- Email: [email protected]
Repeated immunization of mice with phosphorylated-tau peptides causes neuroinflammation
Experimental Neurology 248 (2013) 451–456
Contents lists available at ScienceDirect
Experimental Neurology journal homepage: www.elsevier.com/locate/yexnr
Repeated immunization of mice with phosphorylated-tau peptides causes neuroinflammation☆ Lea Rozenstein-Tsalkovich a, Nikolaos Grigoriadis b, Athanasios Lourbopoulos b, Evangelia Nousiopoulou b, Ibrahim Kassis a, Oded Abramsky a, Dimitrios Karussis a, Hanna Rosenmann a,⁎ a b
The Department of Neurology, The Agnes Ginges Center for Human Neurogenetics, Hadassah Hebrew University Medical Center, Jerusalem, Israel The B' Department of Neurology, AHEPA University Hospital, Thessaloniki, Macedonia, Greece
a r t i c l e
i n f o
Article history: Received 7 May 2013 Revised 2 July 2013 Accepted 12 July 2013 Available online 20 July 2013 Keywords: Tauopathy Alzheimer's-disease Tau protein Phosphorylated-tau Neurofibrillary-tangles Immunotherapy
a b s t r a c t The recent studies of others and of us showing robust efficacy of anti-tangle immunotherapy, directed against phosphorylated (phos)-tau protein, may pave the way to clinical trials of phos-tau immunotherapy in Alzheimer's-disease and other tauopathies. At this stage addressing the safety of the phos-tauimmunotherapy is highly needed, particularly since we have previously shown the neurotoxic potential of tau-immunotherapy, specifically of full-length unphosphorylated-tau vaccine under a CNSproinflammatory milieu [induced by emulsification in complete-Freund's-adjuvant (CFA) and pertussistoxin (PT)] in young wild-type (WT)-mice. The aim of our current study was to address safety aspects of the phos-tau-immunotherapy in both neurofibrillary-tangle (NFT)-mice as well as in WT-mice, under challenging conditions of repeated immunizations with phos-tau peptides under a CNS-proinflammatory milieu. NFT- and WT-mice were repeatedly immunized (7 injections in adult-, 4 in aged-mice) with phos-tau peptides emulsified in CFA–PT. A paralytic disease was evident in the phos-tau-immunized adult NFT-mice, developing progressively to 26.7% with the number of injections. Interestingly, the WT-mice were even more prone to develop neuroinflammation following phos-tau immunization, affecting 75% of the immunized mice. Aged mice were less prone to neuroinflammatory manifestations. Anti-phos-tau antibodies, detected in the serum of immunized mice, partially correlated with the neuroinflammation in WT-mice. This points that repeated phos-tau immunizations in the frame of a proinflammatory milieu may be encephalitogenic to tangle-mice, and more robustly to WT-mice, indicating that – under certain conditions – the safety of phos-tau immunotherapy is questionable. © 2013 Elsevier Inc. All rights reserved.
Introduction There is accumulating evidence that targeting selectively pathological tau, particularly the phosphorylated (phos)-tau isoforms (Asuni et al., 2007; Bi et al., 2011; Boimel et al., 2010; Boutajangout et al., 2010; Boutajangout et al., 2011; Chai et al., 2011; Troquier et al., 2012) or the tau-oligomers (Castillo-Carranza et al., 2010), reduces the taupathology and improves the symptoms of dementia in different animal models for tauopathies using various immunization protocols. Emanating basically from the lessons learned from amyloid immunization studies, particularly the fact that adverse effects were first detected in clinical studies (Orgogozo et al., 2003), and it was only afterwards that the adverse effect (neuroinflammation) was remodeled in animals (Furlan et al., 2003; Monsonego et al., 2006), there is a great need in addressing not only the issue of efficacy but also that of the safety of
☆ This study was supported (in part) by grant no. 300000-4895 from the Chief Scientist Office of the Ministry of Health, Israel. ⁎ Corresponding author at: Department of Neurology, Hadassah Hebrew University Medical Center, Ein Karem, Jerusalem 91120, Israel. Fax: + 972 3 7256022. E-mail address: [email protected] (H. Rosenmann). 0014-4886/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.expneurol.2013.07.006
anti-tangle vaccine already at the preclinical stage, before moving to clinical trials. With this approach we have tested tau vaccines emulsified in complete-Freund's-adjuvant (CFA), a T-helper 1 (Th1)-polarizing adjuvant, and pertussis-toxin (PT), a toxin which increases the BBB permeability. This immunization protocol provides a CNS proinflammatory milieu used by us as a setup to detect possible tau vaccine biohazards of developing neuroinflammation in mice. Such proinflammatory conditions have been reported to remodel neuroinflammation in amyloidmice immunized with amyloid vaccine (Furlan et al., 2003), possibly mimicking the condition in the brain of certain elderly AD patients, who developed neuroinflammation in response to amyloid-vaccination. Under this CNS proinflammatory milieu we have previously demonstrated that the full-length unphosphorylated-tau protein vaccine was encephalitogenic in wild-type (WT)-mice, pointing to the biohazardous potential of anti-tau immunotherapy (Rosenmann et al., 2006). Our subsequent results showed that vaccination (1 injection with 1 booster a week later) with phos-tau peptides (in CFA–PT), peptides which selectively targets NFT-related phos-tau protein, showed a higher safety profile (no signs of encephalitis), accompanied by robust efficacy, as evidenced by the reduced NFT-burden and improved cognition of the NFT-mice (Boimel et al., 2010; Rosenmann, 2013). In our current
452
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
study we studied the safety profile of repeated vaccinations with phostau in NFT-mice in order to evaluate whether such repeated immunization protocols (aiming to increase efficacy) can be safely applied in future clinical trials, similarly to the repeated amyloid vaccinations that have been tried in AD patients (Gilman et al., 2005). Repeated phos-tau immunizations have been previously reported to reduce the tangle-burden without significant adverse effects. However in the latter studies a different immunization protocol was used (utilizing a Th2polirizing alum adjuvant without PT), in which the CNS was less challenged by a detrimental proinflammatory milieu (Asuni et al., 2007; Bi et al., 2011; Boutajangout et al., 2010; Troquier et al., 2012). In addition, we also investigated here the safety of phos-tau vaccination in WT-healthy mice, in order to address the potential of early preventive immunization in high-risk healthy individuals. These studies were conducted in both adult- and aged-animals, in order to test a possible impact of age or of disease phase. Our findings here show that repeated phos-tau immunization in the frame of a proinflammatory milieu induced encephalitogenicity in tangle-mice and even more, in WT-mice, indicating that under certain conditions, such as in individuals with inflammatory diseases, phostau immunotherapy might be unsafe, especially in adult clinically intact (non-demented) individuals.
Microglia were stained using the biotinylated lectin-lycopersicon esculentum immunostaining protocol as reported previously (Rosenmann et al., 2006). Microglia were evaluated on 4 paraffin sections under 20× optical fields and data were expressed as cells/mm2. Detection of antibodies in serum by ELISA Sera of immunized-mice and controls were tested for the presence of antibodies against each of the 3 phos-tau peptides by ELISA, according to the protocol reported previously (Boimel et al., 2010). Data analysis Results are presented as Mean ± SEM or as medians. For comparisons between study groups, the unpaired t-test, median test and Chi-square analysis were used. To assess linear associations between two continuous variables, the Pearson correlation coefficient was calculated. Results Mice immunized repeatedly with phos-tau in CFA–PT develop neurological deficits
Methods Mice The NFT-mice (E257T/P301S-tau-tg-mice) generated by us were crossed with C57Bl mice (Rosenmann et al., 2008). Tg offspring were screened to identify NFT-mice, while the non-Tg mice served as WTmice. Experiments were performed in accordance with NIH-guidelines for the care and use of laboratory-animals. Immunizations Six-month-old NFT- and WT-mice (“adult mice”, at about the age at disease onset) (12–15 mice/group) as well at 12 months of age (“aged mice”, at an age when NFTs are prominent) (5–6/group) were subcutaneously immunized repeatedly with a mixture of 3 phos-tau peptides {Tau195–213[P202/205], Tau207–220[P212/214], Tau224–238 [P231]} previously reported by us (Boimel et al., 2010), as follows: 100 μg of each peptide emulsified in non-enriched CFA and PT, with one booster following 2 weeks; followed by monthly repeated injections of 50 μg of each peptide, in non-enriched CFA. The controls (“CFA–PT” groups) received vehicle only, as follows: non-enriched CFA and PT in the first injection, while CFA-alone in the following injections. In the adult mice a total of 7 injections were delivered, while 4 injections in the aged mice. Clinical follow-up Since the paralytic disease that developed in some of the immunized animals is similar to that seen in experimental-autoimmuneencephalomyelitis (EAE), the clinical evaluation during the 7–8 months (from first immunization) follow-up, was performed using the widely used EAE 0–6 score for paralysis (Kaye et al., 2000). Histology and immunohistochemistry (IHC) Brains were removed, post-fixed in 4% paraformaldehyde for 16–20 h at 4 °C and further processed for paraffin sectioning at 6 μm. Sections were stained with hematoxylin–eosin using standard protocols. Inflammation was evaluated on 6–8 randomly selected brain sections (60 μm apart) under 20× optical fields; the total number of infiltrating inflammatory cells per each brain hemisphere (hemisection) was counted and was expressed as cells/hemisection.
Adult mice The clinical follow-up revealed that adult phos-tau-WT-mice immunized repeatedly with 7 injections of phos-tau in CFA–PT developed neurological deficits with the number of affected mice progressively increasing from 5/12 (41.7%), and a mean maximal EAE score of 0.54 ± 0.25 after the 2nd injection, to 7/12 (58.3%), and a mean maximal EAE score of 1.38 ± 0.4 after the 4th- and 9/12 (75%), and mean maximal score 1.88 ± 0.49, (1 dead animal) after the 7th-injection. Two animals, which developed some deficits following the 2nd injection, subsequently recovered. Contrarily, all 12 mice in the adult CFA–PT control group were free of neurological deficits (p = 0.001 for % affected, p = 0.002 for score) (Fig. 1A–B). Adult phos-tau-NFT-mice also developed neurological deficits following the repeated immunization protocol. Two out of 15 mice (13%) developed paralytic signs (mean score of 0.13 ±0 .65) following the 2nd injection and 4/15 (26.7%), with a mean maximal score of 1.07 ± 0.54, (including 2 dead animals) following the 4th- till the 7th-injection. The incidence of the clinical syndrome of paralysis in the adult-immunized NFT-mice was significantly lower than that in the WT-immunized mice (26.7% vs 75%, respectively, p = 0.027), paralleled by a similar difference in the mean maximal EAE scores (1.06 ± 0.55 vs 1.88 ± 0.49, respectively) (Fig. 1C–D). Interestingly, the NFT-mice did develop neurological deficits following injections of CFA–PT alone, at high proportions: 8 out of 13 animals developed an EAE-like disease (61.5%, score of 1.19 ± 0.36) already after the 2nd injection, with 2 dead animals after the 5th injection. The percentage of affected mice further increased to 9/13 (69.2%) with a score of 2.75 ± 0.6, (including 4 dead mice) after the 7th injection. This incidence was significantly higher than that in the phos-tauNFT-mice (p = 0.009 for the incidence and p = 0.04 for clinical score) (Fig. 1C–D). There was also a trend of higher mortality in the group of the CFA–PT–NFT-mice relative to the phos-tau-NFT-mice (4/13 and, 2/15, respectively). This was significantly different from the group of WT-mice who did not develop any deficits in response to the CFA–PT injections (as is actually expected from healthy mice) (p b 0.001). Aged mice The repeated immunization protocol was also tested in aged animals. However, due to the small number of animals (5–6/group) the animals were vaccinated only with 4 injections, in order to allow histopathological examinations. Among the phos-tau-WT-mice 1/6
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
453
Fig. 1. Neurological deficits in adult WT- and NFT-mice immunized (7 times repeatedly) with phos-tau peptides in CFA–PT. (A) Cumulative % of affected WT-immunized-mice. EAE score was examined in time intervals between injections of phos-tau peptides in CFA–PT or of CFA–PT alone. Adult phos-tau-WT-mice developed neurological deficits, while WT-control (CFA–PT only) animals stayed free of deficits [p = 0.001] (n = 12/group). (B) Mean EAE score in WT-immunized-mice. Adult phos-tau WT-mice developed neurological deficits while CFA–PT–WT-mice stayed free of deficits [p = 0.002]. (C) Cumulative % of affected NFT-immunized-mice. Both adult phos-tau NFT-mice and CFA–PT control NFT-mice developed neurological deficits, with a significantly higher % of affected animals among the CFA–PT mice relative to the phos-tau immunized NFT-mice [p = 0.009] (n = 13–15/group). (D) Mean EAE score in NFT-immunized mice. The adult CFA–PT–NFT-mice showed higher mean maximal score than the phos-tau immunized NFT-mice [p = 0.04]. Among the phos-tau immunized mice the % of affected mice was higher in the WT- than in the NFT-mice [p = 0.03]. Among the CFA–PT controls the NFT-mice showed significantly higher % of affected mice than the WT-mice [p b 0.001].Among the phos-tau immunized mice there was a trend of a higher mean EAE score in the WT-mice than in the NFT-mice. Among the CFA–PT controls the NFT-mice showed significantly higher mean score than the WT-mice [p = 0.0008].
developed deficits (and died) (16.7%, mean maximal EAE score 1) after the full protocol of 4 injections. Interestingly, neurological deficits were also detected in a similar percentage among the aged CFA–
PT–WT-mice (1/6, 16.7%), but with a trend of a lower maximal EAE score (0.33 ± 0.33) (Fig. 2A–B), making the response of aged WT-mice to CFA–PT a reminiscence to that of the NFT-mice.
Fig. 2. Neurological deficits in aged NFT- and WT-mice immunized (4 times repeatedly) with phos-tau peptides in CFA–PT. (A) Cumulative % of affected WT-immunized-mice. EAE score was examined in time intervals between injections of phos-tau peptides in CFA–PT or of CFA–PT alone. Among the WT-mice both the phos-tau immunized and the CFA–PT controls developed neurological deficits with similar % of affected mice (n = 6/group). (B) Mean EAE score in WT-immunized-mice. There was a trend showing that the aged phos-tau WT-mice developed neurological deficits with a higher EAE score than the CFA–PT WT-mice. (C) Cumulative % of affected NFT-immunized-mice. The phos-tau NFT-mice stayed free of deficits, while the CFA–PT NFT-mice did develop neurological deficits (n = 5/group). (D) Mean EAE score in NFT-immunized-mice. There is a trend of a higher % maximal EAE score in the CFA–PT NFT-mice relative to the phos-tau NFT-mice.
454
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
The cumulative incidence of EAE-like paralytic disease among the phos-tau-WT-mice was lower in the aged mice relative to the adult mice (0%, 0%, 16.7% vs 41.7%, 58.3%, 58.3%) but this trend did not reach statistical significance (p = 0.06). None of the 5 phos-tau-NFTmice developed neurological deficits. Similar neurological deficits observed in the adult CFA–PT–NFT-mice were also noticed in 1/5 (20%, score 0.4 ± 0.4) of the aged CFA–PT–NFT-mice following the 3rd injection, and in 2/5 (40%, score 1.6 ± 0.42, 1/5 dead animal) following the 4th injection (Fig. 2C–D), though, with lower incidence relatively to the adult-mice [0% (p = 0.019), 20% and 40% for the 2th, 3th, and 4th injections, respectively, as compared to 61.5% in the adult mice].
Histopathological evidence of neuroinflammation in animals immunized with repeated tau-peptide vaccines Monocytic infiltrates in brains of mice immunized repeatedly with phos-tau in CFA–PT As shown in Fig. 3A–C, monocytic infiltrates in the brain were detected in 8/11 (72.7%) adult phos-tau-WT-mice tested (not including the deceased animals). The infiltrates were detected in the cerebellum, the corpus callosum, and even in the fibria and cortex. Only 1/12 animals among the CFA–PT controls (8.3%) (p = 0.002) had evidence of such CNS infiltrates. These results are in good correlation with the clinical manifestations (66.7% and 0% in the phos-tau-WT-mice and CFA–PT–WT-controls, respectively). Among the mice presenting clinical deficits 88% had monocytic infiltrates in their brain. Similar results were obtained when comparing the infiltrate burden (80.05 ± 22.27 vs 10.00 ± 6.98 mean infiltrates/hemisection, p = 0.007; corresponding medians: 45.0 vs 0.0 cells/hemisection, p = 0.01, respectively),
with some correlation between the total burden of inflammation and the clinical score (r = 0.214). Among the adult NFT-mice, monocytic infiltrates were detected in the brains of 5/13 phos-tau-immunized-mice (38.4%), and in 4/9 CFA–PT control animals (44.4%); the infiltrate burden medians were: 33.0 vs 56.0 cells/hemisection, respectively (p = NS). The latter finding indicates a trend of correlation with the proportion of animals with clinical neurological deficits (26.7% vs 75% respectively). Among the mice presenting clinical deficits 68% had also brain infiltrates. The percentage of the affected mice with infiltrates was higher in phos-tau-WT-mice than in phos-tau-NFT-mice (72.7% vs 38.4%, p = 0.09, infiltrate burden medians: 45 and 33 respectively, p = NS) and lower in CFA–PT–WT-mice than jn CFA–PT–NFT-mice (8.3% vs 44.4%, p = 0.05; infiltrate medians: 0 and 56, p = 0.041; infiltrate means: 10.00 ± 6.98 vs 87.67 ± 27.44, respectively, p = 0.011). No infiltrates were noticed in the brains of the aged mice tested. Microglial burden did not change in mice immunized with phos-tau in CFA–PT As presented in Fig. 3A–C, lectin staining for microglia revealed that phos-tau immunization in the adult mice did not increase the overall microglial burden relative to the CFA–PT controls, both in the NFTand in the WT-mice. Specifically the microglial burden was: 6.74 ± 0.27 vs 7.07 ± 0.25 microglia/mm2 in the phos-tau-WT- and CFA– PT–WT-mice, respectively; 6.14 ± 0.19 vs 6.49 ± 0.24 microglia/mm2 in phos-tau-NFT- and CFA–PT–NFT-mice respectively (p = NS). Similar results were also observed in the aged mice groups. However, it should be noted that local inflammatory infiltrations were associated with an increased local microglial/macrophage (lectin + cells) reaction, as illustrated in Fig. 3A.
Fig. 3. Infiltrates and microglia in adult NFT- and WT-mice immunized (7 times repeatedly) with phos-tau peptides in CFA–PT. (A) Lectin immunostaining with hematoxylin counterstaining displaying microglia/macrophages (brown cells) and perivascular infiltrates (aggregates of blue monocytes, arrows in A), in the brains of WT- and NFT-mice immunized with phos-tau peptides in CFA–PT (A1 and A3 respectively), as well as in control NFT-mice injected with CFA–PT only (A2). Very few infiltrates, if at all, were detected in the CFA–PT–WTmice (not shown). (B) Quantitative analysis of the infiltrates is presented as % of mice affected by brain infiltrates. Phos-tau-immunized WT animals were significantly more affected compared to the control (CFA–PT-immunized) WT mice (p = 0.002); more affected animals in the CFA–PT NFT-group vs CFA–PT WT mice were also observed (p = 0.05), with some trend of higher % of affected mice in the CFA–PT NFT than in phos-tau-immunized NFT-mice. (C) Quantitative analysis of the infiltrates is presented as median infiltrate burden per brain hemisection. Boxplot graph illustrates the median values (dots and diamonds indicate extreme values). Phos-tau-immunized WT animals showed significantly more brain infiltrates than control (CFA–PT-immunized) WT mice (p = 0.01); higher burden of brain infiltrates in the CFA–PT NFT-group than in the CFA–PT WT mice were also observed (p = 0.041), with some trend of more infiltrates in the CFA–PT NFT than in phos-tau-immunized NFT-mice.
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
Antibodies against phos-tau peptides in the serum of immunized mice Immunized animals developed antibodies against the phos-tau peptides of the vaccine (Fig. 4A–B). Some positive correlation was noticed between the antibody levels in the phos-tau-WT-mice and the quantity of brain infiltrates: r = 0.457 for Tau207–220[P212/214], and r = 0.485 for the Tau224–238[P231] peptide. Due to the lower number (n = 3) of affected mice among the phos-tau-NFT-mice (2 of them died without being available for pathological examination of infiltrates) the possibility of such a correlation could not be confirmed, concerning the immunized NFT-mice. Discussion We have shown here that repeated immunization of adult WT-mice with phos-tau in CFA–PT causes neuroinflammation that is dependent on the number of immunizations, reaching an incidence of 75% after 7 injections. Adult NFT-mice immunized repeatedly with the same protocol were significantly more resistant to the development of neuroinflammation (incidence: 26.7%). Additionally, aged mice also developed neuroinflammation in response to repeated immunizations (4 injections), but to a lower degree. A moderate association between serum anti-phos-tau antibodies and monocytic infiltrates in the brains was observed in phos-tau immunized WT-mice. Interestingly, almost 70% of the CFA–PT–NFT-mice developed neurological disability following the repeated injections, whereas CFA–PT–WT-mice remained symptoms-free. The finding that almost three fourths of the WT-mice, and particularly the adult ones, develop neuroinflammation in response to repeated phos-tau immunizations (evident in N40% of the animals as early as following the 2th injection), illustrates the dangers of phos-tau-immunization in healthy individuals. To the best of our knowledge the effect of phos-tau immunization on WT-mice has not yet been investigated and reported in the literature, especially under the experimental setting used in our study that induced a proinflammatory CNS milieu. These findings may have important implications for future human applications of such immunization techniques, since therapeutic intervention in neurodegenerative
Fig. 4. Antibodies against phos-tau peptides in the serum of mice immunized repeatedly with phos-tau peptides. Serum samples (1:100) from the adult phos-tau immunized and CFA–PT control mice were analyzed by ELISA. (A) Antibodies against each of the pho-tau peptides (213[P202/205], Tau207–220[P212/214] and Tau224–238[P231]) were detected in the serum of the phos-tau immunized mice (subtracting the levels at the first injection from that after the last injection), while almost undetectable in the CFA–PT-control-mice. (A) WT-mice (p = 0.07 for antibodies against each peptide). (B) NFT-mice (p = 0.05, p = 0.02, and p = 0.1, respectively).
455
diseases should be started at the earliest possible stages, before the neurodegenerative processes have caused the death of a high number of neurons, and “healthy” adults at high risk could therefore represent the ideal candidates for preventive immunization. Our results raise a significant safety issue concerning repeated preventive tau-vaccinations in adults. Moreover, based on our finding that adult NFT-mice immunized repeatedly with phos-tau peptides developed neuroinflammation, it seems, that also individuals affected by AD/tauopathy are at risk to develop neuroinflammation. The lower risk of neuroinflammation detected in the phos-tau immunized NFT-mice relative to that in the healthy (WT) group of mice may point to a “protective” effect related to the NFT pathology (composed of phos-tau), a pathology which can theoretically stimulate an accelerated immune response mediated by the anti phos-tau peptides antibodies present in the serum of immunized mice which in turn, probably prevent additional bystander immune reactions targeting other CNS Ags. In the absence of NFTs in the WT-mice, the immune response induced by the vaccination could be broader and may attack additional CNS targets and cause encephalitis. It has been previously shown that circulating blood antibodies, including anti-phos-tau antibodies can get access to the CNS either through the extracellular pathways or via retrograde axonal transport (Asuni et al., 2007; Banks, 2004; Fabian, 1990). The entrance of antibodies to the CNS may be accelerated under pathological conditions that cause neuroinflammation and damage of the integrity of the BBB (Boettger et al., 2010); a similar condition was simulated in our experiments using CFA adjuvant in the vaccine, along with administration of PT. The results of our current study pointing to the high susceptibility of WT-mice to neuroinflammation in response to phos-tau vaccination in CFA–PT are in accord with our previous studies showing that immunization of (young) WT-mice with full-length-unphosphorylated tau in CFA–PT (1 injection with 1 booster a week later) was encephalitogenic in 55% of the mice (Rosenmann et al., 2006), suggesting that WT-mice are at high risk for neuroinflammation in response to various protocols of tau-immunotherapy. As for the NFT-mice, while in our current study neuroinflammation was evident in 13.3% following the 2nd injection (delivered 2 weeks following the 1st one), and not at all in the aged group, in our previous study no inflammation was noticed in phos-tau immunized young NFT-mice (1 injection with a booster a week later) (Boimel et al., 2010). This difference (although not robust) could be explained by the different protocol (repeated vaccinations) used in the current study as compared to a single/double immunization in our previous report. In any case, our combined results may indicate a higher safety profile of tau-vaccination in NFT-mice as compared to WT ones, probably depending on the age of immunized animals, the proinflammatory milieu and the protocol of vaccination used. It is possible that vaccination may be associated with higher risk in individuals in whom there is a prominent inflammatory state in their CNS, while in others the vaccine may be safe. Such a notion that the vaccine maybe of risk only in part of the patients, seems to be in accord with the results of the clinical trials of amyloid immunotherapy, where part of the patients, particularly the apoE4 carriers, developed neuroinflammation (Salloway et al., 2009). Further characterization and identification of the risk factors are needed in order to define the population at risk and classify the candidates that may benefit from safe vaccination. An additional finding of our study was that while repeated injection of CFA–PT (only) to adult WT-mice did not induce any neurological deficits, the same vaccination induced neurological deficits in a high percentage of NFT mice (69.2%). This may be related to CNS changes caused by NFT-pathology, presumably a disruption of the BBB. Indeed, a defective/malfunctioning BBB in the presence of NFT pathology has been reported by Asuni et al. in a tangle model (Asuni et al., 2007). Interestingly, aged WT-mice reacted similarly to aged NFT-mice (presence of neurological deficits in a proportion of CFA–PT injected mice); this may
456
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
be explained by the possibility that aged WT-mice may have some baseline subclinical brain pathology at a certain degree, such as BBB impairment and therefore develop neurological deficits in response to repeated CFA–PT injections. The lower cumulative percentage of affected aged mice as well as the later appearance of neurological deficits following phos-tau immunization, as well as in the CFA–PT–NFT-mice, may be related to the impaired immune response that is observed in aging animals (Makinodan et al., 1976). The lack of deficits developed in response to repeated phos-tau immunization in the aged NFT-mice, may be explained by the combination of the presence of tangle pathology along with the abovementioned changes in the elder animals, which may cumulatively be protective. Further confirmation of these findings in a larger number of aged animals is needed. In summary, our results show that repeated immunization with phos-tau peptides under a proinflammatory milieu, and particularly in adult WT-mice and NFT tangle-mice, may be encephalitogenic. Acknowledgment This study was supported (in part) by grant no. 300000-4895 from the Chief Scientist Office of the Ministry of Health, Israel. References Asuni, A.A., Boutajangout, A., Quartermain, D., Sigurdsson, E.M., 2007. Immunotherapy targeting pathological tau conformers in a tangle mouse model reduces brain pathology with associated functional improvements. J. Neurosci. 27, 9115–9129. Banks, W.A., 2004. Are the extracellular [correction of extracelluar] pathways a conduit for the delivery of therapeutics to the brain? Curr. Pharm. Des. 10, 1365–1370. Bi, M., Ittner, A., Ke, Y.D., Gotz, J., Ittner, L.M., 2011. Tau-targeted immunization impedes progression of neurofibrillary histopathology in aged P301L tau transgenic mice. PLoS One 6, e26860. Boettger, M.K., Weishaupt, A., Geis, C., Toyka, K.V., Sommer, C., 2010. Mild experimental autoimmune encephalitis as a tool to induce blood–brain barrier dysfunction. J Neural Transm 117, 165–169. Boimel, M., Grigoriadis, N., Lourbopoulos, A., Haber, E., Abramsky, O., Rosenmann, H., 2010. Efficacy and safety of immunization with phosphorylated tau against neurofibrillary tangles in mice. Exp. Neurol. 224, 472–485. Boutajangout, A., Quartermain, D., Sigurdsson, E.M., 2010. Immunotherapy targeting pathological tau prevents cognitive decline in a new tangle mouse model. J. Neurosci. 30, 16559–16566.
Boutajangout, A., Ingadottir, J., Davies, P., Sigurdsson, E.M., 2011. Passive immunization targeting pathological phospho-tau protein in a mouse model reduces functional decline and clears tau aggregates from the brain. J. Neurochem. 118, 658–667. Castillo-Carranza, D., Lasagna-Reeves, C., Estes, D., Barrett, A., Dineley, K., Jackson, G., Kayed, R., 2010. Modulation of tau oligomers by passive vaccination. Soc. Neurosci. Abstr. 347, 11. Chai, X., Wu, S., Murray, T.K., Kinley, R., Cella, C.V., Sims, H., Buckner, N., Hanmer, J., Davies, P., O'Neill, M.J., Hutton, M.L., Citron, M., 2011. Passive immunization with anti-Tau antibodies in two transgenic models: reduction of Tau pathology and delay of disease progression. J. Biol. Chem. 286, 34457–34467. Fabian, R.H., 1990. Uptake of antineuronal IgM by CNS neurons: comparison with antineuronal IgG. Neurology 40, 419–422. Furlan, R., Brambilla, E., Sanvito, F., Roccatagliata, L., Olivieri, S., Bergami, A., Pluchino, S., Uccelli, A., Comi, G., Martino, G., 2003. Vaccination with amyloid-beta peptide induces autoimmune encephalomyelitis in C57/BL6 mice. Brain 126, 285–291. Gilman, S., Koller, M., Black, R.S., Jenkins, L., Griffith, S.G., Fox, N.C., Eisner, L., Kirby, L., Rovira, M.B., Forette, F., Orgogozo, J.M., 2005. Clinical effects of Abeta immunization (AN1792) in patients with AD in an interrupted trial. Neurology 64, 1553–1562. Kaye, J.F., Kerlero de Rosbo, N., Mendel, I., Flechter, S., Hoffman, M., Yust, I., Ben-Nun, A., 2000. The central nervous system-specific myelin oligodendrocytic basic protein (MOBP) is encephalitogenic and a potential target antigen in multiple sclerosis (MS). J. Neuroimmunol. 102, 189–198. Makinodan, T., Albright, J.W., Good, P.I., Peter, C.P., Heidrick, M.L., 1976. Reduced humoral immune activity in long-lived old mice: an approach to elucidating its mechanisms. Immunology 31, 903–911. Monsonego, A., Imitola, J., Petrovic, S., Zota, V., Nemirovsky, A., Baron, R., Fisher, Y., Owens, T., Weiner, H.L., 2006. Abeta-induced meningoencephalitis is IFN-gamma-dependent and is associated with T cell-dependent clearance of Abeta in a mouse model of Alzheimer's disease. Proc. Natl. Acad. Sci. U.S.A. 103, 5048–5053. Orgogozo, J.M., Gilman, S., Dartigues, J.F., Laurent, B., Puel, M., Kirby, L.C., Jouanny, P., Dubois, B., Eisner, L., Flitman, S., Michel, B.F., Boada, M., Frank, A., Hock, C., 2003. Subacute meningoencephalitis in a subset of patients with AD after Abeta42 immunization. Neurology 61, 46–54. Rosenmann, H., 2013. Immunotherapy for targeting tau pathology in Alzheimer's disease and tauopathies. Curr. Alzheimer Res. 18, 217–228. Rosenmann, H., Grigoriadis, N., Karussis, D., Boimel, M., Touloumi, O., Ovadia, H., Abramsky, O., 2006. Tauopathy-like abnormalities and neurologic deficits in mice immunized with neuronal tau protein. Arch. Neurol. 63, 1459–1467. Rosenmann, H., Grigoriadis, N., Eldar-Levy, H., Avital, A., Rozenstein, L., Touloumi, O., Behar, L., Ben-Hur, T., Avraham, Y., Berry, E., Segal, M., Ginzburg, I., Abramsky, O., 2008. A novel transgenic mouse expressing double mutant tau driven by its natural promoter exhibits tauopathy characteristics. Exp. Neurol. 212, 71–84. Salloway, S., Sperling, R., Gilman, S., Fox, N.C., Blennow, K., Raskind, M., Sabbagh, M., Honig, L.S., Doody, R., van Dyck, C.H., Mulnard, R., Barakos, J., Gregg, K.M., Liu, E., Lieberburg, I., Schenk, D., Black, R., Grundman, M., 2009. A phase 2 multiple ascending dose trial of bapineuzumab in mild to moderate Alzheimer disease. Neurology 73, 2061–2070. Troquier, L., Caillierez, R., Burnouf, S., Fernandez-Gomez, F., Grosjean, M., Zommer, N., Sergeant, N., Schraen-Maschke, S., Blum, D., Buée, L., 2012. Targeting phospho-Ser422 by active tau immunotherapy in the THY-22Tau mouse model: a suitable therapeutic approach. Curr. Alzheimer Res. 9, 397–405.
Contents lists available at ScienceDirect
Experimental Neurology journal homepage: www.elsevier.com/locate/yexnr
Repeated immunization of mice with phosphorylated-tau peptides causes neuroinflammation☆ Lea Rozenstein-Tsalkovich a, Nikolaos Grigoriadis b, Athanasios Lourbopoulos b, Evangelia Nousiopoulou b, Ibrahim Kassis a, Oded Abramsky a, Dimitrios Karussis a, Hanna Rosenmann a,⁎ a b
The Department of Neurology, The Agnes Ginges Center for Human Neurogenetics, Hadassah Hebrew University Medical Center, Jerusalem, Israel The B' Department of Neurology, AHEPA University Hospital, Thessaloniki, Macedonia, Greece
a r t i c l e
i n f o
Article history: Received 7 May 2013 Revised 2 July 2013 Accepted 12 July 2013 Available online 20 July 2013 Keywords: Tauopathy Alzheimer's-disease Tau protein Phosphorylated-tau Neurofibrillary-tangles Immunotherapy
a b s t r a c t The recent studies of others and of us showing robust efficacy of anti-tangle immunotherapy, directed against phosphorylated (phos)-tau protein, may pave the way to clinical trials of phos-tau immunotherapy in Alzheimer's-disease and other tauopathies. At this stage addressing the safety of the phos-tauimmunotherapy is highly needed, particularly since we have previously shown the neurotoxic potential of tau-immunotherapy, specifically of full-length unphosphorylated-tau vaccine under a CNSproinflammatory milieu [induced by emulsification in complete-Freund's-adjuvant (CFA) and pertussistoxin (PT)] in young wild-type (WT)-mice. The aim of our current study was to address safety aspects of the phos-tau-immunotherapy in both neurofibrillary-tangle (NFT)-mice as well as in WT-mice, under challenging conditions of repeated immunizations with phos-tau peptides under a CNS-proinflammatory milieu. NFT- and WT-mice were repeatedly immunized (7 injections in adult-, 4 in aged-mice) with phos-tau peptides emulsified in CFA–PT. A paralytic disease was evident in the phos-tau-immunized adult NFT-mice, developing progressively to 26.7% with the number of injections. Interestingly, the WT-mice were even more prone to develop neuroinflammation following phos-tau immunization, affecting 75% of the immunized mice. Aged mice were less prone to neuroinflammatory manifestations. Anti-phos-tau antibodies, detected in the serum of immunized mice, partially correlated with the neuroinflammation in WT-mice. This points that repeated phos-tau immunizations in the frame of a proinflammatory milieu may be encephalitogenic to tangle-mice, and more robustly to WT-mice, indicating that – under certain conditions – the safety of phos-tau immunotherapy is questionable. © 2013 Elsevier Inc. All rights reserved.
Introduction There is accumulating evidence that targeting selectively pathological tau, particularly the phosphorylated (phos)-tau isoforms (Asuni et al., 2007; Bi et al., 2011; Boimel et al., 2010; Boutajangout et al., 2010; Boutajangout et al., 2011; Chai et al., 2011; Troquier et al., 2012) or the tau-oligomers (Castillo-Carranza et al., 2010), reduces the taupathology and improves the symptoms of dementia in different animal models for tauopathies using various immunization protocols. Emanating basically from the lessons learned from amyloid immunization studies, particularly the fact that adverse effects were first detected in clinical studies (Orgogozo et al., 2003), and it was only afterwards that the adverse effect (neuroinflammation) was remodeled in animals (Furlan et al., 2003; Monsonego et al., 2006), there is a great need in addressing not only the issue of efficacy but also that of the safety of
☆ This study was supported (in part) by grant no. 300000-4895 from the Chief Scientist Office of the Ministry of Health, Israel. ⁎ Corresponding author at: Department of Neurology, Hadassah Hebrew University Medical Center, Ein Karem, Jerusalem 91120, Israel. Fax: + 972 3 7256022. E-mail address: [email protected] (H. Rosenmann). 0014-4886/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.expneurol.2013.07.006
anti-tangle vaccine already at the preclinical stage, before moving to clinical trials. With this approach we have tested tau vaccines emulsified in complete-Freund's-adjuvant (CFA), a T-helper 1 (Th1)-polarizing adjuvant, and pertussis-toxin (PT), a toxin which increases the BBB permeability. This immunization protocol provides a CNS proinflammatory milieu used by us as a setup to detect possible tau vaccine biohazards of developing neuroinflammation in mice. Such proinflammatory conditions have been reported to remodel neuroinflammation in amyloidmice immunized with amyloid vaccine (Furlan et al., 2003), possibly mimicking the condition in the brain of certain elderly AD patients, who developed neuroinflammation in response to amyloid-vaccination. Under this CNS proinflammatory milieu we have previously demonstrated that the full-length unphosphorylated-tau protein vaccine was encephalitogenic in wild-type (WT)-mice, pointing to the biohazardous potential of anti-tau immunotherapy (Rosenmann et al., 2006). Our subsequent results showed that vaccination (1 injection with 1 booster a week later) with phos-tau peptides (in CFA–PT), peptides which selectively targets NFT-related phos-tau protein, showed a higher safety profile (no signs of encephalitis), accompanied by robust efficacy, as evidenced by the reduced NFT-burden and improved cognition of the NFT-mice (Boimel et al., 2010; Rosenmann, 2013). In our current
452
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
study we studied the safety profile of repeated vaccinations with phostau in NFT-mice in order to evaluate whether such repeated immunization protocols (aiming to increase efficacy) can be safely applied in future clinical trials, similarly to the repeated amyloid vaccinations that have been tried in AD patients (Gilman et al., 2005). Repeated phos-tau immunizations have been previously reported to reduce the tangle-burden without significant adverse effects. However in the latter studies a different immunization protocol was used (utilizing a Th2polirizing alum adjuvant without PT), in which the CNS was less challenged by a detrimental proinflammatory milieu (Asuni et al., 2007; Bi et al., 2011; Boutajangout et al., 2010; Troquier et al., 2012). In addition, we also investigated here the safety of phos-tau vaccination in WT-healthy mice, in order to address the potential of early preventive immunization in high-risk healthy individuals. These studies were conducted in both adult- and aged-animals, in order to test a possible impact of age or of disease phase. Our findings here show that repeated phos-tau immunization in the frame of a proinflammatory milieu induced encephalitogenicity in tangle-mice and even more, in WT-mice, indicating that under certain conditions, such as in individuals with inflammatory diseases, phostau immunotherapy might be unsafe, especially in adult clinically intact (non-demented) individuals.
Microglia were stained using the biotinylated lectin-lycopersicon esculentum immunostaining protocol as reported previously (Rosenmann et al., 2006). Microglia were evaluated on 4 paraffin sections under 20× optical fields and data were expressed as cells/mm2. Detection of antibodies in serum by ELISA Sera of immunized-mice and controls were tested for the presence of antibodies against each of the 3 phos-tau peptides by ELISA, according to the protocol reported previously (Boimel et al., 2010). Data analysis Results are presented as Mean ± SEM or as medians. For comparisons between study groups, the unpaired t-test, median test and Chi-square analysis were used. To assess linear associations between two continuous variables, the Pearson correlation coefficient was calculated. Results Mice immunized repeatedly with phos-tau in CFA–PT develop neurological deficits
Methods Mice The NFT-mice (E257T/P301S-tau-tg-mice) generated by us were crossed with C57Bl mice (Rosenmann et al., 2008). Tg offspring were screened to identify NFT-mice, while the non-Tg mice served as WTmice. Experiments were performed in accordance with NIH-guidelines for the care and use of laboratory-animals. Immunizations Six-month-old NFT- and WT-mice (“adult mice”, at about the age at disease onset) (12–15 mice/group) as well at 12 months of age (“aged mice”, at an age when NFTs are prominent) (5–6/group) were subcutaneously immunized repeatedly with a mixture of 3 phos-tau peptides {Tau195–213[P202/205], Tau207–220[P212/214], Tau224–238 [P231]} previously reported by us (Boimel et al., 2010), as follows: 100 μg of each peptide emulsified in non-enriched CFA and PT, with one booster following 2 weeks; followed by monthly repeated injections of 50 μg of each peptide, in non-enriched CFA. The controls (“CFA–PT” groups) received vehicle only, as follows: non-enriched CFA and PT in the first injection, while CFA-alone in the following injections. In the adult mice a total of 7 injections were delivered, while 4 injections in the aged mice. Clinical follow-up Since the paralytic disease that developed in some of the immunized animals is similar to that seen in experimental-autoimmuneencephalomyelitis (EAE), the clinical evaluation during the 7–8 months (from first immunization) follow-up, was performed using the widely used EAE 0–6 score for paralysis (Kaye et al., 2000). Histology and immunohistochemistry (IHC) Brains were removed, post-fixed in 4% paraformaldehyde for 16–20 h at 4 °C and further processed for paraffin sectioning at 6 μm. Sections were stained with hematoxylin–eosin using standard protocols. Inflammation was evaluated on 6–8 randomly selected brain sections (60 μm apart) under 20× optical fields; the total number of infiltrating inflammatory cells per each brain hemisphere (hemisection) was counted and was expressed as cells/hemisection.
Adult mice The clinical follow-up revealed that adult phos-tau-WT-mice immunized repeatedly with 7 injections of phos-tau in CFA–PT developed neurological deficits with the number of affected mice progressively increasing from 5/12 (41.7%), and a mean maximal EAE score of 0.54 ± 0.25 after the 2nd injection, to 7/12 (58.3%), and a mean maximal EAE score of 1.38 ± 0.4 after the 4th- and 9/12 (75%), and mean maximal score 1.88 ± 0.49, (1 dead animal) after the 7th-injection. Two animals, which developed some deficits following the 2nd injection, subsequently recovered. Contrarily, all 12 mice in the adult CFA–PT control group were free of neurological deficits (p = 0.001 for % affected, p = 0.002 for score) (Fig. 1A–B). Adult phos-tau-NFT-mice also developed neurological deficits following the repeated immunization protocol. Two out of 15 mice (13%) developed paralytic signs (mean score of 0.13 ±0 .65) following the 2nd injection and 4/15 (26.7%), with a mean maximal score of 1.07 ± 0.54, (including 2 dead animals) following the 4th- till the 7th-injection. The incidence of the clinical syndrome of paralysis in the adult-immunized NFT-mice was significantly lower than that in the WT-immunized mice (26.7% vs 75%, respectively, p = 0.027), paralleled by a similar difference in the mean maximal EAE scores (1.06 ± 0.55 vs 1.88 ± 0.49, respectively) (Fig. 1C–D). Interestingly, the NFT-mice did develop neurological deficits following injections of CFA–PT alone, at high proportions: 8 out of 13 animals developed an EAE-like disease (61.5%, score of 1.19 ± 0.36) already after the 2nd injection, with 2 dead animals after the 5th injection. The percentage of affected mice further increased to 9/13 (69.2%) with a score of 2.75 ± 0.6, (including 4 dead mice) after the 7th injection. This incidence was significantly higher than that in the phos-tauNFT-mice (p = 0.009 for the incidence and p = 0.04 for clinical score) (Fig. 1C–D). There was also a trend of higher mortality in the group of the CFA–PT–NFT-mice relative to the phos-tau-NFT-mice (4/13 and, 2/15, respectively). This was significantly different from the group of WT-mice who did not develop any deficits in response to the CFA–PT injections (as is actually expected from healthy mice) (p b 0.001). Aged mice The repeated immunization protocol was also tested in aged animals. However, due to the small number of animals (5–6/group) the animals were vaccinated only with 4 injections, in order to allow histopathological examinations. Among the phos-tau-WT-mice 1/6
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
453
Fig. 1. Neurological deficits in adult WT- and NFT-mice immunized (7 times repeatedly) with phos-tau peptides in CFA–PT. (A) Cumulative % of affected WT-immunized-mice. EAE score was examined in time intervals between injections of phos-tau peptides in CFA–PT or of CFA–PT alone. Adult phos-tau-WT-mice developed neurological deficits, while WT-control (CFA–PT only) animals stayed free of deficits [p = 0.001] (n = 12/group). (B) Mean EAE score in WT-immunized-mice. Adult phos-tau WT-mice developed neurological deficits while CFA–PT–WT-mice stayed free of deficits [p = 0.002]. (C) Cumulative % of affected NFT-immunized-mice. Both adult phos-tau NFT-mice and CFA–PT control NFT-mice developed neurological deficits, with a significantly higher % of affected animals among the CFA–PT mice relative to the phos-tau immunized NFT-mice [p = 0.009] (n = 13–15/group). (D) Mean EAE score in NFT-immunized mice. The adult CFA–PT–NFT-mice showed higher mean maximal score than the phos-tau immunized NFT-mice [p = 0.04]. Among the phos-tau immunized mice the % of affected mice was higher in the WT- than in the NFT-mice [p = 0.03]. Among the CFA–PT controls the NFT-mice showed significantly higher % of affected mice than the WT-mice [p b 0.001].Among the phos-tau immunized mice there was a trend of a higher mean EAE score in the WT-mice than in the NFT-mice. Among the CFA–PT controls the NFT-mice showed significantly higher mean score than the WT-mice [p = 0.0008].
developed deficits (and died) (16.7%, mean maximal EAE score 1) after the full protocol of 4 injections. Interestingly, neurological deficits were also detected in a similar percentage among the aged CFA–
PT–WT-mice (1/6, 16.7%), but with a trend of a lower maximal EAE score (0.33 ± 0.33) (Fig. 2A–B), making the response of aged WT-mice to CFA–PT a reminiscence to that of the NFT-mice.
Fig. 2. Neurological deficits in aged NFT- and WT-mice immunized (4 times repeatedly) with phos-tau peptides in CFA–PT. (A) Cumulative % of affected WT-immunized-mice. EAE score was examined in time intervals between injections of phos-tau peptides in CFA–PT or of CFA–PT alone. Among the WT-mice both the phos-tau immunized and the CFA–PT controls developed neurological deficits with similar % of affected mice (n = 6/group). (B) Mean EAE score in WT-immunized-mice. There was a trend showing that the aged phos-tau WT-mice developed neurological deficits with a higher EAE score than the CFA–PT WT-mice. (C) Cumulative % of affected NFT-immunized-mice. The phos-tau NFT-mice stayed free of deficits, while the CFA–PT NFT-mice did develop neurological deficits (n = 5/group). (D) Mean EAE score in NFT-immunized-mice. There is a trend of a higher % maximal EAE score in the CFA–PT NFT-mice relative to the phos-tau NFT-mice.
454
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
The cumulative incidence of EAE-like paralytic disease among the phos-tau-WT-mice was lower in the aged mice relative to the adult mice (0%, 0%, 16.7% vs 41.7%, 58.3%, 58.3%) but this trend did not reach statistical significance (p = 0.06). None of the 5 phos-tau-NFTmice developed neurological deficits. Similar neurological deficits observed in the adult CFA–PT–NFT-mice were also noticed in 1/5 (20%, score 0.4 ± 0.4) of the aged CFA–PT–NFT-mice following the 3rd injection, and in 2/5 (40%, score 1.6 ± 0.42, 1/5 dead animal) following the 4th injection (Fig. 2C–D), though, with lower incidence relatively to the adult-mice [0% (p = 0.019), 20% and 40% for the 2th, 3th, and 4th injections, respectively, as compared to 61.5% in the adult mice].
Histopathological evidence of neuroinflammation in animals immunized with repeated tau-peptide vaccines Monocytic infiltrates in brains of mice immunized repeatedly with phos-tau in CFA–PT As shown in Fig. 3A–C, monocytic infiltrates in the brain were detected in 8/11 (72.7%) adult phos-tau-WT-mice tested (not including the deceased animals). The infiltrates were detected in the cerebellum, the corpus callosum, and even in the fibria and cortex. Only 1/12 animals among the CFA–PT controls (8.3%) (p = 0.002) had evidence of such CNS infiltrates. These results are in good correlation with the clinical manifestations (66.7% and 0% in the phos-tau-WT-mice and CFA–PT–WT-controls, respectively). Among the mice presenting clinical deficits 88% had monocytic infiltrates in their brain. Similar results were obtained when comparing the infiltrate burden (80.05 ± 22.27 vs 10.00 ± 6.98 mean infiltrates/hemisection, p = 0.007; corresponding medians: 45.0 vs 0.0 cells/hemisection, p = 0.01, respectively),
with some correlation between the total burden of inflammation and the clinical score (r = 0.214). Among the adult NFT-mice, monocytic infiltrates were detected in the brains of 5/13 phos-tau-immunized-mice (38.4%), and in 4/9 CFA–PT control animals (44.4%); the infiltrate burden medians were: 33.0 vs 56.0 cells/hemisection, respectively (p = NS). The latter finding indicates a trend of correlation with the proportion of animals with clinical neurological deficits (26.7% vs 75% respectively). Among the mice presenting clinical deficits 68% had also brain infiltrates. The percentage of the affected mice with infiltrates was higher in phos-tau-WT-mice than in phos-tau-NFT-mice (72.7% vs 38.4%, p = 0.09, infiltrate burden medians: 45 and 33 respectively, p = NS) and lower in CFA–PT–WT-mice than jn CFA–PT–NFT-mice (8.3% vs 44.4%, p = 0.05; infiltrate medians: 0 and 56, p = 0.041; infiltrate means: 10.00 ± 6.98 vs 87.67 ± 27.44, respectively, p = 0.011). No infiltrates were noticed in the brains of the aged mice tested. Microglial burden did not change in mice immunized with phos-tau in CFA–PT As presented in Fig. 3A–C, lectin staining for microglia revealed that phos-tau immunization in the adult mice did not increase the overall microglial burden relative to the CFA–PT controls, both in the NFTand in the WT-mice. Specifically the microglial burden was: 6.74 ± 0.27 vs 7.07 ± 0.25 microglia/mm2 in the phos-tau-WT- and CFA– PT–WT-mice, respectively; 6.14 ± 0.19 vs 6.49 ± 0.24 microglia/mm2 in phos-tau-NFT- and CFA–PT–NFT-mice respectively (p = NS). Similar results were also observed in the aged mice groups. However, it should be noted that local inflammatory infiltrations were associated with an increased local microglial/macrophage (lectin + cells) reaction, as illustrated in Fig. 3A.
Fig. 3. Infiltrates and microglia in adult NFT- and WT-mice immunized (7 times repeatedly) with phos-tau peptides in CFA–PT. (A) Lectin immunostaining with hematoxylin counterstaining displaying microglia/macrophages (brown cells) and perivascular infiltrates (aggregates of blue monocytes, arrows in A), in the brains of WT- and NFT-mice immunized with phos-tau peptides in CFA–PT (A1 and A3 respectively), as well as in control NFT-mice injected with CFA–PT only (A2). Very few infiltrates, if at all, were detected in the CFA–PT–WTmice (not shown). (B) Quantitative analysis of the infiltrates is presented as % of mice affected by brain infiltrates. Phos-tau-immunized WT animals were significantly more affected compared to the control (CFA–PT-immunized) WT mice (p = 0.002); more affected animals in the CFA–PT NFT-group vs CFA–PT WT mice were also observed (p = 0.05), with some trend of higher % of affected mice in the CFA–PT NFT than in phos-tau-immunized NFT-mice. (C) Quantitative analysis of the infiltrates is presented as median infiltrate burden per brain hemisection. Boxplot graph illustrates the median values (dots and diamonds indicate extreme values). Phos-tau-immunized WT animals showed significantly more brain infiltrates than control (CFA–PT-immunized) WT mice (p = 0.01); higher burden of brain infiltrates in the CFA–PT NFT-group than in the CFA–PT WT mice were also observed (p = 0.041), with some trend of more infiltrates in the CFA–PT NFT than in phos-tau-immunized NFT-mice.
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
Antibodies against phos-tau peptides in the serum of immunized mice Immunized animals developed antibodies against the phos-tau peptides of the vaccine (Fig. 4A–B). Some positive correlation was noticed between the antibody levels in the phos-tau-WT-mice and the quantity of brain infiltrates: r = 0.457 for Tau207–220[P212/214], and r = 0.485 for the Tau224–238[P231] peptide. Due to the lower number (n = 3) of affected mice among the phos-tau-NFT-mice (2 of them died without being available for pathological examination of infiltrates) the possibility of such a correlation could not be confirmed, concerning the immunized NFT-mice. Discussion We have shown here that repeated immunization of adult WT-mice with phos-tau in CFA–PT causes neuroinflammation that is dependent on the number of immunizations, reaching an incidence of 75% after 7 injections. Adult NFT-mice immunized repeatedly with the same protocol were significantly more resistant to the development of neuroinflammation (incidence: 26.7%). Additionally, aged mice also developed neuroinflammation in response to repeated immunizations (4 injections), but to a lower degree. A moderate association between serum anti-phos-tau antibodies and monocytic infiltrates in the brains was observed in phos-tau immunized WT-mice. Interestingly, almost 70% of the CFA–PT–NFT-mice developed neurological disability following the repeated injections, whereas CFA–PT–WT-mice remained symptoms-free. The finding that almost three fourths of the WT-mice, and particularly the adult ones, develop neuroinflammation in response to repeated phos-tau immunizations (evident in N40% of the animals as early as following the 2th injection), illustrates the dangers of phos-tau-immunization in healthy individuals. To the best of our knowledge the effect of phos-tau immunization on WT-mice has not yet been investigated and reported in the literature, especially under the experimental setting used in our study that induced a proinflammatory CNS milieu. These findings may have important implications for future human applications of such immunization techniques, since therapeutic intervention in neurodegenerative
Fig. 4. Antibodies against phos-tau peptides in the serum of mice immunized repeatedly with phos-tau peptides. Serum samples (1:100) from the adult phos-tau immunized and CFA–PT control mice were analyzed by ELISA. (A) Antibodies against each of the pho-tau peptides (213[P202/205], Tau207–220[P212/214] and Tau224–238[P231]) were detected in the serum of the phos-tau immunized mice (subtracting the levels at the first injection from that after the last injection), while almost undetectable in the CFA–PT-control-mice. (A) WT-mice (p = 0.07 for antibodies against each peptide). (B) NFT-mice (p = 0.05, p = 0.02, and p = 0.1, respectively).
455
diseases should be started at the earliest possible stages, before the neurodegenerative processes have caused the death of a high number of neurons, and “healthy” adults at high risk could therefore represent the ideal candidates for preventive immunization. Our results raise a significant safety issue concerning repeated preventive tau-vaccinations in adults. Moreover, based on our finding that adult NFT-mice immunized repeatedly with phos-tau peptides developed neuroinflammation, it seems, that also individuals affected by AD/tauopathy are at risk to develop neuroinflammation. The lower risk of neuroinflammation detected in the phos-tau immunized NFT-mice relative to that in the healthy (WT) group of mice may point to a “protective” effect related to the NFT pathology (composed of phos-tau), a pathology which can theoretically stimulate an accelerated immune response mediated by the anti phos-tau peptides antibodies present in the serum of immunized mice which in turn, probably prevent additional bystander immune reactions targeting other CNS Ags. In the absence of NFTs in the WT-mice, the immune response induced by the vaccination could be broader and may attack additional CNS targets and cause encephalitis. It has been previously shown that circulating blood antibodies, including anti-phos-tau antibodies can get access to the CNS either through the extracellular pathways or via retrograde axonal transport (Asuni et al., 2007; Banks, 2004; Fabian, 1990). The entrance of antibodies to the CNS may be accelerated under pathological conditions that cause neuroinflammation and damage of the integrity of the BBB (Boettger et al., 2010); a similar condition was simulated in our experiments using CFA adjuvant in the vaccine, along with administration of PT. The results of our current study pointing to the high susceptibility of WT-mice to neuroinflammation in response to phos-tau vaccination in CFA–PT are in accord with our previous studies showing that immunization of (young) WT-mice with full-length-unphosphorylated tau in CFA–PT (1 injection with 1 booster a week later) was encephalitogenic in 55% of the mice (Rosenmann et al., 2006), suggesting that WT-mice are at high risk for neuroinflammation in response to various protocols of tau-immunotherapy. As for the NFT-mice, while in our current study neuroinflammation was evident in 13.3% following the 2nd injection (delivered 2 weeks following the 1st one), and not at all in the aged group, in our previous study no inflammation was noticed in phos-tau immunized young NFT-mice (1 injection with a booster a week later) (Boimel et al., 2010). This difference (although not robust) could be explained by the different protocol (repeated vaccinations) used in the current study as compared to a single/double immunization in our previous report. In any case, our combined results may indicate a higher safety profile of tau-vaccination in NFT-mice as compared to WT ones, probably depending on the age of immunized animals, the proinflammatory milieu and the protocol of vaccination used. It is possible that vaccination may be associated with higher risk in individuals in whom there is a prominent inflammatory state in their CNS, while in others the vaccine may be safe. Such a notion that the vaccine maybe of risk only in part of the patients, seems to be in accord with the results of the clinical trials of amyloid immunotherapy, where part of the patients, particularly the apoE4 carriers, developed neuroinflammation (Salloway et al., 2009). Further characterization and identification of the risk factors are needed in order to define the population at risk and classify the candidates that may benefit from safe vaccination. An additional finding of our study was that while repeated injection of CFA–PT (only) to adult WT-mice did not induce any neurological deficits, the same vaccination induced neurological deficits in a high percentage of NFT mice (69.2%). This may be related to CNS changes caused by NFT-pathology, presumably a disruption of the BBB. Indeed, a defective/malfunctioning BBB in the presence of NFT pathology has been reported by Asuni et al. in a tangle model (Asuni et al., 2007). Interestingly, aged WT-mice reacted similarly to aged NFT-mice (presence of neurological deficits in a proportion of CFA–PT injected mice); this may
456
L. Rozenstein-Tsalkovich et al. / Experimental Neurology 248 (2013) 451–456
be explained by the possibility that aged WT-mice may have some baseline subclinical brain pathology at a certain degree, such as BBB impairment and therefore develop neurological deficits in response to repeated CFA–PT injections. The lower cumulative percentage of affected aged mice as well as the later appearance of neurological deficits following phos-tau immunization, as well as in the CFA–PT–NFT-mice, may be related to the impaired immune response that is observed in aging animals (Makinodan et al., 1976). The lack of deficits developed in response to repeated phos-tau immunization in the aged NFT-mice, may be explained by the combination of the presence of tangle pathology along with the abovementioned changes in the elder animals, which may cumulatively be protective. Further confirmation of these findings in a larger number of aged animals is needed. In summary, our results show that repeated immunization with phos-tau peptides under a proinflammatory milieu, and particularly in adult WT-mice and NFT tangle-mice, may be encephalitogenic. Acknowledgment This study was supported (in part) by grant no. 300000-4895 from the Chief Scientist Office of the Ministry of Health, Israel. References Asuni, A.A., Boutajangout, A., Quartermain, D., Sigurdsson, E.M., 2007. Immunotherapy targeting pathological tau conformers in a tangle mouse model reduces brain pathology with associated functional improvements. J. Neurosci. 27, 9115–9129. Banks, W.A., 2004. Are the extracellular [correction of extracelluar] pathways a conduit for the delivery of therapeutics to the brain? Curr. Pharm. Des. 10, 1365–1370. Bi, M., Ittner, A., Ke, Y.D., Gotz, J., Ittner, L.M., 2011. Tau-targeted immunization impedes progression of neurofibrillary histopathology in aged P301L tau transgenic mice. PLoS One 6, e26860. Boettger, M.K., Weishaupt, A., Geis, C., Toyka, K.V., Sommer, C., 2010. Mild experimental autoimmune encephalitis as a tool to induce blood–brain barrier dysfunction. J Neural Transm 117, 165–169. Boimel, M., Grigoriadis, N., Lourbopoulos, A., Haber, E., Abramsky, O., Rosenmann, H., 2010. Efficacy and safety of immunization with phosphorylated tau against neurofibrillary tangles in mice. Exp. Neurol. 224, 472–485. Boutajangout, A., Quartermain, D., Sigurdsson, E.M., 2010. Immunotherapy targeting pathological tau prevents cognitive decline in a new tangle mouse model. J. Neurosci. 30, 16559–16566.
Boutajangout, A., Ingadottir, J., Davies, P., Sigurdsson, E.M., 2011. Passive immunization targeting pathological phospho-tau protein in a mouse model reduces functional decline and clears tau aggregates from the brain. J. Neurochem. 118, 658–667. Castillo-Carranza, D., Lasagna-Reeves, C., Estes, D., Barrett, A., Dineley, K., Jackson, G., Kayed, R., 2010. Modulation of tau oligomers by passive vaccination. Soc. Neurosci. Abstr. 347, 11. Chai, X., Wu, S., Murray, T.K., Kinley, R., Cella, C.V., Sims, H., Buckner, N., Hanmer, J., Davies, P., O'Neill, M.J., Hutton, M.L., Citron, M., 2011. Passive immunization with anti-Tau antibodies in two transgenic models: reduction of Tau pathology and delay of disease progression. J. Biol. Chem. 286, 34457–34467. Fabian, R.H., 1990. Uptake of antineuronal IgM by CNS neurons: comparison with antineuronal IgG. Neurology 40, 419–422. Furlan, R., Brambilla, E., Sanvito, F., Roccatagliata, L., Olivieri, S., Bergami, A., Pluchino, S., Uccelli, A., Comi, G., Martino, G., 2003. Vaccination with amyloid-beta peptide induces autoimmune encephalomyelitis in C57/BL6 mice. Brain 126, 285–291. Gilman, S., Koller, M., Black, R.S., Jenkins, L., Griffith, S.G., Fox, N.C., Eisner, L., Kirby, L., Rovira, M.B., Forette, F., Orgogozo, J.M., 2005. Clinical effects of Abeta immunization (AN1792) in patients with AD in an interrupted trial. Neurology 64, 1553–1562. Kaye, J.F., Kerlero de Rosbo, N., Mendel, I., Flechter, S., Hoffman, M., Yust, I., Ben-Nun, A., 2000. The central nervous system-specific myelin oligodendrocytic basic protein (MOBP) is encephalitogenic and a potential target antigen in multiple sclerosis (MS). J. Neuroimmunol. 102, 189–198. Makinodan, T., Albright, J.W., Good, P.I., Peter, C.P., Heidrick, M.L., 1976. Reduced humoral immune activity in long-lived old mice: an approach to elucidating its mechanisms. Immunology 31, 903–911. Monsonego, A., Imitola, J., Petrovic, S., Zota, V., Nemirovsky, A., Baron, R., Fisher, Y., Owens, T., Weiner, H.L., 2006. Abeta-induced meningoencephalitis is IFN-gamma-dependent and is associated with T cell-dependent clearance of Abeta in a mouse model of Alzheimer's disease. Proc. Natl. Acad. Sci. U.S.A. 103, 5048–5053. Orgogozo, J.M., Gilman, S., Dartigues, J.F., Laurent, B., Puel, M., Kirby, L.C., Jouanny, P., Dubois, B., Eisner, L., Flitman, S., Michel, B.F., Boada, M., Frank, A., Hock, C., 2003. Subacute meningoencephalitis in a subset of patients with AD after Abeta42 immunization. Neurology 61, 46–54. Rosenmann, H., 2013. Immunotherapy for targeting tau pathology in Alzheimer's disease and tauopathies. Curr. Alzheimer Res. 18, 217–228. Rosenmann, H., Grigoriadis, N., Karussis, D., Boimel, M., Touloumi, O., Ovadia, H., Abramsky, O., 2006. Tauopathy-like abnormalities and neurologic deficits in mice immunized with neuronal tau protein. Arch. Neurol. 63, 1459–1467. Rosenmann, H., Grigoriadis, N., Eldar-Levy, H., Avital, A., Rozenstein, L., Touloumi, O., Behar, L., Ben-Hur, T., Avraham, Y., Berry, E., Segal, M., Ginzburg, I., Abramsky, O., 2008. A novel transgenic mouse expressing double mutant tau driven by its natural promoter exhibits tauopathy characteristics. Exp. Neurol. 212, 71–84. Salloway, S., Sperling, R., Gilman, S., Fox, N.C., Blennow, K., Raskind, M., Sabbagh, M., Honig, L.S., Doody, R., van Dyck, C.H., Mulnard, R., Barakos, J., Gregg, K.M., Liu, E., Lieberburg, I., Schenk, D., Black, R., Grundman, M., 2009. A phase 2 multiple ascending dose trial of bapineuzumab in mild to moderate Alzheimer disease. Neurology 73, 2061–2070. Troquier, L., Caillierez, R., Burnouf, S., Fernandez-Gomez, F., Grosjean, M., Zommer, N., Sergeant, N., Schraen-Maschke, S., Blum, D., Buée, L., 2012. Targeting phospho-Ser422 by active tau immunotherapy in the THY-22Tau mouse model: a suitable therapeutic approach. Curr. Alzheimer Res. 9, 397–405.