Mouse Prion Protein (PrP) Segment 100 to 104 Regulates Conversion of PrP^C to PrP^Sc in Prion-Infected Neuroblastoma Cells

Abstract

Prion diseases are characterized by the replicative propagation of disease-associated forms of prion protein (PrP^Sc; PrP refers to prion protein). The propagation is believed to proceed via two steps; the initial binding of the normal form of PrP (PrP^C) to PrP^Sc and the subsequent conversion of PrP^C to PrP^Sc. We have explored the two-step model in prion-infected mouse neuroblastoma (ScN2a) cells by focusing on the mouse PrP (MoPrP) segment 92-GGTHNQWNKPSKPKTN-107, which is within a region previously suggested to be part of the binding interface or shown to differ in its accessibility to anti-PrP antibodies between PrP^C and PrP^Sc. Exchanging the MoPrP segment with the corresponding chicken PrP segment (106-GGSYHNQKPWKPPKTN-121) revealed the necessity of MoPrP residues 99 to 104 for the chimeras to achieve the PrP^Sc state, while segment 95 to 98 was replaceable with the chicken sequence. An alanine substitution at position 100, 102, 103, or 104 of MoPrP gave rise to nonconvertible mutants that associated with MoPrP^Sc and interfered with the conversion of endogenous MoPrP^C. The interference was not evoked by a chimera (designated MCM2) in which MoPrP segment 95 to 104 was changed to the chicken sequence, though MCM2 associated with MoPrP^Sc. Incubation of the cells with a synthetic peptide composed of MoPrP residues 93 to 107 or alanine-substituted cognates did not inhibit the conversion, whereas an anti-P8 antibody recognizing the above sequence in PrP^C reduced the accumulation of PrP^Sc after 10 days of incubation of the cells. These results suggest the segment 100 to 104 of MoPrP^C plays a key role in conversion after binding to MoPrP^Sc.

INTRODUCTION

Transmissible spongiform encephalopathies (TSEs), or prion diseases, are fatal neurodegenerative disorders that include Creutzfeldt-Jakob disease (CJD), variant CJD (vCJD), Gerstmann-Sträussler-Scheinker syndrome (GSS), fatal familial insomnia, and kuru in humans; scrapie in sheep; bovine spongiform encephalopathy (BSE) in cattle; and chronic wasting disease (CWD) in deer. The diseases are characterized by intense neuronal cell loss and vacuolation and an accumulation of the disease-associated form(s) of prion protein (PrP^Sc; PrP refers to prion protein) in the central nervous system, though these pathological features do not always correlate with the severity of symptoms (41). While it is still under debate whether the accumulation of PrP^Sc in neurons is a direct cause of TSEs (17), the “protein-only hypothesis” (39) claims that the causative infectious agent is PrP^Sc, which propagates by conformational conversion of the normal form of cellular prion protein (PrP^C) encoded by the host gene prnp (6).

PrP^C is an N-glycosylated and glycosylphosphatidylinositol-anchored protein and is sensitive to proteolytic digestion by proteinase K (PK). Nuclear magnetic resonance (NMR)-based analyses of a recombinant mouse PrP (MoPrP) (42) and a recombinant hamster PrP (11) in solution showed that their amino-terminal (N-terminal) halves are flexible, whereas the carboxyl-terminal (C-terminal) domains (residues 121 to 231 for MoPrP) are highly ordered, consisting of three α-helices and two antiparallel β-sheets. A similar structural architecture was determined for recombinant chicken PrP (3). Such structural features likely reflect the structure of PrP^C. In contrast, PrP^Sc is partially resistant to PK digestion, with the carboxyl-terminal half left undegraded (32, 40).

The propagation of PrP^Sc takes place through a catalytic process requiring the presence of preexisting PrP^Sc seeds and recruitment of PrP^C to the preexisting PrP^Sc. In prion-infected cells, this process is believed to occur in the lipid rafts of the plasma membrane and/or in the early endocytic pathway (51). Although details of the propagation process are not well understood, lines of evidence have suggested a two-step model in which the initial binding of PrP^C to PrP^Sc and the subsequent conformational conversion to PrP^Sc are considered distinct events (1, 8, 18, 19). A similar two-step model is argued for the amyloid fibers of the yeast Sup35 protein, known to be the molecular basis for the yeast prion [PSI⁺] (46). In MoPrP, the amino acid sequences 95 to 105, 132 to 156, and 220 to 231 have been considered to be in the putative PrP^C-PrP^Sc replicative-binding interface, because the accumulation of PrP^Sc in mouse neuroblastoma cells infected with scrapie prion (ScN2a cells) (2) was blocked by recombinant Fab fragment antibodies, D13, D18, and R1, raised against these sequences (35, 53). It was also shown that peptides corresponding to the MoPrP residues 88 to 111 and 135 to 157 bound to PrP^Sc in an in vitro binding assay (29, 30, 48). (The numbering that appeared in references 29, 30, 48, and 53 has been changed to match that of MoPrP in the UniProt database [http://www.uniprot.org, accession number P04925] for ease of comparison. The original numbering in these references has been shifted +1 relative to that in the database.)

From a structural viewpoint, on the other hand, a conformational change was suggested in the region around residues 90 to 120 of hamster PrP (corresponding to 89 to 119 in MoPrP) by the observation that antibodies whose epitopes were mapped to these sequences were able to access PrP^C but not PrP^Sc (27, 34, 53). To date, direct structural elucidation of PrP^Sc has not been achieved. However, an analysis by infrared spectroscopy indicated that PrP^Sc was abundant in β-sheets (4), and a recent molecular-fitting approach using electron crystallographic density maps suggested an amyloidotic assembly in a trimeric, left-handed parallel, β-helical fold (15, 52). Later, this model was further modified to include two β-helical turns per PrP^Sc molecule, based on molecular dynamics (26). Alternatively, molecular dynamics followed by an experimental assessment indicated a spiral model (9, 10). PrP^Sc might also fold into a β-sandwich structure (45) like the cross-β spine architecture determined by X-ray crystallography for an amyloid model heptapeptide (GNNQQNY) of the yeast Sup35 protein (31).

In the present study, we focused on residues 92 to 107 of MoPrP (92-GGTHNQWNKPSKPKTN-107) and the corresponding hexadecapeptide of chicken PrP (106-GGSYHNQKPWKPPKTN-121; underlining indicates conserved amino acids), which show modest amino acid sequence homology even though mammalian PrPs and avian PrPs are only distantly related in their whole sequences (36). First, we generated chimeras of MoPrP and chicken PrP to start a framework for the study. Then, the results obtained with mouse-chicken chimeric PrPs were further examined by alanine substitution assays. We show that the MoPrP sequence 100 to 104 (100-KPSKP-104) is critical to the conversion of MoPrP into a PK-resistant form in ScN2a cells and that the segment is an auxiliary binding interface between PrP^C and PrP^Sc in the transitional stage of the conversion process.

MATERIALS AND METHODS

Nomenclature, numbering of residues, and plasmid construction.

The nomenclature for chimeras used in the present study is based on the terminology of Scott et al. (47), in which, for example, an open reading frame (ORF) of chimeric MoPrP having a Syrian hamster PrP ORF cassette between the KpnI and BglI sites (corresponding to MoPrP amino acid residues 95 to 124) was designated MHM2 (Fig. 1) (47). MHM2 possesses an epitope for the antibody 3F4 (Met-Lys-His-Met, derived from the hamster PrP sequence) and an NdeI site (47). MHM2-Q218K (a gift from K. Kaneko, Tokyo Medical University) is a derivative in which Gln²¹⁸ of MHM2 was replaced by Lys²¹⁸ (21). The numbering of amino acid residues is according to the sequences in the UniProt database (http://www.uniprot.org), accession numbers http://www.uniprot.org/uniprot/P04925 for MoPrP and http://www.uniprot.org/uniprot/P27177 for chicken PrP, respectively. The oligonucleotide primers used in the present study are summarized in Table 1.

Fig 1.

Schematic drawing of chimeric PrP constructs. Protein sequences derived from MoPrP are represented by white bars and those derived from chicken PrP by gray bars. The solid lines with residues in parentheses indicate the locations of continuous epitopes of the anti-P8 and 3F4 antibodies, and the dashed line indicates the location of a discontinuous epitope of the 44B1 antibody. The epitope for antibody 3F4 was introduced into all constructs except MoPrP. The names C4, C2M2, and MCM2 shown on the left refer to the derivation of four domains bordered by restriction enzyme cleavage sites (regions I, II, III, and IV) according to the definition of Scott et al. (47), with M for mouse, H for hamster, and C for chicken PrP, respectively. C4 is equivalent to chicken PrP. The KpnI and NdeI sites were used to generate the chimeras. The boundary amino acid positions shown above the bars are the mouse residue numbers, and those under the bars are the chicken residue numbers. The N-terminal signal sequences for sorting into the endoplasmic reticulum (ER) lumen and the C-terminal signal sequences for addition of the glycosylphosphatidylinositol (GPI) anchor are shown by broken bars. The α-helical and β-sheet structures in recombinant MoPrP (42) are represented by black boxes. The CHO labels denote the potential sites for N-glycosylation.

Table 1.

Primers used for cloning and mutagenesis of PrP ORFs^a

No.	ORF	Forward primer (5′–3′)	Reverse primer (5′–3′)
1	MHM2	GgaattcATCATGGCGAACCTTGG	ACGCgtcgacTCATCCCACGATCAGGA
2	Chicken PrP	GgaattcCCGCAGCCATGGCTAGGCTCCTC	ACGCgtcgacGCGAGGACAAGGAACACCCC
3	C4	AAAACCAACATGAAGcatatgGCGGGGGCA	TGCCCCCGCcatatgCTTCATGTTGGTTTT
4	MCM2	CCATCCAGCGGAggtaccTACCACAACCAG	CTGGTTGTGGTAggtaccTCCGCTGGATGG
5	MCM2/c95-102	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTGGGTTTTTTCCATGG
6	MCM2/c95-101	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTGGGTTTGCTCCATGGCTTCTG
7	MCM2/c95-100	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTGGGTTTGCTGGGTGGCTTCTGGTTGTG
8	MCM2/c95-99	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTGGGTTTGCTGGGCTTCTTCTGGTTGTGGTA
9	MCM2/c95-98	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTGGGTTTGCTGGGCTTGTTCTGGTTGTGGTA
10	MCM2/c100-103	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTGGGGGGTTTCCATGGGTTCCACTGATTATG
11	MHM2-N99A	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTTGGTTTGCTGGGCTTAGCCCACTGATTAT
12	MHM2-K100A	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTTGGTTTGCTGGGAGCGTTCCACTG
13	MHM2-P101A	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTTGGTTTGCTAGCCTTGTTCCA
14	MHM2-S102A	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTTGGTTTAGCGGGCTTGTT
15	MHM2-K103A	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTTGGAGCGCTGGGCTT
16	MHM2-P104A	GgaattcATCATGGCGAACCTTGG	GGCcatatgCTTCATGTTGGTTTTAGCTTTGCTGGG

^aThe restriction enzyme sites are shown in lowercase; the mutagenized sequences are underlined.

For the construction of expression plasmids for MHM2 and MHM2-Q218K, the ORFs of MHM2 and MHM2-Q218K (21) were amplified by PCR using primer set 1, and the PCR products were inserted into a pCI-neo expression vector (Promega, Madison, WI) between the EcoRI and SalI sites. The ORF of chicken PrP was obtained by PCR using primer set 2 and genomic DNA from a broiler chicken as the template. The sequence obtained was identical to that in the GenBank database (http://www.ncbi.nlm.nih.gov/genbank) (accession number M95404.1). After subcloning of the chicken PrP ORF into a pCR2.1 vector (Invitrogen, Carlsbad, CA), an epitope for the antibody 3F4 and an NdeI cleavage site were introduced by PCR mutagenesis using primer set 3, and the ORF of chicken PrP tagged by the 3F4 epitope (designated C4) (Fig. 1) was inserted into the pCI-neo vector using EcoRI and SalI sites. In order to create pCI-MCM2 (Fig. 1 and 2A), in which the 10 amino acid residues 95 to 104 of MHM2 were replaced by the corresponding chicken PrP residues 109 to 118, the ORF of MHM2 was first subcloned into the pIB/V5-His vector (Invitrogen) between the EcoRI and NotI sites, and the vector obtained, pIB/V5-MHM2, was cut by EcoRI and NdeI to delete the N-terminal MoPrP sequence (MoPrP residues 1 to 110). Then, the EcoRI-NdeI gene cassette derived from pCI-C4, encoding the N-terminal part of C4 (residues 1 to 124) (Fig. 1), was ligated to generate a mouse-chicken chimera construct, pIB-C2M2 (Fig. 1). A KpnI cutting site was introduced into the C2M2 sequence by PCR mutagenesis using primer set 4, and a mixture of the KpnI-SalI fragment of pIB-C2M2, the EcoRI-KpnI fragment of MHM2, and the EcoRI-SalI-digested pCI-neo vector was subjected to a ligation reaction all at once to yield pCI-MCM2. For the creation of other chimeras (MCM2/c95-102, MCM2/c95-101, MCM2/c95-100, MCM2/c95-99, MCM2/c95-98, and MCM2/c100-103) or single-amino-acid substitutions (MHM2-N99A, MHM2-K100A, MHM2-P101A, MHM2-S102A, MHM2-K103A, and MHM2-P104A), gene cassettes having mutagenized sequences with an EcoRI site at the 5′ end and an NdeI site at the 3′ end were prepared by PCR by using primer sets 5 to 16 and pCI-MCM2 or pIB/V5-MHM2 as the template. The cassettes were then ligated into EcoRI/NdeI-digested pIB/V5-MHM2, and the resultant cDNAs were transferred into the pCI-neo expression vector using EcoRI and SalI sites. The DNA sequences of all constructs were confirmed using an ABI PRISM 310 genetic analyzer (Applied Biosystems, Foster City, CA).

Fig 2.

Replacement of the MoPrP segment 95 to 104 with chicken PrP sequence affects the conversion of the chimeras to PrP^Sc. (A) Alignment of amino acid residues 92 to 107 of MHM2, the corresponding sequence of chicken PrP (C4), and the chimeric PrPs. The residues derived from C4 are shown in blue. (B) Western blot analysis of cell lysate digested by PK (top and bottom) or lysate without digestion (middle). All the exogenous PrPs possessed the 3F4 epitope. The Western blotting membranes detected by antibody 3F4 show relative amounts of PrP^Sc derived from the exogenous PrPs (top) and the relative expression levels of the exogenous PrPs (middle). The antibody 44B1 was used to evaluate the total amount of PrP^Sc (i.e., endogenous and exogenous PrP^Sc) (bottom). Molecular masses are shown. (C) Immunofluorescence analysis of the localization of the exogenously expressed PrPs (MHM2, C4, or the chimeras) in ScN2a cells. The cells were treated with 1 M guanidine hydrochloride for 3 min and stained for PrP using antibody 3F4 and Alexa Fluor 488-conjugated anti-mouse IgG (green). The nuclei were stained with DAPI (blue).

Cell culture and transfection of cells with plasmids.

Mouse neuroblastoma N2a cells, and ScN2a cells persistently infected by Chandler/RML scrapie prion (2), were cultured at 37°C in 5% CO₂ in Dulbecco's modified Eagle Medium (Invitrogen) supplemented with 4 mM l-glutamine, 100 U/ml penicillin G, 100 μg/ml streptomycin, and 10% heat-inactivated fetal calf serum (FCS) (Invitrogen) (33). The cells were split at a 10- to 14-fold dilution every 4 to 5 days by dissociation in phosphate-buffered saline (PBS) (SAFC Biosciences, Lenexa, KS) containing 0.05% trypsin and 0.53 mM EDTA (Invitrogen). Plasmids were introduced into the cells (8 μg of DNA per 6-cm dish) using Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer's instructions. At 72 h after transfection, the cells were harvested for Western blot analysis. All procedures were carried out according to the biosafety guidelines of the National Institute of Infectious Diseases.

Proteinase K treatment of the cell lysate.

All procedures were carried out at 4°C. The cells were rinsed three times with 5 ml of PBS (SAFC Biosciences), dissolved with 1 ml of lysis buffer (20 mM Tris-HCl buffer, pH 7.4, containing 0.5% Triton X-100, 0.2% sodium deoxycholate, and 100 mM NaCl) by pipetting, and then centrifuged at 1,500 × g for 5 min in an ARO15-24 rotor (Tomy Seiko, Tokyo, Japan) to remove insoluble debris. The concentration of protein in the lysate was determined with a bicinchoninic acid (BCA) protein assay kit (Pierce, Rockford, IL) and bovine serum albumin (Pierce) as a standard, and the lysate was adjusted to 1 mg of protein/ml with the lysis buffer. To prepare samples for the analysis of PrP^Sc, aliquots of 800 μl of the lysate were digested with 10 μg of PK (Roche Applied Science, Basel, Switzerland) at 37°C for 1 h. Then, the reaction was terminated with the addition of 20 μl of 20 mM phenylmethylsulfonyl fluoride (Sigma, St. Louis, MO) in methanol. The insoluble materials were collected as precipitates by centrifugation (90,700 × g; 1 h) in a TLA-55 rotor (Beckman Coulter, Fullerton, CA), and the precipitates were washed with 500 μl of ice-cold methanol by centrifugation (13,000 × g; 30 min) in the ARO15-24 rotor. The pellet was dissolved in 40 μl of a sodium dodecyl sulfate (SDS) sample buffer (62.5 mM Tris-HCl buffer, pH 6.8, containing 5% SDS, 2 M urea, 4% β-mercaptoethanol, 5% glycerol, 0.04% bromophenol blue, and 3 mM EDTA) and heated at 95°C for 10 min. Aliquots of 5 μl of the PK-digested lysate were subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) for Western blot analysis. To examine the total amount of PrP in the cell lysate, aliquots of 5 μl of the non-PK-digested lysate (containing 5 μg of protein) were also subjected to Western blotting.

Western blot analysis.

Proteins were resolved by SDS-PAGE using 12% NuPAGE Bis-Tris gels in morpholinopropanesulfonic acid (MOPS)-SDS running buffer (Invitrogen) and transferred to polyvinylidene difluoride membranes (Invitrogen). After blocking treatment with 10 ml of 1% skim milk (Difco-BD Biosciences, Franklin Lakes, NJ) in PBST (PBS containing 0.1% Tween 20) for 1 h, the membranes were washed three times with 50 ml of PBST for 5 min and probed with 10 ml of anti-PrP antibody 3F4 (1:10,000; Millipore, Billerica, MA), anti-PrP antibody 44B1 (1:10,000) (22), or anti-actin antibody (1:10,000; Chemicon-Millipore) overnight at 4°C in PBST containing 0.5% skim milk. The membranes were then washed three times with 50 ml of PBST for 5 min each time and probed with 10 ml of a horseradish peroxidase-labeled anti-mouse IgG (H+L) (1:20,000; KPL, Gaithersburg, MD) for 1 h in PBST containing 0.5% skim milk. After three washes with 50 ml of PBST for 5 min, immunoreactive proteins were detected using ECL Plus Western Blotting Detection Reagents (GE Healthcare, Uppsala, Sweden) with a LAS-1000plus chemiluminescence imaging system (Fuji Film, Tokyo, Japan). Signal intensities were determined by Image Gauge software (Fuji Film) using the integrated peak volume quantification mode.

Binding of PrP^C to PrP^Sc.

The efficacy with which PrP^C bound to PrP^Sc was examined by a previously described method with a slight modification (38). The cells were rinsed with PBS at 72 h after transfection and dissolved in lysis buffer supplemented with a proteinase inhibitor cocktail (Complete Mini; Roche Applied Science, Indianapolis, IN). After the removal of insoluble debris by centrifugation (1,500 × g; 5 min) at 4°C in the ARO15-24 rotor, the protein concentration in the lysate was determined with the BCA protein assay kit, and the lysate was adjusted to 1 mg of protein/ml with lysis buffer. Then, aliquots of the lysate (100 μl) were incubated at 37°C overnight and centrifuged at 15,000 × g for 10 min at room temperature in the ARO15-24 rotor. The supernatant was transferred to a new tube, and the pellet was suspended in 100 μl of lysis buffer. The supernatant and pellet were then added to 10 volumes of ice-cold methanol, and proteins were precipitated by centrifugation at 13,000 × g for 30 min at 4°C in the ARO15-24 rotor. The precipitated proteins were dissolved in 20 μl of SDS sample buffer, heated for 10 min at 95°C, and subjected to Western blot analysis.

Immunofluorescence analysis.

The cells were cultured in 35-mm glass bottom dishes (12 mm in diameter; Asahi Glass Co., Ltd., Chiba. Japan) to 50% confluence at 37°C. All manipulations were done at room temperature unless otherwise stated. The cells were washed three times with 2 ml of PBS and fixed with 150 μl of Cytofix fixation buffer (BD Biosciences, Franklin Lakes, NJ) for 15 min. The cells were then washed three times with 2 ml of PBS, permeabilized by incubation with 150 μl of 0.1% Triton X-100 in PBS for 10 min, washed three times with 2 ml of PBS, and treated with 150 μl of 1 M guanidine hydrochloride for 0 to 10 min to retrieve the immunoreactivity of PrP^Sc (50). After three washes with 2 ml of PBS, the cells were incubated with 150 μl of 10% FCS in PBS for 30 min for blocking and then with either of the following antibodies in 150 μl of PBS: the anti-PrP antibody 3F4 (Millipore) at 1:1,000 for 1 h at room temperature or anti-P8 (44) at 1:200 overnight at 4°C. After removal of the excess antibodies by washing with 2 ml of PBS three times, the cells were incubated with the following secondary antibodies and 4′,6-diamidino-2-phenylindole (DAPI) (1 μg/ml; Dojindo Laboratories, Kumamoto, Japan) in 150 μl of PBS for 1 h; an Alexa Fluor 488-conjugated goat anti-mouse IgG antibody (Invitrogen) at 1:2,000 for the antibody 3F4 or an Alexa Fluor 488-conjugated goat anti-rabbit IgG antibody (Invitrogen) at 1:2,000 for the anti-P8 antibody. The cells were mounted in Fluoromount-G (SouthernBiotech, Birmingham, AL) and observed using an FV-1000 confocal microscope (Olympus, Tokyo, Japan) with Fluoviewer software (Olympus).

Treatment of the cells with anti-PrP antibody or peptides.

To the culture medium of ScN2a and N2a cells in 6-well plates, anti-P8 antibody (raised in a rabbit in our laboratory; the epitope is included in residues 92 to 109 of MoPrP) (44) was added at a final concentration of 0 to 10 μg/ml. The cells were cultured for 10 days with two passages on days 3 and 6 before harvest. As a control, the cells were treated with rabbit normal IgG (R&D Systems, Inc., Minneapolis, MN) in the same way. Peptides derived from the amino acid sequence (positions 93 to 107) of MoPrP were synthesized by 9-fluorenylmethyloxycarbonyl solid-phase chemistry (Operon Biotechnologies, Tokyo, Japan). Their amino acid sequences were GTHNQWNKPSKPKTN (PrP_93-107), GTHNQWNAPSKPKTN (PrP_K100A), GTHNQWNKPAKPKTN (PrP_S102A), GTHNQWNKPSAPKTN (PrP_K103A), and GTHNQWNKPSKAKTN (PrP_P104A). The purity of the peptides was verified by reversed-phase high-performance liquid chromatography (RP-HPLC) and matrix-assisted laser desorption time-of-flight mass spectrometry (purity, >90%). The peptides were dissolved in distilled water at a concentration of 5 mM and added to the culture medium of ScN2a cells in 6-well plates at a final concentration of 10 or 100 μM (33). The cells were cultured for 10 days with two passages on days 3 and 6 before harvest. The stability of the peptides during the cell treatment was examined as follows by using PrP_93-107 as a representative peptide. ScN2a cells were seeded in 24-well plates containing 500 μl of the culture medium per well, and the peptide was added to the medium on day 0 at a final concentration of 100 μM. After cultivation of the cells in triplicate for 0, 1, 2, 3, and 4 days at 37°C, the medium and the cells were retrieved by pipetting up and down. Aliquots of 470 μl of the cell suspension in the medium were added with 30 μl of 100% (wt/vol) trichloroacetic acid (TCA) (Nacalai Tesque, Inc., Kyoto, Japan); then, the samples were centrifuged at 12,000 × g for 10 min at 4°C for deproteinization, and the supernatant was stored at −80°C until analysis. The peptide was also incubated in the medium, but without the cells, and was processed in the same way. For quantification of the peptide, aliquots of the supernatant were diluted 10 times with distilled water, and 20 μl of the diluted supernatant (equivalent to 188 pmol of the peptide if no degradation had occurred) was subjected to RP-HPLC on an LC-2000Plus system (Jasco, Tokyo, Japan) equipped with a fluorescence detector (FP-2020Plus; Jasco) and an Inertsil ODS-2 column (4.6 mm by 150 mm; GL Sciences, Tokyo, Japan). Elution was carried out at 40°C by a linear gradient from 100% solution A (5% acetonitrile-water, 0.05% trifluoroacetic acid [TFA]) to 10% solution B (60% acetonitrile-water, 0.05% TFA) in 32 min at a flow rate of 0.8 ml/min. PrP_93-107 was detected by fluorescence (excitation at 280 nm/emission at 320 nm) derived from a tryptophan residue in its amino acid sequence. PrP_93-107 was eluted at approximately 22 min. The amounts of PrP_93-107 were determined by peak height in the chromatogram in comparison with the peak height of 188 pmol of PrP_93-107 as a standard.

RESULTS

Mutagenic analysis with chicken PrP gene cassettes identified MoPrP residues 99 to 104 as a critical segment.

To explore the roles of region 95 to 104 of MoPrP in prion propagation, we took advantage of the limited homology in the region between MoPrP and chicken PrP and used gene cassette-based mutagenesis to generate a series of mouse-chicken chimeric PrPs (Fig. 1 and 2A). In order to distinguish the exogenously expressed PrPs from the endogenous MoPrP in N2a and ScN2a cells (2), the exogenous PrPs were tagged with the epitope for the antibody 3F4 (Met-Lys-His-Met) (28, 47). MHM2 (MoPrP possessing the 3F4 epitope) was converted efficiently to PrP^Sc in ScN2a cells (Fig. 2B, top, lane 1), consistent with a previous report (47). Under these conditions, the expression of C4 (i.e., 3F4 epitope-tagged chicken PrP) and all seven chimeras (Fig. 2A) was comparable to that of MHM2 (Fig. 2B, middle). As expected given the species barrier to prion infection (41), C4 could not be converted to a PK-resistant form (Fig. 2B, top, lane 2). Among the chimeras with chicken sequences of various lengths, MCM2/c95-98, in which the MoPrP residues 95 to 98 (HNQW) were replaced by the chicken sequence YHNQ but residues 99 to 104 were unchanged, was converted to a PK-resistant form almost as efficiently as MHM2 (Fig. 2B, top, lane 8). Conversely, MCM2/c100-103, in which MoPrP residues 100 to 103 were replaced with the chicken sequence, failed to be converted (Fig. 2B, top, lane 9). The other five chimeras, in which the chicken cassettes extended further to the C-terminal side (up to residues 99, 100, 101, 102, and 103), were also incapable of changing into PK-resistant forms (Fig. 2B, top, lanes 3 to 7). The nonconvertibility of the chimeras was unlikely to have been due to mislocalization of the chimeric PrPs in the cells, since immunofluorescence analysis showed similar distributions of these chimeras and MHM2 on the plasma membrane and in intracellular punctate compartments (Fig. 2C). Accordingly, we concluded that residues 99 to 104 of MoPrP (99-NKPSKP-104) include key residues for inducing the conversion of PrP^C to the PrP^Sc state.

Alanine scanning refined the key residues in MoPrP 99 to 104.

Based on the results obtained with the mouse-chicken chimeras, we next narrowed down the area crucial for the conversion of PrP^C to PrP^Sc by replacing single amino acids in the MoPrP 99-to-104 segment with alanine (Fig. 3A). As shown in Fig. 3, Western blot analysis (Fig. 3B, middle; probed by the antibody 3F4) and immunofluorescence analysis (Fig. 3C) indicated the expression levels and cellular localization of the 3F4 epitope-tagged alanine-substituted mutants were comparable to those observed for MHM2. Under these conditions, the replacement of Pro¹⁰¹ with Ala (designated the P101A mutant), a mutation analogous to P102L identified in patients with the inherited prion disease GSS (20), gave rise to PK-resistant PrP^Sc as efficiently as MHM2 (Fig. 3B, top, lane 5). Also, N99A was a moderately convertible mutant. In contrast, the K100A mutant was almost nonconvertible, though a small amount of PrP^Sc was still detectable by Western blot analysis (Fig. 3B, top, lane 4). The replacement of Ser¹⁰², Lys¹⁰³, or Pro¹⁰⁴ (S102A, K103A, and P104A, respectively) completely prevented conversion into PrP^Sc in the cells (Fig. 3B, top, lanes 6 to 8). Although P104A of MoPrP was analogous to the human PrP Pro¹⁰⁵-to-Leu mutation identified in patients with another type of GSS (24, 25, 54), it was nonconvertible in the present experimental setting. The conversion efficacy of this type of mutation might be synergistic with additional factors, for example, the co-occurrence of polymorphic variation at codon 129 (methionine or valine) (25, 54). The K100A, S102A, K103A, and P104A mutants not only were nonconvertible themselves, but also blocked conversion of the endogenous PrP^C to PrP^Sc (Fig. 3B, top and bottom, lanes 4 and 6 to 8) at levels similar to those of MHM2-Q218K, which was known as a “dominant-negative” mutant (21, 23) (Fig. 3B, top and bottom, lane 9). Taken together, the amino acid residues Lys¹⁰⁰, Ser¹⁰², Lys¹⁰³, and Pro¹⁰⁴ of MoPrP were critical for inducing the conversion of PrP^C to the PrP^Sc state.

Fig 3.

Replacement of the amino acid residue at position 100, 102, 103, or 104 of MoPrP with alanine inhibits the conversion of the mutants and endogenous PrP^C to PrP^Sc. (A) Alignment of residues 92 to 107 of MHM2, MCM2, and the alanine-substituted mutants. The residues derived from C4 are shown in blue. The substituted alanine is shown in red. (B) Western blot analysis of cell lysate digested by PK (top and bottom) or lysate without digestion (middle). All the exogenous PrPs possessed the 3F4 epitope. The membranes represent the amounts of PrP^Sc converted from the exogenous PrPs (top; detected by antibody 3F4), the expression levels of the exogenous PrP (middle; antibody 3F4), and the total amount of PrP^Sc (i.e., endogenous and exogenous PrP^Sc) (bottom two images; antibody 44B1) in ScN2a cells. The two bottom images of the same membrane with different exposure times show that the signals detected by antibody 44B1 are not saturated but are within the range of the chemiluminescence imaging system. The signal intensities of PrP^Sc bands detected by antibody 44B1 were quantified and expressed as percentages of that of MHM2, with 0% denoting an undetectable level of PrP^Sc (invisible band). The data represent the means and standard errors of the mean (SEM) for three independent experiments. (C) Immunofluorescence analysis of the localization of the alanine-substituted PrPs and Q218K in ScN2a cells. Null denotes untransfected ScN2a cells. The cells were treated with 1 M guanidine hydrochloride for 3 min and stained for PrP using antibody 3F4 and Alexa Fluor 488-conjugated anti-mouse IgG (green). The nuclei were stained with DAPI (blue).

Residues 100 to 104 play a pivotal role in the conversion.

Next, we examined whether changes in the amino acid sequence 100 to 104 of MoPrP affects the binding of PrP^C to PrP^Sc or the subsequent conversion step (1, 8, 18, 19). To distinguish the two steps, we performed a biochemical assay to detect the formation of a complex of PrP^C with PrP^Sc. This assay is based on the notion that the newly synthesized PrP^C bound to preexisting PrP^Sc is converted to PrP^Sc and finally assimilated into the PrP^Sc aggregates, so that all such temporary complexes and resultant aggregates may be recovered in the pellet after centrifugation of the cell lysate.

In N2a cells, approximately 70 to 82% of the exogenously expressed PrPs were recovered in the supernatant (Fig. 4A and B), while 20 to 30% of them were recovered in the pellet at background levels in the absence of PrP^Sc. When the lysate of ScN2a cells expressing MHM2 (the convertible control) was fractionated by centrifugation, as much as 75% of exogenously expressed MHM2 was retrieved in the pellet (Fig. 4C and D, lanes 1). MCM2 (Fig. 4C, lane 2, and D, lane 4) and the alanine-substituted mutants (Fig. 4C, lanes 3 to 8) were retrieved in the pellet in a manner similar to that of MHM2. Interestingly, 60% of C4 was recovered in the supernatant in ScN2a cells, whereas C2M2 and MCM2 (a nonconvertible chimera, as shown in Fig. 2B, lane 3) were retrieved in the pellet at levels comparable to that of MHM2 (the convertible control) (Fig. 4D). The difference between C2M2 and C4 was in their C-terminal halves: C2M2 had the chicken PrP sequence residues 1 to 118 followed by the MoPrP sequence residues 105 to 231 (Fig. 1). This difference was also present for MCM2 and C4 (Fig. 1). Accordingly, the data shown in Fig. 4D suggest a predominant binding interface for PrP^Sc in the C-terminal half of PrP^C (residues 105 to 231 in the MoPrP sequence). It also should be noted that as much as 40% of C4 expressed in ScN2a cells was fractionated into the pellet (Fig. 4D), approximately 20% of which was ascribed to the background level independent of PrP^Sc (Fig. 4B) while the remaining 20% was due to the association of C4 and MoPrP^Sc. The question remains whether such association of C4 with MoPrP^Sc occurs by interaction of particular segments of chicken PrP (C4) with MoPrP^Sc. The results obtained by these chimeras, together with the nonconvertibility of the alanine-substituted mutants in binding to PrP^Sc (Fig. 3 and 4), suggested that segment 100 to 104 was involved in the conversion and, consequently, in the growth of PrP^Sc aggregates.

Fig 4.

Replacement of MoPrP segment 99 to 104 by the chicken sequence does not affect binding of the chimeric PrP to PrP^Sc. (A and B) N2a cells transfected with the expression vectors encoding MHM2, MCM2, the alanine-substituted mutants, or Q218K (A) and C4 or C2M2 (B) were subjected to a fractionation assay, followed by Western blot analysis. PrPs recovered in the supernatant (Sup.) (A and B, top) and the pellet (A and B, middle) were detected by antibody 3F4. MHM2 (A and B, lanes 1) and Q218K (A, lane 9) were examined as references, while null (A, lane 10) represents untransfected N2a cells. The signal intensities of the PrP bands detected in the supernatant and pellet were quantified and expressed as percentages of the total amount of 3F4-positive PrP in the cells (A and B, bottom). The data represent the means ± SEM for four independent experiments. (C and D) The same experiment as in panels A and B was carried out using ScN2a cells. Exogenous PrPs associated with PrP^Sc aggregates were recovered in the pellet (C and D, middle), while those free from PrP^Sc aggregates were found in the supernatant (C and D, top). The signal intensities of the PrP bands detected in the supernatant and pellet were quantified and expressed as percentages of the total amount of 3F4-positive PrP in the cells (C and D, bottom). The data represent the means ± SEM for four independent experiments.

Residues 92 to 109 of MoPrP are cryptic in the PrP^Sc form.

A previous study using a panel of antibodies in a cell-free enzyme-linked immunosorbent assay suggested a structural change in residues 90 to 120 of hamster PrP between PrP^C and PrP^Sc (34). This was based on the observation that the antibodies, for example, D4, D13, and R10, whose epitopes were mapped in this region (35), were reactive to PrP^C but not to PrP^Sc unless it was denatured by guanidine thiocyanate (34, 53). Following this observation, we examined if an antibody raised against residues 92 to 109 of MoPrP (designated the anti-P8 antibody) (44) would bind to PrP^C but not to PrP^Sc in N2a and ScN2a cells. In an immunofluorescence analysis, the anti-P8 antibody gave an unchanged level of immunopositive signals for PrP^C in N2a cells irrespective of pretreatment of the cells with 1 M guanidine hydrochloride (Fig. 5, top row). Such a staining pattern for N2a cells is similar to that previously reported for PrP^C in N2a cells using the antibody 6H4 (51). In ScN2a cells, the anti-P8 antibody gave much weaker signals for cells without the guanidine treatment (Fig. 5, compare the left column). However, it gave rise to an increased intensity of staining after guanidine treatment for 5 min (Fig. 5, bottom row), though it was uncertain whether the effect of 1 M guanidine was in unfolding PrP^Sc aggregates or in dispersion of an unidentified molecule(s) associated with PrP^Sc to unmask the epitope. These results suggest that the epitope for anti-P8 antibody is accessible in PrP^C, and the epitope might be hindered in access in the PrP^Sc state unless the cells are treated with guanidine (see Discussion).

Fig 5.

Anti-P8 antibody can bind to PrP^C but not PrP^Sc in ScN2a cells under native conditions. N2a cells or ScN2a cells were pretreated with 1 M guanidine hydrochloride for 0 to 10 min as specified below and then incubated with the anti-P8 antibody and Alexa Fluor 488-conjugated secondary antibodies for detection of endogenous MoPrPs (green). The nuclei were stained with DAPI (blue). The images of confocal microscopy were collected by setting the detector gain at the same level.

Anti-P8 antibody inhibited the propagation of PrP^Sc in ScN2a cells, but synthetic peptides derived from MoPrP residues 93 to 107 did not.

As described above, the anti-P8 antibody is expected to bind to PrP^C but not to PrP^Sc in ScN2a cells. Thus, to further support the involvement of Lys¹⁰⁰, Ser¹⁰², Lys¹⁰³, and Pro¹⁰⁴ in the conversion step, we cultivated ScN2a cells in the presence of the anti-P8 antibody for 10 days, during which time the antibody should bind to PrP^C in the cells. The amount of PrP^Sc decreased depending on the dosage of the antibody (Fig. 6A, top, lanes 1 to 4), whereas rabbit normal IgG (the control) was ineffective (Fig. 6A, top, lanes 5 to 8). Western blot analysis of the samples not digested by PK (Fig. 6A, middle) detected 25- to 40-kDa bands that corresponded to the full-length forms of the di-, mono-, and nonglycosylated PrP (f-PrP), including PrP^Sc and newly synthesized PrP^C, and 19- to 25-kDa bands that were attributed to remnant PrP^Sc (r-PrP) partially resistant to cellular endogenous proteases (33). Since the anti-P8 antibody did not affect the amount of f-PrP (Fig. 6A, middle, lane 4) but predominantly eliminated r-PrP, it was considered not to interfere with the de novo synthesis of PrP^C, and thus, the clearance of PrP^Sc in the cells was not due to a reduction in the supply of PrP^C to be converted. The inhibitory effect of the anti-P8 antibody on the conversion of PrP^C to PrP^Sc might be ascribed either to the masking of a putative primary binding site or to clamping of the epitope of PrP^C to hamper the structural change. To examine this point, we incubated ScN2a cells with a synthetic peptide composed of residues 93 to 107 of the wild-type MoPrP, which corresponded to the core sequence of the anti-P8 antibody's epitope, or a cognate peptide in which either Lys¹⁰⁰, Ser¹⁰², Lys¹⁰³, or Pro¹⁰⁴ was replaced with Ala (Fig. 6C). The peptides were added to the culture medium, and the cells were cultured for 10 days, with two passages on days 3 and 6 in fresh medium containing peptides before harvest. The peptides, however, did not reduce the accumulation of PrP^Sc in the cells (Fig. 6B). We asked if the ineffectiveness of the peptides was ascribed to their rapid degradation. As examined with a representative peptide, PrP_93-107, the loss of the peptide at the initial concentration of 100 μM proceeded slightly faster in the presence of the cells than in the medium only (Fig. 6D). We observed that about 60% of PrP_93-107 remained on day 3, and its half-life was 4 days in the cell culture (Fig. 6D, solid line). This excluded the possibility of rapid degradation and loss of the peptides during the treatment of the cells. Analysis of recombinant MoPrP by NMR spectroscopy suggested the region of the amino acid sequence from residues 93 to 107 of MoPrP^C is not constrained in solution (42), and we expect the synthetic peptides corresponding to this region are structurally as flexible as the region in the PrP^C molecule. Accordingly, if this region of the PrP^C molecule interacts with and is accommodated in particular sites of PrP^Sc, we suppose that the synthetic peptides similarly could mimic the corresponding region of the PrP^C molecule. The lack of competitive inhibition of the propagation of PrP^Sc by the peptides suggested that the predominant binding interface of PrP^C-PrP^Sc was outside residues 93 to 107 of MoPrP, conceivably in the C-terminal domain of MoPrP, as described above (Fig. 4). Although residues 93 to 107 are not involved in the primary binding interface, we see a possibility that they participate in an auxiliary binding interface. The effect of the anti-P8 antibody might be ascribed to either clamping of the epitope of PrP^C or masking of any putative auxiliary binding interfaces near the epitope, preventing the conversion of PrP^C to PrP^Sc from proceeding.

Fig 6.

Anti-P8 antibody inhibits the accumulation of PrP^Sc in ScN2a cells, but a synthetic peptide derived from MoPrP residues 93 to 107 and its cognate peptides do not. (A) Western blot analysis of the lysate from ScN2a cells after 10 days of culture in the presence of the anti-P8 antibody or rabbit normal IgG at 0.1 to 10 μg/ml. PrP was detected by antibody 44B1. The lysate with PK digestion (top) shows the amount of PrP^Sc, and the lysate without digestion (middle) shows the total amount of PrP (i.e., PrP^C and PrP^Sc). The membrane was also probed with anti-actin antibody (bottom) for normalization of the protein content. The signal intensity of each PrP^Sc band detected in the upper membrane was quantified and expressed as a percentage of that in the absence of the antibody, with 0% denoting an undetectable level of PrP^Sc (invisible band). The data represent the means ± SEM for three independent experiments. (B) Western blot analysis of the lysate from ScN2a cells after 10 days of culture in the presence of the peptides at 10 or 100 μM. PrP was detected by antibody 44B1. The lysate with PK digestion (top) shows the amount of PrP^Sc, and the lysate without digestion (middle) shows the total amount of PrP (i.e., PrP^C and PrP^Sc). The membrane was also probed with the anti-actin antibody (bottom) for normalization of the protein content. The signal intensity of each PrP^Sc band detected in the upper membrane was quantified and expressed as a percentage of that in the absence of the peptides, with 0% denoting an undetectable level of PrP^Sc (invisible band). The data represent the means ± SEM for three independent experiments. (C) Sequences of peptides PrP_93-107, PrP_K100A, PrP_S102A, PrP_K103A, and PrP_P104A. The substituted alanine residues are underlined. (D) Time course of relative amounts of PrP_93-107 during the cell culture. The peptide was added to the medium with ScN2a cells (solid line) or without the cells (dashed line) at a concentration of 100 μM on day 0 and incubated at 37°C for up to 4 days. The amounts of PrP_93-107 were determined by reversed phase HPLC and are expressed as percent recovery. A 100% rating denotes no loss of PrP_93-107. The data represent means ± SEM for three independent experiments.

DISCUSSION

Replicative propagation of disease-associated forms of prion protein (PrP^Sc) through conversion of the normal form of prion protein, PrP^C, is a hallmark of transmissible spongiform encephalopathies. This process is considered to occur in two kinetically distinguishable steps (1, 8, 18, 19): the initial binding of PrP^C to preexisting PrP^Sc and the subsequent conversion of PrP^C into PrP^Sc. In the present study, we focused on residues 92 to 107, part of the conformationally heterogeneous and flexible region of a recombinant MoPrP (42) but found at the distal N-terminal end of PrP^Sc after PK digestion (32, 40).

A previous report speculated that residues 90 to 104 and 106 to 115 in hamster PrP (corresponding to 89 to 103 and 105 to 114 in MoPrP, respectively) might not participate in the initial binding because two antibodies (one against 90 to 104 and the other, antibody 3F4, against 106 to 115) did not inhibit the generation of a PK-resistant form(s) of PrP in an in vitro cell-free conversion system (18). The results of the present study, indicating that MoPrP residues 100 to 104 (KPSKP) play a role in the conversion rather than as a predominant binding interface during prion propagation, seem to support the above notion. However, in contrast to the cell-free conversion assay, which showed that the antibody against hamster PrP residues 90 to 104 did not block the formation of a PK-resistant form(s) of PrP (18), the incubation of ScN2a cells with the anti-P8 antibody whose epitope was in MoPrP residues 92 to 109 decreased the amount of PrP^Sc in the present study (Fig. 6A, top, lanes 1 to 4). In this regard, Peretz et al. (35) reported that the incubation of ScN2a cells with the Fab fragment antibody D13 (10 μg/ml), which was specific to MoPrP residues 95 to 105 (27, 35, 53), disrupted the propagation of PrP^Sc. They ascribed this inhibition to the D13 antibody's binding to the PrP^C molecule on the cell surface, thereby masking a putative PrP^C-PrP^Sc binding interface in the proximity of the epitope (35). Like the antibody D13, we supposed that the anti-P8 antibody preferentially binds to PrP^C rather than to PrP^Sc on the cell surface (Fig. 5). However, in an immunofluorescence analysis of ScN2a cells infected with the scrapie 22L strain, Veith et al. (51) reported that immunopositive signals for PrP^Sc by staining with antibody 6H4 after PK digestion were noticeable in intracellular areas than at the plasma membrane and that the immunopositive signals at the plasma membrane without PK digestion could be caused by both PrP^C and PrP^Sc. Thus, we might not be clearly discriminating PrP^C and PrP^Sc at the plasma membrane shown in Fig. 5 under our experimental conditions. Further study is needed to refine this point.

Approaches using synthetic peptides and peptide libraries also implicated several segments of PrP^C in possible binding interfaces. For example, a peptide corresponding to residues 109 to 141 of hamster PrP was reported to inhibit the generation of a PK-resistant form of the PrP in an in vitro cell-free conversion assay (5). This led to speculation that the residues in the vicinity of positions 109 to 141 of either PrP^C or the PK-resistant form of PrP were involved in the intermolecular association (5). In screening assays utilizing MoPrP-derived peptide libraries grafted into immunoglobulin G molecules, the peptides corresponding to residues 23 to 33, 97 to 109, and 135 to 157 of MoPrP strongly bound to PrP^Sc (29, 30, 48). While peptide-based studies have demonstrated the putative binding interfaces between PrP^C and PrP^Sc, the antibodies against residues 143 to 151 (12), 143 to 153 (16), 132 to 156 and 220 to 231 (35), and 218 to 232 (18) of MoPrP or the amino acid substitution at residue 138 (37) interfered with the propagation of PrP^Sc. However, the incubation of prion-infected cells (RK13 cells) with the 1C5 antibody against residues 118 to 129 of MoPrP (7) for a week was ineffective at reducing the accumulation of PrP^Sc in the cells (16). We consider that the dominant and high-affinity binding interface for PrP^Sc is in the C-terminal half of PrP^C, as shown in Fig. 4B, and that these regions could be candidates for the initial binding interface between PrP^C and PrP^Sc.

It should be pointed out that the alanine-substituted mutants eliminated the endogenous PrP^Sc in ScN2a cells (Fig. 3B, bottom), while the mouse-chicken chimeras were unable to reduce the amounts of endogenous PrP^Sc (Fig. 2B, bottom). We reasoned that this was due to a difference in the affinity of the alanine-substituted mutants and the mouse-chicken chimeras for PrP^Sc, assuming that residues 95 to 104 serve as an auxiliary binding interface or in the coordination of the complex to strengthen the interaction between PrP^C and preexisting PrP^Sc. The single alanine substitution in the segment 100 to 104 (KPSKP) would not affect the affinity for preexisting PrP^Sc significantly, and the eventual “dead-end” complex hampered the growing amyloidotic assembly of PrP^Sc in the cells (Fig. 3B, bottom). The mouse-chicken chimeric PrPs, however, associated with preexisting PrP^Sc through their C-terminal regions (Fig. 4B, lanes 3 and 4), but the additional binding to or coordination with PrP^Sc was disrupted due to more than two amino acid mismatches between mouse and chicken sequences in the region of residues 95 to 104 of MoPrP^C. Thus, unlike the alanine-substituted mutants, the chimeras would be less stably associated with the preexisting PrP^Sc and, accordingly, incapable of competition and interference (Fig. 3B, bottom).

We found the replacement of Lys¹⁰⁰, Ser¹⁰², Lys¹⁰³, or Pro¹⁰⁴ in MoPrPs with Ala generated nonconvertible dead-end mutants. This finding, however, is inconsistent with a report that a MoPrP mutant in which the lysine residues at positions 100 and 103 were replaced with Ala was still converted to the PK-resistant form with an efficacy similar to that of the wild-type MoPrP in ScN2a cells (1). The reason for the discrepancy is unknown. In the above-mentioned report, the replacement of both Lys¹⁰⁵ and Lys¹⁰⁹ with Ala gave rise to a nonconvertible mutant, and the authors considered a PrP^C-PrP^Sc-interacting domain to be present between residues 106 and 110 of MoPrP (1).

Transgenic mice expressing a mutant MoPrP lacking residues 32 to 93 were shown to be susceptible to scrapie-prion agent, though they showed a decrease in the deposition of PrP^Sc and no histopathological features of scrapie (13). In contrast, transgenic mice expressing a mutant MoPrP lacking residues 32 to 106 did not succumb to scrapie (14). These results suggest the functional importance of the segment examined in the present study. Recent studies using molecular fitting or molecular dynamics have predicted the amyloidotic assembly of PrP^Sc in trimeric, left-handed parallel β-helical folds (15, 43, 45, 49). The models predict that segment 100-KPSKP-104 of MoPrP forms an outward loop extending from the core β-helical architecture (15, 26, 49). Our results showing that the alanine substitutions at Lys¹⁰⁰, Ser¹⁰², Lys¹⁰³, and Pro¹⁰⁴ in MoPrP disrupted the conversion of PrP^C to PrP^Sc (Fig. 3) and that the replacement of the MoPrP sequence 95 to 98 (HNQW) with the corresponding chicken sequence (YHNQ) did not affect conversion efficacy (Fig. 2) highlight the functional importance of the predicted loop structure in the β-helical models.

In conclusion, segment 100 to 104 of MoPrP plays a critical role in promoting conversion. Also, while the primary binding interface seems to reside in the C-terminal region of PrP^C, this segment might be involved in the auxiliary binding interface of PrP^C to PrP^Sc. Further study is needed to clarify if the binding of the C-terminal region of PrP^C to PrP^Sc allosterically activates the segment 100 to 104 for conversion in the PrP^C-PrP^Sc complex.