Efficacy and safety of a liposome-based vaccine against protein Tau, assessed in tau.P301L mice that model tauopathy

Abstract

Progressive aggregation of protein Tau into oligomers and fibrils correlates with cognitive decline and synaptic dysfunction, leading to neurodegeneration in vulnerable brain regions in Alzheimer's disease. The unmet need of effective therapy for Alzheimer's disease, combined with problematic pharmacological approaches, led the field to explore immunotherapy, first against amyloid peptides and recently against protein Tau. Here we adapted the liposome-based amyloid vaccine that proved safe and efficacious, and incorporated a synthetic phosphorylated peptide to mimic the important phospho-epitope of protein Tau at residues pS396/pS404. We demonstrate that the liposome-based vaccine elicited, rapidly and robustly, specific antisera in wild-type mice and in Tau.P301L mice. Long-term vaccination proved to be safe, because it improved the clinical condition and reduced indices of tauopathy in the brain of the Tau.P301L mice, while no signs of neuro-inflammation or other adverse neurological effects were observed. The data corroborate the hypothesis that liposomes carrying phosphorylated peptides of protein Tau have considerable potential as safe and effective treatment against tauopathies, including Alzheimer's disease.

Figure 1. CD spectrum of ACI-35 corresponds to β-sheet secondary structure.

The CD spectrum of ACI-35 at (1∶9) dilution in PBS. The spectrum of liposomes lacking the phospho-peptide was subtracted to the signal of ACI-35 for baseline correction. The spectrum shows a maximum around 199.5 nm and a broad minimum around 218 nm.

Figure 2. ACI-35 elicits robust and specific antisera against Tau in wild-type and Tau.P301L mice.

(A) Vaccination schedule with ACI-35 in wild-type mice is shown schematically with s.c. injections represented by the syringes and the bleedings by the letter B with a number. Antisera titers were measured by ELISA on the phosphorylated antigenic sequence incorporated in the vaccine (pTau peptide), and on the non-phosphorylated peptide of the same primary amino acid sequence (Tau peptide) (see text for details). Data are presented as mean± SD. Statistical analysis: one-way ANOVA followed by Bonferroni multiple comparison test (**p<0.01, **** p<0.0001) and by unpaired student's t-test (**** p<0.0001). (B) Similar vaccination with ACI-35 of Tau.P301L mice and analysis by ELISA. Data are presented as mean± SD. Statistical analysis by unpaired student's t-test (** p<0.01; **** p<0.0001) (C) TAUPIR with antisera from ACI-35 vaccinated wild-type mice and Tau.P301L mice demonstrated a specific reaction with neurofibrillary tangles and neuropil threads in forebrain of biGT mice. IHC with Mab AT100 is included for comparison. IHC with sera from Tau.P301L mice injected with PBS or from Tau.P301L mice that were not vaccinated, were devoid of specific antibodies and auto-antibodies against human protein Tau. Scale bars: 50 µm.

Figure 3. Biochemical analysis of brain from Tau.P301L mice vaccinated with ACI-35.

(A) Fractionation scheme of total brain homogenates from Tau.P301L mice to generate soluble (S1) and sarcosyl insoluble fractions (SInT). (B) Representative western blots of S1 and SInT fractions from forebrain and from brainstem of two untreated terminal Tau.P301L mice (age 10 months) developed for total protein Tau (Mab Tau5), for total human protein Tau (Mab HT7) and for phosphorylated Tau (antibodies pT231, pS396 and pS404 as indicated). (C) Reduction of pS396 in soluble fraction of brainstem and forebrain (p = 0.026 and 0.0523, respectively, Student's t-test) from ACI-35 vaccinated Tau.P301L mice, relative to PBS injected mice. (D) Reduction of pS396 (p = 0.0091, Mann Whitney test) and HT7 (p = 0.0706, Mann Whitney test) in SInT in forebrain by ACI-35 vaccination of Tau.P301L mice. (E) Reduction of tangled neurons, marked by IHC for AT100 or pS422, in the forebrain of ACI-35 vaccinated Tau.P301L mice after 3 months of treatment. Data: mean± SEM.

Figure 4. Improved clinical parameters of Tau.P301L mice vaccinated with ACI-35.

(A) The left panel shows the evolution of the bodyweight of Tau.P301L mice over 3 months of ACI-35 vaccination relative to PBS injected Tau.P301L mice. The panel on the right presents the loss in bodyweight of the Tau.P301L mice at day 91 relative to the start of the study (day -7). Data: mean± SEM (p = 0.094, unpaired Student's t-test). (B) Clasping score of vaccinated Tau.P301L mice compared to placebo injected Tau.P301L mice. (C) Rotarod performance of ACI-35 vaccinated Tau.P301L mice compared to PBS-injected Tau.P301L mice, tested over 300 sec on the accelerating rod (see Methods section for details). Data: mean± SEM. (E) Mortality of Tau.P301L mice vaccinated with ACI-35 versus placebo Tau.P301L mice.

Figure 5. No increased inflammatory response in ACI-35 treated Tau.P301L mice.

(A) IHC did not reveal marked differences in inflammation-related parameters in forebrain of ACI-35 vaccinated Tau.P301L mice, relative to PBS-injected Tau.P301L mice. IHC reaction with the different specific antibodies, specified in the captions, was analyzed by image analysis (details see Methods section) and presented as mean± SEM. Scale bars: 50 µm. (B) Western blotting for GFAP of total brain homogenates from Tau.P301L mice vaccinated with ACI-35 (n = 34) or injected with PBS (n = 33). Data presented as mean± SEM. (C) ELISPOT analysis of IFN-γ and IL-4 production by T cells isolated from spleens of naive mice or from mice immunized by either ACI-35 or with recombinant protein Tau. Splenocytes were re-stimulated with medium (cells alone), recombinant Tau protein (100 µg/ml) or with the phosphorylated peptide used in the ACI-35 vaccine and with its un-phosphorylated counterpart (1 µg/ml). Results are expressed as the number of foci (spots per million cells) +SD (n = 10 mice). Statistical analysis: two-way ANOVA followed by Bonferroni multiple comparison test (*** p<0.001).