1. Introduction
Alzheimer’s disease (AD) is the most common form of age-associated neurodegenerative disorder, clinically characterized by a decline in cognitive function or dementia [
1]. Pathologically, it is defined by the accumulation of two main lesions, amyloid (Aβ) plaques and intracellular neurofibrillary tangles (NFTs) [
2]. In 2006, evidence was presented that (Aβ) binds to Tau in solution and that this may be a precursor event to AD [
3].
The amyloid hypothesis has dominated the AD field and the development of AD therapeutics has focused mostly on removing Aβ from the brain. However, so far there are no positive results in human clinical trials of drugs that can revert or arrest the progression of AD [
4]. The amyloid cascade hypothesis states that first Aβ becomes abnormal and this drives pathology and mediates neurodegeneration [
5]. There is now a search for other therapeutic targets including anti amyloid and anti Tau interventions [
4]. Evaluations of human neuropathological data show that Tau pathology begins about 10 years before Aβ plaques appear. Therefore, it is unlikely that Aβ initiates the cascade of AD [
6].
The hypothesis behind our research is based on the findings that an interaction between APP and Tau plays a role in the induction and/or progression of AD [
7,
8] and a disruption of this association may therapeutically prevent the cognitive ability deterioration seen in AD.
In our previous publication [
9], we demonstrated that the association or binding between the two AD-involved proteins, APP and Tau, can be disrupted by a mixture of two peptides APP1 (aa 390–412 HFQKAKERLEAKHRERMSQVMRE) and Tau1 (aa19–34 GLGDRKDQGGYTMHQD). These peptides were predicted to associate with each other using several algorithm programs [
10,
11], and were actually demonstrated to crosslink to each other by identifying areas in both APP and Tau proteins as possible areas of interaction between the two proteins. Upon analysis of the crosslinked material which was processed for LC-MS/MS, only one crosslink was identified between APP and Tau, between lysine 370 on APP and lysine 387 on Tau. The crosslinked lysine for APP resides very close to the accordingly predicted APP1 (390–412) peptide, as can be seen by the crystal structure of the region [
12]. As for Tau, we selected the peptide Tau1 (residues 19–34), which is in the N-terminal end of the Tau protein, since phospho-Tyr-hTau located in the N-terminal was reported to accompany AD progression and Tauopathy [
13]. We also selected peptide Tau2 (residues 331–348) from the microtubule area of Tau protein, which is proximal to the crosslinked lysine 387. We tested in addition to APP1 and Tau1 two other peptides as controls for binding to each other both by dot blot and by Eliza testing and found that APP1 and Tau1 Mix were the best candidates for reducing plaques and improving cognitive ability.
A mixture of APP1 and Tau1 peptides administered intranasally to four-month-old transgenic mice (relating to either APP or Tau Tg), exhibiting declined cognitive functions, prevented this decline as well as reduced the formation of amyloid plaques in the brain [
9]. The interference with the association of APP and Tau proteins was demonstrated only by a mixture of the two specific peptides and not by either peptide alone. The effect of the peptide mixture on the cognitive functions in the murine AD model 5xFAD mice [
14] was attained when treatment was initiated either at an early age (three months), preventing the deterioration of cognition, or at the age of six months, when severe cognitive function loss and plaque load were already evident. These findings, namely improvement in cognitive functions and reduction in the (Aβ) plaque load, indicated the potential of this peptide mixture as therapeutic agent.
The present study sought to evaluate whether a mixture of these two peptides is essential for the reported beneficial effects, or whether the peptides can retain their biological effect when linked to each other via a peptide bridge to yield a single longer peptide. There are two advantages to this bridging approach: firstly, the use of a single defined product is preferable to a mixture of two components. The second advantage is that it can be expressed recombinantly, making it a less expensive procedure than chemical synthesis. In the present study, two versions of such linked peptides were prepared, employing either a “flexible” (3×GGGGS or a “rigid” (3×EAAAK) peptide bridge for the linkage [
15,
16]. Both versions were consequently evaluated for their beneficial effect in the 5xFAD mouse AD model. To that end, the current study evaluated the treatment effects of the linked peptides on cognitive functions through Y-maze experiments and on plaque load using brain sections labelled with anti Aβ antibodies.
3. Discussion
The current study demonstrates that APP1 (390–412) and Tau1 (19–34) peptides, linked together with either a flexible or a rigid peptide bridge, are able to inhibit, in vitro, the interaction between APP and Tau proteins. Furthermore, these linked peptides concomitantly reduce the Aβ plaque burden in treated 5xFAD Tg mice, a model for AD, and improve their cognitive functions. APP and Tau are involved in AD development, likely by binding to each other. The linked peptides are able to inhibit, in vitro, the interaction between APP and Tau, indicating the therapeutic potential of the linked peptides toward drug development. Additionally, our ~six-month oral treatment with the Flex linked peptide showed a full recovery of cognitive functions, reaching identical levels to those of the non-Tg mice. Taken together, the nasal and oral results strongly indicate the peptides’ potential toward drug development.
In our previous publication [
9], we described that a mixture of two peptides (APP1 390–412 and Tau1 19–34) prevents the association of the two brain proteins APP and Tau and improves the cognitive ability in a transgenic mouse model of AD.
In the context of drug development, using a mixture rather than a single component is a complication due to the varying pharmaco-kinetics of the different agents [
21]. To avoid this complication, several approaches have been suggested, including the use of nano particles [
22]. In the present study, since the two components are peptides, a simpler solution is linking the two peptides into a single longer peptide. There are two advantages to this approach: firstly, the use of a single defined component is preferable to the use of the two-peptide mixture. The second advantage is that it can be expressed recombinantly, a less expensive procedure than chemical synthesis. The linkage could be provided by either a flexible (3×GGGGS) or a rigid (3×EAAAK) peptide bridge [
15,
16]. It should be noted that, in both versions of the linked peptide, each of the two individual peptides is exposed and available to interact with the other, thus simulating the effect of the mixture of the peptides. Since it was not possible to predict in advance which combined peptide is expected to be more effective, both versions were prepared and evaluated for their biological effect, namely for their effects on plaque burden and cognitive functions in a mouse model of AD.
The delivery procedure is a significant consideration in drug administration. In certain cases, the delivery system is mainly intravenous transfusion (e.g., monoclonal antibodies). However, smaller molecules of various compositions and sizes may be delivered non-invasively (e.g., orally or nasally) [
23]. In general, non-invasive drug delivery is the preferred method of delivery and, in AD patients specifically, it may be easier for them to take a pill. In the treatment of multiple sclerosis, there are several drugs in use, some of them administered orally [
24] and some by injection [
25]. The latter include beta interferon, as well as Copaxone
®, which is widely used due to its high efficacy and safety profile [
26]. Nevertheless, many patients prefer taking orally delivered drugs [
27].
The results of the biotin-labelled conjugated peptide delivered nasally show that it indeed reaches the brain of transgenic mice, leading to a substantial reduction in the plaque load. These results are indeed promising for a peptide that is ~50 amino acids long. Furthermore, nasal administration leads not only to delivery of the peptide to its target organ, but also and mainly, to its efficacy in preventing the loss of cognitive functions in the treated mice. It is important to emphasize that no accumulation of this product was observed in similarly treated non-Tg mice, thus providing reassuring safety considerations.
Similarly, and of even higher practical consideration for drug development, is the effectiveness of the linked Flex peptide when delivered orally by gavage treatment. This effectiveness, protecting cognitive functions, was evident following 5–6 months of treatment (every other day, three times per week) that started at the age of three months, before cognition deficiency occurred. Such treatment may be considered as a preventive treatment. It should be noted that, in our previous study, using a mixture of the two individual APP and Tau peptides, we demonstrated a similar protective effect on cognition not only when administrating the Mix at three months of age, before cognitive deficiency is observed, but also at the age of six months, when cognitive impairments were already evident. In both cases, upon completion of the repeated series of treatment, which was continued for 150 days, full cognitive ability was restored [
9].
Although the results presented here relate to a transgenic mouse model of AD, they may have implications toward drug development for the treatment of AD in humans, for which no drug is available yet. Hence, the finding that a peptide, which theoretically should not present any safety issue, is efficient in preventing cognitive loss when administered repeatedly, either intranasally or orally, is encouraging. Furthermore, the results of our previous study [
9] demonstrate that the effect of the APP1 and Tau1 peptide mixture can be observed not only as a preventive means, but also when initiated at the age of six months, when cognitive impairments are already observed, leading to the recovery of cognitive functions. If pertinent to AD, these combined data may be indicative of possible effectivity, either as preventive, at the very early stages of the disease or at later stages, to alleviate continued cognitive deficiency.
4. Materials and Methods
4.1. Proteins, Chemicals and Antibodies
Recombinant human APP770, (BLG843201), recombinant human Tau-441 (2N4R) (BLG6842501) and purified anti-amyloid 1–16 6E10 (RRID:RB_2564653) were all purchased from Bio-Legend, San Diego, CA, 92121, USA. Anti-mouse HRP was obtained from Jackson Immunoresearch, Westgrove PA, USA 19390 (Cat#715-035-151). Mouse anti-glyceraldehyde 3-phosphate dehydrogenase antibody (GAPDH) Cat#MAB374 was purchased from Merck-Millipore, MA, USA. Chemicals and antibodies were purchased between 2019–2021.
4.2. Mouse Lines
The following mice were used: 5xFAD male and female double transgenic mice (Tg6799 line APP/PS1, JAX: 034848, Jackson, ME, USA), co-expressing the human amyloid precursor protein carrying five familial AD mutations: the Swedish, Florida and London mutations and two mutations of the human presenilin-1. All mouse lines were maintained on a C57Bl/6 background (Jackson, RRID:IMSR JAX:000664). Genotyping was performed by PCR amplification of ear DNA, as previously described [
14]. The mice were housed in individually ventilated cages (no more than five mice per cage) in a temperature-controlled facility with a 12-h light/dark cycle. All animal care and experimental use were in accordance with the Weizmann Institute of Science guidelines and were approved by the Weizmann Animal Care Committee IACUC #18691119-2 (31 December 2019–2 January 2023). Animal weight was 20–25 g, and mice were given food and water ad libitum.
4.3. Synthetic Peptides
Two synthetically linked peptides were prepared by linking APP1 and Tau1 peptides with a flexible linker or with a rigid linker. Another synthetic peptide of Biotin-flexible peptide was prepared as well, all by GL Biochem Ltd. (Shanghai, China). Custom-made materials can be shared upon reasonable request.
APP1: HFQKAKERLEAKHRERMSQVMRE;
Tau1: GLGDRKDQGGYTMHQD
Flex: HFQKAKERLEAKHRERMSQVMREGGGGSGGGGSGGGGSGLGDRKDQGGYTMHQD
Rigid: HFQKAKERLEAKHRERMSQVMREEEAAAKEAAAKEAAAKGLGDRKDQGGYTMHQD
4.4. In Vitro Tests
An enzyme-linked immunosorbent assay (ELISA) plate (96 wells, flat bottom) was coated with Tau protein 1 µg/mL bicarbonate PH = 8.2 (50 µL/well), then incubated overnight at 4 °C. At the same time, Flex and Rigid peptides (1 µg/mL PBS) and Mix peptide which is a combination of APP1 and Tau1 (0.5 + 0.5 µg/mL PBS), were also incubated overnight at 4 °C in PBS. The next day, the plate was washed 3× with PBS and blocked with 3% BSA/PBS (10 µL) for two hours at RT. The plate was washed again 3× with PBS, and peptides or their combinations (samples incubated the night before at 4 °C) 50 µL/well were added for four hours at RT. The plate was washed again 3× with PBS, and APP protein, 1 µg/mL per 50 µL/well was added for an ON incubation at 4 °C.
The next day, the plate was washed 3× with PBS, following which anti-Aβ 6E10 (1 mg/mL) was added (1:500 dilution, 50 µL/well) for two hours at RT. The plate was washed again and anti-mouse HRP antibodies (1:10,000 dilution in PBS 1% BSA) were added at 50 µL/well. Color reaction was stopped with 50 µL/well of 1M H2S04 and read at O.D. 450.
4.5. Histological Staining and Quantitation of Amyloid
Mice were euthanized with a Pantol injection 20mg/100u, (CTS Israel), and saline-perfused at the end of the experiment as approved by IACUC #18691119-2 (31 December 2019–2 January 2023). The right brain hemisphere of each mouse was stored at −70 °C to quantify Aβ levels. The left hemispheres were fixed (4% paraformaldehyde) for one week, paraffin-embedded and sectioned sagitally (6 microns). Paraffin sections were deparaffinized and rehydrated. Antigen retrieval was performed in 10 mM citric acid pH 6 for 10 min in a microwave to break protein crosslinks and to unmask the antigens. After pre-incubation with 20% normal horse serum and 0.2% Triton X-100 for 90 min, sections were incubated with 6E10 anti Aβ antibody (1:200) at RT for 18hr. The second antibody step was performed by labelling with species highly cross-absorbed cy3 conjugated anti mouse antibody (1:100 Jackson Immune Research, West Grove, PA, USA) for 30–60 min.
For the biotin control animals, the signal was detected by incubation of the sections taken from this group, with cy2 conjugated streptavidin (1:100, Jackson Immuno-Research) for 30 min. All sections were counterstained with Hoechst 33258 (Molecular Probes, Eugene, OR, USA) for nuclear labelling. Stained sections were examined and photographed with a fluorescence microscope (Eclipse Ni-U, Nikon, Tokyo, Japan) equipped with Plan Fluor objectives, connected to camera (DS-Qi1, Nikon).
Quantification analysis of amyloid depositions number, sum of the deposit area and percentage area of deposits (from total measured area), was performed using Image-Pro V-10 software.
The measurements were performed in a fixed rectangle area (3.05 × 1.8) mm, covering all the hippocampus (5.5 mm2) in each section, with three sections per animal.
4.6. In Vivo Animal Treatment and Behavior Tests
The in vivo research design employed in this study is illustrated in
Figure 3. Mice used were heterozygous 5xFAD Tg mice (Tg 6799), and as controls, non-Tg littermates. No predetermined sample calculation was performed and no exclusion criteria were predetermined. A total number of 70 mice were used for the in vivo studies (41 mice in the nasal treatments and 29 mice in the gavage feeding experiment). The mice were kept on a reversed cycle light regime (12–12:lights on at 20:00) to allow for testing during the dark-active diurnal phase. Mice were arbitrarily chosen, but no official randomization method was used. Mice were arbitrarily assigned to experimental treatment groups of peptide treatment or PBS-treated controls (3–9 animals each). No behavioral differences were observed between the sexes when tested in the Y-maze experiment, in terms of response to treatment. Treatment in four different experiments started when the mice were 3 to 5 months old. A mixture of APP1 peptide 5 µg/5 µL and Tau1 peptide 5 µg/5 µL as well as Flex and Rigid peptide (10 µg/10 µL) was administered nasally every other day for 5–7 months. The control 5xFAD-Tg mice, as well as non-Tg littermates, were used as controls, treated with PBS. In the gavage feeding experiment, mice were fed 100ug Flex peptide, diluted in 250 µL PBS per treatment every other day, three times per week. No anesthetics/analgesics were used as neither treatment nor behavioral testing caused discomfort to the mice. The mice were verified three times a week, and no mortality or weight loss occurred. At the end of the experiment, the mice were euthanized by Pantol injection (20 mg/100 µL, CTS Israel) as approved by the ethical protocol, and their brains were excised. One hemisphere was prepared for histology and the other was frozen in −70 °C for processing Aβ 1–42 content.
The Y-maze test [
28] consisted of two trials, the first “training” and second “retention”, separated by an inter-trial interval (ITI). Each arm of the Y-maze is equipped with a guillotine door that could be operated manually. The three identical arms were randomly designated as follows: the “start” (steam) arm, in which the mouse began to explore the maze (always open); the “novel” arm, which was closed during the first trial (“training”), but open in the second one (“retention”) and the “other” arm (always open). The first (‘training”) trial lasted five minutes and allowed the mouse to explore only two arms (“start” and “other”) of the maze. Access to the third (“novel”) arm was blocked. The second trial (“retention”) was conducted after a three-minute ITI. During the three-minute trial, all three arms were accessible. The mouse was returned to the same starting arm and was allowed to explore all three arms. Retention was indicated by a preferential index to the “novel” arm, which was calculated as follows: time spent in the “novel” arm minus the time spent in the “other” arm divided by the sum of time spent in both those arms: “0” indicates no preference. Values significantly higher than “0” indicate a preference to the “novel” arm, i.e., adequate retention [
28].
We added another behavior test assessing nest building to support the Y-maze results in 5xFAD mice [
18]. Mice were placed individually in a cage with two cotton pressed squares. Cages were housed in reversed cycle light regime 12–12: starting from 20 p.m. to 8:00 a.m. We then scored the condition of the nest created using a five-point nest rating scale. All behavioral testing and analysis were conducted in a blinded manner.
4.7. Statistical Analysis
Statistical analysis was performed in R, using the rstatix package. For the in vivo studies, a repeated measures ANOVA was performed to assess the significance of the time factor (pre-treatments vs. follow-up measurements), treatment factor (Rigid/Flex/Mix/PBS/non-transgenic) and their interaction. Follow-up post hoc tests were performed by pairwise t-tests, where relevant. One-sample t-test was used to compare Y-maze results to chance levels (). A probability value (p) of less than 0.05 was considered significant.