Abs to the prion protein (PrP) can protect against experimental prion infections, but efficient Ab responses are difficult to generate because PrP is expressed on many tissues and induces a strong tolerance. We previously showed that immunization of wild-type mice with PrP peptides and CpG oligodeoxynucleic acid overcomes tolerance and induces cellular and humoral responses to PrP. In this study, we compared Ab and T cell repertoires directed to PrP in wild-type and PrP knockout (Prnpo/o) C57BL/6 mice. Animals were immunized with mouse PrP-plasmid DNA or with 30-mer overlapping peptides either emulsified in CFA or CpG/IFA. In Prnpo/o mice, Abs raised by PrP-plasmid DNA immunization recognized only N-terminal PrP peptides; analyses of Ab responses after PrP peptide/CFA immunization allowed us to identify six distinct epitopes, five of which were also recognized by Abs raised by PrP peptides/CpG. By contrast, in wild-type mice, no Ab response was detected after PrP-plasmid DNA or peptide/CFA immunization. However, when using CpG, four C-terminal peptides induced Abs specific for distinct epitopes. Importantly, immune sera from Prnpo/o but not from wild-type mice bound cell surface PrP. Abs of IgG1 and IgG2b subclasses predominated in Prnpo/o mice while the strongest signals were for IgG2b in wild-type mice. Most anti-PrP Th cells were directed to a single epitope in both Prnpo/o and wild-type mice. We conclude that endogenous PrPC expression profoundly affects the Ab repertoire as B cells reactive for epitopes exposed on native PrPC are strongly tolerized. Implications for immunotherapy against prion diseases are discussed.

Transmissible spongiform encephalopathies (TSE)4 are neurodegenerative disorders that are associated with the conversion of the normal host-encoded cellular prion protein (PrPC) into an abnormal conformer, PrPSc. TSE are transmitted through unconventional agents termed “prions,” the major component of which is PrPSc (1, 2). The immune system participates in the transport and replication of prions in peripheral lymphoid organs (3), but no acquired immune response can be evidenced in the course of TSE infections (4). In contrast, Abs to prion protein (PrP) were shown to inhibit PrPSc accumulation in stably infected neuroblastoma cells (5, 6, 7, 8) and epithelial Rov cells (9) or to prolong survival when passively transferred into mice infected with scrapie (8, 10) and to protect transgenic mice producing anti-PrP Abs (11).

In a previous work, we demonstrated that PrP knockout (Prnpo/o) mice immunized with mouse PrP plasmid DNA (PrP-pDNA) encoding full-length PrP or with 30-mer PrP peptides in CFA generated PrP-specific T cells and Abs that recognized native PrPC (12). No immune response to PrP could be detected in wild-type (wt) mice using the same immunization protocols, likely due to natural tolerance since PrPC is expressed in the thymus and many peripheral tissues (13). The lack of a response in wt mice could relate to anergy or deletion of B cell and/or Th precursor cells expressing specific Ag receptors for PrP epitopes. However, self-proteins carrying a foreign T cell epitope often induce autoantibodies in normal mice, showing that B cell tolerance is generally incomplete and that autoreactive B cells can be activated when appropriate T cell help is provided (14). Heppner et al. (11) found that transgenic mice expressing an anti-PrP Ig H chain displayed high titers of Abs to PrP along with a normal level of PrPC, suggesting that tolerance to PrPC is unlikely to be entirely confined to B cells, but could also be attributed to unresponsiveness or deletion of specific CD4+ T cell clones. Indeed, the development of an immune response is determined not only by exposure of immunogenic epitopes on the injected Ags but also by the presence of functional primary B and T cell repertoires.

Tolerance of Th cells may be overcome with the use of suitable adjuvants. Using three PrP peptides in combination with CpG oligodeoxynucleotides (referred to as CpG) (15), we generated T and B cell responses to PrP peptides in C57BL/6 wt mice (16).

In the present study, we compared the anti-PrP T and B cell responses in Prnpo/o and wt mice with the same genetic background. Distinct Ab repertoires were generated after immunization of wt and Prnpo/o mice with PrP-pDNA or 30-mer PrP peptides in CFA or CpG. Notably, Abs that bound to native PrPC were not detected in immunized wt mice. The only anti-PrP Abs raised in peptide-immunized wt mice bound peptide segments that are probably not accessible on cellular PrPC. Most of the anti-PrP Th cell repertoire of Prnpo/o and wt mice was directed to a single epitope located around residues 158–172 in the PrP molecule. These results have important implications for the design of immunotherapeutic strategies against TSE.

Six-week-old female C57BL/6 wt mice were purchased from Janvier. Prnpo/o mice (provided by C. Weissmann, Imperial College, London, U.K. (17)) have been backcrossed 10 times into the C57BL/6 background. All mice were bred in specific pathogen-free conditions in accordance with European recommendations on animal ethics.

The 30-mer PrP peptides with 15 aa overlaps on each side were synthesized by the Neosystem; some were shorter (especially P10) because of high hydrophobicity which rendered the synthesis difficult. They were purified (>80%) by HPLC on a C8 reverse-phase column and controlled by electrospray mass spectrometry (Table I). A library of 15-mer overlapping peptides covering PrP sequence 140–189 was synthesized under the same conditions (Table II). Oligonucleotide-CpG 1826 (18) was synthesized by Sigma-Aldrich.

Table I.

Sequence of 30-mer overlapping peptides covering the entire murine PrP

PeptideSequence
P1 = 23–52 KKRPKPGGWNTGGSRYPGQGSPGGNRYPPQ 
P2 = 39–67 YPGQGSPGGNRYPPQGGTWGQPHGGGWGQPH 
P3 = 68–97 HGGSWGQPPGGSWGQPHGGGWGQGGGTHNQ 
P4 = 83–112 PHGGGWGQGGGTHNQWNKPSKPKTNLKHVA 
P5 = 98–127 WNKPSKPKTNLKHVAGAAAAGAVVGGLGGY 
P6 = 118–142 GAVVGGLGGYMLGSAMSRPMIHFGN 
P7 = 128–157 MLGSAMSRPMIHFGNDWEDRYYRENMYRYP 
P8 = 143–172 DWEDRYYRENMYRYPNQVYYRPVDQYSNQN 
P9 = 158–187 NQVYYRPVDQYSNQNNFVHDCVNITIKQHT 
P10 = 173–189 NFVHDCVNITIKQHTVT 
P11 = 193–218 KGENFTETDVKMMERVVEQMCVTQYQK 
P12 = 212–232 MCVTQYQKESQAYYDGRRSSS 
P13 = 218–240 QKESQAYYDGRRSSSTVLFSSPP 
PeptideSequence
P1 = 23–52 KKRPKPGGWNTGGSRYPGQGSPGGNRYPPQ 
P2 = 39–67 YPGQGSPGGNRYPPQGGTWGQPHGGGWGQPH 
P3 = 68–97 HGGSWGQPPGGSWGQPHGGGWGQGGGTHNQ 
P4 = 83–112 PHGGGWGQGGGTHNQWNKPSKPKTNLKHVA 
P5 = 98–127 WNKPSKPKTNLKHVAGAAAAGAVVGGLGGY 
P6 = 118–142 GAVVGGLGGYMLGSAMSRPMIHFGN 
P7 = 128–157 MLGSAMSRPMIHFGNDWEDRYYRENMYRYP 
P8 = 143–172 DWEDRYYRENMYRYPNQVYYRPVDQYSNQN 
P9 = 158–187 NQVYYRPVDQYSNQNNFVHDCVNITIKQHT 
P10 = 173–189 NFVHDCVNITIKQHTVT 
P11 = 193–218 KGENFTETDVKMMERVVEQMCVTQYQK 
P12 = 212–232 MCVTQYQKESQAYYDGRRSSS 
P13 = 218–240 QKESQAYYDGRRSSSTVLFSSPP 
Table II.

Sequence of 15-mer overlapping peptides covering residues 140–189

PeptideSequence
140–154 FGNDWEDRYYRENMY 
144–158 WEDRYYRENMYRYPN 
148–162 YYRENMYRYPNQVYY 
152–166 NMYRYPNQVYYRPVD 
156–170 YPNQVYYRPVDQYSN 
160–174 VYYRPVDQYSNQNNF 
164–178 PVDQYSNQNNFVHDC 
168–182 YSNQNNFVHDCVNIT 
172–186 NNFVHDCVNITIKQH 
176–189 HDCVNITIKQHTVT 
PeptideSequence
140–154 FGNDWEDRYYRENMY 
144–158 WEDRYYRENMYRYPN 
148–162 YYRENMYRYPNQVYY 
152–166 NMYRYPNQVYYRPVD 
156–170 YPNQVYYRPVDQYSN 
160–174 VYYRPVDQYSNQNNF 
164–178 PVDQYSNQNNFVHDC 
168–182 YSNQNNFVHDCVNIT 
172–186 NNFVHDCVNITIKQH 
176–189 HDCVNITIKQHTVT 

pcDNA3.1 plasmid (Invitrogen Life Technologies) mouse PrP DNA (PrP-pDNA) was prepared and purified as previously described (12). Control DNA consisted in an empty pcDNA3.1 plasmid. Mice were injected three times at weekly intervals with 100 μg DNA divided between the two tibialis anterior muscles, treated 5 days before with 50 μl of cardiotoxin (10 mM solution in PBS) from Naja nigricollis venom (Latoxan). PrP peptides (100 μg) were mixed with CpG (50 μg) and emulsified in IFA (v/v) and then injected s.c. at the base of the tail. A boost was performed 10–14 days later under identical conditions. Spleen and blood were collected 10- 14 days after the last injection.

The number of IFN-γ-producing cells from spleens of immunized mice was evaluated by ELISPOT as previously described (19). Briefly, nitrocellulose-based 96-well plates (Millipore) were coated with anti-mouse IFN-γ Abs (1/500; BD Biosciences) for 2 h at 37°C followed by overnight incubation at 4°C. Plates were washed with PBS-0.05% Tween 20 (PBS-T) and blocked with RPMI 1640 containing 10% FCS for 2 h at 37°C. Responder splenocytes from individual mice were seeded at 1 × 106 cells/well and restimulated with PrP peptides at 10 μg/ml or with medium alone. Plates were incubated at 37°C in 5% CO2 during 24 h, washed with PBS-T, and incubated for 2 h at 37°C with biotinylated anti-mouse IFN-γ (BD Biosciences). After washing with PBS-T, alkaline phosphatase-conjugated streptavidin (Roche) was added and left for 2 h; IFN-γ-secreting cells were revealed using nitroblue tetrazolium/5-bromo-4-chloro-indolyl phosphate substrate (Promega), and spots were counted using an ELISPOT plate counter (AID). Test wells were assayed in triplicates and the frequency of peptide-specific T cells was calculated by subtracting the mean number of spots obtained after incubation of medium without peptide. An experimental value was considered significantly positive when over the mean control value plus 3 SD.

Flat-bottom 96-well plates (Maxisorp; Nunc) were coated with 10 μg/ml PrP peptides in sodium carbonate buffer (0.05 M, pH 9.6) overnight at 4°C. Plates were washed with PBS and blocked with 1% nonfat milk in PBS-T for 2 h at 37°C. Serially diluted sera from immunized and control mice were added in duplicate and reacted for 2 h at room temperature. Plates were washed again and 200 μl of peroxidase-conjugated anti-mouse Ig (Roche) was added and left for 2 h at room temperature. Plates were then washed and 200 μl/well of freshly prepared H2O2/o-phenylenediamine substrate solution (Sigma-Aldrich) were added. The reaction was stopped with 2 N sulfuric acid and OD was measured at 492 nm. An experimental value was considered significantly positive when exceeding the mean control value plus 3 SD.

Sera were serially diluted and tested in duplicates in ELISA as described above. Detection of IgG1, IgG2a, and IgG2b isotypes was performed using specific peroxidase-conjugated goat anti-mouse Abs (Southern Biotechnology Associates).

EL-4 cells overexpressing murine PrPC (allotype s7) were obtained after transfection as described elsewhere (12). Cells were activated with plastic-coated anti-CD3 mAb (2-C11, 10 μg/ml) 24 h before testing immune sera for maximal expression of PrPC. The level of PrPC expression was checked on EL-4 cells using FITC-conjugated anti-PrP mAb SAF83 (from Dr. J. Grassi (Commissariat à l’Energie Atomique, Saclay, France). After blocking Fc receptors with Ab 2.4G2 for 20 min at 4°C in FACS buffer, cells were incubated with control or immune sera diluted 1/10 for 20 min at 4°C, washed, and analyzed on a FACSCalibur flow cytometer using CellQuest software (BD Biosciences). A serum was considered significantly positive for native PrPC binding when mean fluorescence intensity (MFI) of peptide-immunized serum was over the MFI + 3 SD of sera from mice treated with CFA or CpG alone.

To have an exhaustive view of Ab repertoires, Prnpo/o and wt mice were immunized with PrP-pDNA. Epitope mapping was performed using a library of overlapping PrP peptides covering the entire sequence of PrP (Table I). Abs generated in Prnpo/o mice are directed against linear epitopes located at the N-terminal part of the PrP molecule (shown by ELISA, Fig. 1,A). Five peptides were most frequently recognized by individual immune sera: P1 (12 of 15, 80%), P2 (9 of 15, 60%), P3 (3 of 14, 21%), P4 (8 of 15, 53.3%), and P5 (3 of 15, 20%). Titration of individual sera on peptides indicated that Abs to P1, P2, and P4 dominated the response elicited by PrP-pDNA (Fig. 1 B). Under the same immunizing conditions, no Abs against peptides were detected in wt mice (data not shown).

FIGURE 1.

Peptide specificities of Abs raised in Prnpo/o mice after immunization with PrP-pDNA. Individual immune sera were tested in ELISA for binding to the PrP-overlapping peptides (1 μg/well) coated on plastic microplates. A, Histograms represent the pattern of reactivity (OD) for each peptide of 11 to 15 individual sera (1/50 diluted) from PrP-pDNA-immunized mice (▪) or the mean OD values of sera from empty pDNA-immunized mice (□). B, Serial dilutions of seven sera tested on peptides P1–P5.

FIGURE 1.

Peptide specificities of Abs raised in Prnpo/o mice after immunization with PrP-pDNA. Individual immune sera were tested in ELISA for binding to the PrP-overlapping peptides (1 μg/well) coated on plastic microplates. A, Histograms represent the pattern of reactivity (OD) for each peptide of 11 to 15 individual sera (1/50 diluted) from PrP-pDNA-immunized mice (▪) or the mean OD values of sera from empty pDNA-immunized mice (□). B, Serial dilutions of seven sera tested on peptides P1–P5.

Close modal

In a second set of experiments, Prnpo/o and wt mice were immunized twice with each of the overlapping PrP peptides emulsified in CFA and serum reactivities to the corresponding peptides were tested by ELISA. Results (Fig. 2 A) indicated that 8 of 12 peptides elicited in Prnpo/o mice Ab responses with variable frequency and intensity. These data confirmed that P1, P2, P4, and P5 contain immunogenic epitopes as shown above with antisera raised against PrP-pDNA. In addition, P8, P9, P11, and P12 also generated Abs.

FIGURE 2.

Ab responses to PrP-overlapping peptides. Individual sera were collected from Prnpo/o mice immunized with CFA only or PrP peptides and CFA (A), from wt mice immunized with CpG only or PrP peptides and CpG (B), and from Prnpo/o mice immunized with CpG only or PrP peptides and CpG (C). Each point represents the reactivity of one individual serum (1/50 diluted) measured in ELISA on the corresponding peptide-coated wells.

FIGURE 2.

Ab responses to PrP-overlapping peptides. Individual sera were collected from Prnpo/o mice immunized with CFA only or PrP peptides and CFA (A), from wt mice immunized with CpG only or PrP peptides and CpG (B), and from Prnpo/o mice immunized with CpG only or PrP peptides and CpG (C). Each point represents the reactivity of one individual serum (1/50 diluted) measured in ELISA on the corresponding peptide-coated wells.

Close modal

In contrast, no Ab response could be obtained by immunizing wt mice with peptides in CFA (Ref.14 , and data not shown). On the basis of our previous findings (16), wt mice were then immunized twice with each of the peptides mixed with CpG. Sera collected 10–12 days later were analyzed for the presence of Abs in peptide ELISA (Fig. 2,B). Results failed to demonstrate a serum reactivity for the PrP portions P1–P4 (positions 23–112). As found in Prnpo/omice, Abs could be raised against P5 and P11 but not P3, P6, P7, and P10. However, in contrast to Prnpo/o mice, wt mice poorly responded to P9 and not at all to P12; only two of eight mice were positive for P8. Because adjuvants may influence immune responses, peptide immunizations of Prnpo/o mice were also performed with CpG in the same experimental conditions as those done in wt mice. Fig. 2 C shows that P1, P2, P5, P8, P9, and P12 could raise specific Abs with variable frequency and intensity. Some differences appeared when compared with Ab responses obtained with CFA: P4 and P11 did not induce Abs whereas P6 raised low but consistent Ab levels.

Because most of the peptides used for specificity determination were 30-mer peptides with 15-aa overlaps, we attempted to determine the number and positions of PrP segments recognized by Abs. We compared the ELISA reactivity of sera from Prnpo/o and wt mice tested against the immunizing peptide and against the upstream and downstream adjacent overlapping peptides.

In Prnpo/o mice (Fig. 3, A and B), several sera raised against one immunizing peptide recognized adjacent peptides. Anti-P1 and -P2 sera obtained with peptides/CFA (Fig. 3,A) cross-reacted with P1 and P2, suggesting that they shared a common epitope (residues 29–57). Sera from mice immunized with P4 or P5 recognized P4 and P5 but not P3 or P6, suggesting the existence of an epitope shared by P4 and P5 (residues 98–112). Sera from mice immunized with P8 recognized two distinct PrP segments, one shared with P7 (residues 143–157) and another with P9 (residues 158–172). The latter segment was also recognized by sera from P9-immunized mice. Finally, Abs induced by P11 and P12 did not cross-react with each other nor with P10 and P13, and thus seemed to recognize two distinct epitopes. The results obtained with Prnpo/o mice after immunization with peptides/CpG indicated that Abs also recognized the epitopes shared by P1/P2, P4/P5, P7/P8, P8/P9, and P12 (Fig. 3 B).

FIGURE 3.

Mapping B cell epitopes: sera from individual Prnpo/o mice immunized with the 30-mer overlapping peptides either with CFA (A) or CpG (B) and from wt mice immunized with peptides and CpG (C) were tested in ELISA on the corresponding peptide (▪) and on upstream (▧) or downstream (□) adjacent overlapping peptides. Gray bars represent the reactivity of sera from mice immunized with CFA or CpG alone. Histograms show the reactivity of two to five sera for each immunizing peptide.

FIGURE 3.

Mapping B cell epitopes: sera from individual Prnpo/o mice immunized with the 30-mer overlapping peptides either with CFA (A) or CpG (B) and from wt mice immunized with peptides and CpG (C) were tested in ELISA on the corresponding peptide (▪) and on upstream (▧) or downstream (□) adjacent overlapping peptides. Gray bars represent the reactivity of sera from mice immunized with CFA or CpG alone. Histograms show the reactivity of two to five sera for each immunizing peptide.

Close modal

In wt mice immunized with peptides/CpG, no significant reactivity to P1, P2, and P4 was detected. P5 elicited Abs that did not cross-react with P4 or P6 and thus likely recognized a PrP segment around residues 112–118, different from the segment 98–112 recognized by P5-immunized Prnpo/o mice sera (Fig. 3 C). P8-immunized sera cross-reacted with P9 and therefore contained Abs recognizing a segment located around residues 158–172. In contrast, Abs to P9 did not cross-react with adjacent peptides and were thus specific for a segment located around residues 173–187, different from the segment 158–172 recognized by P9-immunized Prnpo/o mice sera. P11 elicited Abs which did not bind to P10 and P12 (residues 193–212).

The PrP portion between residues 143 and 187 appeared to contain three distinct immunogenic segments which were recognized differently in Prnpo/o and wt mice after immunization with P8 and P9: one shared by P7/P8, another by P8/P9, and one on P9 not shared by P8. To localize more precisely these epitopes, we used a library of 15-mer peptides with 4 aa overlaps covering sequence 140–189 (Table II). Sera from Prnpo/o mice immunized with P8 and CFA (Fig. 4 A) recognized one segment shared by peptides D and E (overlapping sequence 144–154) and another shared by G and H (overlapping sequence 156–166), whereas wt mice recognized only segment 156–168 (G–J).

FIGURE 4.

Mapping B cell epitopes: sera from individual Prnpo/o and wt mice immunized with the 30-mer overlapping P8 (A) or P9 (B) with CFA or CpG, respectively, were tested in ELISA on 15-mer overlapping peptides covering PrP sequence 140–189. Results represent the mean reactivity (±SD) of five sera from P8- or P9-immunized Prnpo/o mice and of two sera from P8- or six from P9-immunized wt mice.

FIGURE 4.

Mapping B cell epitopes: sera from individual Prnpo/o and wt mice immunized with the 30-mer overlapping P8 (A) or P9 (B) with CFA or CpG, respectively, were tested in ELISA on 15-mer overlapping peptides covering PrP sequence 140–189. Results represent the mean reactivity (±SD) of five sera from P8- or P9-immunized Prnpo/o mice and of two sera from P8- or six from P9-immunized wt mice.

Close modal

Fig. 4 B shows that the PrP segments recognized by anti-P9 antisera are different when immunization was performed in Prnpo/o mice (peptide H: 156–170) as compared with wt mice (peptides K–L: 178–186). Identical patterns of reactivity were observed in Prnpo/o mice after immunization with P8 and P9 when CpG was used instead of CFA (data not shown).

All sera from Prnpo/o mice immunized three times with PrP-pDNA strongly bound native PrPC expressed at the surface of activated Prnp-transfected EL-4 cells, with a MFI ranging from 34 to 72 (Fig. 5 A).

FIGURE 5.

Ab binding to native PrPC: A, sera from Prnpo/o mice immunized with PrP-pDNA; B, with PrP peptides with CFA (⋄) or CpG (♦); and C, sera from wt mice immunized with PrP peptides and CpG were incubated (1/10 diluted) with EL-4 cells overexpressing PrPC. Abs were revealed with PE-conjugated anti-mouse Ig and fluorescence was analyzed by FACS. Each point represents the MFI of an individual serum.

FIGURE 5.

Ab binding to native PrPC: A, sera from Prnpo/o mice immunized with PrP-pDNA; B, with PrP peptides with CFA (⋄) or CpG (♦); and C, sera from wt mice immunized with PrP peptides and CpG were incubated (1/10 diluted) with EL-4 cells overexpressing PrPC. Abs were revealed with PE-conjugated anti-mouse Ig and fluorescence was analyzed by FACS. Each point represents the MFI of an individual serum.

Close modal

Peptide-specific sera from Prnpo/o mice were also analyzed for their binding capacity to native PrPC (Fig. 5 B). All sera from mice immunized with P5 bound to PrPC with a high MFI whether mice were immunized emulsified in CFA or in IFA/CpG. Two of six P1 immune sera obtained with CFA and one of four P1 immune sera obtained with CpG displayed low but significant binding to EL-4 cells. Surprisingly, four of eight positive sera from mice immunized with P8 and CFA significantly bound also to EL-4 cells while none of those obtained with CpG bound to EL-4 cells. Sera specific for P2, P4, P6, P7, P9, P11, and P12 did not bind native PrPC.

In contrast to Prnpo/o mice, none of the sera collected from wt mice, whether immunized with PrP-pDNA or peptides with CFA or CpG, significantly stained PrPC on EL-4 cells (Fig. 5 C).

Because IgG subclasses display diverse effector properties, we compared the IgG isotype patterns of Abs raised in Prnpo/o and wt mice after immunization with P5, P8, or P9 emulsified in IFA/CpG. The strongest signal was found for IgG1 and IgG2b subclasses in sera from Prnpo/o mice whatever peptide used for immunization, whereas sera from wt mice mostly contained IgG2b subclass Abs (Fig. 6). IgG2a levels varied according to peptides but not significantly between Prnpo/o and wt mice. Identical isotype patterns were observed in Prnpo/o mice when immunization was performed with CFA (data not shown).

FIGURE 6.

Ab IgG subclass determination: sera (1/50 diluted) from Prnpo/o and from wt mice immunized with peptides and CpG were tested in ELISA plates coated with the corresponding peptide and revealed with peroxidase-conjugated goat anti-mouse IgG1, IgG2a, and IgG2b.

FIGURE 6.

Ab IgG subclass determination: sera (1/50 diluted) from Prnpo/o and from wt mice immunized with peptides and CpG were tested in ELISA plates coated with the corresponding peptide and revealed with peroxidase-conjugated goat anti-mouse IgG1, IgG2a, and IgG2b.

Close modal

The quality of Ab responses is also dependent on specific Th activity. In contrast to Prnpo/o mice, the T cell repertoire specific for PrP was not stimulated in wt mice after immunization with PrP-pDNA or PrP peptides mixed with CFA (data not shown and Refs.12 and 16). However, we previously demonstrated that P9 and, to a lesser extent, P8 administered with CpG induced T cell responses in wt mice (16). Thus, we compared the T cell repertoire of Prnpo/o and wt mice generated after immunization with each peptide emulsified in IFA/CpG and we measured the frequency of specific IFN-γ-secreting T cells in ELISPOT. Fig. 7,A shows that in Prnpo/o mice, P9 appears as the most immunogenic, yielding a frequency (mean ± SE) of 306 ± 64 IFN-γ-secreting T cells/106 splenic cells. P8 also stimulated a significant but lower number of IFN-γ-secreting T cells (97 ± 32). No other tested peptide consistently generated a T cell response although a few mice yielded significant numbers of IFN-γ-secreting peptide-specific T cells (Fig. 7,A). In wt mice, P9 was also the most immunogenic (201 ± 26 IFN-γ-secreting T cells/106 spleen cells) but only one of five mice responded to P8 (Fig. 7 B). P8 or P9 immune spleen cells cross-reacted with both P8 and P9 and thus were specific for a shared epitope located between residues 158–172. This was true for wt mice (16) and also for Prnpo/o mice immunized with PrP-pDNA as well as with peptides in CFA (12) or CpG (data not shown).

FIGURE 7.

Frequency of IFN-γ-secreting peptide-specific T cells in Prnpo/o and wt mice measured in ELISPOT. Mice were immunized s.c. and boosted once with 100 μg of each PrP peptide with CpG. Splenocytes from immunized mice were restimulated for 24 h in vitro with 10 μg of the corresponding peptide per ml. Each point represents the frequency of IFN-γ-secreting peptide-specific T cells per 106 spleen cells for individual immunized mouse.

FIGURE 7.

Frequency of IFN-γ-secreting peptide-specific T cells in Prnpo/o and wt mice measured in ELISPOT. Mice were immunized s.c. and boosted once with 100 μg of each PrP peptide with CpG. Splenocytes from immunized mice were restimulated for 24 h in vitro with 10 μg of the corresponding peptide per ml. Each point represents the frequency of IFN-γ-secreting peptide-specific T cells per 106 spleen cells for individual immunized mouse.

Close modal

Studies of Ab responses in wt and Prnpo/o mice with the same C57BL/6 genetic background indicates that the endogenous expression of PrP qualitatively affects the specific B cell repertoire: in wt mice, no Ab response was detected to N-terminal epitopes of PrP. In addition, Abs raised in wt mice recognized PrP segments not exposed on membrane-bound PrPC and were distinct from those recognized by Abs raised in Prnpo/o mice.

Immunization of Prnpo/o mice with PrP-pDNA activated an Ab repertoire preferentially directed to peptides P1–P5, which are located in the N-terminal region and may thus be considered as immunodominant epitopes (Fig. 8). In addition, such sera contained Abs that bound native PrPC on activated Prnp-transfected EL-4 cells. However, in a competitive assay, addition of peptides P1–P5 lowered the reactivity of PrP-pDNA sera for native PrPC only by 30–50% (data not shown), strongly suggesting the presence of additional Abs specific for conformational epitopes. Other PrP segments (P8, P9, P11, and P12 emulsified in CFA) elicited Abs after immunization of Prnpo/o mice. Epitope mapping using adjacent 30-mer overlapping peptides and a panel of 15-mer overlapping peptides covering the 140–189 (P8/P9) PrP region revealed that Prnpo/o B cells produced Abs specific for at least six distinct immunogenic segments of PrP. Because the use of CpG instead of CFA might cause differences in immune responses, Prnpo/o mice were also immunized with peptides/CpG. Abs raised in these conditions recognized six distinct epitopes (Fig. 3, A and B), five of which seemed identical to those recognized by Abs induced by immunization with peptides/CFA (Fig. 8).

FIGURE 8.

Regions of the PrP molecule recognized by Abs raised in Prnpo/o and wt mice. Immunization was performed either with PrP-pDNA or overlapping PrP peptides with CpG. Plain bars represent the PrP regions recognized by Abs which bind to plastic-adsorbed peptides only. Hatched bars represent the PrP regions recognized by Abs which bind to native PrPC in addition to plastic-adsorbed peptides.

FIGURE 8.

Regions of the PrP molecule recognized by Abs raised in Prnpo/o and wt mice. Immunization was performed either with PrP-pDNA or overlapping PrP peptides with CpG. Plain bars represent the PrP regions recognized by Abs which bind to plastic-adsorbed peptides only. Hatched bars represent the PrP regions recognized by Abs which bind to native PrPC in addition to plastic-adsorbed peptides.

Close modal

By contrast, injection of PrP-pDNA or peptides with CFA did not elicit detectable Ab or T cell responses in wt mice (12). CpG oligodeoxynucleotides, which are known to stimulate dendritic cells (20) and B cells via TLR9 (21), allowed us to generate immune responses in wt mice (16). When peptides were injected with CpG, wt mice produced Abs only against four segments located at the C-terminal part of the PrP molecule (Fig. 8). Characterization of epitope specificities further demonstrated that Abs to segments P5 and P9 recognized epitopes distinct from those recognized by Abs raised in Prnpo/o mice.

These findings suggest that the lower Ab responses from wt mice are not due to the inability of CpG to stimulate Ab production but rather to unresponsiveness of B cells specific for the N-terminal peptides. However, Abs to P6 could be stimulated in Prnpo/o mice immunized in the presence of CpG while Abs to P11 could not. These results may be attributed to the type of response elicited by CFA or CpG: although both adjuvants promote T cell responses toward Th1 (22), CpG stimulated a greater number of CD4+ IFN-γ-secreting Th cells (23) and was able to break the tolerance (16, 17, 18, 19, 20, 21, 22, 23).

Another interesting observation was that none of the Abs elicited in wt mice recognized membrane-bound PrPC, whereas several Abs elicited in Prnpo/o mice did (particularly those specific for the 98–112 segment). Yet, we previously showed that Abs to P5, P8, and P9 from both Prnpo/o or wt mice reacted with plastic-bound recombinant PrP (12, 16). Our observations are in accordance with those of Polymenidou et al. (24), who found that Abs produced by wt mice after immunization with recombinant PrP failed to recognize native membrane PrPC despite reactivity for recombinant PrP in ELISA. They postulated that membrane PrPC and recombinant PrP differed in some structural features or that the microenvironment of PrPC with associated proteins or lipids masked or distorted epitopes that were exposed on recombinant PrP.

That Abs to native PrPC were not detected in wt mice’s immune sera might relate to a lower sensitivity of the cytofluorometric assay as compared with ELISA. However, Ab levels measured by ELISA in sera positive for P5 or P8 were similar in both Prnpo/o and wt mice. It is also possible that in PrP+ mice, Abs would be produced but spontaneously form complexes with endogenous PrPC, thus preventing their detection in the serum (Ref.25 and C. Féraudet, manuscript in preparation). Although no Ab-PrP complexes could be detected in tested wt sera (data not shown), Abs may also be adsorbed on cell surface PrPC, which is expressed on several hemopoietic cell lineages (13).

Alternatively, Abs to cellular PrPC might not be produced in wt mice because B cells reacting against segments that are exposed on PrPC are either tolerized or lack T cell help. Anti-P5 Abs raised in Prnpo/o mice recognized segment 98–112 and bound to native PrPC while in wt mice, they were specific for segment 112–118, and failed to bind native PrPC, strongly suggesting that B cells specific for 98–112 are anergized or deleted in wt mice. In P8-immunized Prnpo/o mice, four of eight sera recognized native PrPC and were specific for segment 143–157 while sera which did not bind native PrPC recognized segment 158–172. In accordance with these data, wt mice immunization by P8 generated Abs only to segment 158–172 and they did not bind to native PrPC. This dual reactivity might explain the lack of binding to native PrPC of sera obtained after immunization of Prnpo/o mice with P8 using CpG as adjuvant. These sera contained Abs specific for two distinct segments, some of them recognizing native PrPC (Abs to 143–157) and others that did not (Abs to 158–172); the proportion of these specificities might dictate the capacity of immune sera to bind native PrPC.

Difference in the location of the epitopes recognized by Abs from wt and Prnpo/o mice after immunization with the same peptide suggest that B cells specific for PrP epitopes exposed on native PrPC are strongly tolerized in wt mice. Only B cells potentially binding epitopes that are not exposed on PrPC would escape tolerance and be the main, if not exclusive, contributors to the autoantibody response. Self-proteins expressed as membrane-bound surface Ags result in the deletion of specific B cells during bone marrow ontogeny (26). In contrast, a substantial number of autoreactive B cells, specific for the unfolded conformation of a self-Ag, escape tolerance (27).

Abs raised by peptide immunization in Prnpo/o and wt mice displayed different patterns of IgG subclasses. Immunization of Prnpo/o mice, whatever the peptide and the adjuvant used, induced predominantly IgG1 and IgG2b subclasses, whereas in wt mice, the strongest signals were found for IgG2b. It was reported that T cells increase the Ig switching process with a hierarchy of IgG subclass Ab production which directly correlated with the 5′ to 3′ IgH-C gene order, i.e., IgG1>IgG2b>IgG2a (28, 29). Consequently, the predominance of IgG2b in wt mice might be related to the skewing toward a more Th1 cell-dependent response than in Prnpo/o mice, possibly because of the tolerant context.

We may exclude the possibility that the observed differences could be related to genetic disparities in antigenic background, since Prnpo/o mice had been backcrossed 10 times with C57BL/6 mice. Furthermore, no proliferation was found when splenocytes from Prnpo/o and wt mice were primed in vivo by each other (data not shown).

Besides a deletion in the B cell repertoire, the Th signal promoting Ab production may be insufficient in wt mice. In general, Ab responses to peptides are weak and highly dependent on T cell help. Only two PrP epitopes could induce a T cell proliferation from Prnpo/o but not from wt mice immunized with PrP-pDNA or peptides with CFA (12). We recently demonstrated that CpG allowed us to induce the generation of IFN-γ-secreting T cells specific for an epitope shared by P8/P9 in wt mice (16). In this study, the T cell repertoires generated by immunization with PrP peptides/CpG were rather similar in Prnpo/o and in wt mice: P9 consistently triggered the strongest T cell responses against a P8/P9 shared epitope. Yet, the magnitude of T cell responses to peptides was generally lower in wt mice than in Prnpo/o mice, which is compatible with the possibility that the T cell repertoire is partially tolerized in wt mice.

If one assumes that only Abs that bind membrane PrPC are efficient in blocking prion pathogenesis (24), the very modest effects of active immunization of wt mice against PrP in protecting against prion infections are not surprising (30, 31). However, it is worth noting that mAbs that target both normal and pathological forms of PrP blocked PrPSc accumulation in vitro more efficiently than mAbs recognizing PrPC only (9). Abs that do not bind PrPC, such as those generated by peptide immunization with CpG, might be efficient in blocking PrPSc replication in vivo by targeting epitopes only exposed on PrPSc. B cell repertoires producing such neutralizing Abs might escape central tolerance and reach the periphery. New perspectives in immunotherapeutic approaches in TSE could rely on the stimulation of such B cells: immunogens using appropriate PrP peptides conjugated with a potent, non-self Th epitope could enhance B cell maturation and sustain long-term and high-affinity Ab production in tolerant mice.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by funds from Groupe d’Intérêt Scientifique “maladies à prions” and European 6th Framework programme network of excellence “Neuroprion.”

4

Abbreviations used in this paper: TSE, transmissible spongiform encephalopathy; PrP, prion protein; PrPC, normal cellular PrP; wt, wild type; pDNA, plasmid DNA; PrPSc, pathological conformer PrP; MFI, mean fluorescence intensity.

1
Prusiner, S. B..
1998
. Prions.
Proc. Natl. Acad. Sci. USA
95
:
13363
.-13383.
2
Prusiner, S. B., M. R. Scott, S. J. De Armond, F. E. Cohen.
1998
. Prion protein biology.
Cell
93
:
337
.-348.
3
Aucouturier, P., C. Carnaud.
2002
. The immune system and prion diseases: a relationship of complicity and blindness.
J. Leukocyte Biol.
72
:
1075
.-1083.
4
Porter, D. D., H. G. Porter, N. A. Cox.
1973
. Failure to demonstrate a humoral immune response to scrapie infection in mice.
J. Immunol.
111
:
1407
.-1410.
5
Enari, M., E. Flechsig, C. Weissmann.
2001
. Scrapie prion protein accumulation by scrapie-infected neuroblastoma cells abrogated by exposure to a prion protein antibody.
Proc. Natl. Acad. Sci. USA
98
:
9295
.-9299.
6
Gilch, S., F. Wopfner, I. Renner-Muller, E. Kremmer, C. Bauer, E. Wolf, G. Brem, M. H. Groschup, H. M. Schatzl.
2003
. Polyclonal anti-PrP auto-antibodies induced with dimeric PrP interfere efficiently with PrPSc propagation in prion-infected cells.
J. Biol. Chem.
278
:
18524
.-18531.
7
Peretz, D. R., A. Williamson, K. Kaneko, J. Vergara, E. Leclerc, G. Schmitt-Ulms, I. R. Mehlhorn, G. Legname, Wormald, P. M. Rudd, et al. 2001. Antibodies inhibit prion propagation and clear cell cultures of prion infectivity. Nature 412: 739–743.
8
White, A. R., P. Enever, M. Tayebi, R. Mushens, J. Linehan, S. Brandner, D. Anstee, J. Collinge, S. Hawke.
2003
. Monoclonal antibodies inhibit prion replication and delay the development of prion disease.
Nature
422
:
80
.-83.
9
Beringue, V., D. Vilette, G. Mallison, F. Archer, M. Kaisar, M. Tayebi, G. S. Jackson, A. R. Clarke, H. Laude, J. Collinge, S. Hawke.
2004
. PrPSc binding antibodies are potent inhibitors of prion replication in cell.
J. Biol. Chem.
279
:
39671
.-39676.
10
Sigurdsson, E. M., M. S. Sy, R. Li, H. Sholtzova, R. J. Kascsak, R. Kascsak, R. Carp, H. C. Meeker, B. Frangione, T. Wisniewski.
2003
. Anti-prion antibodies for prophylaxis following prion exposure in mice.
Neurosci. Lett.
336
:
185
.-187.
11
Heppner, F. L., C. Musahl, I. Arrighi, M. A. Klein, T. Rulicke, B. Oesch, R. M. Zinkernagel, U. Kalinke, A. Aguzzi.
2001
. Prevention of scrapie pathogenesis by transgenic expression of anti-prion protein antibodies.
Science
294
:
178
.-182.
12
Grégoire, S., C. Logre, P. Metharom, E. Loing, J. Chomillier, M. Bruley Rosset, P. Aucouturier, C. Carnaud.
2004
. Identification of two immunogenic domains of the prion protein—PrP—which activate class II-restricted T cells and elicit antibody responses against the native molecule.
J. Leukocyte Biol.
76
:
125
.-134.
13
Bendheim, P. E., H. R. Brown, R. D. Rudelli, L. J. Scala, N. L. Goller, G. Y. Wen, R. J. Kascsak, N. R. Cashman, D. C. Bolton.
1992
. Nearly ubiquitous tissue distribution of the scrapie agent precursor protein.
Neurology
42
:
149
.-160.
14
Steinhoff, U., C. Burkhartt, H. Arnheiter, H. Hengartner, R. Zinkernagel.
1994
. Virus or hapten-carrier complex can activate autoreactive B cells by providing linked T help.
Eur. J. Immunol.
24
:
773
.-776.
15
Krieg, A..
2003
. CpG motifs: the active ingredient in bacterial extracts?.
Nat. Med.
9
:
831
.-835.
16
Bruley Rosset, M., C. Ballerini, S. Grégoire, P. Metharom, C. Carnaud, P. Aucouturier.
2004
. Breaking immune tolerance to the prion protein (PrP) with PrP peptides plus oligo-CpG in mice.
J. Immunol.
172
:
5168
.-5174.
17
Bueler, H., M. Fischer, Y. Lang, H. Bluethmann, H. P. Lipp, S. J. De Armond, S. B. Prusiner, M. Aguet, C. Weissmann.
1992
. Normal development and behaviour of mice lacking the neuronal cell-surface PrP protein.
Nature
356
:
577
.-582.
18
Ballas, Z. K., A. M. Krieg, T. Warren, W. Ramussen, H. L. M. Davis Waldschmidt, G. J. Weiner.
2001
. Divergent therapeutic and immunologic effects of oligodeoxynucleotides with distinct CpG motifs.
J. Immunol.
167
:
4878
.-4886.
19
Stern, B. V., B. O. Boehm, M. Tary-Lehmann.
2002
. Vaccination with tumor peptide in CpG adjuvant protects via IFN-γ-dependent CD4 cell immunity.
J. Immunol.
168
:
6099
.-6105.
20
Krug, A., A. Towarowski, S. Britsch, S. Rothenfusser, V. Hornung, R. Bals, T. Giese, H. Engelmann, S. Endres, A. M. Krieg, G. Hartmann.
2001
. Toll-like receptor expression reveals CpG DNA as a unique microbial stimulus for plasmacytoid dendritic cells which synergizes with CD40 ligand to induce high amount of IL-12.
Eur. J. Immunol.
31
:
3026
.-3037.
21
Krieg, A. M., A. K. Yi, S. Matson, T. J. Waldschmidt, G. A. Bishop, R. Teasdale, G. A. Koretzeky, D. M. Klinman.
1995
. CpG motifs in bacterial DNA trigger direct B-cell activation.
Nature
374
:
546
.-549.
22
Yip, H. C., A. Y. Karulin, M. Tary-Lehmann, M. D. Hesse, H. Radeke, P. S. Heeger, R. P. Trezza, F. P. Heinzel, T. Forsthuber, P. V. Lehmann.
1999
. Adjuvant-guided type-1 and type-2 immunity: infectious/non-infectious dichotomy defines the class of response.
J. Immunol.
162
:
3942
.-3949.
23
Ichikawa, H. T., L. P. Williams, B. M. Segal.
2002
. Activation of APCs through CD40 or Toll-like receptor 9 overcomes tolerance and precipitates autoimmune disease.
J. Immunol.
169
:
2781
.-2787.
24
Polymenidou, M., F. L. Heppner, E. C. Pellicioli, E. Urich, G. Miele, N. Braun, F. Wopfner, H. M. Shätzl, B. Becher, A. Aguzzi.
2004
. Humoral immune response to native eukaryotic prion protein correlates with anti-prion protection.
Proc. Natl. Acad. Sci. USA
101
:
14670
.-14676.
25
Volland, H., P. Pradelles, P. Ronco, M. Azizi, D. Simon, C. Creminon, J. Grassi.
1999
. A solid-phase immobilized epitope immunoassay (SPIE-IA) permitting very sensitive and specific measurement of angiotensin II in plasma.
J. Immunol. Methods
228
:
37
.-47.
26
Fulcher, D. A., A. Basten.
1997
. B-cell activation versus tolerance. The central role of immunoglobulin receptor engagement and T-cell help.
Int. Rev. Immunol.
15
:
33
.-52.
27
Tsujihata, Y., Y. Mizukami, T. Ueda, T. So, T. Imoto.
2002
. B-cell repertoire specific for unfolded self-determinant of mouse lysozyme escape tolerance and dominantly participate in the autoantibody response.
Immunology
107
:
394
.-402.
28
Mongini, P. K., D. L. Longo, W. E. Paul.
1984
. T cell regulation of immunoglobulin class expression in the B cell response to TNP-Ficoll: characterization of the T cell responsible for preferential enhancement of the IgG2a response.
J. Immunol.
132
:
1647
.-1653.
29
Mongini, P. K., W. E. Paul, E. S. Metcalf.
1982
. T cell regulation of immunoglobulin class expression in the antibody response to trinitrophenyl-Ficoll: evidence for T cell enhancement of the immunoglobulin class switch.
J. Exp. Med.
155
:
884
.-902.
30
Schwarz, A., O. Kratke, M. Burwinkel, C. Riemer, J. Schultz, P. Henklein, T. Bamme, M. Baier.
2003
. Immunisation with a synthetic prion protein-derived peptide prolongs survival times of mice orally exposed to the scrapie agent.
Neurosci. Lett.
350
:
187
.-189.
31
Sigurdsson, E. M., D. R. Brown, M. Daniels, R. J. Kascsak, R. Kascsak, R. Carp, H. C. Meeker, B. Frangione, T. Wisniewski.
2002
. Immunization delays the onset of prion disease in mice.
Am. J. Pathol.
161
:
13
.-17.