Tipler C. Laszlo L. Arnold J. Lowe J. Landon M. Biomed. Pharmacother. 1994; 48: 282-286Crossref PubMed Scopus (9) Google Scholar,41Lowe J. Fergusson J. Kenward N. Laszlo L. Landon M. Farquhar C. Brown J. Hope J. Mayer R.J. J. Pathol. 1992; 168: 169-177Crossref PubMed Scopus (37) Google Scholar, 42Arnold J.E. Tipler C. Laszlo L. Hope J. Landon M. Mayer R.J. J. Pathol. 1995; 176: 403-411Crossref PubMed Scopus (150) Google Scholar, 43Vey M. Pilkuhn S. Wille H. Nixon R. DeArmond S.J. Smart E.J. Anderson R.G. Taraboulos A. Prusiner S.B. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14945-14949Crossref PubMed Scopus (489) Google Scholar, 44Prusiner S.B. Peters P. Kaneko K. Taraboulos A. Lingappa V. Cohen F.E. DeArmond S.J. Prusiner S.B. Prion Biology and Diseases. 1st Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1999: 349-391Google Scholar). Some data suggest that conversion occurs in caveolae-like lipid rafts (30Taraboulos A. Scott M. Semenov A. Avrahami D. Laszlo L. Prusiner S.B. Avraham D. J. Cell Biol. 1995; 129: 121-132Crossref PubMed Scopus (518) Google Scholar, 44Prusiner S.B. Peters P. Kaneko K. Taraboulos A. Lingappa V. Cohen F.E. DeArmond S.J. Prusiner S.B. Prion Biology and Diseases. 1st Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1999: 349-391Google Scholar), non-acidic microdomains on the plasma membrane. However, in view of the endosomal recycling activity of PrPC, it is possible that cell surface PrPC may be composed of two populations: one directly from the Golgi and another recycled from the endosome as acidified PrP. The data presented in this manuscript demonstrate that, after acidification, PrP possesses properties physically distinct from the un-acidified molecules, which are retained even when the protein is returned to physiological pH. Consistent with the notion of recruitable and "non-recruitable" pools of PrPC is the finding that only ∼5–10% PrPC has been found to be converted into PrPScin scrapie-infected neuroblastoma cells (37Taraboulos A. Raeber A.J. Borchelt D.R. Serban D. Prusiner S.B. Mol. Biol. Cell. 1992; 3: 851-863Crossref PubMed Scopus (240) Google Scholar, 38Borchelt D.R. Taraboulos A. Prusiner S.B. J. Biol. Chem. 1992; 267: 16188-16199Abstract Full Text PDF PubMed Google Scholar). In prion diseases, it is thought that PrPC is recruited to PrPSc by a template-directed process that can be mimicked in vitro(3Kocisko D.A. Come J.H. Priola S.A. Chesebro B. Raymond G.J. Lansbury P.T. Caughey B. Nature. 1994; 370: 471-474Crossref PubMed Scopus (792) Google Scholar, 4Saborio G.P. Soto C. Kascsak R.J. Levy E. Kascsak R. Harris D.A. Frangione B. Biochem. Biophys. Res. Commun. 1999; 258: 470-475Crossref PubMed Scopus (61) Google Scholar, 5Saborio G.P. Permanne B. Soto C. Nature. 2001; 411: 810-813Crossref PubMed Scopus (1022) Google Scholar). Our finding that acidic pH/GdnHCl-treated human brain PrP constitutes a superior substrate for this reaction might be exploited to detect PrPSc in tissues and fluids in which the template is present in extremely low concentrations. It is also possible that the conversion of acidified PrP into a PK-resistant form may more closely resemble PrPSc propagation in vivo, compared with two other in vitro PrP conversion systems previously reported (3Kocisko D.A. Come J.H. Priola S.A. Chesebro B. Raymond G.J. Lansbury P.T. Caughey B. Nature. 1994; 370: 471-474Crossref PubMed Scopus (792) Google Scholar, 4Saborio G.P. Soto C. Kascsak R.J. Levy E. Kascsak R. Harris D.A. Frangione B. Biochem. Biophys. Res. Commun. 1999; 258: 470-475Crossref PubMed Scopus (61) Google Scholar) in which a 50-fold or a 10-fold molar excess of PrPSc are required for the conversion to occur. Saborioet al. (5Saborio G.P. Permanne B. Soto C. Nature. 2001; 411: 810-813Crossref PubMed Scopus (1022) Google Scholar) have also reported an in vitroconversion system in which trace concentrations of hamster brain PrPSc can be detected by a process called PMCA utilizing sequential incubation-sonication to enhance conversion. However, the hamster brain PMCA system optimized by Saborio et al. (5Saborio G.P. Permanne B. Soto C. Nature. 2001; 411: 810-813Crossref PubMed Scopus (1022) Google Scholar) does not appear to support significant amplification of human PrPSc. Our data indicate that acidic/GdnHCl-treated PrPC is a superior substrate to untreated PrPCin amplification detection of human PrPSc and also obviates the requirement for lengthly incubation-sonication cycles. It is possible that acidic pH/GdnHCl-treated human brain PrP may be useful to determine the conformational events of underlying prion protein conversion in disease, the molecular co-factors and posttranslational modifications critical in conversion, and pharmaceutical agents that might prevent PrPSc formation in vitro andin vivo. We thank Dr. Catherine Bergeron and Sharon Bauer for providing the human brain tissues, Drs. Jennifer Griffin and Marty Lehto for helpful comments on the manuscript, and Jue Yuan for technical assistance. We thank Drs. Qingzhong Kong and Pierluigi Gambetti for provision of mouse brain.
Referência(s)