An Immunomodulatory Therapeutic Vaccine Targeting Oligomeric Amyloid-β 1

Materials

All peptides used in the study were purchased from Biomer Tech Inc. (CA, USA).

Human wild type Aβ_1–42 (WT42), Aβ_1–35 (WT35), and Aβ_1–25 (WT25) sequences are as follows (the length of each amino acid type is delineated by the subscript numbers): D₁AEFRHDSGYEVHHQKLVFFAEDVG₂₅SNKGAIIGLM₃₅VGGVVIA₄₂.

GenBank access number: Q29149

Mutant Aβ_1–42 (E22W42), E22W1-25 (E22W25), and E22W1-35 (E22W35) sequences are as follows (the length of each amino acid type is delineated by the subscript numbers): D₁AEFRHDSGYEVHHQKLVFFAWDVG₂₅SNKGAIIGLM₃₅VGGVVIA₄₂.

Primary Aβ antibody A8 was obtained from Dr. Ying Zhang’s laboratory at Beijing Jiaotong University (Beijing China); A8 targets oligomer Aβ. The 6E10 antibody was purchased from Bio Legend (CA, USA, Cat# 803001).

Animals

Double transgenic APP_swe/PS1 _ΔE9 (Tg) and C57BL/6J control mice were used in this study, and the number of mice in each group is no fewer than nine. APP_swe/PS1 _ΔE9 mice with C57BL/6J background were originally purchased from JAX MMRRC (Stock# 034829) and bred in our facility. All APP/PS1 mice were initially genotyped by PCR at the time of grouping. The PCR result was further confirmed by using blood Aβ_1–40 measurements. Only the mice expressing Aβ_1–40 were used as transgenic mice. Pre-behavior results of RAWM (error and latency) and the plasma Aβ_1–40 levels were balanced among Tg groups. Mice were caged individually and maintained on a 12/12 h light/dark cycle and in a temperature-controlled room. Food and water were available ad libitum. Mice were 10 months of age when treatment began.

All described procedures were conducted in accordance with the United States Public Health Service’s Policy on Humane Care and approved by the Institutional Animal Care and Use Committee (IACUC) of the University of South Florida. All animals were housed in the vivarium at the USF College of Medicine.

B and T cell epitope prediction

The Immune Epitope Database (IEDB, https://tools.iedb.org/) was used to analyze the T cell and B cell epitope of wild type and mutated Aβ peptides. After the performance of B cell epitope prediction using DNAstar and IEDB, there were no differences among the wild type and mutant peptides [37]. Peptides were then subjected to MHC binding prediction. Binding affinity was predicted for a set of random natural peptides of 8-14 amino acid lengths for the 27 most frequent HLA class I alleles. This set of alleles is considered to cover >97% of the global population. The binding prediction was made using the NetMHCpan-4.0 method as implemented in the IEDB MHC binding prediction tool TepiTool [41 –43]. The prediction tool provides the predicted binding affinity in terms of percentile ranks ranging from 0 to 100; a peptide that has a lower percentile ranking is a better binder.

Monomer Aβ preparation

The purchased Aβ peptides (E22W42, WT42) were dissolved in HFIP at 1 mg/ml and incubated at 37°C for 3 days on a rotator then sonicated in a water bath for 1 h to disrupt any pre-existing aggregates and then aliquoted and lyophilized. Samples were then stored at –80°C for future applications.

Oligomer Aβ preparation

HFIP pretreated Aβ (E22W42, WT42) film was diluted at 5 mg/ml with DMSO and further diluted to 20μM with Ham’s F12 medium for aggregation, which was achieved by 72 h of incubation at 37°C. The oligomer formation was detected and confirmed with western blot.

Bone marrow and immature DC preparation

Mouse bone marrow dendritic cells (mDCs) were harvested from the femur and tibia of 10-week-old non-transgenic mice (C57BL/6J). On day 0, after removing all muscle tissue with gauze from the femurs and tibias, the bones were washed with 1XPBS and soaked in 70% ethanol for 30 s. The bones were then washed three times with 1XPBS, and both ends were cut in a dish. The marrow was flushed with RPMI 1640 with a syringe, then gently resuspended and passed through a 70μm strainer into a centrifuge tube. Following centrifugation at 300×g for 10 min, the cell pellets were loosened by gentle vortex. Then red blood cells were lysed by incubation with 5 ml ACK buffer (0.15 M NH₄Cl, 1 mM KHCO₃, 0.1 mM EDTA, pH 7.3) for 30 s while shaken at room temperature, and 45 ml 1XHBSS was added to stop lysis. After centrifugation, cells were resuspended with RPMI 1640 (10% FBS, and 1% penicillin-streptomycin, 50μM of 2-mercaptoethanol, 10 ng/ml of recombinant mouse GM-CSF (BD, Cat: 554586), and 10 ng/ml of recombinant mouse IL-4 (BD Cat:550067)), then brought to 1×10⁶ cells/ml and seeded in 6-well plates with 3 ml/well. The medium was replaced on day 1, and cell treatment was initiated on day 4 by aspirating off 1ml of medium and then adding back 1 ml/well fresh medium containing 60μg/ml WT42 or E22W42. The cells were harvested on day 8, and the concentration was adjusted to 5×10⁶ cells/ml for injection.

Immunization procedure

10-month-old APP_swe/PS1 _ΔE9 mice received a dose of 1.5×10⁶ DCs per mouse (0.3 ml in 1×PBS) administered by intraperitoneal (IP) injection bi-weekly for the first three injections. For the following two injections, DCs dosage decreased to 1×10⁶ cells in 0.2 ml 1×PBS administered monthly by IP injection. A total of five injections were administered. Blood was collected 10 days after each vaccination [36].

Radial arm water maze (RAWM)

The radial arm water maze test was used to monitor the cognitive function of the mice before and after vaccinations. It contained 6 swim paths (arms) radiating from an open central area with a hidden escape platform located at the end of one of the arms. The pool was surrounded by several extra maze cues to allow spatial navigation. On each trial, the mouse was allowed to swim for up to 60 s to find the escape platform. The platform was located in the same arm on each trial. On day one, mice were given 15 trials alternating between a visible platform (above the water) and a hidden platform (below the water). On day two, mice were given 15 additional trials, with all the trials using a hidden platform. The start arm varied for each trial so that mice relied upon spatial cues to solve the task instead of learning motor rules. The goal arm for each mouse was different to avoid odor cues from revealing the goal arm. Entry into an incorrect arm (all four limbs within the arm) was scored as an error. Failure to make an arm entry within 15 s was also scored as an error. The errors for blocks of three consecutive trials were averaged for data analysis. Mice averaging 1 error or less by the end of day two are considered to have reached the learning criterion. On the third day, a reversal trial was performed, with the goal platform placed in the arm 180° from the original location. Mice were given 15 trials, all with a hidden platform. Latency to find and ascend the platform was recorded (60 s maximum) [44].

Brain tissue preparation

After the RAWM test, mice were anesthetized with SomnaSol (Henry Schein Animal Health, Cat#024352) and intracardially perfused with 50 ml of saline. The brains were carefully removed, and the right hemisphere was frozen at –80°C. On the day of final brain tissue preparation, frozen tissue was thawed and homogenized in the RIPA buffer containing proteinase inhibitor (100 mM Tris [45], 150 mMNaCl, 0.5% DOC, 1% NP-40, 0.2% SDS,1 mM Na3VO4, 10 mM NaF, 1 mM PMSF, 20μM Leupeptin) with a pellet pestle motor and 10 s sonication, then centrifuged for 20 min at 21,000 g at 4°C. Crude protein concentrations were determined by Bio-rad DC protein assay (Bio-Rad Cat:5000112) and adjusted to the same level for all the samples. The supernatants obtained from this protocol were stored at –80°C. The left hemispheres were transferred into a 4% paraformaldehyde solution for future immunohistochemistry tests.

Blood sample collection and detection

Blood samples were collected in EDTA tubes by submandibular phlebotomy bleeding before the vaccination and 10 days after every injection. On the euthanasia day, mice were anesthetized and blood was drawn from each mouse via cardiac puncture. All plasma samples were analyzed for plasma Aβ level, plasma anti-Aβ antibody purifications, plasma antibody titer, plasma cytokine, and plasma antibody Ig isotyping detection.

Purification of plasma antibodies

Mouse plasma antibodies were purified by SulfoLink Coupling Resin (Thermo Scientific Prod # 20402) and conjugated with synthesized cysteine residue Aβ peptide according to the manufacturer’s instructions. Briefly, after equilibrating the resin with binding buffer, the diluted plasma was passed through the resin five times. The antibody was then washed with binding buffer and eluted with the Elution Buffer. The eluted antibodies were immediately adjusted to physiologic pH by adding Neutralization Buffer, and the concentration was determined by Bio-rad DC protein assay (Bio-Rad Cat:5000112).

Antibody titer determination

Plasma anti-Aβ antibody levels were determined by ELISA analysis. A 96-well plate was coated with 50μL of Aβ_1–42 peptides at 10μg/mL in carbonate bicarbonate (CBC) buffer. A CBC buffer plate was also used in order to measure control background binding. The Aβ_1–42 peptide experimental plate and CBC-bound control plate were incubated overnight at 4°C. The plates were then washed with 1XPBST and blocked with 1.5% BSA in PBST for 1 h at room temperature, followed by extensive washing. Plasma samples were then diluted serially in the blocking buffer and loaded to both the Aβ_1–42 peptide plate and CBC coated plates. Plates were then incubated for 1 h at 37°C and washed, then HRP-conjugated anti-mouse IgG (Sigma Cat: A9044) at 1 : 5000 dilutions with blocking buffer was added into each well and incubated for 1 h at 37°C. After washing, TMB peroxidase substrate (Surmodics Cat: TMBS-1000) was added to each well and incubated at room temperature for 10 min. The reaction was stopped by adding 100μL/well of 0.4 M H₂SO4. Absorbance at 450 nm was read with a BioTek Synergy H4 microplate reader. Samples with OD450 nm absorbance 3 times higher than controls were considered as positive, the highest dilution being designated as the endpoint titer.

Antibody Ig isotyping

Isotyping of immunoglobulin in plasma was performed using magnetic bead-based multiplex assay (Millipore Cat: MGAMMAG-300K). The assay was performed according to the manufacturer’s instructions. The results were obtained by reading the plates on the Bioplex MAGPIX Multiplex Reader, and the concentration was calculated by using Bio-plex manager 6.1 software. A total of 6 markers (IgA, IgG1, IgG2a, IgG2b, IgG3, and IgM) were tested and quantified upon each standard.

Plasma cytokine expression profile detection

Plasma cytokine expression was performed using a magnetic bead panel 96-well plate assay (Millipore Cat: MCYTOMAG-70K) according to the manufacturer’s instructions. The results were obtained and analyzed by Bio-plex MAGPIX Multiplex Reader and Bio-plex manager 6.1 software, and each cytokine was calculated upon each standard.

Plasma Aβ_1–40 and Aβ_1–42 level detection

The concentrations of Aβ_40/42 were measured by the Aβ_1–40 and Aβ_1–42-specific sandwich ELISA kit (Mega Nano Biotech, FL, USA). Each well of a 96-well plate was coated with 50μl G1–42 (goat anti-human Aβ_1–42) antibody diluted to 10μg/mL in 1XPBS and incubated overnight at 4°C. The plate was washed 5 times and blocked by adding a 200μl blocking buffer at 37°C for 1 h. After washing and drying the plate, the plate was kept at room temperature ready for assay.

In a 96-well sample preparation plate, 50μl diluted detection antibody was added into each well and 50μl diluted peptide standard (Aβ_1–40 or Aβ_1–42) or 50μl diluted plasma samples were added at different dilutions to each designated well. Then the solutions were immediately transferred into the antibody-coated plates. The plates were incubated at 4°C for overnight. After washing, 100μl diluted secondary antibody was added to each well and incubated for 45 min on an orbital shaker at room temperature. The plates were washed, and TMB peroxidase substrate (Surmodics Cat: TMBS-1000) was added to each well and incubated at room temperature for 10 min. The reaction was stopped by adding 100μl/well of 0.4 M H₂SO4. Absorbance at 450 nm was read with a BioTek Synergy H4 microplate reader. The concentration was calculated upon the peptide standard.

Western blot detection for protein expression

Equal amounts of mouse brain protein samples were denatured with a loading buffer (Invitrogen Cat: NP0007) containing 6% β-mercaptoethanol and heated at 70°C for 10 min. Protein samples were then loaded to each well and separated using a 10% Bis-Tris gel. Precision plus protein dual color standards (Bio-rad, #1610374) were used as molecular standards. The separated samples were transferred with semi-dry assay to the PVDF membrane (Millipore Cat: IPFL00010). Membranes were first blocked with 0.2% Iblock buffer for 1 h at room temperature, and then incubated with the primary antibody at designated dilutions in blocking buffer on a shaker overnight at 4°C. After washing with 1XPBST three times for 5 min, blots were incubated with the appropriate horseradish peroxidase-conjugated secondary antibody in blocking buffer for 1 h. The enhanced chemiluminescence substrate (Thermo Scientific Prod # 34078) was used to develop the blots. Image J software was used for gel quantification.

The primary antibodies used were phospho-GSK-3β (Cell signaling Cat: 9336S), GSK-3β (Cell signaling Cat: 9315S), phospho-Erk (Cell signaling Cat:4370S), Erk (Cell signaling Cat: 4695S), LC3I/II (Cell signaling Cat: 4108S), A8 [46] (Beijing Jiaotong University), and β-Actin (Sigma Cat: A5441).

Statistical analysis

The data were analyzed by using GraphPad Prism 6 software. Statistical evaluation of the results was initially performed using a one-way ANOVA involving all groups. This was then followed by a post-hoc pairwise analysis of group differences, and then Fisher LSD test was used. Pearson correlation was used to analyze the correlation. The level of statistical significance was set at α= 0.05. Results were presented as mean±SEM.

Antigenicity prediction and immune response of the DC preparation in vivo

For optimal vaccine function, both B cell and T cell responses must be induced. The B cell response results in antibody production. In addition, the T cell response results in cytokine production and T cell surface marker expression of both CD4 and CD8. These T cells are activated through peptides presented by MHC-II and MHC-I, respectively. The TepiTool tools predict peptide binding to MHC class I and class II molecules. When compared to WT42, 13 novel T cell epitopes were discovered when analyzing E22W42 (Table 1). These 13 new peptides were all from mutations from E to W at amino acid number 22. The results demonstrate that these new peptides have a strong affinity to MHC class I and class II molecules. Moreover, the point mutation can help break Aβ-related immune-tolerance.

WT42 has a major T-cell epitope located at the central regions encompassing residues 16–24 [33]. Mice administered E22W35 peptide-sensitized DC vaccine can induce a much higher antibody response compared to E22W25, WT25, and WT35 (Fig. 1). Therefore, the optimized minimum length for the DC vaccine activation should be 1–35 or longer.

OPEN IN VIEWER

Fig. 1

E22W35 is the minimum peptide length to sensitize DCs to induce the immune response. Antibody titers of the C57/BL6 mice after two injections by using E22W25, E22W35, WT25, and WT35 sensitized DCs vaccine at 6-month-old. Only E22W35 can induce an antibody response (n = 4 per group).

E22W42-sensitized DC vaccine targeting oligomeric Aβ

Aβ₄₂ plays a key role in the pathogenesis of AD and is a core biomarker for the diagnosis of AD. In addition, oligomeric Aβ is well accepted as the major toxic isoform to neurons and may play a role in AD development [40]. Therefore, it is essential to determine whether E22W42 can generate an oligomer-specific antibody compared to other groups. As demonstrated in Fig. 2a, E22W42 group mice showed a good antibody response and these antibodies bound non-aggregated Aβ₄₂ peptide even at 1 : 800 dilutions. Moreover, our data did not show a linear decrease in absorbance when the antibody was diluted from 1 : 200 to 1 : 800 (1 : 4 dilution). Furthermore, the data in Fig. 2a also demonstrates that there was a high level of anti-oligomer Aβ antibody in plasma (1 : 800 dilution) because there was an excellent response to dilution from 1 : 200 to 1 : 800 (coated antigen is pre-aggregated WT42). The antibody isolated from the E22W42 DCs vaccinated mice plasma was then compared with the other anti-Aβ antibodies by western blot assay. In Fig. 2c, the blot was probed with the A8, an oligomer-specific antibody [46]. In Fig. 2d, the blot was probed with a purified antibody from plasma of the E22W42 DCs vaccinated mice. In Figure 2e, the blot was probed with 6E10 antibodies. Monoclonal antibody 6E10 is among the first anti-Aβ monoclonal antibodies. The antibody 6E10 targets both monomeric and oligomeric Aβ [47]. The significance is that the E22W42 DCs vaccine produced an oligomer-specific antibody. Both ELISA assay and western blot show this vaccine targets the toxic isoform of Aβ (Fig. 2). Figure 2f shows the antibody production and time duration for the immunization. The results show that the antibody level can be continuously boosted up until the 5th immunization injection, and no significant changes in level occurred, even after 2 months.

OPEN IN VIEWER

Fig. 2

E22W42 DCs vaccines can successfully induce high level antibody binding to monomeric and oligomeric Aβ. Characterization of antisera from E22W42 DCs vaccine. a) The binding ability of plasma collected from mice treated by different peptide-sensitized DCs vaccine to non-aggregated Aβ peptide. DCs vaccine is dendritic cells sensitized by E22W42, WT42, or DC-only control. b) The same sera bound to pre-aggregated Aβ peptide (three days). c-e) Western blot results. Lane 1 is hexafluoroisopropanol (HFIP)-treated wild Aβ1-42 (WT42, monomer), lane 2 is HFIP-treated WT42 aggregated for 3 days (oligomer), lane 3 is E22W42-treated with HFIP (assumed monomer), lane 4 is E22W42 peptide treated with HFIP under aggregation for 3 days (oligomer). Descriptions are as follows: c) Probed with mouse monoclonal anti-Aβ antibody named A8 (an oligomer specific antibody), d) Probed with purified antibody from plasma collected from E22W42 DCs vaccinated mice, and e) Probed with 6E10 antibody. f) Antibody duration post-vaccination. T1 is the blood collected 10 days after the second immunization. T2 is the blood collected 10 days after the 5^th immunization. T3 is the blood collected 2 months after T2. There is no significant difference in antibody titers after 2 months, between T2 and T3 (p > 0.05).

The immature DC cell-based vaccine retains the properties of DCs

DCs are considered a bridge of the immune system, connecting innate immunity and adaptive immunity. A DC vaccine can induce a moderate antibody response that helps avoid adverse effects. Table 2 shows the correlation analysis results of the cytokine profile post-vaccination among all vaccinated groups. The horizontal rows are correlation results of IL-6 versus GM-CSF, IL-10 versus GM-CSF, and IL-6 versus IL-10. The vertical columns are the DCs control and WT42 and E22W42-treated groups of mice. The correlation results of IL-6 versus GM-CSF, IL-10 versus GM-CSF, and IL6 versus IL-10 results indicate that the WT peptide group is significantly different from the DCs only and E22W42 groups. This indicates that the WT42-sensitized DCs vaccine can change the DCs’ properties. However, the E22W42-sensitized DC vaccine retains the property of DCs (p < 0.01). This means that the natural dendritic cell immune function will remain the same, which makes the E22W42 DC vaccine a better approach for AD treatment.

The DC vaccine promotes a Th2 response

Due to the failure of many Aβ-based vaccines in clinical trials, we need to further characterize the immunotype of the vaccine by using the IgG1/IgG2 ratio. Our result indicated that the E22W42-sensitized DCs vaccine initiates the Th2 type response (Fig. 3). It is well known that Th1 cells are involved in cellular immunity and a pro-inflammatory response. Th2 cells can help B cells differentiate into antibody-secreting cells and normally generate a less inflammatory response. A study showed IFN-γ production by Aβ-specific Th1 cells promotes microglial activation and increases plaque burden in a mouse model of AD [48], and Aβ-specific Th2 cells have been shown to reverse cognitive impairment [49]. Meanwhile, Th1 cytokines and Th2 cytokines induce isotype switching to IgG2a (Fig. 3a) and IgG1 (Fig. 3b), respectively [50, 51]. In order to exclude individual differences, we used the ratio of post/pre IgG1 or IgG2a. The results show that the E22W42 groups only increased the ratio of IgG1 level compared to the WT42 and control group. However, the ratio of post/pre IgG2a level remained the same, indicating that the effect of the E22W42 DC vaccine generated a favorable immune response.

OPEN IN VIEWER

Fig. 3

E22W42 DCs vaccine promotes Th2 response. Ig Isotyping changes post-DCs vaccination injection. The IgG1 (a), IgG2a (b) levels in the mice plasma of control, WT42, and E22W42 DCs vaccination pre- and post-injection, respectively, and the post/pre ratio level. There is increased IgG1 in the E22W42 vaccinated group post-vaccination versus pre-vaccination, but there are no differences among all groups in the ratio of IgG1/IgG2a.

This DCs vaccine does not cause inflammation

Active immunotherapies (anti-Aβ vaccines) have failed in the past because of the abnormal immune responses [17, 19]. AN-1792 (an Aβ-based adjuvant vaccine) exhibited an increased risk of Th1-mediated meningoencephalitis [52]. Meanwhile, increased inflammation is a primary symptom of AD, so treatment will need to avoid causing any inflammation. Th1 (pro-inflammatory) cytokines IL-2, TNF-α, and IFN-γ cause stimulation of CD8-positive cytotoxic T lymphocytes [53]. We tested the IL-2, IFN-γ, and TNF-α levels in the plasma of E22W42 vaccine-treated mice and found no significant differences. This indicates that the E22W42-sensitized DCs vaccine has little potential for over priming the immune system (Fig. 4).

OPEN IN VIEWER

Fig. 4

E22W42 DCs vaccine has no inflammatory response. Cytokine expression profile detection post-vaccination. Levels of IFN-γ (a), IL-2 (b), and TNF-α (c) in mouse plasma of the control, WT42 DCs, and E22W42 DCs vaccine for pre and post-vaccination, respectively, and also the post/pre ratio level of each group. There is no significant difference across all inflammatory cytokines (a-c).

E22W-sensitized DCs vaccine can benefit memory in APP/PS1 mouse model

Memory benefit is the major readout for any AD treatment method. We have tested the change in spatial working memory and showed that this vaccine generates significant memory benefits compared to non-sensitized DC-injected transgenic mice (Fig. 5). The E22W42 DC vaccine-treated old APP/PS1 mice showed memory improvement when tested by a radial arm water maze. Figures 5a and 5b show that the working memory performance of old APP/PS1 mice was significantly worse compared to non-transgenic mice (NTG) controls for both Errors and Latency, respectively. This demonstrates that the APP/PS1 transgenic mice showed memory loss as characteristic of AD. Working memory performance lost in old APP/PS1 mice is similar to AD patients. Meanwhile, errors of working memory are reduced in the E22W42 group mice. The latency of working memory is shortened in WT42 group mice compared to the DC control group. This implies that this type of DCs vaccine may be an ideal treatment for AD patients.

OPEN IN VIEWER

Fig. 5

E22W42 DCs vaccine improved memory in APP/PS1 mice. Spatial Working Memory improved after post-vaccination (bar graphs). In both a and b, the first bar (from left to right) is the non-transgenic mice, the second (from the left) bar is the DC control group, the third bar (from the left) is the WT42 DCs vaccine treated group, and the last bar (from the left) is the E22W42 DCs treated group. a) Number of errors that mice made in testing working memory performance. b) Latency of mice to find the platform in testing working memory performance. c) ELISA are the results of brain anti-Aβ antibody level for both E22W42 DC and DC control. The plate is coated with re-aggregated Aβ, and then brain tissue lysates are loaded as samples at a 1:50 dilution (protein concentration is 1mg/ml). E22W42 DCs vaccine group is significantly higher than the DC control group (n = 6 per group, p < 0.01). d) The blood anti-Aβ antibody level for both E22W42 DCs vaccine and DC control is demonstrated. The plate is coated with pre-aggregated Aβ_1-42, and plasma is diluted at 1:800 then detected with an anti-mouse antibody. The Anti-Aβ antibody level in E22W42 DC vaccinated group is significantly higher than the DCs control group (n = 6 per group, p < 0.01).

Significantly, the E22W42 DC vaccine treatment elevated the level of brain Aβ antibody, compared to DC control (Fig. 5c). In addition, the E22W42 DC vaccine treatment elevated the blood Aβ antibody level, compared to the DC control (Fig. 5d).

The vaccine can increase brain LC3-II level

Mapping the mechanism is very critical for a safe and effective therapeutic, so we further analyzed autophagy in brain tissue and discovered that LC3-II levels increased in the vaccinated group compared to the control group (Fig. 6). LC3 is a protein involved in autophagy. In normal conditions, the LC3 protein exists in the cytosol as type I (LC3-I). LC3-I is converted to LC3-II by lipidation and is recruited to autophagosomes when autophagy is activated. A study showed that LC3-II is also associated with endocytosis, facilitates Aβ clearance, and mitigates neurodegeneration in murine AD [54]. E22W42 group mice show a higher level of LC3-II (Fig. 6a) and LC3-II /LC3-I (Fig. 6c) in APP/PS1 mice compared with the WT and DC only group (control group). E22W42 can enhance autophagy activity and neuroprotection; no differences were observed in the LC3-I levels among the groups (Fig. 6b).

OPEN IN VIEWER

Fig. 6

Increased level of LC3-II and LC3-II /LC3-I level in E22W42 DCs vaccine group may have a role in neuroprotection. The detection of protein expression levels in the brain by western blot. There is a total of 28 APP/PS1 mice that are individually label-located above in three different groups in (a), (b), and (c). DCs control, WT42 DCs, and E22W42 DCs vaccine groups are labeled in the blot (DC ctr=9, WT42 DCs=9, E22W42 DC=10). The first horizontal row of bands is detected with the anti-mouse LC3-I antibody and the second horizontal row of bands is probed with LC3-II antibody. The third horizontal row of bands is probed with β-actin (loading control) antibody. (d), (e), and (f) are the quantification data of (a)-(c). (d) is the relative quantitative level of LC3-II (the second row versus the third row of (a)-(c). The first bar from the left is the E22W42 DCs treated group, the second bar from the left is the DC control group, and the third bar from the left is the WT42 DCs treated group. The LC3-II expression level in the E22W42 group is significantly higher than the other two groups (E22W42 DC versus WT42 DC p < 0.05, and E22W42 DC versus DC control p < 0.01). (e) is the quantitative level of LC3-I (the first row versus the third row of (a)-(c), the same order as (d), and there are no significant differences among all three groups. (f) is the level of LC3-II /LC3-I ratio (second row versus the first row of (a)-(c), the same order as (d). The E22W42 DC group is significantly higher than the other two groups (p < 0.05).

Brain GSK-3β level

We have analyzed pGSK-3β and total GSK-3β, as well as the ratio of pGSK-3β/GSK-3β (Fig. 7). Increased GSK-3β activity has been proposed to play a central role in the development of AD, and its overexpression has been shown to induce neuronal death [50, 51]. The p-GSK3β level is associated with memory decline in AD, so it is pivotal to determine any changes during anti-Aβ treatment. The E22W42 DC vaccine-injected mice did not show an effect on the level of p-GSK-3β, but longer treatment times need to be conducted. The average GSK-3β level in brain tissue of E22W42-vaccinated mice stayed the same as the DC injected group. The average GSK-3β level of the WT42 group is 98% of that of the DC control; the average p-GSK-3β levels of E22W42 and WT42 are 91% and 95% of that of the DC control, respectively. Meanwhile, the average pGSK-3β/GSK-3β level of E22W42 and WT42 is 93% and 99% of that of the DC control.

OPEN IN VIEWER

Fig. 7

E22W42 DC vaccine mice effects on the level of p-Gsk-3β. Brain GSK-3β expression post-vaccination. There were a total of 28 APP/PS1 mice, with group labels located in three different blots in (a), (b), and (c). GSK-3β, β-actin of GSK-3β, and p-GSK-3β bands are clearly labeled. In both (d) and (e), bar 1 from left is the E22W42 DCs group, bar two from left is the DC control group, and bar three is the WT42 DCs group. (d) is the quantification of the level of p-GSK-3β/actin ratio. (e) is the quantification of the level of GSK-3β/actin ratio. There are no significant differences among all three groups (p > 0.05).

Conclusion

Our results lead to the following important conclusions: 1) Only Aβ peptides with mutations in the T cell epitope, such as E22W42, can sensitize DCs as a vaccine. The major advantage of using E22W42 is that it can generate several novel T cell epitopes and silence auto T cell epitopes. These novel T cell epitopes may lay a foundation for identifying T cell activators for cell-based immunotherapy. 2) E22W42 can sensitize DCs to target biologically relevant oligomeric forms of Aβ. The sensitized DCs vaccine mediated the amelioration of memory deficits that occurred in the mouse AD model. 3) This is the first time that the E22W42 DCs vaccine has been tested on APP/PS1 transgenic mice. The E22W42 DCs-based vaccine has immunomodulatory effects that target oligomeric Aβ. 4) The E22W42 DCs helped to maintain dendritic cells’ immunomodulatory properties. Though the E22W42-sensitized DC vaccine is being developed for AD patients, it can potentially help strengthen the immune system of elderly patients as well. E22W42 may possibly lay the foundation for future immunotherapies for aging-related disorders.