Tyrosol Reduces Amyloid-β Oligomer Neurotoxicity and Alleviates Synaptic,Oxidative,and Cognitive Disturbances in Alzheimer’s Disease Model Mice

Abstract

Soluble amyloid-β (Aβ) oligomers (AβOs), which elicit neurotoxicity and synaptotoxicity, are thought to play an initiating role in the pathology of Alzheimer’s disease (AD). Since AβOs are a key therapeutic target, we attempted to identify natural agents that reduce AβO neurotoxicity. Using an assay system in which primary cultured neurons are treated with AβOs, we found that Rhodiola rosea extracts and one of its main constituents, tyrosol, significantly inhibited AβO-induced caspase-3 activation. We then assessed the in vivo efficacy of tyrosol by oral administration of the compound into AD model (5XFAD) transgenic and non-transgenic mice from either 2 or 4 to 7 months of age. In both paradigms, tyrosol treatment did not affect body weights of mice. Immunohistochemical analysis revealed that the immunoreactivity of spinophilin, a dendritic synaptic protein, was significantly reduced in three hippocampal subregions of vehicle-treated AD mice compared with non-transgenic mice, which was reversed in tyrosol-treated AD mice. Tyrosol treatment also prevented the enhancement of 4-hydroxy-2-nonenal immunoreactivity in the hippocampal CA3 region of AD mice. By contrast, tyrosol administration did not affect Aβ accumulation, as evaluated by immunohistochemical and biochemical analyses. Moreover, the Barnes maze test showed that tyrosol administration modestly mitigated spatial memory impairment in AD mice. These findings collectively indicate that the natural agent tyrosol protects neurons against AβO neurotoxicity in vitro and ameliorates synaptic disturbance, oxidative stress responses, and cognitive impairment in vivo. We thus suggest that tyrosol is potentially an effective, safe, and unique drug candidate for AD.

Keywords

Alzheimer’s disease amyloid-β neuron oligomer oxidative stress synapse

INTRODUCTION

Alzheimer’s disease (AD) is a major dementia disorder that has exhibited a dramatic increase in incidence owing to aging of the population [1]. Although the pathogenesis of AD is not yet fully understood, the amyloid cascade hypothesis has been widely accepted, which posits that the abnormal accumulation of amyloid-β (Aβ) plays a central role in the pathomechanism of AD [2]. In recent years, soluble Aβ oligomers (AβOs) have been identified as neurotoxic and synaptotoxic species, after when a modified Aβ cascade hypothesis has been proposed in which AβOs trigger the deleterious cascades involved in the pathology of AD [3, 4]. This is an attractive and reasonable theory supported by a large number of studies, which have demonstrated that AβOs induce various neuronal abnormalities, including synaptic deficits, oxidative stress, apoptosis, and abnormal alterations of the microtubule-associated protein, tau [3, 4]. These abnormalities apparently reflect the characteristic features of AD pathology.

Numerous studies have underscored the importance of synaptic dysfunction in AD, in which AβOs may play a major role [5, 6]. In fact, AβOs are present at synaptic sites and induce synaptic disorganization, affecting synaptic plasticity (e.g., inhibiting long-term potentiation as well as synaptic structures (e.g., reducing spine density) [5, 6]. It is believed that the synaptic dysfunction and synapse loss contribute to the cognitive deficits in patients with AD. Oxidative stress of lipids, proteins, and DNA is extensively present in the AD brain. Therefore, oxidative stress appears to play a detrimental role in the pathogenesis of AD [7]. Recent evidence strongly suggests that AβOs are responsible for such oxidative damage [4 –6]. Furthermore, oxidative stress mechanisms are suggested to underlie synapse dysfunction in AD [8].

From a therapeutic standpoint, disease-modifying therapies, such as Aβ immunotherapies and inhibition of Aβ-producing proteases, including β-site Aβ precursor protein (AβPP) cleaving enzyme 1 (BACE1), which catalyzes the amyloidogenic processing of AβPP, have recently been developed to prevent Aβ accumulation [9 –11]. However, no such disease-modifying drugs have proven to be clinically effective, and considering different therapeutic strategies may be required to break this impasse. Targeting the neurotoxicity of AβOs itself is one such alternative therapeutic approach. Here, we attempted to identify natural agents that reduce AβO neurotoxicity using a primary neuron model and identified tyrosol, a plant-derived small molecule compound, as such an agent. Our subsequent in vivo preclinical studies demonstrated that chronic oral administration of tyrosol significantly reversed synaptic and oxidative abnormalities and alleviated cognitive impairment in AD model mice. Our results thus highlight tyrosol as a safe and effective drug candidate for AD.

MATERIALS AND METHODS

Cell culture

Primary cerebral cortical neurons were obtained from 17-day-old embryos of a Wistar rat as described previously [12, 13]. Neurons were plated on poly-L-lysine-coated plates or dishes at a density of 680 cells/mm². Cells were maintained in a humidified atmosphere of 5% CO₂/95% air in Macs Neuro Medium (Miltenyi Biotec, Auburn, CA, USA) containing 0.5 mM L-glutamine, NeuroBrew-21 (Miltenyi Biotec), and penicillin-streptomycin. Half of the medium was replaced with fresh medium every 3-4 days.

Antibodies and chemicals

The following primary antibodies were used in this study: anti-Aβ (82E1, IBL, 10323, Gunma, Japan), anti-BACE1 (D10E5, Cell Signaling, 5606, Danvers, MA, USA), anti-AβPP (R37 [14]), anti-presenilin 1 (PS1) (Cell Signaling, 5643), anti-spinophilin (Cell Signaling, 14136), anti-4-hydroxy-2-nonenal (4-HNE) (Japan Institute for the Control of Aging, MHN-020P, Shizuoka, Japan), anti-hexanoyl-lysine adduct (HEL) (Japan Institute for the Control of Aging, MHL-021P), anti-cleaved caspase-3 (Asp175) (Cell Signaling, 9664), and anti-actin (Wako, 017-24556). Tyrosol and epigallocatechin gallate (EGCG) were acquired from Tokyo Chemical Industry (Tokyo, Japan), and salidroside was from Sigma. Rosarian, rosavin, rosin, and rosiridin were purchased through Namiki Shoji Co., Ltd (Tokyo, Japan).

Plant extracts

Plant extracts, including extracts of Arachis hypogaea (skin), Kaempferia parviflora (rhizome), Apocynum venetum (leaf), Vitis spp. (seed), Ginkgo biloba (leaf), Rhodiola rosea (R. rosea) (rhizome), and other three undisclosed plant extracts, were provided by Tokiwa Phytochemical Co., Ltd (Chiba, Japan). The rhizomes of R. rosea were extracted with water at 80°C for 1–2 h, and then centrifuged. The residue was again extracted with water in the same manner. The solutions were combined, concentrated, and dried to obtain the R. rosea extract.

Aβ preparation and treatment

Aβ₄₂ oligomers and fibrils were prepared as described previously [13 , 16]. Briefly, human Aβ_1 - 42 peptide (Peptide Institute, Osaka, Japan) was dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP; Sigma) to obtain 1 mM solution. HFIP was evaporated overnight in a hood and further under vacuum for 1 h, and dried peptide films stored at –30°C. Prior to use, dried Aβ peptide was resolved in dimethyl sulfoxide (DMSO) to prepare 5 mM stock, and sonicated in an ultrasonic bath sonicator for 10 min. To prepare oligomers, 5 mM Aβ DMSO stock was diluted in DMEM/F12 without phenol red and left for 1 day at 4°C. Immediately before addition to neurons (9 days in vitro (DIV)), Aβ preparations were diluted in regular medium and used to replace the entire medium. Control cultures were treated with the same concentration of DMSO.

Thioflavin T assay

The effects of tyrosol on Aβ₄₂ aggregation were assessing using thioflavin T (ThT) assays. Aβ₄₂, prepared as dried peptide films as described above, was dissolved in DMSO with ultrasonic bath sonication to yield a 5 mM stock. This Aβ₄₂ solution was diluted to 50 μM with 50 mM phosphate buffer (pH 7.4) containing 100 mM NaCl, to which different amounts of tyrosol (in DMSO) were added, giving final concentrations of tyrosol ranging from 10 to 100 μM. DMSO only and EGCG were used as negative and positive controls, respectively. The mixture was incubated at 37°C and monitored for 72 h. A portion of the mixture was collected after appropriate times and diluted 100-fold with 5 μM ThT (Abcam, Cambridge, MA, USA) in 50 mM glycine-NaOH buffer (pH 8.7). Fluorescence measurements were obtained at excitation and emission wavelengths of 450 and 486 nm, respectively, using a microplate reader (BioTek Instruments, Inc., Winooski, VT, USA).

Mouse models and treatment

Heterozygous 5XFAD mice (C57BL/6 background), which overexpress mutant AβPP (K670 N/M671 L, I716 V, and V717I) and PS1 (M146 L and L286 V) [17], were obtained from the Mutant Mouse Resource Research Center (MMRRC, Bar Harbor, ME, USA), and maintained by crossing with C57BL/6J mice. Littermate transgenic (Tg) and non-Tg mice were used in this study. For tyrosol treatment, tyrosol was dissolved in ultra-pure water to a concentration of 0.5 mM and sterilized by filtration. This tyrosol solution was administered orally by diluting in drinking water and supplying ad libitum in the feed-water bottle. The bottle was exchanged once a week. The amount of tyrosol administered was estimated to be ∼12.5 mg/kg/day, assuming an average body weight of ∼25 g and daily water consumption of ∼5 ml. This amount was chosen by reference to previous studies in which tyrosol was orally administered to rodents [18, 19]. Control mice were administered only water. Four groups of mice—Tg-Veh, Tg-Tyr, Non-Veh, and Non-Tyr (as described in Fig. 3)—were treated as above for 12 weeks (from 4 to 7 months old) or 20 weeks (from 2 to 7 months old). Although both male and female mice were used, the proportions of male to female were almost similar in the four groups.

All animal experiments were approved by the Animal Investigation Committee of the National Institute of Neuroscience, NCNP and performed in accordance with international and Japanese guidelines for animal care.

Immunohistochemistry

After euthanizing mice with ether, brains were rapidly removed and bisected through the midsagittal plane into left and right hemispheres. For immunohistochemical analysis, left hemisphere samples were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) overnight at 4°C, and cryoprotected in 15∼30% sucrose in PBS, mounted with O.C.T. compound (Sakura Finetek Japan, Tokyo, Japan) and stored at –80°C. The cerebral cortex and hippocampus were separately excised from the right hemisphere samples, frozen in liquid nitrogen, and stored at –80°C for subsequent use in enzyme-linked immunosorbent assay (ELISA) and western blot analysis procedures. Serial coronal sections (14-μm thick) were cut by a cryostat (Leica CM 1900; Leica Biosystems, Wetzlar, Germany) and mounted on slide glasses. Three to four sections per each mouse were taken that included the hippocampus and piriform cortex, from –1.8 mm to –2.0 mm to bregma according to the mouse brain atlas of Franklin and Paxinos [20].

Immunostaining of Aβ was performed as described previously [21]. Three sections per each mouse were treated with 0.3% hydrogen peroxide in PBS for 20 min and 1% horse serum in PBS for 5 min, and then incubated with anti-Aβ antibody diluted with PBS containing 1% bovine serum albumin and 0.1% Triton X-100 overnight at 4°C. After washing with PBS, sections were incubated with biotinylated universal antibody (Vectastain Universal Elite ABC kit; Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature. Sections were washed with PBS again and reacted with avidin-biotin-horseradish peroxidase (HRP) complex (Vectastain Universal Elite ABC kit) for 30 min. Labeling was developed with 0.05% diaminobenzidine tetrahydrochloride (DAB) and 0.03% hydrogen peroxide, and sections were counterstained with hematoxylin.

For immunofluorescence staining, three sections per each mouse were permeabilized and blocked with PBS containing 0.3% Triton X-100 and 1% horse serum for 20 min. Sections were then washed with PBS and incubated with anti-spinophilin antibody (Cell Signaling Technology, MA, USA) diluted in PBS containing 0.1% TritonX-100 and 0.2% horse serum or 4-HNE antibody diluted in Can Get Signal^® Immunostain Immunoreaction Enhancer Solution A (TOYOBO, Osaka, Japan) overnight at 4°C. After washing with PBS, sections were incubated with Alexa488-conjugated anti-rabbit or anti-mouse IgG secondary antibodies (Molecular Probes, Eugene, OR, USA) for 1 h. Sections were subsequently stained with 1 μg/ml DAPI for 15 min in some occasions.

Image acquisition and analysis

Aβ burden was estimated using a previously described method [21]. Images were obtained and digitally captured using a microscope with a 2x objective lens. The hippocampus and cortex were outlined, and areas of plaques in the outlined structures were measured by outlining the boundaries of all Aβ-positive plaques using Image J software (National Institute of Health, Bethesda, MD, USA). We chose the piriform cortex, including posterolateral cortical amygdaloid nuclei, as a region of interest (ROI), as illustrated in Fig. 4; This region has been considered suitable for evaluation of Aβ deposition [22]. The percent plaque areas obtained in each section were averaged to calculate Aβ burden in each mouse. Immunofluorescence-stained specimens were observed under an LSM 780 laser-scanning confocal microscope (Carl Zeiss, Jena, Germany). For quantitative analyses, images of antibody-labeled specimens were acquired using a 20x dry objective lens. The signal intensity of immuno-labeled proteins in a ROI was measured using ZEN software (Zeiss). ROIs for spinophilin were determined as illustrated in Fig. 5 (∼35,000, ∼25,000, and ∼40,000 μm² in dentate gyrus (DG) [hilus], CA3 [stratum lucidum], and CA1 [stratum radiatum] regions of the hippocampus, respectively). ROIs for 4-HNE were determined as illustrated in Fig. 6 (∼11,000 μm² in the CA3 region [pyramidal cell layer]). For quantitative analysis, mean values of fluorescence intensity were calculated.

Immunocytochemistry

Primary neurons cultured on poly-L-lysine-coated coverslips were fixed with 4% paraformaldehyde in PBS. Fixed cells were permeabilized and blocked as described previously [13 , 23] and incubated with anti-HEL antibodies, followed by incubation with Alexa488-conjugated anti-mouse IgG secondary antibodies. Specimens were observed under an LSM 780 laser-scanning confocal microscope (Carl Zeiss, Jena, Germany).

Western blot analysis

Cells were lysed on ice in RIPA buffer (10 mM Tris pH 8.0, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 5 mM EDTA) containing protease inhibitors (aprotinin, leupeptin, pepstatin, PMSF). After rocking for 1 h at 4°C, samples were centrifuged at 100,000×g for 30 min, and the supernatants used as cell lysates. The protein content in cell lysates was estimated with the bicinchoninic acid assay (Pierce, Rockford, IL, USA). Western blotting of cell lysates was performed using a standard procedure, as described previously [13]. Protein expression was detected with a chemiluminescence reagent (Perkin-Elmer, Boston, MA, USA), and the resulting images examined with a LAS-1000 image analyzer (Fuji Film, Tokyo, Japan). Actin was used as the internal control to normalize the levels of proteins of interest.

Aβ measurement

Mouse brain tissues were processed as described previously [21]. Dissected brain tissues were homogenized in Tris-buffered saline (TBS) (50 mM Tris-HCl pH 7.4, 150 mM NaCl) containing protease inhibitors. The homogenate was divided into two tubes and centrifuged at 100,000×g for 30 min. The supernatants were combined and used as Tris-soluble fraction. The pellet in the first tube was then resuspended in guanidine-HCl buffer (6M guanidine-HCl, 50 mM Tris-HCl pH 7.4), and subsequently sonicated with an ultrasonic homogenizer and incubated in a bath sonicator for 30 min. The samples were centrifuged at 280,000×g for 30 min, and the supernatant solution (guanidine-soluble fraction) was stored at –80°C for Aβ measurement. The pellet in the second tube was extracted by incubating with RIPA buffer containing protease inhibitors for 1 h, followed by centrifugation at 100,000×g for 30 min. The resultant supernatant solution (RIPA-soluble fraction) was used for western blotting.

The guanidine-soluble and TBS-soluble fractions described above were diluted with water, and the amounts of Aβ₄₀ and Aβ₄₂ were measured using sandwich ELISA kits (Wako, Osaka, Japan), as described previously [21]. Briefly, samples and Aβ standard solutions were applied to BNT77-coated 96-well plates and incubated overnight, followed by incubation with HRP-conjugated BA27 or BC05 for 1 h. Bound enzyme activity was measured using the TMB Microwell Peroxidase Substrate System (Seracare, Gaithersburg, MD, USA).

Behavioral test

Spatial memory of mice was tested in the Barnes maze system (O’Hara & Co., Ltd., Tokyo, Japan) according to a previously described protocol, with slight modifications [24]. The Barnes task was conducted on a brightly lit (900 lx), white, 1.0-m diameter, circular plate with 12 holes equally spaced around the periphery. This circular open field was elevated 96 cm from the floor. A black removal escape box (17×13×7 cm) was located under one of the holes. The location of the target was consistent for each mouse. Four colored shapes were placed in the room as visual cues. Animals were habituated to the maze for 2 days prior to the beginning of testing. In the training session, mice were placed inside a cylinder in the center of the plate. After the cylinder was removed, the mice were allowed to explore the maze for a maximum of 5 min. Two trials per day were conducted on five successive days. On day 6, a probe trial was conducted without the escape box. Each mouse was given 3 min to explore the maze. The latency to enter the target hole (during the training sessions) and the time spent around each hole (during the probe trial) were recorded using video tracking Time BCM software (O’Hara & Co., Ltd.).

Statistical analysis

All results are presented as means±SEM. Behavioral analysis data were statistically analyzed using repeated two-way or factorial one-way analysis of variance (ANOVA) followed by a Tukey or a Bonferroni multiple comparison test. Other data were analyzed using factorial one-way ANOVA followed by a Tukey or a Bonferroni multiple comparison test or two-tailed Student’s t-test. p-values < 0.05 were considered statistically significant. Analyses were performed using GraphPad Prism5 J (GraphPad Software Inc., San Diego, CA, USA).

RESULTS

Identification of tyrosol as a natural agent that protects against AβO neurotoxicity

Using a neuronal culture system, we screened natural plant extracts for those that were capable of decreasing the neurotoxic effect of AβOs. For this purpose, rat primary neurons were exposed to 2.5 μM Aβ₄₂ oligomers [13, 16], alone or together with each of the extracts, for 3 days, followed by western blot analysis of cell lysates for cleaved caspase-3 (Fig. 1A). AβO treatment induced a significant increase in the level of cleaved caspase-3, consistent with our previous studies [13, 16]. Notably, a few extracts, including R. rosea and G. biloba extracts and another undisclosed extract, appeared to exert inhibitory effects on caspase-3 activation (Fig. 1B, C). Repeat experiments showed that R. rosea extracts were most effective in reducing AβO-induced increases in cleaved caspase-3 (Fig. 1C). R. rosea is known to contain six main biologically active constituents: salidroside, tyrosol, rosarin, rosavin, rosin, and rosiridin [25, 26]. We thus tested these compounds in the above assay and found that tyrosol most consistently exerted a protective effect against caspase-3 activation (Fig. 1D, E). Subsequent confirming experiments showed that tyrosol significantly inhibited caspse-3 activation at concentrations of 5 to 10 μM (Fig. 1F, G); it produced a lessor inhibitory effect at 20 μM and appeared ineffective at 2.5 μM. In addition, cell survival assays indicated that tyrosol itself was non-toxic, even at higher concentrations (up to 200 μM) (data not shown). Collectively, these observations identify tyrosol as a natural agent that protects neurons against AβO toxicity.

Fig.1

Identification of tyrosol as an agent that protects against AβO-induced neurotoxicity. A) Experimental design. Primary neurons at 9 days in vitro (DIV9) were treated with 2.5 μM AβOs or AβOs plus plant extracts or a plant constituent for 3 days. Cells were then harvested for western blot analysis. B) Neurons were treated with AβOs (O) only or AβOs plus plant extracts (10 μg/ml), followed by western blotting with anti-cleaved caspase-3 antibody. R. rosea (E1) and G. biloba (E2) were dissolved in 0.1 M Tris-HCl buffer (pH 9.0) and DMSO, respectively, at 5 mg/ml and then diluted with culture medium. C) Quantitative analysis of cleaved caspase-3 levels in neurons treated as described in B. Data represent means±SEM from three separate experiments. One-way ANOVA followed by a Tukey’s test was used for significance. D) Neurons were treated with AβOs only or AβOs plus one of the six main constituents of R. rosea (5 μM), and analyzed as in B. E) Quantitative analysis of cleaved caspase-3 levels in D. Data represent means±SEM from three separate experiments. T, tyrosol; S, salidroside; R1, rosarian; R2, rosavin; R3, rosin; R4, rosiridin. *p < 0.05, compared with control cells; ^#p < 0.05, compared with AβO-treated cells. One-way ANOVA followed by a Tukey’s test was used for significance. F) Neurons were treated with AβOs only or AβOs plus the indicated concentration (μM) of tyrosol (T), and analyzed as in B. G) Quantitative analysis of cleaved caspase-3 levels in F. Data represent means±SEM from three separate experiments. *p < 0.05, compared with control cells; ^#p < 0.05, compared with AβO-treated cells. One-way ANOVA followed by a Tukey’s test was used for significance.

We further tested whether tyrosol affects Aβ aggregation using ThT assays. Unlike the positive control EGCG, which significantly inhibited Aβ aggregation [27], tyrosol had no significant effect on Aβ aggregation over 72 h (Fig. 2). Moreover, EGCG, but not tyrosol, exerted a disaggregating effect on preformed Aβ aggregates (data not shown). In addition, tyrosol did not influence secretion of Aβ₄₀ and Aβ₄₂ from primary neurons at 5–10 μM (data not shown).

Fig.2

Effect of tyrosol on Aβ₄₂ aggregation. Reaction mixtures containing 50 μM Aβ₄₂, 50 mM phosphate buffer (pH 7.4), 100 mM NaCl, different concentrations of tyrosol [0 (□), 10 (◊), 25 (▵), 50 (*), or 100 μM (○)], or 50 μM EGCG (×) were incubated at 37°C for the indicated times, followed by ThT assay, as described in Materials and Methods. Data represent means±SEM from three separate experiments.

Chronic oral administration of tyrosol is safe in5XFAD mice

To determine whether tyrosol has anti-AD effects in vivo, we used 5XFAD mice, which display relatively rapid disease progression [17, 28], as an animal model of AD. Because tyrosol is water soluble, it was administered to 5XFAD mice continuously via drinking water (Tg-Tyr group). The amount of tyrosol delivered was estimated to be ∼12.5 mg/kg/day. 5XFAD mice were administered only water as controls (Tg-Veh group). Non-transgenic (non-Tg) wild-type mice were similarly treated with only water or tyrosol-containing water (Non-Veh and Non-Tyr groups).

As depicted in Fig. 3A, two series of experiments were performed. In one, tyrosol was administered for 12 weeks (from 4 to 7 months of age); in the other, it was administered for 20 weeks (from 2 to 7 months of age). These experimental schedules were chosen because 2 and 4 months are ages before and after the appearance of Aβ plaques, respectively, and abundant Aβ plaques have formed by 7 months of age. At 7 months of age, we assessed cognitive function using the Barnes maze test, which evaluates spatial memory in mice [28]. Immediately after this test, mice were sacrificed for pathological and biochemical analyses. 5XFAD and non-transgenic mice were weighed monthly during the treatment period, and in both cases, body weights were found to be unaffected by treatment with tyrosol (Fig. 3B). At the time of sacrificing, blood was collected and levels of albumin, blood urea nitrogen, and alanine aminotransferase in serum were measured. The values of each of these parameters were found to be within normal ranges in both treated and untreated 5XFAD and non-Tg mice (data not shown).

Fig.3

Schematic representation of the treatment paradigm. A) 5XFAD (Tg) and Non-Tg (Non) mice were treated orally with tyrosol or vehicle for 12 weeks (from 4 to 7 months old) or 20 weeks (from 2 to 7 months old) as described in Materials and Methods. Four groups of mice were defined, as indicated in the figure. Tyr, mice treated with tyrosol; Veh, mice treated with vehicle. Following completion of oral treatment, cognitive function was evaluated using behavioral studies, after which brain samples were collected for immunohistochemistry (IHC) and biochemical analyses. B) Changes in body weight during treatment, measured monthly. Data represent mean body weights in each group±SEM.

Tyrosol does not affect Aβ accumulation in AD mice

We examined whether tyrosol administration affected Aβ accumulation in 5XFAD mice using both immunohistochemical and biochemical analyses. Immunohistochemical staining with anti-Aβ antibodies showed that Aβ plaques in the hippocampus and cortex were similarly abundant in both Tg-Veh and Tg-Tyr groups after 12 or 20 weeks of tyrosol administration, as confirmed by quantification of Aβ burdens in these areas (Fig. 4A, B). Consistent with this, sandwich ELISA analysis revealed that the amounts of Aβ₄₀ and Aβ₄₂ in the guanidine-soluble fraction of cortical tissues were comparable between Tg-Veh and Tg-Tyr groups after chronic treatment for either 12 or 20 weeks (Fig. 4C, D). The amounts of Aβ₄₀ and Aβ₄₂ in the TBS-soluble fractions of cortical tissues were also similar between the two groups (Fig. 4E, F). In addition, western blot analysis of cortical extracts indicated that the levels of AβPP, BACE1, and PS1 were similar between Tg-Veh and Tg-Tyr groups (Supplementary Figure 1). It was further noticed that Aβ plaque burdens in 2- and 4-month-old 5XFAD mice were 0 and ∼20–25% of those in 7-month-old mice, respectively.

Fig.4

Tyrosol administration does not affect Aβ accumulation in 5XFAD mice. A) Brain sections of 5XFAD mice treated with vehicle or tyrosol for 20 weeks were immunostained with anti-Aβ antibody. Representative photomicrographs of the hippocampus and piriform cortex are shown. Scale bar, 500 μm. B) The Aβ plaque burden in the hippocampus and piriform cortex was measured as described in Materials and Methods. Note that vehicle-treated and tyrosol-treated mice exhibited comparable levels of Aβ plaque burden. Data are expressed as means±SEM (n = 6). C, D) Aβ₄₀ and Aβ₄₂ in guanidine soluble (C) and TBS-soluble (D) cortical extracts were measured by sandwich ELISA, as described in Materials and Methods. Note that amounts of Aβ₄₀ and Aβ₄₂ were similar in mice treated for 12 or 20 weeks. Data represent means±SEM (n = 6).

Tyrosol reverses synaptic abnormalities in AD mice

A major consequence of AβO neurotoxicity is synaptotoxicity, and it has been reported that synaptic functions and structures are disturbed in AD model mice [5, 6]. In our investigation of potential beneficial effects of tyrosol on synaptic abnormalities, we focused on spinophilin, a scaffold protein in dendritic synapses [29]. Immunostaining revealed spinophilin expression in the hippocampus, particularly in CA3, CA1, and dentate hilar regions. A comparison of spinophilin immunoreactivity among groups in the 12- or 20-week treatment cohort showed reduced intensity of spinophilin immunoreactivity in these hippocampal regions in vehicle-treated 5XFAD mice compared with non-Tg mice; however, tyrosol treatment of 5XFAD mice appeared to restore spinophilin immunoreactivity in these regions to intensity comparable to that in non-Tg mice (Fig. 5A). Quantitative analyses demonstrated that in the first cohort, spinophilin intensity was reduced by ∼30% in the Tg-Veh group compared with the Non-Veh group, but was comparable between Tg-Tyr and Non-Veh groups; the difference between Tg-Veh and Tg-Tyr groups was statistically significant in the dentate region (Fig. 5B). Very similar results with statistical significance between Non-Veh and Tg-Veh and between Tg-Veh and Tg-Tyr groups were obtained in analyses of mice in the second cohort (Fig. 5C). No differences in spinophilin immunoreactive intensity in the hippocampal regions were observed between Non-Veh and Non-Tyr groups (Supplementary Figure 2). An additional examination of total levels of spinophilin by western blotting of hippocampal tissue extracts revealed no significant differences among the four groups of mice (data not shown).

Fig.5

Tyrosol administration prevents the reduction in spinophilin immunoreactivity in the hippocampus of 5XFAD mice. A) Brain sections of Non-Tg and 5XFAD mice treated with vehicle and 5XFAD mice treated with tyrosol for 20 weeks were stained with anti-spinophilin antibody. ROIs used for quantitative analysis of spinophilin immunofluorescence intensity are depicted by polygons in the upper photograph. Scale bar, 500 μm. Representative images of spinophilin immunofluorescence staining in the hippocampal DG, CA3, and CA1 regions are shown. Non-Veh, vehicle-treated Non-Tg mice; Tg-Veh, vehicle-treated 5XFAD mice; Tg-Tyr, tyrosol-treated 5XFAD mice. Scale bar, 100 μm. B) Quantification of relative fluorescence intensity of spinophilin in DG, CA3, and CA1 ROIs in Non-Veh, Tg-Veh, and Tg-Tyr groups treated for 12 or 20 weeks. Values are means±SEM (n = 5). *p < 0.05, **p < 0.01, compared with Non-Veh mice; ^#p < 0.05, ^##p < 0.01, compared with Tg-Veh mice. One-way ANOVA followed by a Bonferroni’s test was used for significance.

Tyrosol reduces oxidative stress responses in in vitro and in vivo models

AβOs are known to induce oxidative stress, and tyrosol possesses antioxidant properties [30, 31]. To gain insight into the mechanism underlying the protective effect of tyrosol on synaptic disturbances, we analyzed the effect of tyrosol administration on 4-HNE, a well-known marker of oxidative stress responses [32, 33], by immunostaining with an anti-4-HNE antibody. Relatively intense punctate immunoreactive signals were detected in neurons in the hippocampal CA3 region. Double staining with anti-4-HNE antibody and DAPI revealed the presence of 4-HNE immunoreactive signals in the neuronal perikarya (Fig. 6A). Interestingly, the intensity of 4-HNE immunoreactivity appeared stronger in vehicle-treated 5XFAD mice (Tg-Veh) than non-Tg mice (Non-Veh), and the signal intensity in tyrosol-treated 5XFAD mice was comparable to that in controls (Fig. 6A). A quantitative analysis showed that the intensity of 4-HNE immunoreactivity in the CA3 region was significantly increased by ∼15–20% in Tg-Veh mice compared with non-Tg mice, but was reversed to control levels in Tg-Tyr mice in both 12- and 20-week treatment cohorts (Fig. 6B). Although the extent of the difference was small, the difference may be underestimated, given that the 4-HNE immunoreactivity appeared to be restricted to the cytosol of neurons. No significant difference in 4-HNE immunoreactivity was observed between Non-Veh and Non-Tyr groups (Supplementary Figure 3).

Fig.6

Tyrosol administration rescues the increase in 4-HNE immunoreactivity in the hippocampal CA3 region of 5XFAD mice. A) Brain sections of Non-Tg and 5XFAD mice treated with vehicle and 5XFAD mice treated with tyrosol for 20 weeks were stained with anti-4-HNE antibody. ROIs used for quantitative analysis of 4-HNE immunofluorescence intensity are depicted by polygons in the upper left photograph. Scale bar, 100 μm. Representative images show immunohistochemical staining for 4-HNE in the hippocampal CA3 region. The upper middle image is a magnification of the boxed area. The magnified image (upper right) indicates that 4-HNE immunoreactive signals are present in the neuronal perikarya. Non-Veh, vehicle-treated Non-Tg mice; Tg-Veh, vehicle-treated 5XFAD mice; Tg-Tyr, tyrosol-treated 5XFAD mice. Scale bar, 50 μm. B) Quantification of relative fluorescence intensity of 4-HNE in the ROI in the CA3 pyramidal cell layer of Non-Veh, Tg-Veh, and Tg-Tyr groups treated for 12 or 20 weeks. Values are means±SEM (n = 5). *p < 0.05, **p < 0.01, compared with Non-Veh mice; ^#p < 0.05, ^##p < 0.01, compared with Tg-Veh mice. One-way ANOVA followed by a Bonferroni’s test was used for significance.

We further examined whether tyrosol is capable of preventing AβO-induced oxidative stress in our neuron model. To this end, we performed an immunocytochemical analysis using an antibody to HEL, a marker of oxidative damage by lipid peroxidation [34]. HEL immunoreactivity was detected in the cell soma and proximal part of neurites and was more intense in AβO-treated neurons than control neurons, reflecting induction of oxidative stress by AβOs. In contrast, HEL immunoreactive intensity in neurons co-treated with AβOs and tyrosol was similar to that in controls, suggesting an antioxidative effect of tyrosol (Supplementary Figure 4).

Tyrosol modestly mitigates cognitive impairment in 5XFAD mice

We performed the Barnes maze test, which evaluates spatial memory in mice [35] to assess whether tyrosol can improve cognitive function in 5XFAD mice. In this test, escape latency was monitored during the training period for 5 days. In the first cohort, treated for 12 weeks, escape latencies were significantly longer in both Tg-Veh and Tg-Tyr groups at days 2 and 3 compared with the Non-Veh group, but were not statistically different at days 4 and 5 (Fig. 7A), despite a trend toward longer latencies in the Tg-Veh group compared with the Tg-Tyr group at day 5. There were no significant differences in escape latency between Non-Veh and Non-Tyr groups. Two-way repeated-measures ANOVA (tyrosol X training day) revealed that an interaction effect between tyrosol and training day (F_4,72 = 2.655, p = 0.0398) and a main effect of training day (F_1,72 = 3.724, p = 0.0001) were significant, although a main effect of tyrosol (F_1,72 = 0.01159, p = 0.9155) was not significant (Fig. 7A). The escape latencies in Tg-Tyr group were significantly shorter at days 4 and 5 compared with at day 1, while there were no significant differences in escape latency over the entire period in Tg-Veh group (Fig. 7A). In the second cohort, treated for 20 weeks, the Tg-Veh group, but not the Tg-Tyr group, displayed significantly longer escape latencies than the Non-Veh group at days 4 and 5 (Fig. 7B). Two-way repeated-measures ANOVA (tyrosol X training day) revealed that a main effect of tyrosol (F_1,76 = 6.964, p = 0.0162) was significant, although a main effect of training day (F_4,76 = 1.241, p = 0.3009) and interaction effect between tyrosol and training day (F_4,76 = 0.1502, p = 0.9624) were not significant (Fig. 7B). In the probe trial of the first cohort, mice in the Tg-Tyr group spent significantly more time around the holes in the target quadrant than in the nontarget quadrants, but those in the Tg-Veh group did not (Supplementary Figure 5A). Though statistically not significant, similar results were obtained in the second cohort (Supplementary Figure 5B). These results suggest that the impairment in spatial memory in 5XFAD mice was modestly mitigated by chronic oral intake of tyrosol.

Fig.7

Chronic tyrosol treatment modestly ameliorates impaired spatial memory in 5XFAD mice. Performance in the Barnes maze test. Four groups of mice treated for 12 weeks (A) or 20 weeks (B) were subjected to the Barnes maze test, performed as described in Materials and Methods. Mean latencies to enter the target hole during acquisition training are shown. Numbers of animals used in (A) and (B) are as follows: Non-V, 11 and 10; Non-T, 10 and 7; Tg-V, 10 and 12; Tg-T, 10 and 9. Data represent means±SEM. *p < 0.05, **p < 0.01, compared with Non-Veh mice. Two-way ANOVA followed by a Tukey’s test was used for significance. ^#p < 0.05, compared with Tg-V mice. Two-way repeated-measures ANOVA (tyrosol X training day) followed by followed by a Tukey’s test.

DISCUSSION

AβOs are considered an important therapeutic target in AD. Major strategies for targeting AβOs include removing them through immunodepletion by specific antibodies or blocking their formation using agents that inhibit Aβ production or aggregation. However, it is also worthwhile to target AβO neurotoxicity itself through developing agents that efficiently prevent it. On the basis of this concept, we searched for natural compounds that can protect neurons from AβO-associated neurotoxicity. In screening experiments using our primary neuron model, we serendipitously found that extracts of R. rosea, which are known to have various physiological effects in humans, including anti-fatigue [25], exert a protective effect against AβOs. We subsequently found that, of the main constituents of R. rosea, tyrosol exhibited the most consistent protective effect, although other constituents, such as salidroside and rosin, may also have weaker effects. Tyrosol is a small molecule compound, present not only in R. rosea but also in olives and white wine, that has been reported to have antioxidant properties [32, 33].

To assess the potential of tyrosol as a drug candidate for AD, we investigated its efficacy against cognitive deficits and pathological abnormalities using 5XFAD mice. In our in vivo experiments, we used two different administration schedules—from 4 to 7 months old in the first, and from 2 to 7 months old in the second—on the premise that it is important to investigate the effects of tyrosol intake both before and after the appearance of Aβ plaques. In general, the former treatment is thought to be suitable for assessing a preventive effect on disease progression and the latter for assessing a prophylactic effect.

We examined whether tyrosol can prevent synaptic abnormalities in 5XFAD mice, as synaptotoxicity is a major effect of Aβ oligomers. Importantly, tyrosol administration significantly reversed the reduction in spinophilin intensity in hippocampal sub-regions of 5XFAD mice in both treatment paradigms. Spinophilin is a scaffold protein that is principally located in the post-synaptic compartment of dendritic synapses and plays important roles in regulating the formation and function of dendritic spines [29, 36]. Spinophilin expression is reported to be downregulated in the brains of 5XFAD mice [37] and AD patients [38]. Given that tyrosol protects against AβO neurotoxicity, it is conceivable that protection of synapses from AβOs in vivo by tyrosol administration prevents synaptic disturbances, thereby accounting for the amelioration of cognitive impairment.

Oxidative stress is one of the pivotal pathological features of AD pathology [7]. Consistent with this, AβOs induce oxidative stress responses in neurons, an effect that likely underlies its neurotoxicity and synaptotoxicity [4–6 , 8]. Our observation that the increase in HEL immunoreactivity in AβO-treated neurons was prevented by tyrosol co-treatment suggests a protective effect of tyrosol against AβO-induced oxidative stress. Furthermore, tyrosol treatment prevented the enhancement of 4-HNE in the hippocampal CA3 region in 5XFAD mice, supporting the view that the neuroprotective effect of tyrosol in vivo reflects its prevention of oxidative stress responses. This protective effect may be attributable to the antioxidant potential of tyrosol; however, other undefined mechanisms might also operate as well. The finding that 4-HNE enhancement was most evident in the CA3 region may be attributable to the selective vulnerability of this region to certain stress stimuli [39, 40]. In fact, it has been reported that 4-HNE immunoreactivity is induced in the CA3 pyramidal layer after experimental brain injury in the rat [41].

The results of Barnes maze tests indicate that either 12 or 20 weeks tyrosol treatment mitigated the cognitive impairment of 5XFAD mice, although the effects appeared to be modest. This ameliorative effect appears to reflect the protective effects of tyrosol against synaptic and oxidative abnormalities in vivo. However, it is a limitation of the present study that we used only one behavioral test to assess cognitive deficits. We found that escape latencies in 5XFAD mice were not abnormal at 4 months of age (data not shown), in keeping with the idea that these mice develop cognitive dysfunction at about 6 months of age [27 , 43], when Aβ plaques develop abundantly. Further research is required to investigate whether administration of greater amounts of tyrosol can ameliorate cognitive function more clearly.

Our immunohistochemical and ELISA analyses showed that Aβ accumulation was unchanged in 5XFAD mice after tyrosol treatment for 12 or 20 weeks, as shown in Fig. 4. This finding is consistent with the fact that tyrosol has little effect on Aβ aggregation as well as Aβ production. It remains to be determined whether levels of AβOs are altered in the mice after chronic treatment with tyrosol; however, our data indicate that at least levels of TBS-soluble Aβ were unaffected by tyrosol treatment. In addition, tyrosol administration had no effect on the expression of AβPP, BACE1, or PS1. It is thus plausible that tyrosol exerts beneficial effects in vivo without influencing Aβ accumulation.

A few studies have described beneficial effects of tyrosol in animal models of brain diseases. For example, intraperitoneal administration of tyrosol reduced infarct volume in a rat model of transient focal cerebral ischemia [44], and intravenous administration of tyrosol showed clear neuroprotective effects in a rat model of global cerebral ischemia [45]. Oral administration of tyrosol and hydroxyl-tyrosol exerted antioxidant effects in in vivo models of rat brain hypoxia-reoxygenation [18]. These compounds were reported to protect neuronal cells against Aβ-induced toxicity and other insults [46 –49]. Furthermore, hydroxy-tyrosol was shown to mildly improve cognitive impairment in APP/PS1 mice without affecting Aβ pathology [50]. The authors of this latter study interpreted this improvement in cognition as being mediated by amelioration of oxidative stress and brain inflammation. Interestingly, a moderate concentration of tyrosol has also been shown to promote stress resistance and increase lifespan in Caenorhabditis elegans [51].

Tyrosol is a very small molecular weight compound (MW = 138) and is considered capable of crossing the blood-brain barrier [45]. A previous study reported that tyrosol sulfate, a tyrosol metabolite, is detectable in the brains of rats after oral administration of a phenolic extract of olive cake [52]. Intriguingly, recent work suggests that the antioxidant properties of tyrosol sulfate are comparable to those of tyrosol [53, 54], implying that tyrosol metabolites also retain beneficial effects in the brain. In our preliminary analysis using a liquid chromatography and mass spectroscopy, both tyrosol and tyrosol sulphate seemed to be more abundantly present in brain tissues of tyrosol-treated C57BL/6 mice relative to vehicle-treated mice (data not shown). Further research is required to quantitatively determine the extent to which tyrosol and its metabolites accumulate in the mouse brain following oral intake.

Another remaining question is whether tyrosol has a protective effect on tau abnormalities driven by AβOs. Our preliminary immunocytochemcial analysis with phospho-specific anti-tau antibodies (AT8) suggested that AT8 immunoreactivity is enhanced in primary neurons treated with AβOs, but not in those co-treated with AβOs and tyrosol, compared with controls (data not shown). Future research is required to clarify whether tyrosol can prevent tau hyperphosphorylation in in vitro and in vivo models of AD.

Notably, tyrosol has a high safety profile, even upon chronic administration, as evidenced by the results of our mouse experiments. In addition to its overall safety, this agent has several properties that make it advantageous for therapeutic use. First, tyrosol is water-soluble and can thus be easily administered orally. Second, tyrosol has a mechanism of action distinct from that of existing drug candidates that target the accumulation of Aβ [9 –11], as depicted in Fig. 8; thus, in principle, it can be used in combination with other drugs, potentially providing superior therapeutic efficacy. Third, tyrosol has a very simple chemical structure, making its pharmaceutical formulation easy. Fourth, tyrosol is considered to be a brain-penetrant molecule, as discussed above. Considering its safety and practical properties, tyrosol is potentially a promising and unique candidate, not only as a disease-modifying drug for treating patients with AD, especially at earlier stages, but also as a prophylactic for elderly individuals to prevent the occurrence of AD. Future clinical trials are warranted to evaluate the clinical efficacy of tyrosol for disease treatment and prevention.

Fig.8

Schematic illustration of the distinct mechanism of action of tyrosol. In contrast to current therapeutic strategies that target Aβ accumulation, including β-secretase inhibition, Aβ aggregation inhibition, and Aβ immunotherapy, tyrosol reduces AβO neurotoxicity and exerts a protective action on neurons, preventing oxidative, synaptic, and other disturbances. In principle, tyrosol can be used in combination with other drugs currently under development.

In conclusion, we identified tyrosol, one of the main constituents of the plant R. rosea, as an agent that protects neurons from the neurotoxicity of AβOs. Chronic oral administration of tyrosol in AD model (5XFAD) mice significantly reversed synaptic and oxidative abnormalities and modestly mitigated cognitive impairment without affecting Aβ accumulation. These results collectively suggest that the natural agent tyrosol is a safe, effective, and unique drug candidate for AD that has a mechanism of action distinct from those of current drugs that target Aβ accumulation.

Footnotes

ACKNOWLEDGMENTS

This work was supported by JSPS KAKENHI Grant Number 16K09688 (W.A.), 16K09664 (A.T.), and an Intramural Research Grant (27-9, 30-3) for Neurological and Psychiatric Disorders from the National Center of Neurology and Psychiatry.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Ballard

, Gauthier

, Corbett

, Brayne

, Aarsland

, Jones

(2011) Alzheimer’s disease. Lancet 377, 1019–1031.

Hardy

, Selkoe

(2002) The amyloid hypothesis of Alzheimer’s disease: Progress and problems on the road to therapeutics. Science 297, 353–356.

Ferreira

, Klein

(2011) The Aβ oligomer hypothesis for synapse failure and memory loss in Alzheimer’s disease. Neurobiol Learn Mem 96, 529–543.

Viola

, Klein

(2015) Amyloid β oligomers in Alzheimer’s disease pathogenesis, treatment, and diagnosis. Acta Neuropathol 129, 183–206.

Pozueta

, Lefort

, Shelanski

(2013) Synaptic changes in Alzheimer’s disease and its models. Neuroscience 251, 51–65.

Cochran

, Hall

, Roberson

(2014) The dendritic hypothesis for Alzheimer’s disease pathophysiology. Brain Res Bull 103, 18–28.

Butterfield

, Drake

, Pocernich

, Castegna

(2001) Evidence of oxidative damage in Alzheimer’s disease brain: Central role for amyloid beta-eptide. Trends Mol Med 7, 548–55.

Kamat

, Kalani

, Rai

, Swarnkar

, Tota

, Nath

, Tyagi

(2014) Mechanism of oxidative stress and dynapse dysfunction in the pathogenesis of Alzheimer’s disease: Understanding the therapeutics strategies. Mol Neurobiol 50, 852–865.

Selkoe

(2013) The therapeutics of Alzheimer’s disease: Where we stand and where we are heading. Ann Neurol 74, 328–336.

10.

Goure

, Krafft

, Jerecic

, Hefti

(2014) Targeting the proper amyloid-beta neuronal toxins: A path forward for Alzheimer’s disease immunotherapeutics. Alzheimers Res Ther 6, 42.

11.

Yan

, Vassar

(2014) Targeting the β secretase BACE1 for Alzheimer’s disease therapy. Lancet Neurol 13, 319–329.

12.

Brewer

, Torricelli

, Evege

, Price

(1993) Optimized survival of hippocampal neurons in B27-supplemented Neurobasal, a new serum-free medium combination. J Neurosci Res 35, 567–576.

13.

Mamada

, Tanokashira

, Hosaka

, Kametani

, Tamaoka

, Araki

(2015) Amyloid β-protein oligomers upregulate the β-secretase, BACE1, through a post-translational mechanism involving its altered subcellular distribution in neurons. Mol Brain 8, 73.

14.

Kametani

, Tanaka

, Ishii

, Ikeda

, Kennedy

, Allsop

(1993) Secretory form of Alzheimer amyloid precursor protein 695 in human brain lacks beta/A4 amyloid immunoreactivity. Biochem Biophys Res Commun 191, 392–398.

15.

Stine

, Jungbauer

, Yu

, LaDu

(2011) Preparing synthetic Aβ in different aggregation states. Methods Mol Biol 670, 13–32.

16.

Tanokashira

, Mamada

, Yamamoto

, Taniguchi

, Tamaoka

, Lakshmana

, Araki

(2017) The neurotoxicity of amyloid β-protein oligomers is reversible in a primary neuron model. Mol Brain 10, 4.

17.

Oakley

, Cole

, Logan

, Maus

, Shao

, Craft

, Guillozet-Bongaarts

, Ohno

, Disterhoft

, Van Eldik

, Berry

, Vassar

(2006) Intraneuronal beta-amyloid aggregates, neurodegeneration, and neuron loss in transgenic mice with five familial Alzheimer’s disease mutations: Potential factors in amyloid plaque formation. J Neurosci 26, 10129–10140.

18.

Chandramohan

, Pari

, Rathinam

, Sheikh

(2015) Tyrosol, a phenolic compound, ameliorates hyperglycemia by regulating key enzymes of carbohydrate metabolism in streptozotocin induced diabetic rats. Chem Biol Interact 229, 44–54.

19.

De La Cruz

, Ruiz-Moreno

, Guerrero

, Reyes

, Benitez-Guerrero

, Espartero

, González-Correa

(2015) Differences in the neuroprotective effect of orally administered virgin olive oil (Olea europaea) polyphenols Tyrosol and Hydroxytyrosol in rats. J Agric Food Chem 62, 5957–5963.

20.

Paxinos

, Franklin

KBJ

(2001) The mouse brain in stereotaxic coordinates, 2nd edition. Oxford: Elsevier, Amsterdam.

21.

Araki

, Oda

, Motoki

, Hattori

, Itoh

, Yuasa

, Konishi

, Shin

, Tamaoka

, Ogino

(2013) Reduction of β-amyloid accumulation by reticulon 3 in transgenic mice. Curr Alzheimer Res 10, 135–142.

22.

Kim

, Onstead

, Randle

, Price

, Smithson

, Zwizinski

, Dickson

, Golde

, McGowan

(2007) Abeta40 inhibits amyloid deposition in vivo. J Neurosci 27, 627–633.

23.

Tanokashira

, Motoki

, Minegishi

, Hosaka

, Mamada

, Tamaoka

, Okada

, Lakshmana

, Araki

(2015) LRP1 downregulates the Alzheimer’s β-secretase BACE1 by modulating its intraneuronal trafficking. eNeuro 2, ENEURO.0006-15.2015.

24.

Tozuka

, Kumon

, Wada

, Onodera

, Mochizuki

, Wada

(2010) Maternal obesity impairs hippocampal BDNF production and spatial learning performance in young mouse offspring. Neurochem Int 57, 235–247.

25.

Panossian

, Wikman

, Sarris

(2010) Rosenroot (Rhodiola rosea): Traditional use, chemical composition, pharmacology and clinical efficacy. Phytomedicine 17, 481–493.

26.

, Wang

, Hou

, Ma

, Luo

, Lu

, Jin

, Chen

, Xu

, Patel

, Gorecki

(2011) Simultaneous quantification of polyherbal formulations containing Rhodiola rosea L. and Eleutherococcus senticosus Maxim. using rapid resolution liquid chromatography (RRLC). J Pharm Biomed Anal 55, 908–915.

27.

Bieschke

, Russ

, Friedrich

, Ehrnhoefer

, Wobst

, Neugebauer

, Wanker

(2010) EGCG remodels mature alpha-synuclein and amyloid-beta fibrils and reduces cellular toxicity. Proc Natl Acad Sci U S A 107, 7710–7715.

28.

Webster

, Bachstetter

, Nelson

, Schmitt

, Van Eldik

(2014) Using mice to model Alzheimer’s dementia: An overview of the clinical disease and the preclinical behavioral changes in 10 mouse models. Front Genet 5, 88.

29.

Sarrouilhe

, di Tommaso

, Métayé

, Ladeveze

(2006) Spinophilin: Fromartners to functions. Biochimie 88, 1099–1113.

30.

de la Puerta

, Ruiz Gutierrez

, Hoult

(1999) Inhibition of leukocyte 5-lipoxygenase by phenolics from virgin olive oil. Biochem Pharmacol 57, 445–449.

31.

Giovannini

, Straface

, Modesti

, Coni

, Cantafora

, De Vincenzi

, Malorni

, Masella

(1999) Tyrosol, the major olive oil biophenol, protects against oxidized-LDL-induced injury in Caco-2 cells. J Nutr 129, 1269–1277.

32.

Di Domenico

, Tramutola

, Butterfield

(2017) Role of 4-hydroxy-2-nonenal (HNE) in the pathogenesis of Alzheimer disease and other selected age-related neurodegenerative disorders. Free Radic Biol Med 111, 253–261.

33.

Dalleau

, Baradat

, Guéraud

, Huc

(2013) Cell death and diseases related to oxidative stress: 4-hydroxynonenal (HNE) in the balance. Cell Death Differ 20, 1615–1630.

34.

Kato

, Miyake

, Yamamoto

, Shimomura

, Ochi

, Mori

, Osawa

(2000) Preparation of a monoclonal antibody to N(epsilon)-(Hexanonyl)lysine: Application to the evaluation of protective effects of flavonoid supplementation against exercise-induced oxidative stress in rat skeletal muscle. Biochem Biophys Res Commun 274, 389–393.

35.

Sharma

, Rakoczy

, Brown-Borg

(2010) Assessment of spatial memory in mice. Life Sci 87, 521–536.

36.

Feng

, Yan

, Ferreira

, Tomizawa

, Liauw

, Zhuo

, Allen

, Ouimet

, Greengard

(2000) Spinophilin regulates the formation and function of dendritic spines. Proc Natl Acad Sci U S A 97, 9287–9292.

37.

Hongpaisan

, Sun

, Alkon

(2011) PKC ɛ activation prevents synaptic loss, Aβ elevation, and cognitive deficits in Alzheimer’s disease transgenic mice. J Neurosci 31, 630–643.

38.

Akram

, Christoffel

, Rocher

, Bouras

, Kövari

, Perl

, Morrison

, Herrmann

, Haroutunian

, Giannakopoulos

, Hof

(2008) Stereologic estimates of total spinophilin-immunoreactive spine number in area 9 and the CA1 field: Relationship with the progression of Alzheimer’s disease. Neurobiol Aging 29, 1296–1307.

39.

Stein-Behrens

, Mattson

, Chang

, Yeh

, Sapolsky

(1994) Stress exacerbates neuron loss and cytoskeletal pathology in the hippocampus. J Neurosci 14, 5373–5380.

40.

Wilde

, Pringle

, Wright

, Iannotti

(1997) Differential vulnerability of the CA1 and CA3 subfields of the hippocampus to superoxide and hydroxyl radicals in vitro. J Neurochem 69, 883–886.

41.

Zhang

, Dhillon

, Mattson

, Yurek

, Prasad

(1999) Immunohistochemical detection of the lipid peroxidation product 4-hydroxynonenal after experimental brain injury in the rat. Neurosci Lett 272, 57–61.

42.

Ohno

, Chang

, Tseng

, Oakley

, Citron

, Klein

, Vassar

, Disterhoft

(2006) Temporal memory deficits in Alzheimer’s mouse models: Rescue by genetic deletion of BACE1. Eur J Neurosci 23, 251–260.

43.

Devi

, Ohno

(2010) Genetic reductions of beta-site amyloid precursor protein-cleaving enzyme 1 and amyloid-beta ameliorate impairment of conditioned taste aversion memory in 5XFAD Alzheimer’s disease model mice. Eur J Neurosci 31, 110–118.

44.

, Rho

, Kim

, Lee

, Kim

, Choi

, Kim

(2007) Neuroprotective effect of tyrosol on transient focal cerebral ischemia in rats. Neurosci Lett 414, 218–221.

45.

Atochin

, Chernysheva

, Smolyakova

, Osipenko

, Logvinov

, Zhdankina

, Sysolyatin

, Kryukov

, Anfinogenova

, Plotnikova

, Plotnikov

(2016) Neuroprotective effects of p-tyrosol after the global cerebral ischemia in rats. Phytomedicine 23, 784–792.

46.

, Kim

, Sung

(2009) Compounds with neuroprotective activity from the medicinal plant Machilus thunbergii. J Enzyme Inhib Med Chem 24, 1117–1121.

47.

Vauzour

, Corona

, Spencer

(2010) Caffeic acid, tyrosol and p-coumaric acid are potent inhibitors of 5-S-cysteinyl-dopamine induced neurotoxicity. Arch Biochem Biophys 501, 106–111.

48.

St-Laurent-Thibault

, Arseneault

, Longpré

, Ramassamy

(2011) Tyrosol and hydroxytyrosol, two main components of olive oil, protect N2a cells against amyloid-β-induced toxicity. Involvement of the NF-κB signaling. Curr Alzheimer Res 8, 543–551.

49.

Dewapriya

, Himaya

, Li

, Kim

(2013) Tyrosol exerts a protective effect against dopaminergic neuronal cell death in in vitro model of Parkinson’s disease. Food Chem 141, 1147–1157.

50.

Peng

, Hou

, Yang

, Li

, Jia

, Liu

, Tang

, Shi

, Li

, Long

, Liu

(2016) Hydroxytyrosol mildly improve cognitive function independent of APP processing in APP/PS1 mice. Mol Nutr Food Res 60, 2331–2342.

51.

Cañuelo

, Gilbert-López

, Pacheco-Liñán

, Martínez-Lara

, Siles

, Miranda-Vizuete

(2012) Tyrosol, a main phenol present in extra virgin olive oil, increases lifespan and stress resistance in Caenorhabditis elegans. Mech Ageing Dev 133, 563–574.

52.

Serra

, Rubió

, Borràs

, Macià

, Romero

, Motilva

(2012) Distribution of olive oil phenolic compounds in rat tissues after administration of a phenolic extract from olive cake. Mol Nutr Food Res 56, 486–496.

53.

Atzeri

, Lucas

, Incani

, Peñalver

, Zafra-Gómez

, Melis

, Pizzala

, Morales

, Deiana

(2016) Hydroxytyrosol and tyrosol sulfate metabolites protect against the oxidized cholesterol pro-oxidant effect in Caco-2 human enterocyte-like cells. Food Funct 7, 337–346.

54.

Muriana

FJG

, Montserrat-de la Paz

, Lucas

, Bermudez

, Jaramillo

, Morales

, Abia

, Lopez

(2017) Tyrosol and its metabolites as antioxidative and anti-inflammatory molecules in human endothelial cells. Food Funct 8, 2905–2914.