|
|
||||||||

* Institut National de la Santé et de la Recherche Médicale UMRS 893, Paris; Université Pierre et Marie Curie Univ Paris 06, Hôpital Saint-Antoine, Paris, France; and
Centre National de la Recherche Scientifique/Unité Mixte de Recherche; Université Pierre et Marie Curie Univ Paris 06, Hôpital La Pitié Salpetrière, Paris, France
| Abstract |
|---|
|
|
|---|
secreting T cells and poor and variable Ab responses. Both treatments impaired early accumulation of PrPres in the spleen and prolonged survival of mice infected with 139A scrapie. Additional P9 boosts after 139A infection sustained the T cell response and partially inhibited PrPres early accumulation but did not improve the survival. Surprisingly, when P9 injections were started 1 mo after infection and repeated subsequently, specific T cell and Ab responses were impaired and no beneficial effect on prion disease was observed. After a single injection of P9, the number of IFN-
secreting CD4+ T cells was also reduced in mice 8- to 10-wk postinfection compared with healthy mice. In vivo and in vitro removal of CD4+CD25+ T cells restored the T cell response to P9 in infected mice. In conclusion, CD4+ T cells as well as Abs might participate to the protection against scrapie. Of importance, the peripheral accumulation of PrPres during infection negatively interferes with the development of T and B cell responses to PrP and regulatory T cells might contribute to this phenomenon. | Introduction |
|---|
|
|
|---|
Besides Abs, other immune effectors, including CD4+ T cells specific for the PrP-derived peptides, might interfere with prion progression. We and others (24, 25, 26, 27, 28, 29) have characterized PrP-specific CD4+ T cells. A major immunogenic MHC Iab-restricted T cell epitope was identified at aa 155–170 (29) and two B cell epitopes including aa 112–116 and 173–187 (23). Injection of P5 (PrP98–127) or P9 (PrP158–187) peptides together with CpG oligodeoxynucleotides in IFA can break tolerance and lead to the recruitment of specific T cells and Abs in C57BL/6 wt mice (30). Peptide P5 mainly induces specific Abs and a very weak CD4+ Th cell response. In contrast, P9 stimulates CD4+ Th1 (IFN-
secreting) and Th2 (IL-4 secreting) cells but few B cells that produce low and variable levels of IgG2b and no IgG1, indicating a strong requirement of T cell help.
In this study, we tested the protective potential of P5 and P9 vaccination in experimental murine scrapie. Although PrP155–170 was characterized as the major T cell epitope in previous studies (29), PrP158–187 was used for immunization because it induced twice more frequent IFN-
secreting T cells, likely due to the presence of a second epitope at aa 167–181 (data not shown). Results indicate that active immunization with both peptides before scrapie infection delays the onset of clinical symptoms significantly, but prolongs moderately the duration of the clinical stage of scrapie infected mice. However, additional boosts abolished the protective effect and impaired T and B cell responses to P9 when treatment was initiated in previously infected mice. Furthermore, in vitro and in vivo depletion of CD4+CD25+ cells completely restored the immune response to P9 of infected mice, which for the first time suggests a connection between PrPSc accumulation and regulatory T cells (Treg).
| Materials and Methods |
|---|
|
|
|---|
Six-week-old female C57BL/6 mice were purchased from Janvier and housed in individual ventilated cages under strict pathogen-free conditions, in compliance with European recommendations. Mice were injected s.c. at the base of the tail with PrP peptides (50 or 100 µg) mixed with CpG (50 µg) and emulsified in IFA (v/v) (23, 30). One or several boosts were performed as described in the Results section, and spleen and blood samples collected for in vitro analysis 10–14 days after the last injection.
Peptides and adjuvants
PrP peptides P5 (98–127) and P9 (158–187) were synthesized by Neosystem and purified by HPLC on a C8 reverse phase column and identified by electrospray mass spectrometry with purity >80%. CpG oligodeoxynucleotide N° 1826 was synthesized by Sigma-Aldrich.
In vivo depletion of CD25+ cells
Mice received two i.p. injections of 100 µg of anti-CD25 mAb (ascites prepared in nude mice with PC61 hybridoma) at days –10 and –4 before immunization with P9.
Scrapie inoculation and disease scoring
Mice were inoculated i.p. with 100 µl of 0.5% homogenate prepared from terminal 139A-scrapie brain, corresponding to 104.33 ID50 (31). Mice were monitored twice a week for clinical disease starting at 20-wk postinfection (p.i.), by observing their activity levels and competence on a set of parallel bars as described previously (32).
ELISPOT assay
The number of IFN-
-producing cells from the spleens of immunized mice was evaluated by ELISPOT assay. In brief, nitrocellulose-based 96-well plates (Millipore) were coated with anti-mouse IFN-
capture Abs (1/500) (BD Biosciences) for 2 h at 37°C, followed by an overnight incubation at 4°C. Plates were blocked with medium containing 10% FCS for 2 h at 37°C. Responders from individual mice were seeded at 106 cells/well for total splenocytes and stimulated with 10 µg/ml peptide or medium. CD4+ T cells were enriched from spleen cells by negative magnetic selection (Dynal and Invitrogen) and added at 2 x 105 cells/well; CD4+CD25– cells were further purified from CD4+, negatively selected with CD25+ beads (Miltenyi Biotec), and added at 2 x 105 cells/well. Spleen dendritic cells (DC) were purified by positive selection on CD11c+ magnetic kit (Miltenyi Biotec), incubated 4 h with 10 µg/ml P9 in the presence of 10 µg/ml GM-CSF, and added to responders at 5 x 104 cells/wells. Plates were incubated at 37°C in 5% CO2 for 24 h, washed with PBS-T, and then incubated 2 h at 37°C with biotinylated anti-mouse IFN-
detection Abs (BD Biosciences). After washing, alkaline phosphatase conjugated to streptavidin was added (Roche) (1/500 dilution, 100 µl/well) for 2 h. Secreting cells were visualized using tetrazolium nitroblue/bromochloro-indolylphosphate substrate (Promega) and spots were counted using an automatic ELISPOT plate reader (ALD). Test wells were assayed in triplicate and the frequency of peptide-specific T cells was calculated after subtracting the mean number of spots obtained in the absence of peptide.
Ab titration (ELISA)
Plates (Maxisorp and Nunc) were coated with 10 µg/ml peptides in sodium carbonate buffer (0.05M; pH 9.6) overnight at 4°C, washed with PBS and blocked with 1% nonfat milk in PBS-Tween 20 for 2h at 37°C. One to fifty diluted sera from immunized and control mice were added in duplicate and incubated overnight at 4°C. After washing, 200 µl of peroxidase conjugated anti-mouse Ig (1/5000) was added and left 2 h at room temperature. To determine the Ab isotype, detection was performed using peroxidase-conjugated goat anti-IgG1, IgG2b, and IgG2a mouse Abs (Southern Biotechnology Associates). Plates were then washed and 200 µl/well of freshly prepared H2O2/O-phenylediamine substrate solution (Sigma-Aldrich) was added. The reaction was stopped with sulfuric acid (2N) and plates were read at 492 nm.
Flow cytometry
EL4 cells overexpressing murine PrPc (allotype s7) were obtained after transfection as described (29). 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 EL4 cells using FITC-conjugated anti-PrP mAb SAF83 (from Dr. Grassi, Commissariat à l'Energie Atomique, 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 Cell Quest software (BD Biosciences). A serum was considered significantly positive for native PrP binding when the mean fluorescence intensity (MFI) of peptide-immunized serum was over the MFI plus 3SD of sera from mice treated with CpG alone.
Western blot analysis of PrPres
Spleen homogenates (10%) were prepared using Fast Prep technique. A total of 500 µl of supernatant was mixed with an equal volume of 4% sarkosyl in PBS (pH 7.4). Samples were digested with 50 u/ml Benzonase (Benzon nuclease, purity 1; Merck) for 30 min at 37°C. Samples were then precipitated with 0.3% sodium phosphotungstic acid at 37°C for 30 min and pellets were resuspended in 20 µl PBS 0.1% sarkosyl.
Samples were digested with 20 µg/ml proteinase K (PK) at 37°C for 15 min, blocked by addition of loading buffer (125 mM Tris (pH 6.8), 20% glycerol, 4% 2-ME, 8 mM 4-(2-aminoethyl)-benzene sulfonyl fluoride, and 0.02% bromophenol blue) and heated at 100°C for 10 min. After electrophoresis in a 12.5% polyacrylamide/SDS gel, proteins were electrotransferred onto a polyvinylidene difluoride membrane and blocked in 5% nonfat milk 0.1% PBS-T. Blots were incubated overnight at 4°C in a 1/30,000 dilution of anti-PrP mAb (SAF83), then for 1h with peroxidased-conjugated goat anti-mouse IgG Ab, and finally proteins were revealed by chemoluminescence (ECL+; Amersham Biosciences).
Statistical analyses
Comparison of mean incubation times, survivals, or durations of clinical stages was performed with the Mann-Whitney U test. Mean precursor frequencies were compared using the Students t test.
| Results |
|---|
|
|
|---|
To examine whether specific immune responses to PrP protect from murine scrapie, C57BL/6 mice were immunized twice at 10- to 14-day interval with 100 µg of P5 or P9 in CpG/IFA 1 wk before i.p. inoculation with 0.5% 139A scrapie. T cell and Ab responses as well as the early accumulation of splenic PrPres were monitored and mice followed for clinical signs of scrapie and survival. The duration of the T cell response was first investigated in healthy C57BL/6 mice: consistent with previous results (30), we observed that 2 wk after the second injection of 100 µg of PrP peptides, P9 treatment induced a strong IFN-
-secreting T cell response while P5 induced a weak response in all mice (Fig. 1A). Six weeks later, three of five P9-treated mice still displayed a significant number of IFN-
-secreting cells upon in vitro stimulation with the corresponding peptide while response to P5 remained low. T cell responses lessened 12 wk after immunization in nearly all mice except one that was treated with P9. We then did the same experiment with mice infected 1 wk after the second injection of 100 µg of peptide/CpG. We observed that a strong IFN-
-secreting T cell response developed 2 wk later upon in vitro stimulation with P9 but not with P5 as observed in healthy mice (Fig. 1B). Six weeks later, only two of six P9-treated and none of the P5-treated mice displayed a significant number IFN-
-secreting cells upon in vitro stimulation with the corresponding peptide. T cell responses were no longer detected 8 wk after immunization. Spleen cells from untreated or CpG-treated infected mice were never found to secrete IFN-
in the same conditions of stimulation.
|
|
|
|
We postulated that long term maintenance of specific immune response, by repeating peptide injections after scrapie inoculation, will improve the treatment efficiency. As the level of Abs to P5 remained high during the 4 mo of observation, following experiments were focused on T cell stimulation by P9 because it was transient. We first tested whether repeated injection of P9 would sustain the response into healthy C57BL/6 mice. Mice received four consecutive P9 injections of 50 or 100 µg/mouse and results on Fig. 4 indicated that a significant T cell response developed immediately after the first injection and was amplified after the 2nd, 3rd, and 4th injection. Both doses of peptide resulted in the same level of immune response.
|
secreting T cells (mean ± SD) was 190 ± 96/106 cells after two injections and remained at 128 ± 81/106 cells after the 4th injection at 6-wk p.i. (Fig. 5a). Serum Abs were detectable after two P9 injections in nearly half of the 11 tested mice (Fig. 5b) and titers increased in several negative mice after the 2nd and 3rd boost. Spleen PrPres level measured by Western blot was lower at 12-wk p.i. in P9-treated mice (Fig. 7). Additional injections of 50 µg of P9 did not result in an increased incubation period or survival (Table II). Of note, injections of CpG before infection significantly increased the incubation period (p = 0.038) but not the survival.
|
|
To see whether treatment with P9 might interfere with prion progression at a later stage of incubation, the first two immunizations with 50 µg/mouse were performed 4 and 6 wk after scrapie inoculation. Two additional boosts at the same dose were subsequently applied at 10- and 12-wk p.i. (protocol B, Fig. 6). A specific T cell response developed with a mean frequency (±SD) of 128 ± 36/106 splenocytes after two immunizations but in contrast with results obtained in healthy mice (Fig. 4), this response significantly declined to 27.5 ± 13.6/106 after the 2nd and 3rd boost (Fig. 6a; p = 0.017). The eight mice treated with P9/CpG produced no or very low levels of Abs even after three boosts (Fig. 6b). In contrast to mice that had been infected after treatment with P9, accumulation of PrPres in the spleen was not decreased in mice similarly treated 4 wk after inoculation (Fig. 7). Incubation periods and duration of overt disease were identical (Table II). Injections of CpG starting after infection had no effect on the incubation period but slightly increased the duration of the clinical phase (p = 0.02).
|
|
The above data showed that in infected mice, decline in T and B cell responses to P9 appeared only 10–12 wk after scrapie inoculation, suggesting a connection with PrPSc splenic accumulation. As anti-P9 Ab production by B cells depends on T cell help, a defective CD4+ compartment is likely to be responsible for their absence. One possible mechanism may relate to CD4+CD25+ Treg. We, thus, examine whether in vivo depletion of CD4+CD25+ T cells using anti-CD25 mAb treatment before a single P9 immunization would influence the response to P9 in both non-infected and 8–10 wk infected mice. For this purpose, mice were infected or not with 139A 8–10 wk before receiving a single injection of 100 µg of P9/CpG or CpG only. First, the frequency (mean ± SE) of specific IFN-
-secreting T cells was lower in spleens from infected mice (53.7 ± 16/106 cells) than in non-infected mice (92.3 ± 15/106 cells) but the difference was not significant (p = 0.10) (Fig. 8A). When anti-CD25 treatment was applied 10 and 4 days before P9 immunization in non-infected mice, a similar frequency of such T cells was found in the spleen (83.0 ± 24 vs 92.3 ± 15/106 cells). In contrast, this response was enhanced by 3-fold (158 ± 18 vs 53.7 ± 16/106 cells; p < 0.01) in infected mice treated in vivo with anti-CD25 before P9 immunization, resulting in restoration of anti-P9 response to a level superior to non-infected mice. These data suggest an inhibitory effect for CD25+ T cells and were confirmed by in vitro depletion of CD25+ cells from the CD4+ compartment (Fig. 8B). Indeed, the CD4+CD25– cell population purified from spleens of immunized infected mice displayed a level of response to P9 superior to that of spleens from immunized non-infected mice (401 ± 23 vs 250 ± 38/106 cells; p < 0.01). In this experiment, the frequency of P9-specific CD4+ cells was significantly lower (p = 0.057) in the spleen of infected (82 ± 24/106 cells) than of non-infected mice (205 ± 18/106 cells).
|
| Discussion |
|---|
|
|
|---|
We immunized with two PrP peptides possessing B and T cell epitopes previously shown to recruit Ab and CD4+ T cell functional repertoires specific for PrP when injected with CpG/IFA as adjuvant in C57BL/6 wt mice (30). Treatment of mice with either P5 or P9 partially inhibited early prion accumulation in the spleen and increased both the incubation period and the duration of the clinical phase, compared with untreated mice or mice having received CpG alone, resulting in total survival prolongation of 20 and 15 days, respectively. The overall effects, although moderate, were statistically significant.
Analyses of immune responses generated by PrP peptides P5 and P9 suggest that humoral and cellular T cell effectors could participate in protection. A long lasting Ab response and few or no IFN-
-secreting T cells developed in all P5 treated-mice at 2, 6, and 8 wk after immunization. This weak T cell response apparently provided sufficient help for isotype switching to IgG1 and IgG2a/b Abs. At variance to P5 peptide, P9 with CpG mostly induced specific T cells that secreted IFN-
, as previously observed (23, 30). In 139A scrapie-infected mice, this response declined 6 wk after immunization and was no longer measurable 8 wk after immunization. Abs to P9 developed in approximately half of infected mice, and titers varied during the incubation period. The kinetic and magnitude of immune responses to these peptides resemble that developed in healthy mice similarly immunized, although the T cell response to P9 seems to decline quicker in infected mice (Fig. 1, B vs A). As already observed in non-infected mice (30), anti-P9 Abs are predominantly of IgG2a/2bsubclasses and no IgG1 were detected suggesting a strong requirement of T cell help. All these results are in agreement with previous data (23, 24, 25, 26, 27, 28, 29, 30) showing that P5/CpG mainly stimulates an humoral response with a weak CD4+ T cell response while P9 evokes CD4+ Th1 (IFN-
secreting) cells but low and variable levels of Abs that bind P9 peptide and not native membrane PrPc.
The delayed pathogenesis of scrapie suggests that Abs to P5 might be efficient while anti- P9 responses rather implicate T cell effectors: indeed, all immunized mice generated T cells specific for P9 but Abs were not always detected and no correlation was found between the levels of Abs and the disease time course. Vaccination trials using other forms of Ags, adjuvants or routes of immunization (15, 16, 17, 21) also resulted in moderate prolongation of disease time course. Only one trial using mucosal vaccination with recombinant Salmonella vaccine expressing tandem copies of PrP succeed in protecting 30% of mice from disease but immunization began 7 wk before scrapie challenge (22). In the present work, the survival was prolonged by 20 days upon P5 immunization, which is close to the 23-day increase observed after vaccination with PrP105–125 of mice exposed to scrapie (15). Such a limited effect might relate to poor binding of raised Abs to native membrane-bound PrPc (33) as PrPc binding might impair conversion to PrPSc (10) or decrease the half-life of PrPc (34). Abs to P5 and P9 recognized the specific peptide and plastic-coated recombinant PrP in an ELISA but very poorly the native membrane PrPc. We previously showed that the buried epitopes were located at residues 112–116 and 173–187, respectively (23).
In some studies (16, 19, 20, 21, 22), cellular responses to PrP were not explored although immunogens were likely to contain T cell epitopes. In the work of Fernandez-Borges (35), PrP DNA fused to lysosomal integral membranes stimulated Ab and CD4+ T cell responses to PrP resulting in a dramatic delay in the onset of early clinical signs after intracerebral challenge; but once the disease appeared, it rapidly killed the mice, possibly due to Ab-mediated toxicity. A recent study reported that immunization with dimeric PrP and CpG-oligodeoxynucleotide coencapsulated in polylactide-coglycolide microspheres enhanced specific CD4+ T cell responses and, more prominent, the induction of CD8+ T cells. However, the influence of this treatment on survival of scrapie-infected mice was not addressed (36). In our experiments, CD4+ T cells specific for the 155–170 and 167–181 epitopes present in P9 (23, 24, 25, 26, 27, 28, 29, 30) may also have contributed by interfering with scrapie agent accumulation, replication, or transport, possibly through cytokine secretion and activation of phagocytic cells. We attempted to sustain the T cell stimulation by repeating immunogen injections, which proved efficient in healthy mice, but did not improve scrapie progression in infected mice: early spleen PrPres accumulation was transiently impaired at 12-wk p.i. and no beneficial effect was induced on disease time course. When started after scrapie infection, P9/CpG immunization neither reduced the early spleen PrPres accumulation nor prolonged survival. A possible explanation is that induced effectors are only active at early stages of prion propagation.
Alternatively, the lack of protection may relate to a defect in the immune response to P9: specific IFN-
-secreting T cells, while detected after two P9 injections, declined after the 3rd and 4th immunizations and Abs were almost undetectable during the whole period of observation. Because of the predominance of IgG2a/2b Abs and the virtual absence of IgG1 Abs in P9 immune sera of healthy mice (Fig. 2B), suggesting requirement of a strong T cell help, one may propose that T cell defect is critical in the impaired generation of Abs. A low frequency of specific T cells was found in mice infected 8 to 10 wk before receiving a single P9 injection but the difference with non-infected mice did not reach significance. However, when measured within the purified CD4+ cells instead of total spleen cells, the difference of anti-P9 IFN-
-secreting cells became significant. Thus, the impaired response may not be due to a decreased number of specific CD4+ spleen T cells. No immunosuppression has been clearly evidenced so far in scrapie infected mice (37, 38) except a reduced B cell response to LPS (39). Our findings suggest that prion accumulation in lymphoid tissues may interfere with the development of immune responses to PrP.
A possible explanation for the observed impaired response, is the exhaustion phenomenon that has been described after chronic viral infection: upon Ag persistence, T cells may modulate their responsiveness by up-regulating inhibitory receptors such as PD-1 that alterate TCR signaling (40). Alternatively, the suppressive activity of natural CD4+CD25+ Treg may control T cell responses (41). Treg intervention is supported by the complete restoration of the anti-P9 T cell response after anti-CD25 mAb treatment of infected mice, to a level even higher than in non-infected mice. Numerous studies reported that during chronic infections and autoimmune diseases, the number and function of specific CD4+CD25+ Treg are increased or stronger at the site of infection (42). Local Ag persistence is crucial for Treg maintenance, survival, and function (43). Thus, because prions accumulate in secondary lymphoid organs, Treg might participate in scrapie pathogenesis. To address this issue, a better characterization of CD4+CD25+Foxp3+ number and function in transmissible spongiform encephalopathies is required. Evaluation of the impact of Treg cells on the natural course of prion infection is now under investigation in Treg-manipulated mice. Of note, a defective function of APC is also possible: PrPSc accumulation in DC might result in deficient Ag processing and presentation to T cells.
The lack of protection when peptide vaccinations were repeated, even when started before infection, is unclear. Paradoxically, good T cell and Ab responses were observed in these conditions. The partial inhibition of splenic PrPres accumulation indicated an early beneficial effect that proved insufficient to impair the clinical course. CpG are known to activate DC through TLR-9 (44), which might favor the propagation of prions from the periphery to the CNS (31). Indeed, activation of mononuclear phagocytes by TLR ligands was recently shown to impair prion degradation and facilitate prion replication (45). Impairment of scrapie by CpG (46) was also reported but it was provoked by repeated much higher doses that destroyed the architecture of lymph nodes (47).
In conclusion, Abs and CD4+ T cells specific for PrP peptides may contribute to possible treatments of transmissible spongiform encephalopathies but optimization of Ag/adjuvant delivery is required for the development of efficient vaccination. Impairment of T cell responses when P9 immunization was applied in previously infected mice points to a possible negative interaction between PrPSc accumulation and the development of a specific immune response to PrP. Preliminary, experiments suggested for the first time a role of CD4+CD25+ Treg cells. An increase in number or function of Treg in the lymphoid organs during prion infection, if demonstrated, is of particular importance because it will strongly indicate that the immune system is not blind to infectious PrPSc.
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by funds from Institut National de la Santé et de la Recherche Médicale and the Neuroprion network of excellence, European FP6. ![]()
2 Address correspondence and reprint requests to Dr. Martine Bruley Rosset, Institut National de la Santé et de la Recherche Médicale UMR 893, Bâtiment Kourilsky, Hôpital Saint-Antoine, 184 rue du faubourg Saint Antoine, F-75571 Paris Cedex 12, France. E-mail address: rosset{at}st-antoine.inserm.fr ![]()
3 Abbreviations used in this paper: PrPc, cellular prion protein; DC, dendritic cell; PrPSc, scrapie PrP; PrPres, resistant PrP; p.i., postinfection; Treg, regulatory T cell; wt, wild type; PK, proteinase K. ![]()
Received for publication January 3, 2008. Accepted for publication May 1, 2008.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Gourdain, S. Gregoire, S. Iken, V. Bachy, G. Dorban, T. Chaigneau, H. Debiec, A.-S. Bergot, I. Renault, P. Aucouturier, et al. Adoptive Transfer of T Lymphocytes Sensitized against the Prion Protein Attenuates Prion Invasion in Scrapie-Infected Mice J. Immunol., November 15, 2009; 183(10): 6619 - 6628. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q. Li, S. Teitz-Tennenbaum, E. J. Donald, M. Li, and A. E. Chang In Vivo Sensitized and In Vitro Activated B Cells Mediate Tumor Regression in Cancer Adoptive Immunotherapy J. Immunol., September 1, 2009; 183(5): 3195 - 3203. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |