Diet Supplementation with Hydroxytyrosol Ameliorates Brain Pathology and Restores Cognitive Functions in a Mouse Model of Amyloid-β Deposition

Abstract

Alzheimer’s disease is the most common form of dementia affecting a large proportion of aged people. Plant polyphenols have been reported to be potentially useful in the prevention of AD due to their multiple pharmacological activities. The aim of the present study was to assess whether the previously reported neuroprotective and anti-inflammatory effects resulting from oleuropein aglycone administration were reproduced by diet supplementation with similar amounts of its metabolite hydoxytyrosol (HT). Four-month-old TgCRND8 and wild type mice were treated for 8 weeks with a low-fat diet (5%) supplemented with HT (50 mg/kg of diet). We found that HT supplementation significantly improved cognitive functions of TgCRND8 mice and significantly reduced Aβ₄₂ and pE3-Aβ plaque area and number in the cortex; in the hippocampal areas of HT-fed TgCRND8 mice, we found a significant reduction in the pE3-Aβ plaque number together with a tendency toward a reduction in Aβ₄₂ load and pE3-Aβ plaque area, associated with a marked reduction of TNF-α expression and astrocyte reaction. Macroautophagy induction and modulation of MAPKs signaling were found to underlie the beneficial effects of HT. Our findings indicate that HT administration reproduces substantially the beneficial effects on behavioral performance and neuropathology previously reported in TgCRND8 mice fed with oleuropein aglycone, resulting in comparable neuroprotection.

Keywords

Amyloid plaques autophagy hydoxytyrosol MAPKs signaling memory function

INTRODUCTION

Alzheimer’s disease (AD), the most common form of dementia affecting a large proportion of aged people, presently still lacks effective therapy; accordingly, several therapeutic approaches are currently under investigation. Among these approaches, the use of a panel of plant molecules, notably polyphenols, enriched in the Mediterranean diet (MD) is gaining increasing interest. Plant polyphenols combat amyloid degeneration in several ways that go beyond their well-known antioxidant properties; for example, they interfere with self-assembly of amyloidogenic proteins/peptides (amyloid-β (Aβ) peptides and tau protein in case of AD) into amyloid oligomers and fibrils, presently considered a key feature of AD pathogenesis (reviewed in [1]).

At the present, many research data and population studies suggest that a safe lifestyle, including a close adherence to the MD enhances cognitive functions, reduces the risk of developing mild cognitive impairment and of its conversion to AD [2, 3]. The outcomes of these studies have been better related to the presence in the diet, as the main lipid, of extra virgin olive oil (EVOO), particularly of its polyphenols [4, 5]. Previous clinical studies indicated that a MD supplemented with olive oil or nuts is associated both with a significant improvement of cognition in a population of aged people [5] and with HDL atheroprotective functions in humans [6]. Overall, these studies support the idea that the MD, with particular emphasis to the intake of polyphenols contained in some of its characteristic foods, provides consistent and effective protection against the risk of aging-associated neurodegeneration (reviewed in [7, 8]). In particular, the daily consumption of EVOO, with its significant content of polyphenols, has been shown to induce beneficial effects, among others, against Aβ/tau pathologies [9, 10] and Parkinson disease. The polyphenols enriched in EVOO include hydroxytyrosol (HT), tyrosol, oleocanthal, and oleuropein, the latter both in the glycated and in the aglycone (OLE) forms. Oleuropein is the major phenolic compound in olive drupes and leaves, while HT, originating from oleuropein hydrolysis during fruit ripening and EVOO storage, is the major polyphenol in olive oil, both in the free form and in the oleuropein molecule.

In addition to their antioxidant power, many beneficial properties of EVOO polyphenols as key regulators of cell/tissue metabolism have been associated to the modulation of signaling pathways, including those focused around mammalian target of rapamycin (mTOR) and adenosine monophosphate-dependent protein kinase (AMPK), two key regulators of the metabolic energy balance involved in longevity and inflammatory states (reviewed in [12]). EVOO polyphenols also appear to modulate epigenetic activities in their target tissues [13]. Finally, plant, including EVOO, polyphenols were reported to interfere with protein/peptide aggregation in several ways, most often skipping the formation of toxic intermediates, presently considered among the main culprits of tissue/cell impairment in amyloid diseases [14]. We have recently shown that OLE triggers macroautophagy in our Tg model of mice [9] following AMPK activation and mTOR inhibition [15]. We also reported that, in the TgCRND8 mouse model, OLE affects the acetylation state of histones, relieves the inflammatory reaction, and reduces the build-up of amyloid aggregates of the Aβ₄₂ peptide and of its pyroglutamylated 3–42 derivative (pE3-Aβ), with strong improvement of memory and behavioral performance.

HT, as the primary degradation product of OLE, has also been reported to be potentially useful in the prevention of different diseases [16], although less information is presently available on the benefits of this compound against neurodegeneration. In addition to OLE, HT is gaining increasing attention due to its multiple pharmacological activities; in particular, its strong antioxidant power, its antiproliferative and apoptotic activities and its anti-inflammatory properties [17, 18]. Similarly to OLE, HT is absorbed by the human intestine after oral administration [19] and, once in the bloodstream, increases the plasma antioxidant capacity and reduces the effect of free radicals in various cellular systems.

In the present study, we extended our previous investigations on the neuroprotective and anti-inflammatory effects of EVOO polyphenols to better assess whether the previously reported effects of OLE administration were, at least in part, reproduced by administration of similar amounts of its metabolite HT [13]. To do this, we supplemented with HT for 8 weeks the normal diet of TgCRND8 mice and investigated the resulting effects on cognitive functions, brain amyloid deposits, inflammatory response, and autophagy, as well as on the signaling pathways mediated by the extracellular regulated kinase (ERK)1/2 and the stress-activated protein kinase c-jun N-terminal kinase (SAPK/JNK).

MATERIALS AND METHODS

Ethics statement

Transgenic (Tg) CRND8 mice and wild type (wt) control littermates were used following the ECC (DL 116/92, Directive 86/609/EEC) and National guidelines (PN: 71/2017-PR).

Animals and diet

Four-month-old Tg mice (n = 6/group/genotype) were used for diet treatment. As previously reported for OLE treatment [20], the animals were treated for 8 weeks with a modified low-fat AIN-76A diet (10 g/day/mouse), either as such (untreated mice) or supplemented with HT at 50 mg/kg of diet. AIN-76A was composed of 50% sucrose, 5.0% fat, 20% casein, 15% corn starch, 5.0% powdered cellulose, 3.5% AIN-76 mineral mix, 1.0% AIN-76A vitamin mix, 0.3% DL-methionine, and 0.2% choline bitartrate (Piccioni, Milan, Italy). The mice were divided into four groups: 1) Tg mice fed with low-fat diet as such (untreated Tg); 2) Tg mice fed with the same diet supplemented with HT (Tg HT); 3) wt mice fed with the same diet supplemented with HT (wt HT); and 4) wt mice fed with low-fat diet as such (untreated wt).

Behavioral experiments

At the end of the treatment, the mice (6-month-old) were tested in the Morris Water Maze (MWM), “Step Down” inhibitory avoidance, and Object Recognition Test (ORT). The MWM apparatus consisted of a circular white plastic pool (1.2 m in diameter and 0.47 m high), filled to a depth of 20 cm with water (24-25°C) made opaque by the addition of non-toxic white paint. A white plastic escape platform (10 cm in diameter) with a grooved surface for a better grip was submerged 0.5 cm under the water level. As previously reported, we applied some modifications to the method described by Janus [22]; briefly, all mice underwent a reference memory training with a hidden platform placed in the center of one quadrant of the pool for four days, with four trials/day, with the four starting locations varying in different trials. When the platform was not localized within the maximum allowed time (60 s), the mouse was guided to the location. The mouse was allowed to remain 20 s on the platform. For each trial, latency to find the platform (maximum 60 s) was recorded by a video-tracking/computer digitizing system (HVS Image, Hampton, UK).

The apparatus and procedures used for the “Step-Down” inhibitory avoidance test were as previously described [9]. The apparatus was an open field Plexiglas box (40×40 cm) with a steel rod floor and a Plexiglas platform (4×4×4 cm) set in the center of the grid floor. Intermittent electric shocks (20 mA, 50 Hz) were delivered to the grid floor. On day 1 (training test, TT), each mouse was placed on the platform and received an electric shock for 3 s when it stepped down placing all paws on the grid floor. Responsiveness to the punishment in the TT was assessed by animal vocalization; only those mice that vocalized touching the grid (∼70%) were used for retention test (RT), 24 h after TT. The latencies were measured, considering 180 s as the upper cut-off, during TT and RT.

For ORT, the apparatus and procedures were as previously described. A session of two trials (T1 and T2) with at 60 min interval was given on the test day. In T1, the time spent by each mouse exploring two identical 8.0 cm side grey cubes presented for 10 min in two opposite corners of the box was recorded. During T2, one of the cubes was replaced by an 8.0 cm side grey cylinder, and the mice were left in the box for 5 min. The time spent exploring the familiar (F) and the new (N) object were recorded and a discrimination score (D = N/N + F) was calculated. In this task, a discrimination score above 0.5 indicates the ability of mice to discriminate between the familiar and the novel object while a score below or equal to 0.5, reflecting an exploration time of the novel object minor than, or equal to, the half of the total time spent between the two objects, indicates memory impairment.

Animal tissue processing

After completing the behavioral tests, the mice were sacrificed by cervical dislocation and the brains were rapidly removed and divided sagittally. For protein analysis, cortical and hippocampal samples from one hemibrain were sectioned, snap-frozen, and stored at –80 °C. The other hemibrain was postfixed in phosphate-buffered, pH 7.4, containing 4.0% paraformaldehyde at 4°C for 48 h, rinsed in PBS and paraffin-embedded for immunohistochemistry.

Immunohistochemistry and western blotting

Immunohistochemical analyses were performed on 5.0μm coronal paraffin-embedded sections, as previously described [9]. For western blotting analysis, tissue samples were homogenized in ice-cold RIPA lysis buffer; then 40μg of proteins was applied to 4–12% Criterion XT Bis-Tris Gel (Novex NuPAGE, Lyfe Technologies, NY, USA) for electrophoresis with XT MOPS running buffer (Lyfe Technologies, NY, USA), as previously reported.

Determination of Aβ₄₂ and pE3-Aβ plaque-load

To quantify plaque burden, cortices and hippocampi of the sections stained for Aβ₄₂ and pE3-Aβ were digitized and acquired with an Olympus BX63 microscope equipped with CellSens Dimensionsoftware (Olympus, Germany) [9]. Six coronal brain sections separated by 60μm interval from each mouse (4-5 animals/group) were analyzed. Plaque number and total area were determined automatically.

Reverse transcript and real-time PCR

Total RNA from tissue was isolated using the Trizol Reagent (Life Technologies) and treated with DNase to remove genomic DNA contaminations. One microgram of RNA was retro-transcribed using iScript (Bio-Rad) and amplified with the following specific primers: for 18 S, forward 5-AAAACCAACCCGGTGAGCTCCCTC-3_and reverse 5-CTCAGGCTCCCTCTCCGGAATCG-3; for TNF-α, forward 5’-GCATGATCCGCGACGTGGAACT-3’ and reverse 5’-CGAATGAGAGGGAGGCCATTTGG-3’. All primers were purchased from IDT (IDT, Germany). PCR amplification was carried out by means of SsoAdvancedTMUniversal SYBR^®Green Supermix (Bio-Rad, USA) according to manual instructions, using a RotorGene 3000 (Qiagen) instrument. The RT-qPCR was performed using the following procedure: 98°C for 1 min, 40 cycles of 98 C for 5 s, 60°C for 10 s. The program was set to reveal the melting curve of each amplicon from 60°C to 95°C with 0.5°C reading intervals.

Data analysis

Two-way-ANOVA plus Bonferroni’s post comparison test were used to evaluate behavioral performance during the MWM training session. One-way ANOVA followed by Bonferroni post hoc test were used to evaluate behavioral performance in the step down inhibitory avoidance and ORT, SAPK/JNK, and ERK 1/2 levels. Statistical significance was defined as P < 0.05. Data are reported as mean±standard error of the mean(S.E.M).

RESULTS

Hydroxytyrosol improves memory deficits in TgCRND8 mice

Diet supplementation with HT was well tolerated by the mice, their body weight was not affected, and no HT-fed animal died during treatment. In the last week of treatment, memory performance was assessed by the MWM task and then by the step-down avoidance and ORT tests. Tg and wt mice from each experimental group were trained for 4 days in the MWM task to learn where the hidden platform was located. The mice were naive to the water maze and showed no deficiencies in swimming abilities, directional swimming toward the platform, or climbing onto a hidden platform during training trials. Statistical analysis was performed by two-way ANOVA with treatment and training time as the variables: Treatment, F _(3,116) = 13.76, p < 0.0001, Training time, F _(3,116) = 12.97, p < 0.0001, Interaction Treatment X Training time, F _(9,116) = 0.5945, p > 0.05, n.s. The Bonferroni post comparison’s test revealed that the time required to find the platform by untreated Tg mice was significantly increased compared to both groups of wt mice (^**p < 0.01, day 2 and day 3, versus untreated wt; ^# #p < 0.01, day 2 and ^# # #p < 0.001 day 3, versus wt HT) (Fig. 1A). Interestingly, the acquisition phase of HT-fed Tg mice was remarkably improved with respect to untreated Tg mice; the treated mice were good swimmers, showed an appropriate swim-search response after being placed in water, and the escape latency during the 2- and 3-day reached that displayed by both groups of wt mice; in particular, on day 2 it appeared significantly improved compared to untreated Tg mice (Fig. 1A, °p < 0.05). These findings indicate that HT administration restored to the level displayed by wt mice the early impairment in spatial memory for platform location induced by the transgene. At day 4 of the acquisition phase, the escape latency displayed by untreated Tg mice was shortened with no significant differences when compared to HT-fed Tg or wt mice (Fig. 1A), likely suggesting that in this experimental set-up, the disruption underlying the spatial memory for platform location in untreated Tg mice was restored oversessions.

Fig.1

HT restores cognitive functions of TgCRND8 mice. A) MWM test; untreated TgCRND8 mice required significantly more time to find the platform ^**p < 0.01 at day 2 and 3, compared to untreated wt and ^# #p < 0.01, day 2 and ^# # #p < 0.001 day 3, versus wt HT); HT-fed Tg mice showed a response similar to controls, and their escape latency on day 2 was significantly improved compared to untreated Tg mice (°p < 0.05). B) step-down inhibitory avoidance test; the training test showed no significant differences between groups. The 24 h retention test showed increased latencies in controls and HT-fed Tg mice versus respective training latencies (°°°p < 0.0001, °°p < 0.001, and °p < 0.05). In untreated Tg mice, the retention latencies were significantly reduced respect to control groups (^***p < 0.0001) and to HT-fed Tg mice (^*p < 0.05) and not significantly different from training latencies. C) ORT test; in the T2 trial, the discrimination index of untreated Tg mice differed from the discrimination index of both groups of wt and HT-fed Tg mice (^***p < 0.0001). The dotted line indicates the chance level performance. Number of animals: n = 6/group.

The same animals were also tested in the step down inhibitory avoidance (Fig. 1B). No significant differences were observed among all groups during the training test. In the 24 h RT, the step-down latencies recorded in Tg mice were significantly reduced respect to both groups of wt (^***p < 0.0001) and not significantly different from the training latency (p > 0.05), indicating that Tg mice were unable to memorize the punishment and to perform the inhibitory avoidance. However, HT administration to Tg mice significantly improved their performance (^*p < 0.05, versus untreated Tg mice) to levels not different from those displayed by HT treated or untreated wt mice (Fig. 1B). Finally, the same mice were tested for ORT. In the T1 trial, the exploration time of the familiar object was comparable in all four groups, where HT-fed and untreated animals showed no deficiencies in exploratory activity, directional movement towards the objects, and locomotor activity. However, in the T2 trial, the untreated Tg mice exhibited impairments in novel object preference compared to both groups of wt mice, as shown by the significant reduction in the discrimination score (^***p < 0.0001). The ability of HT-fed Tg mice to discriminate between the familiar and the novel object was significantly improved with respect to that displayed by untreated Tg mice (^***p < 0.0001) and undistinguishable from that of HT treated or untreated wt mice (Fig. 1C).

Altogether, the results of the memory performance tests indicate that, in our Tg mouse model, cognitive impairment is completely prevented/rescued following HT administration. These data confirm those previously reported both for OLE-fed Tg mice [9] and for Tg mice fed with the mix of polyphenols present in olive mill wastewater, suggesting that the protective effect of OLE against behavioral disturbances resides in its polyphenolic HT moiety.

Hydroxytyrosol modifies Aβ burden in TgCRND8 mice

At the end of the behavioral performance we checked whether cognitive improvement in HT-fed mice resulted from the well-known antioxidant power of the molecule or, rather, it implied any altered amyloid load in this model of plaque deposition. Therefore, the mice were analyzed for Aβ₄₂ and pE3-Aβ load, on the basis of our previous findings showing that OLE reduces Aβ₄₂ and pE3-Aβ deposition in this mouse model [9, 13]. We found that the differently shaped and sized Aβ₄₂-positive plaques detected in the cortex and hippocampal areas of untreated Tg mice were reduced in the brains of HT-fed Tg mice (stained in black in Fig. 2A). In particular, the quantitative analysis revealed a significant reduction of total Aβ plaque area and number in the parietal cortex and an apparent tendency towards some reduction in the hippocampus of HT-fed Tg mice (Fig. 2A). These data indicate that HT administration partially protects against Aß deposition. We also checked whether pE3-Aβ load was modified in HT-treated Tg mice in respect to untreated animals. As shown in representative photomicrographs (Fig. 2B), only a few radiating small-sized pE3-Aβ positive plaques (stained in black) were detected in the parietal cortex and hippocampal areas of HT-fed Tg mice, as compared to untreated animals. The quantitative analysis of pE3-Aβ load in the parietal cortex and in the hippocampus (Fig. 2B) confirmed the imaging results; in fact, plaque load (as plaque number and total plaque area) was significantly reduced in the parietal cortex of HT-fed animals while, in the hippocampus, a significant reduction in plaque number and an apparent tendency towards reduction of total plaque area was seen, compared to untreated animals. Altogether, these data indicate that protection by HT against amyloid load is stronger in the cortex than in the hippocampus.

Fig.2

HT administration reduces Aβ₄₂ and pE3-Aβ burden in the brains of TgCRND8 mice. Representative photomicrographs of Aβ₄₂ (A) and pE3-Aβ (B) immunopositive deposits and quantitative analysis of Aβ₄₂ (A) and pE3-Aβ (B) plaque area and number in the parietal cortex and in the hippocampus of untreated and HT-fed Tg mice (n = 4-5/group, six sections/mouse). ^*p < 0.05 and ^**p < 0.01 versus untreated Tg mice. Scale bars = 100 and 400μm apply to both the parietal cortex and the hippocampus images, respectively.

Hydroxytyrosol reduces TNF-α expression in the hippocampus of TgCRND8 mice

Many preclinical and clinical studies have reported that deposition of senile plaques in the brain is associated to an inflammatory response driven by the activated glial cells following the expression of inflammatory cytokines (see review, [24]). We previously reported that astrocytes activation in the brain of TgCRND8 mice is strongly attenuated by a diet supplemented with OLE [9, 13]. Here, we examined whether HT administration affected the expression and cell localization of tumor necrosis factor-α (TNF-α), a main pro-inflammatory cytokine, in hippocampal areas of Tg mice. As exemplified in Fig. 3A, immunofluorescence analysis with anti-TNF-α antibody revealed strong TNF-α immunoreactivity (TNF-α IR) (red) throughout hippocampal areas of untreated Tg mice that was markedly diminished in HT-fed Tg mice. TNF-α IR was almost undetectable in hippocampal areas of both untreated and HT-fed wt mice (not shown). We also investigated the expression levels of TNF-α in Tg mice normally fed or supplemented with HT by checking the levels of TNF-α mRNA. As expected, the latter were over 6 times higher in the hippocampus of Tg mice than in the hippocampus of both HT-treated or untreated wt mice; however, they were significantly reduced in Tg mice upon HT treatment, where they rescued close to the TNF-α mRNA levels found in wt mice (Fig. 3B). These data agree with those previously reported in OLE-fed Tg mice [9] and confirm that the HT moiety of OLE is the main responsible for the anti-inflammatory effects of the latter.

Fig.3

HT administration reduces TNF-α expression in the hippocampus of TgCRND8 mice. A) Representative images of TNF-α immunoreactivity (red) plus DAPI (blue) in the hippocampus of untreated and HT-fed Tg mice. Scale bar = 20μm applies to both images B) quantitative analysis of the TNF-α mRNA levels in the hippocampus of untreated or HT-fed wt and Tg mice. ^*p < 0.05 versus untreated Tg mice; C) representative photomicrographs of TNF-α (red) and GFAP (green) immunoreactivity plus DAPI (blue) in the hippocampus of untreated or HT-treated Tg mice. Yellow color (arrows) in the merged image of untreated Tg mice indicates that TNF-α immunoreactivity co-localizes with GFAP staining. Scale bar = 20μm applies to all images (n = 4-5/group, six sections/mouse).

Hydroxytyrosol modulates intracellular signaling pathways

It is widely recognized that some modulation of intracellular signaling pathways is involved in the beneficial effects of different phenols and flavonoids on neuronal processes [15 , 25]. Flavonoids and their metabolites have been reported to interact with a number of signaling cascades triggered by protein kinases and the inhibition/activation of these pathways is likely to greatly affect neuronal function (see references in [25]). Mitogen-activated protein kinases (MAPKs) play a pivotal role in the prevention/pathogenesis of AD [26], and we previously reported a differential activation of p38MAPK, SAPK/JNK, and ERK1/2 in the TgCRND8 mouse brain [27, 28]. Here we evaluated whether HT treatment was involved of in the reversal of the transgene-induced alteration of SAPK/JNK and ERK1/2 signaling pathways as a possible mechanism of neuroprotection in TgCRND8 mice. The levels of phospho-SAPK/JNK, higher in the cortex of untreated Tg mice, with respect to both wt groups, were significantly reduced in the cortex of HT-treated Tg mice, as shown by western blotting (Fig. 4A). In contrast, the transgene-reduced levels of phospho-ERK 1/2 in the cortex of untreated Tg mice were significantly enhanced in the cortex of HT-treated Tg mice (Fig. 4B). These data are incomplete and need further investigation; however, they suggest that HT does interfere, though differently in different brain regions, with specific signaling pathways, which contribute to explain some of its effects.

Fig.4

HT administration modulates SAPK/JNK and ERK1/2 expression in the cortex of TgCRND8 mice. Western blotting analysis of phospho-SAPK/JNK (A) and phospho-ERK 1/2 protein (B) levels in the cortex of HT-fed or untreated wt and Tg mice. A): ^*p < 0.05 and ^**p < 0.001 versus untreated Tg mice; B) left panel: ^**p < 0.001 versus untreated Tg mice and °p < 0.05 versus untreated wt mice; right panel: ^*p < 0.05 versus untreated Tg mice and °p < 0.05 versus wt HT mice.

Hydroxytyrosol induces autophagy in the cortex of TgCRND8 mice

We recently reported an intense activation of the autophagosome-lysosome system in the cortex of TgCRND8 mice fed with different doses of OLE, showing the presence of a low dose (0.5 mg/kg of diet) unable to provide protection for what plaque deposition and behavioral improvements are concerned [20]. In OLE-fed Tg mice, autophagy was triggered following activation of the Ca^2 +-CAMKKβ-AMPK axis [15] and was considered a major responsibility for both the marked reduction of number and compactness of Aβ plaques and their disaggregation in these mice [20]. Here we report that the same intense activation of autophagy was also found in the cortex of HT-fed Tg mice. In fact, a bright and punctate immunoreactivity for LC3, the lipidated LC3 form directly involved in the initiation and execution of autophagy, was detected in the soma, perikaryal, and dendrites of neurons in different layers of somatosensory/parietal cortex of HT-fed Tg mice (Fig. 5A), whereas it was absent in the same regions of untreated Tg mice. Similarly to OLE-fed wt mice (50 mg/kg of diet) [9], some slight activation of autophagy occurred also in HT-fed wt mice, as shown by the stronger and bright LC3 IR (Fig. 5A).

Fig.5

HT induces macroautophagy in the cortex of TgCRND8 mice. A) representative images of LC3 immunoreactivity (green) plus DAPI (blue) showing a strong and bright LC3 staining (arrows) in the neuronal cell bodies and processes of neurons in the parietal cortex of HT-fed Tg, and, to a lesser extent, in HT-fed wt mice, as compared to the untreated animals. B) Representative images of LAMP-2A staining (green) plus DAPI (blue) (n = 5/group). Scale bar = 20μm applies to all images.

We also sought to assess whether, in addition to macroautophagy, some activation of the chaperone-mediated autophagy (CMA) also contributed to the molecular mechanisms underlying the beneficial effects of HT and other olive oil polyphenols. Therefore, we investigated whether the transgene and the HT-supplemented diet affected the expression levels of both the lysosome-associated membrane protein type 2A (LAMP-2A), a CMA receptor at the lysosomal membrane, and hsp70, a heat shock protein involved in cell defense against proteotoxicity in different neurodegenerative diseases (refs in [29]). Western blotting analysis did not reveal any difference in the cortical levels of either hsp70 (data not shown) or LAMP-2A (Fig. 5B) in normally fed and in HT-fed Tg mice respect to similarly treated wt mice. Apparently, these data lead to exclude, in our experimental model, any effect of the transgene on the CMA system and that CMA activities are involved in HT-induced Aβ plaque disaggregation and reduction, supporting the idea that only macroautophagy is responsible for the activation of the autophagic flux in HT-fed Tg mice.

DISCUSSION

We have previously reported that dietary supplementation of either OLE or a mix of olive oil polyphenols to TgCRND8 mice results in beneficial behavioral, molecular, and histo-pathological effects, suggesting that these natural molecules could be useful against pre-symptomatic or early AD in humans [20]. Moreover, OLE and the mix of polyphenols remarkably activated neuronal autophagy and OLE increased histone 3/4 acetylation following decrease of histone deacetylase-2 expression, with improvement of synaptic functions; in addition, the lowered pE3-Aβ levels in OLE-fed mice were ascribed to the reduced expression of glutaminyl cyclase [13]. Finally, we showed that the Aβ levels in cortical and hippocampal neurons were significantly reduced in OLE-fed Tg mice, in agreement with previous data showing that OLE activates the non-amyloidogenic pathway of APP metabolism [30].

The OLE molecule results from the esterification of elenolic acid with HT, the polyphenol moiety of the molecule, leading to the idea that the true responsible for the biological effects of OLE was its HT component either in the whole molecule or free, after hydrolysis by tissue/cellular esterases. This idea was reinforced by our finding showing the presence of two OLE metabolites, HT and homovanillic acid (but not OLE), in the brains of OLE-fed Tg mice [13]. However, this finding was not conclusive since OLE could have been processed in tissue/cells to its metabolites prior to detection. That the whole OLE molecule or its components may be differently responsible for their in vitro and in vivo properties was also supported by a recent study carried out on diabetic rats, showing that HT was the OLE moiety mostly responsible for inhibition of amylin aggregation whereas the antidiabetic effect of OLE, associated with stimulation of insulin secretion, required the whole molecule [31]. Therefore, to better define the protective power of OLE, and the role performed by its HT moiety, against neurodegeneration, we extended our previous study on OLE to the dietary administration to TgCRND8 mice of HT at the same dose as the highest dose of OLE previously administered to theTg mice.

The HT-supplemented diet resulted in significant functional and histopathological benefits comparable to those previously reported for OLE-fed Tg mice. We also sought to decipher the molecular mechanisms underlying the beneficial effects of HT to assess whether they were comparable to those previously described in OLE-fed Tg mice. As already reported for Tg mice fed with OLE or with the mix of polyphenols from olive mill wastewater [9, 20], HT administration also resulted in autophagy activation; in HT-fed mice, this effect was accompanied by a significant reduction of TNF-α expression and a modulation of MAPK signaling pathways, thus disclosing new biochemical and biological effects of HT. However, in HT-fed mice we did not investigate the presence of any epigenetic effect. Taken together, our data support a multifunctional activity of HT and other natural polyphenols underlying their beneficial effects.

The cognitive functions of TgCRND8 mice were evaluated in the MWM and, as in our previous studies with OLE, in the step-down and ORT tasks. Our findings clearly show that TgCRND8 mice fed for 8 weeks with a diet supplemented with HT displayed significant improvements of spatial and working memory by ameliorating both cortical and hippocampal memory circuitry. Memory is believed to be the product of dynamic interactions among multiple systems in the brain [32]. Notably, working and spatial memory for platform location in the MWM appears to be mainly hippocampus-dependent as this brain area is necessary for acquisition and retrieval of spatial information. The step-down inhibitory avoidance and ORT memory depends mainly on the integrated activity of the entorhinal/parietal cortex, since the inability to acquire the step down-inhibitory response and to explore a novel object over afamiliar one reflects dysfunction of cortical areas [9, 33]. Functional disruption in the neuronal network has repeatedly been reported in AD mouse models and aberrant neuronal activity, with significant reduction of the number of active neurons particularly present near amyloid plaques, whose presence causes neuronal disturbances. These alterations result in abnormalities of whole neuronal networks both in animal models [34] and in asymptomatic patients with amyloid deposits [35]. We previously showed significant behavioral and memory improvements in Tg mice dietary supplemented with OLE or the mix of polyphenols found in olive mill waste water [20]; similarly, here we report that the cognitive amelioration in HT-fed, in respect to normally-fed, Tg mice was paralleled by a remarkable reduction of Aβ₄₂ and pE3-Aβ deposits both in the cortex and in the hippocampus, likely reflecting a functional neuronal recovery. These findings agree with similar data showing that HT administration significantly improves spatial reference/working memory dysfunctions induced by intracerebral injection of Aβ₄₂ plus ibotenic acid in the C57BL/6 mouse model ofAD [36].

We here also report that an HT-supplemented diet modulates MAPKs signaling by activating ERK and down regulating SAPK/JNK expression, a mechanism that may underlie memory improvements in HT-fed Tg mice. These data agree with other findings suggesting an involvement of ERK stimulation in memory formation and synaptic plasticity (references in [37]). A strong glia reaction has been reported around and within amyloid plaques in the brain of TgCRND8 mice [9]; furthermore, oxidative stress and neuroinflammation associated with the presence of amyloid plaques appear to play a crucial role in AD progression [24]. Brain inflammation is strongly reduced by EVOO polyphenols; in particular, we previously reported that OLE administration to Tg mice remarkably reduces astrocyte activation [9] and NF-κB-mediated inflammation [23], yet without any significant reduction of the oxidative stress [9]. Here we found that HT supplementation to the same mouse model of the same age relieved the inflammatory response by reducing astrocyte reaction and the expression of TNF-α in hippocampal areas. TNF-α plays a central role as a trigger of the inflammatory state in the brain and is the only cytokine consistently implicated as detrimental in AD [38]. Altogether, our data show that olive polyphenols are similarly beneficial and that their effects are comparable when administered either as pure components (OLE), their metabolites (HT), or as a mixture of these and other minor molecules in concentrated olive mill wastewater.

OLE, HT, and other natural phenols activate the autophagy-lysosome system that, as previously suggested [7, 11], might represent a key mechanism for removal of toxic intra/extracellular amyloid aggregates, thus providing further support to the beneficial effects of these molecules. Various types of autophagy have been described including microautophagy, macroautophagy, and CMA. Once shown that OLE and HT trigger macroautophagy, we sought to assess whether they were also able to foster CMA. Our findings show that CMA delivery cargo to lysosome is not involved in the beneficial effects of HT. On the other hand, CMA does not appear to be impaired in our transgenic mouse strain, as indicated by the lack of differences in LAMP-2A immunoreactivity and hsp70 protein levels between untreated and HT-fed Tg mice. We conclude that macroautophagy is the only autophagic flux impaired in our Tg model and that it is significantly and similarly rescued by OLE or HT administration, further supporting the idea that the cellular effects of OLE is largely mediated by its HT moiety.

In conclusion, our findings indicate that the beneficial effects of EVOO polyphenols, including HT and other natural polyphenols, can largely be traced back to a common molecular scaffold but arise from multifunctional activities. These include, among others, their ability to interfere with the aggregation path of disease-associated peptides/proteins, their antioxidant/anti-inflammatory power, and their pro-autophagic effect. Taken together, the molecular determinants of these activities provide a solid, yet still incomplete, rationale supporting the suggested protection against amyloid-associated neurodegeneration by olive polyphenols.

Footnotes

ACKNOWLEDGMENTS

We thank Dr. P. St. George-Hyslop and Dr. D. Westaway for supplying the TgCRND8 mouse strain and Prof. Maurizio Servili (PG, Italy) for supplying hydroxytyrosol. Supported by University of Florence and ECRF 2015.

Authors’ disclosures available online ().

References

Stefani

, Rigacci

(2013) Protein folding and aggregation into amyloid: The interference by natural phenolic compounds. Int J Mol Sci 14, 12411–12457.

Scarmeas

, Stern

, Mayeux

, Manly

, Schupf

, Luchsinger

(2009) Mediterranean diet and mild cognitive impairment. Arch Neurol 66, 216–225.

Cheng

, Gong

, Xiao

, Petersen

, Zheng

, Huang

(2013) Inhibiting toxic aggregation of amyloidogenic proteins: A therapeutic strategy for protein misfolding diseases. Biochim Biophys Acta 1830, 4860–4871.

Martínez-Lapiscina

, Clavero

, Toledo

, Estruch

, Salas-Salvadó

, San Julián

, Sanchez-Tainta

, Ros

, Valls-Pedret

, Martinez-Gonzalez

MÁ

(2013) Mediterranean diet improves cognition: The PREDIMED-NAVARRA randomised trial. J Neurol Neurosurg Psychiatry 84, 1318–1325.

Valls-Pedret

, Sala-Vila

, Serra-Mir

, Corella

, de la Torre

, Martínez-González

MÁ

, Martínez-Lapiscina

, Fitó

, Pérez-Heras

, Salas-Salvadó

, Estruch

, Ros

(2015) Mediterranean diet and age-related cognitive decline: A randomized clinical trial. JAMA Intern Med 175, 1094–1103.

Hernáez

, Castañer

, Elosua

, Pintó

, Estruch

, Salas-Salvadó

, Corella

, Arós

, Serra-Majem

, Fiol

, Ortega-Calvo

, Ros

, Martínez-González

MÁ

, de la Torre

, López-Sabater

, Fitó

(2017) Mediterranean diet improves high-density lipoprotein function in high-cardiovascular-risk individuals: A randomized controlled trial. Circulation 135, 633–643.

Stefani

, Rigacci

(2014) Beneficial properties of natural phenols: Highlight on protection against pathological conditions associated with amyloid aggregation. Biofactors 40, 482–493.

Casamenti

, Stefani

(2017) Olive polyphenols: New promising agents to combat aging-associated neurodegeneration. Expert Rev Neurother 17, 345–358.

Grossi

, Rigacci

, Ambrosini

, Ed Dami

, Luccarini

, Traini

, Failli

, Berti

, Casamenti

, Stefani

(2013) The polyphenol oleuropein aglycone protects TgCRND8 mice against Aß plaque pathology. PLoS One 8, e71702.

10.

Daccache

, Lion

, Sibille

, Gerard

, Slomianny

, Lippens

, Cotelle

(2011) Oleuropein and derivatives from olives as Tau aggregation inhibitors. Neurochem Int 58, 700–707.

11.

, Deng

, Tang

, Ma

, Yuan

, Ma

(2016) Hydroxytyrosol induces phase II detoxifying enzyme expression and effectively protects dopaminergic cells against dopamine- and 6-hydroxydopamine induced cytotoxicity. Neurochem Int 96, 113–120.

12.

Rigacci

, Stefani

(2016) Nutraceutical properties of olive oil polyphenols. An itinerary from cultured cells through animal models to humans. Int J Mol Sci 17, E843.

13.

Luccarini

, Grossi

, Rigacci

, Coppi

, Pugliese

, Pantano

, la Marca

, Ed Dami

, Berti

, Stefani

, Casamenti

(2015) Oleuropein aglycone protects against pyroglutamylated-3 amyloid-ß toxicity: Biochemical, epigenetic and functional correlates. Neurobiol Aging 36, 648–663.

14.

Ladiwala

, Mora-Pale

, Lin

, Bale

, Fishman

, Dordick

, Tessier

(2011) Polyphenolic glycosides and aglycones utilize opposing pathways to selectively remodel and inactivate toxic oligomers of amyloid β. Chembiochem 12, 1749–1758.

15.

Rigacci

, Miceli

, Nediani

, Berti

, Cascella

, Pantano

, Nardiello

, Luccarini

, Casamenti

, Stefani

(2015) Oleuropein aglycone induces autophagy via the AMPK/mTOR signalling pathway: A mechanistic insight. Oncotarget 6, 35344–35357.

16.

Gonzáles-Correa

, Navas

, Lopez-Villodres

, Trujillo

, Espartero

, De La Cruz

(2008) Neuroprotective effect of hydroxytyrosol and hydroxytyrosol acetate in rat brain slices subjected to hypoxia-reoxygenation. Neurosci Lett 446, 143–146.

17.

Bernini

, Gilardini Montani

, Merendino

, Romani

, Velotti

(2015) Hydroxytyrosol-derived compounds: A basis for the creation of new pharmacological agents for cancer prevention and therapy. J Med Chem 58, 9089–9107.

18.

Peng

, Zhang

, Yao

, Duan

, Fang

(2015) Dual protection of hydroxytyrosol, an olive oil polyphenol, against oxidative damage in PC12 cells. Food Funct 6, 2091–2100.

19.

D’Antuono

, Garbetta

, Ciasca

, Linsalata

, Minervini

, Lattanzio

, Logrieco

, Cardinali

(2016) Biophenols from table olive cv Bella di Cerignola: Chemical characterization, bioaccessibility, and intestinal absorption. J Agric Food Chem 64, 5671–5678.

20.

Pantano

, Luccarini

, Nardiello

, Servili

, Stefani

, Casamenti

(2017) Oleuropein aglycone and polyphenols from olive mill waste water ameliorate cognitive deficits and neuropathology. Br J Clin Pharmacol 83, 54–62.

21.

Grossi

, Francese

, Casini

, Rosi

, Luccarini

, Fiorentini

, Gabbiani

, Messori

, Moneti

, Casamenti

(2009) Clioquinol decreases amyloid-beta burden and reduces working memory impairment in a transgenic mouse model of Alzheimer’s disease. J Alzheimers Dis 17, 423–440.

22.

Janus

, Welzl

, Hanna

, Lovasic

, Lane

, St George-Hyslop

, Westaway

(2004) Impaired conditioned taste aversion learning in APP transgenic mice. Neurobiol Aging 25, 1213–1219.

23.

Luccarini

, Pantano

, Nardiello

, Cavone

, Lapucci

, Miceli

, Nediani

, Berti

, Stefani

, Casamenti

(2016) The polyphenol oleuropein aglycone modulates the PARP1-SIRT1 interplay: An in vitro and in vivo study. J Alzheimers Dis 54, 737–750.

24.

Heppner

, Ransohoff

, Becher

(2015) Immune attack: The role of inflammation in Alzheimer disease. Nat Rev Neurosci 16, 358–372.

25.

Baptista

, Henriques

, Silva

, Wiltfang

, da Cruz eSilva

(2014) Flavonoids as therapeutic compounds targeting key proteins involved in Alzheimer’s disease. ACS Chem Neurosci 5, 83–92.

26.

Kawamata

, Shimohama

(2011) Stimulating nicotinic receptors trigger multiple pathways attenuating cytotoxicity in models of Alzheimer’s and Parkinson’s diseases. J Alzheimers Dis 24(Suppl 2), 95–109.

27.

Bellucci

, Rosi

, Grossi

, Fiorentini

, Luccarini

, Casamenti

(2007) Abnormal processing of tau in the brain of aged TgCRND8 mice. Neurobiol Dis 27, 328–338.

28.

Giovannini

, Cerbai

, Bellucci

, Melani

, Grossi

, Bartolozzi

, Nosi

, Casamenti

(2008) Differential activation of mitogen-activated protein kinase signalling pathways in the hippocampus of CRND8 transgenic mouse, a model of Alzheimer’s disease. Neuroscience 153, 618–633.

29.

Kaushik

, Cuervo

(2012) Chaperone-mediated autophagy: A unique way to enter the lysosome world. Trends Cell Biol 22, 407–417.

30.

Kostomoiri

, Fragkouli

, Sagnou

, Skaltsounis

, Pelecanou

, Tsilibary

, Tzinia

(2013) Oleuropein, an anti-oxidant polyphenol constituent of olive promotes α-secretase cleavage of the amyloid precursor protein (AβPP). Cell Mol Neurobiol 33, 147–154.

31.

, Velander

, Liu

, Xu

(2017) Olive component oleuropein promotes β-cell insulin secretion and protects β-cells from amylin amyloid-induced cytotoxicity. Biochemistry 56, 5035–5039.

32.

Izquierdo

, Barros

, Mello e Souza

, de Souza

, Izquierdo

, Medina

(1998) Mechanisms for memory types differ. Nature 393, 635–636.

33.

Bellucci

, Luccarini

, Scali

, Prosperi

, Giovannini

, Pepeu

, Casamenti

(2006) Cholinergic dysfunction, neuronal damage and axonal loss in TgCRND8 mice. Neurobiol Dis 23, 260–272.

34.

Grienberger

, Rochefort

, Adelsberger

, Henning

, Hill

, Reichwald

, Staufenbiel

, Konnerth

(2012) Staged decline of neuronal function in vivo in an animal model of Alzheimer’s disease. Nat Commun 3, 774.

35.

Sperling

, Laviolette

, O’Keefe

, O’Brien

, Rentz

, Pihlajamaki

, Marshall

, Hyman

, Selkoe

, Hedden

, Buckner

, Becker

, Johnson

(2009) Amyloid deposition is associated with impaired default network function in older persons without dementia. Neuron 63, 178–188.

36.

Arunsundar

, Shanmugarajan

, Ravichandran

(2015) 3,4-dihydroxyphenylethanol attenuates spatio-cognitive deficits in an Alzheimer’s disease mouse model: Modulation of the molecular signals in neuronal survival-apoptotic programs. Neurotox Res 27, 143–155.

37.

Giovannini

, Lana

, Pepeu

(2015) The integrated role of ACh, ERK and mTOR in the mechanisms of hippocampal inhibitory avoidance memory. Neurobiol Learn Mem 119, 18–33.

38.

Janelsins

, Mastrangelo

, Oddo

, LaFerla

, Federoff

, Bowers

(2005) Early correlation of microglial activation with enhanced tumor necrosis factor-alpha and monocyte chemoattrnt protein-1 expression specifically within the entorhinal cortex of triple transgenic Alzheimer’s disease mice. J Neuroinflammation 2, 23–acta.

Diet Supplementation with Hydroxytyrosol Ameliorates Brain Pathology and Restores Cognitive Functions in a Mouse Model of Amyloid-β Deposition

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Ethics statement

Animals and diet

Behavioral experiments

Animal tissue processing

Immunohistochemistry and western blotting

Determination of Aβ42 and pE3-Aβ plaque-load

Reverse transcript and real-time PCR

Data analysis

RESULTS

Hydroxytyrosol improves memory deficits in TgCRND8 mice

Hydroxytyrosol modifies Aβ burden in TgCRND8 mice

Hydroxytyrosol reduces TNF-α expression in the hippocampus of TgCRND8 mice

Hydroxytyrosol modulates intracellular signaling pathways

Hydroxytyrosol induces autophagy in the cortex of TgCRND8 mice

DISCUSSION

Footnotes

ACKNOWLEDGMENTS

References

Determination of Aβ₄₂ and pE3-Aβ plaque-load