Journal of Alzheimer’s Disease 28 (2012) 81–92DOI 10.3233/JAD-2011-110662IOS Press

81

Extra Virgin Olive Oil Improves Learningand Memory in SAMP8 Mice

Susan A. Farra,b,∗, Tulin O. Pricec, Ligia J. Dominguezd, Antonio Motisid, Filippo Saianoe,Michael L. Niehoffb, John E. Morleya,b, William A. Banksg,h, Nuran Ercali and Mario Barbagallod

aGeriatric Research Educational and Clinical Center (GRECC), VA Medical Center, St. Louis, MO, USAbDepartment of Internal Medicine, Division of Geriatric Medicine, St. Louis University School of Medicine,St. Louis, MO, USAcDepartment of Internal Medicine, Division of Endocrinology, St. Louis University School of Medicine, St. Louis,MO, USAdDivision of Geriatric Medicine, University of Palermo, Palermo, ItalyeDepartment DEMETRA, Agronomy Faculty, University of Palermo, Palermo, Italyf Department of Agro-Environmental Systems, University of Palermo, Palermo, ItalygGeriatric Research Educational and Clinical Center (GRECC), Veterans Affairs Puget Sound Health Care System,Seattle, WA, USAhDepartment of Internal Medicine, Division of Gerontology and Geriatric Medicine, University of WashingtonSchool of Medicine, Seattle, WA, USAiDepartment of Chemistry, Missouri University of Science and Technology, Rolla, MO, USA

Accepted 14 August 2011

Abstract. Polyphenols are potent antioxidants found in extra virgin olive oil (EVOO); antioxidants have been shown to reverseage- and disease-related learning and memory deficits. We examined the effects of EVOO on learning and memory in SAMP8mice, an age-related learning/memory impairment model associated with increased amyloid-� protein and brain oxidativedamage. We administered EVOO, coconut oil, or butter to 11 month old SAMP8 mice for 6 weeks. Mice were tested in T-mazefoot shock avoidance and one-trial novel object recognition with a 24 h delay. Mice which received EVOO had improvedacquisition in the T-maze and spent more time with the novel object in one-trial novel object recognition versus mice whichreceived coconut oil or butter. Mice that received EVOO had improve T-maze retention compared to the mice that receivedbutter. EVOO increased brain glutathione levels suggesting reduced oxidative stress as a possible mechanism. These effectsplus increased glutathione reductase activity, superoxide dismutase activity, and decreased tissue levels of 4-hydroxynoneal and3-nitrotyrosine were enhanced with enriched EVOO (3× and 5× polyphenols concentration). Our findings suggest that EVOOhas beneficial effects on learning and memory deficits found in aging and diseases, such as those related to the overproductionof amyloid-� protein, by reversing oxidative damage in the brain, effects that are augmented with increasing concentrations ofpolyphenols in EVOO.

Keywords: Extra virgin olive oil, learning, memory, object recognition, oxidative stress, SAMP8, T-maze

∗Correspondence to: Susan A. Farr, Ph.D., Associate Profes-sor/ Research Health Scientist, St. Louis University School ofMedicine/VA Medical Center, St. Louis, 915 North Grand Blvd.151/JC, St. Louis, MO 63106, USA. Tel.: +314 289 7608; Fax: +314289 7046; E-mail: farrsa@slu.edu.

INTRODUCTION

Olive oil is a staple of the diet in the Mediter-ranean region. Studies indicate that the inhabitants ofthis region have increased lifespans compared to otherregions of Europe [1, 2]. The Mediterranean diet is

ISSN 1387-2877/12/$27.50 © 2012 – IOS Press and the authors. All rights reserved

82 S.A. Farr et al. / EVOO Improves Learning and Memory

associated with a reduced incidence of cardiovasculardisease, cancer, and obesity [2–5]. Studies in rodentshave found that foods high in polyphenols improvecognition and reverse oxidative damage in the brain [6,7]. In humans, greater adherence to the Mediterraneandiet is associated with a slower rate of decline in theMini Mental Status Exam [8], reduced risk of conver-sion from mild cognitive impairment to Alzheimer’sdisease [9], and a reduction in the risk for Alzheimer’sdisease [10].

Consumption of various foods high in antioxidantshas been found to prevent loss of and improve learn-ing and memory in rodents [7]. Blackberries improvemotor and cognitive function in aging rats [11]. Mul-berry extract improves learning and memory in 9month old SAMP8 mice [12]. Chronic green tea con-sumption prevents learning and memory deficits in19 month old rats and morphological changes in thehippocampus [13]. Red wine improved cognition andattenuated AD-type pathology in the Tg2576 mice[15].

Among the potent antioxidants found in berries,wine, and green tea as well as extra virgin olive oil(EVOO) are polyphenols. Polyphenols prevent loss ofand improve deficits in learning and memory associ-ated with aging and disease [6] in part by reducingoxidative damage in the brain. EVOO decreases glu-tathione reductase (GR) activity and expression in thebrain of aged rats [14].

In the current study, we examined the effects ofEVOO enhanced with additional polyphenols on learn-ing and memory in SAMP8 mice. SAMP8 mice have anage-related impairment in learning and memory medi-ated by an age-related increase in amyloid-� protein(A�) [16–20]. SAMP8 mice have oxidative damage inthe brain [21–23]. The antioxidants alpha lipoic acidand n-acetyl cysteine as well as a diet supplementedwith n-3 polyunsaturated fatty acids have been found toimprove learning and memory in 12 month old SAMP8mice [24]. In these studies, mice were fed EVOO for 4weeks and then tested in T-maze footshock avoidanceand object recognition tests. We then measured levelsof oxidative stress in the brain.

MATERIALS AND METHODS

Mice

At the start of treatment, the subjects for the exper-iments were 11 month old SAMP8 mice from ourbreeding colony. Sentinels from the facility were tested

regularly to ensure our facility is virus- and pathogen-free. Food (Richland 5001) and water were available onan ad libitim basis and the rooms had a 12 h light-darkcycle with lights on at 0600 h. Experiments were con-ducted between 0730 and 1100 h. The number of miceper group in study 1 and study 2 was 10. These stud-ies were conducted with the approval of the AnimalCare and Use Committee at the VA Medical Center,St. Louis, MO.

Phenolic fraction extraction

EVOO from the olive-growing area of western Sicily(Trapani, Italy; cultivars “Cerasuola” and “Nocellaradel Belice”) was used as the source for extractingthe phenolic fraction and subsequent enrichments.This EVOO is rich in the dialdehydic form ofdeacetoxy-ligstroside aglycone, an established pheno-lic compound [25] that inhibits the cyclooxygenaseenzymes COX-1 and COX-2, similar to the inhibi-tion attained with ibuprofen [26]. This phenol hasbeen identified as one of the main substances respon-sible for the bitter taste of EVOO [27]. The extractionof the phenolic fraction was obtained by a two-phase continuous system. We slightly modified themethod described by Morello et al. [28]. Briefly,methanol–water (80 : 20 v/v) (2 × 50 ml) was added to100 g of virgin olive oil and homogenized for 5 minwith a Polytron. After that, the two phases were sep-arated by centrifugation at 2000 × g for 15 min. Thehydrolcoholic extracts were concentrated under vac-uum at 35◦C until a syrupy consistency was reached.This concentrated phenolic extract was dissolved in10 ml of acetonitrile and washed with 3×40 ml ofn-hexane. The apolar phases were purified with 10 mlof acetonitrile. The n-hexane phase was discarded andthe two acetonitrile phases were combined and rotatoryevaporated to dryness under vacuum. The residue wasdissolved in 5 ml of acetonitrile and evaporated undera nitrogen stream. Total phenolic analyses were car-ried out by a colorimetric method previously described[29]. The extracts were dissolved in ethanol/water(50 : 50, v/v) and a polyton used for 1 min to fully incor-porate them into the lipid matrix of fresh EVOO. Thefinal EVOO had a 2×, 3×, or 5× enrichment in itspolyphenolic concentration (total polyphenolic con-centration of the original EVOO: 210 mg/Kg, gallicacid equivalent).

The predominant cultivar in the EVOO used forthe experiments is “Cerasuola” with minor proportionof “Nocellara del Belice”. The most abundant phe-nolic compounds in this type of EVOO are tyrosol

S.A. Farr et al. / EVOO Improves Learning and Memory 83

(4-hydroxyphenylethanol) (19 ppm); hydroxytyrosol(3,4-dihydroxyphenylethanol) (42 ppm); dialdehydicform of deacetoxy-ligstroside aglycon (77 ppm); ace-toxypinoresinol (36 ppm), and oleuropein aglycon-dialdehydic form (18 ppm), measured by HPLC.Many other minor compounds for these cultivarsare present at lower concentrations, for exam-ple, vanillic acid (3-methoxy-4-hydroxybenzoic acid),and hydroxytyrosol acetate (4-Acetoxyethyl-1,2-dihydroxybenzene), among others.

Treatment

In our first study, SAMP8 mice were divided into3 groups that received either EVOO (75 �l), but-ter (100 �l), or coconut oil (75 �l) daily. Mice wereadministered doses of EVOO, butter, or coconut oil viaoral gavage in an amount to obtain equal caloric con-tent. Our second study consisted of 5 groups: 4 groupsreceived 75 �l daily of either unenriched EVOO or anEVOO enriched to 2 × s, 3 × s, or 5 × s the polyphe-nolic concentration of the unenriched EVOO; the 5thgroup acted as control and received 75 �l daily of tapwater. Treatment in both studies lasted for a total of6 weeks. After the 4th week of treatment, behavioraltesting was performed.

BEHAVIORAL TESTING

T-maze training and testing procedures

The T-maze is both a learning task based onworking-memory and a reference-memory task. TheT-maze consisted of a black plastic alley with a startbox at one end and two goal boxes at the other. The startbox was separated from the alley by a plastic guillo-tine door that prevented movement down the alley untilraised at the onset of training. An electrifiable floor ofstainless steel rods ran throughout the maze to delivera mild scrambled foot-shock.

Mice were not permitted to explore the maze prior totraining. A block of training trials began when a mousewas placed into the start box. The guillotine door wasraised and a cue buzzer sounded simultaneously; 5 slater foot-shock was applied. The arm of the mazeentered on the first trial was designated “incorrect”and the mild foot-shock was continued until the mouseentered the other goal box, which in all subsequenttrials was designated as “correct” for the particularmouse. At the end of each trial, the mouse was returnedto its home cage until the next trial.

Mice were trained until they made 1 avoidance.Training used an intertrial interval of 35 s, the buzzerwas of the door-bell type sounded at 55 dB, and shockwas set at 0.35 mA (Coulbourn Instruments scram-bled grid floor shocker model E13-08). Retention wastested one week later by continuing training until micereached the criterion of 5 avoidances in 6 consecutivetrials. The results were reported as the number of trialsto criterion for the retention test.

Object-place recognition

Object-place recognition is a declarative memorytask that involves the hippocampus when, as performedhere, the retention interval is 24 h after initial exposureto the objects [30]. Mice were habituated to an emptyapparatus for 5 min a day for 3 days prior to entry ofthe objects. During the training session, the mouse wasexposed to two similar objects (plastic frogs) which itwas allowed to examine for 5 min. The apparatus andthe objects were cleaned between each mouse. 24 hlater, the mouse was exposed to one of the originalobjects and a new novel object in a new location andthe percent of time spent examining the new object wasrecorded. The novel object was made out of the samematerial as the original object and of the same size,but a different shape. This eliminated to possibility ofsmell associated with a particular object being a factor.The underlying concept of the task is based on thetendency of mice to spend more time exploring new,novel objects than familiar objects. Thus, the greaterthe retention/memory at 24 h, the more time spent withthe new object.

Strength testing and body weights

Strength testing was performed on the mice in study2. String Hang: A mouse was allowed to grip a stringsuspended 12 inches above a 3/4 inch foam pad. Thetime that a mouse was able to hang on to the string wasrecorded with a maximum hang time of 3 min. CageHang: The mouse was allowed to grip the top of a wiremess cage which was then inverted 12 inches over a3/4 inch foam pad. The time the mouse was able to hangwas recorded with the test ended after a maximum of3 min. Pole Balance: A mouse was placed on a pole1/4 inch in diameter 12 inches over a 3/4 inch foam andthe time the mouse could balance on the pole recorded.The test was stopped after 3 min. Strength Meter: Fore-limb grip strength was measured by allowing the mouseto grasp a wire mesh connected to a grip strength meter

84 S.A. Farr et al. / EVOO Improves Learning and Memory

(Columbus Instruments, Columbus, OH). The valueswere corrected by dividing by body weight in kg.

OXIDATIVE STRESS

Acetonitrile, acetic acid, water, and phosphoric acid(all HPLC grade) were purchased from Fisher Sci-entific (St. Louis, MO). All other chemicals wereobtained from Sigma (St. Louis, MO). N-(1-pyrenyl)-maleimide (NPM) was purchased from Aldrich (Mil-waukee, WI).

HPLC system

The HPLC system (Thermo Electron Corpora-tion) consists of a Finnigan TM Spectra SYSTEMSCM1000 Vacuum Membrane Degasser, Finni-gan TM SpectraSYSTEM P2000 Gradient Pump,Finnigan™ SpectraSYSTEM AS3000 Autosampler,and Finnigan™ SpectraSYSTEM FL3000 Fluores-cence Detector (λex = 330 nm and λem = 376 nm). TheHPLC column was a Reliasil ODS-1 C18 column (Col-umn Engineering, Ontario, CA, USA).

Glutathione determination

Glutathione (GSH) concentrations were determinedby reverse phase HPLC according to the method devel-oped by Winters et al. [31]. Briefly, brain tissue sampleswere homogenized in serine-borate buffer (100 mMTris-HCl, 10 mM boric acid, 5 mM L-serine, 1 mMDETAPAC (diethylenetriaminepentaacetic acid), pH7.4) on ice. Brain tissue homogenates were deriva-tized with 1.0 mM NPM (N-(1-pyrenyl)-maleimide) inacetonitrile. Briefly, sufficient HPLC grade water wasadded to each sample to make a volume of 250 �l andthen 750 �l NPM (1 mM in acetonitrile) added. Thismixture was incubated for 5 min at room temperaturethen acidified with 2 N HCl. The derivatized sampleswere filtered through a 0.2-�m acrodisc filter (Advan-tec MFS, Inc. Dublin, CA, USA) and injected onto a5-�m C18 column (Column Engineering, Ontario, CA,USA) in a reverse-phase HPLC system. The mobilephase was 70% acetonitrile and 30% water and wasadjusted to a pH 2.5 through the addition of 1 ml/L ofboth acetic and o-phosphoric acids. The NPM deriva-tives were eluted from the column isocratically at aflow rate of 1 ml/min [31, 32].

All sample quantitation was determined from stan-dard curves using purified glutathione as described[31].

Lipid peroxidation determination by HPLC

Malondialdehyde (MDA), a byproduct of lipidperoxidation, content was determined according themethod described by Draper et al. [33]. Briefly, 550 �lof 5% tricholoroacetic acid (TCA) and 550 �l of500 ppm butylatedhydroxytoluene (BHT) in methanolwere added to brain homogenates. The samples werethen heated in a boiling water bath for 30 min. Aftercooling on ice, the samples were centrifuged. Thesupernatant fractions were mixed 1 : 1 with saturatedaqueous thiobarbituric acid (TBA) solution and placedin a boiling water bath for an additional 30 min.After cooling on ice, 0.50 ml of each sample wasextracted with 1 ml n-butanol and centrifuged to facil-itate the separation phases. The resulting organiclayers were first filtered through 0.45 acrodisc andthen chromatographed as described above. The con-centration of the thiobarbituric acid-malondialdehyde(TBA-MDA) complex in samples was determined byusing the calibration curve obtained from a 1,1,3,3-tetraethoxypropane standard solution.

Glutathione peroxidase activity

Glutathione peroxidase (GPx) activity was deter-mined using a test kit (OxisResearch, Portland,Oregon, USA). The GPx assay is an indirect measureof the activity of cellular GPx. The oxidized form ofglutathione (GSSG), produced upon reduction of anorganic peroxide by cellular GPx, is recycled to itsreduced state by the enzyme GR. To assay cellularGPx, a cell sample is added to a solution containingGSSG, GR, and NADPH. The oxidation of NADPH toNADP+ is accompanied by a decrease in absorbancefor 3 min at 340 nm, providing a spectrophotometricmeans for monitoring GPx enzyme activity followingaddition of 350 �l of tert-butyl hydroperoxide as theworking substrate.

Glutathione reductase activity

Glutathione reductase (GR) activity was determinedusing a commercial kit from OxisResearch (Portland,Oregon, USA). The assay is based on the oxidation ofNADPH to NADP+ catalyzed by a limiting concentra-tion of glutathione reductase. One GR activity unit isdefined as the amount of enzyme catalyzing the reduc-tion of one micromole of GSSG per minute at pH 7.6and 25◦C.

S.A. Farr et al. / EVOO Improves Learning and Memory 85

Superoxide dismutase activity determination

Total superoxide dismutase (SOD) activity was mea-sured according to the McCord et al. method [34] basedon the production of superoxide radicals during theconversion of xanthine to uric acid by xanthine oxi-dase and the inhibition of cytochrome C reduction. Oneunit (U) of SOD activity was defined as the amount ofSOD that produces 50% inhibition of cytochrome Creduction. The calculated SOD activity was expressedas U/mg protein in the tissue.

Protein determination

The Bradford method was used to determine theprotein content of brain tissue samples using concen-trated Coomassie Blue (Bio-Rad, Hercules, CA, USA)on optical density determinations at 595 nm [35]. Astandard curve using bovine serum albumin was con-structed. The homogenized tissues were subject toappropriate dilutions before protein was determined.

Western blot assessment of 4-hydroxynonealprotein adducts and 3-NT formation

The tissue samples were weighed and frozen at−80◦C until they were processed. Tissue sampleswere then homogenized on ice in 5 volumes (5 × theweight in grams) of a buffer containing 20 mM TrisHCl (pH 7.4), 0.15 M NaCl, 2 mM EDTA, 1 mMEGTA, 0.5% Triton X-100 and a protease inhibitorcocktail (Sigma, P2714). To remove cellular debris,the samples were centrifuged at 1,000 × g for 10 minat 4◦C and the supernatant saved. The supernatantsamples were allowed to shake for 30 min at 4◦Cfollowed by centrifugation at 20,000 × g for 40 minat 4◦C. The supernatants were collected into freshtubes. Samples of 25 � g total protein were heated at95◦C for 5 min after using a reducing agent (Invitro-gen NP0009) and then electrophoresed in NuPAGENovex 4–12% Bis-Tris precast gels with the aid ofan XCell SureLock Mini-Cell gel running apparatus(Invitrogen, Carlsbad, CA). Proteins were transferredfrom the gel onto nitrocellulose membranes (0.45 �mpore size). The membranes were washed in TBS-T(Tris-buffered saline, 10 mM Tris HCl, 150 mM NaCl,pH = 7.5) supplemented with 0.05% Tween 20, andincubated in freshly prepared blocking buffer (5% non-fat dry milk in TBS-T) for 1 h at room temperature.To probe the membrane with anti-hydroxynonenal(HNE) antibody (Anti-HNE; HNE11-S Alpha Diag-nostics, San Antonio, TX) (1 : 2,000), the 5% milk

solution was removed and the membrane was incu-bated with Anti-HNE in 5% milk (diluted in TBS-T)overnight at 4◦C. After three 5 min washes in TBS-T,the membranes were probed with the secondary anti-body (horseradish peroxidase-linked goat anti-rabbitIgG; 1 : 10,000; Santa Cruz Biotechnologies, SantaCruz, CA) in 5% milk solution diluted in TBS-T for1 h at room temperature. The blots were finally washedthree times (5 min each) with TBS-T and a 1 : 1 mix-ture of Supersignal West Pico Stable Peroxide Solutionand Supersignal West Pico Luminol/Enhancer Solu-tion (Pierce, Rockford, IL) was added. Specific proteinbands were visualized by exposure to BioMax XARScientific Imaging Film (Kodak) and optical den-sity was quantified using Image J analysis software(National Institute of Health, USA). Optical densityvalues of individual protein-bound HNE bands withineach lane were summed to obtain a total value. Tovalidate equal protein loading among various lanes,nitrocellulose membranes were stripped and reprobedwith �-actin antibody (Sigma, St.Louis, MO).

3-NT-modified proteins were detected with the samemethod as HNE-modified proteins, with a few alter-ations. The primary anti-nitrotyrosine antibody (3-NT)was a polyclonal rabbit antibody (AB5411, Chemicon-Millipore, Billerica, MA) (1 : 1,000). The secondaryantibody was a polyclonal anti-rabbit IgG antibody(horseradish peroxidase-linked goat anti-rabbit IgG;1 : 10,000; Santa Cruz Biotechnologies, Santa Cruz,CA).

Statistics

Results were analyzed using analysis of variance(ANOVA) to examine the effect among groups. Themeasure of acquisition and retention in the t-maze werethe number of trials to reach criterion. The results forobject recognition are presented in percent time spentexploring the novel object out of total exploration time.Results are expressed as means with their standarderrors. Tukey’s post hoc analysis was used to comparemeans between groups.

RESULTS

Comparison of butter, coconut oil, and EVOO onT-maze foot shock avoidance in 12 month oldSAMP8 mice

In the first study, we examined the effect of butter,coconut oil, and EVOO on T-maze foot shock avoid-ance. The one-way ANOVA with trials to criterion as

86 S.A. Farr et al. / EVOO Improves Learning and Memory

the independent variable showed a significant treat-ment effect F(2,26) = 12.41, p < 0.001. Tukey’s posthoc analysis indicated that the mice that receivedEVOO took significantly fewer trials to reach criterionthan the mice that received butter or coconut oil (seeFig. 1A). A one-way ANOVA for trials to criterion onthe retention test showed a significant treatment effectF(2,25) = 5.28, p < 0.01. Tukey’s post hoc analysisindicated that the mice that received EVOO took sig-nificantly fewer trials to reach criterion than the micethat received butter. There was no difference betweenthe mice that received coconut oil and the mice whichreceived butter (see Fig. 1B).

Effect of EVOO on object-place recognition

The one-way ANOVA for object-place recognitionproduced a significant effect for treatment F(2,26)= 5.567, p < 0.01. Post hoc analysis indicated that themice that received EVOO spent significantly moretime exploring the novel object than the mice thatreceived butter and the mice that received coconutoil (see Fig. 1C). The one-way ANOVA for total

exploration time during acquisition was not significantF(2,26) = 0.3205, p ns. The mean total exploration timefor each group was butter 28.56 ± 5.04, coconut oil24.20 ± 3.19, and EVOO 28.44 ± 5.84. The one-wayANOVA for total exploration time during the 24 hoursretention test was not significant F(2,26) = 0.8686,p ns. The mean total exploration time for each groupwas butter 18.78 ± 2.77, coconut oil 19.60 ± 4.38, andEVOO 26.60 ± 6.44.

Effect of EVOO on food intake and body weight

The one-way ANOVA for body weight change fromthe start of treatment to the end of treatment was notsignificant F(2,25) = 2.02, p ns. The mice that receivedbutter had a mean change of −1.52 gm, the mice thatreceived coconut oil had a mean change of −0.47 gmand the mice that received EVOO had a mean changeof −0.47 gm. The one-way ANOVA for daily foodintake was not significant F(2,25) = −0.81, p ns. Themice that received butter had a mean daily food intakeover the course of the study of 5.74 ± 0.34, coconut oil5.17 ± 0.28, and EVOO 5.16 ± 0.25.

A B

C

Fig. 1. SAMP8 mice administered extra virgin olive oil (EVOO) and coconut oil daily for 4 weeks had improved acquisition of T-maze footshock avoidance compared to the mice that received butter (A). SAMP8 mice that received EVOO had improved retention when tested oneweek after training (B). SAMP8 mice that received chronic EVOO had improved memory in object recognition when the retention test wasadministered 24 h after acquisition (C). *p < 0.05, **p < 0.01.

S.A. Farr et al. / EVOO Improves Learning and Memory 87

A

B

Fig. 2. SAMP8 mice administered coconut oil and EVOO had sig-nificantly greater levels of GSH than the mice that received butter(A). There were no significant differences between groups in MDAlevels (B). *p < 0.05.

Effects of EVOO on GSH and MDA levels.

The one-way ANOVA for GSH levels produceda significant effect for treatment F(2, 25) = 12.00,p < 0.001 (see Fig. 2A). Post hoc analysis indicatedthat the mice that received coconut oil and EVOO hadsignificantly greater levels of GSH than the controlmice. There was no difference in GSH levels betweenthe mice that received coconut oil and the mice thatreceived EVOO. The one-way ANOVA for MDA levelswas not significant F(2, 21) = 0.47, p ns. (see Fig. 2B).

Effect of EVOO with enhanced concentrations ofpolyphenols on T-maze

The one-way ANOVA for trials to criterion on acqui-sition showed a significant effect for treatment F(4,45)= 4.814, p < 0.003. Tukey’s post hoc analysis indicated

that the mice that received the EVOO enhanced with2×,3×, and 5xs the polyphenols took significantlyfewer trials than the mice that received regular diet(see Fig. 3A). There were no differences between anyof the other groups. The one-way ANOVA for trialsto criterion on the retention test produced a significanteffect for treatment F(4,46) = 6.092, p < 0.001. Tukey’spost hoc analysis indicated that the mice that receivedEVOO at 3× and 5× the concentration of polyphenolstook significantly fewer trials to reach criterion thanthe mice that received regular diet (see Fig. 3B).

Effect of EVOO with enhanced polyphenols onobject-place recognition

The one-way ANOVA for percent time exploringthe novel object on the retention test showed a sig-nificant effect for treatment F(3,38) = 3.13, p < 0.05.Tukey’s post hoc analysis indicated that the micethat received the 2× or 5× -enriched EVOO spentsignificantly more time with the novel object com-pared to the mice that received unenriched EVOO(see Fig. 3C). The ANOVA for total exploration timeon day one was not significant F(3,38) = 1.874, pns. The mean total exploration time for each groupwas regular EVOO 21.70 ± 3.55, 2 × 23.60 ± 3.95,3 × 20.20 ± 3.29, and 5 × 32.80 ± 6.11. The ANOVAfor total exploration time on the 24 h retention testwas not significant F(3,38) = 1.532. The mean totalexploration time for each group was regular EVOO17.10 ± 1.97, 2 × 20.14 ± 3.65, 3 × 12.90 ± 2.83, and5 × 23.30 ± 5.57.

Effect of enriched EVOO on strength

The one-way ANOVA for string hang was not sig-nificant F(4,29) = 0.7678, p ns. The one-way ANOVAfor pole balance was not significant F(4,29) = 0.9627,p ns. The one-way ANOVA for cage hang was not sig-nificant F(4,29) = 0.3332, p ns. The one-way ANOVAfor strength was not significant F(4,29) = 2.314, p ns.

Effects of enriched EVOO on GSH, GR, SOD,HNE, and 3-NT

The one-way ANOVA for GSH levels showed a sig-nificant effect for treatment F(4,40) = 5.74, p < 0.001.Post hoc analysis indicated that mice that receivedany of the EVOOs had significantly higher levels ofGSH than the control mice (see Fig. 4A). The one-way ANOVA for GR levels showed a significant effectfor treatment F(4,36) = 5.47, p < 0.005. Tukey’s post

88 S.A. Farr et al. / EVOO Improves Learning and Memory

A B

C

Fig. 3. SAMP8 mice given chronic EVOO with 2, 3, or 5 times the polyphenolic concentration of regular EVOO had improved acquisitioncompared to the mice on regular rodent chow (A). SAMP8 mice that received chronic EVOO with 3 or 5 times the polyphenol content hadimproved retention in T-maze on the one-week retention test (B). The mice that received chronic EVOO with 2 or 5 times the polyphenolconcentration had improved memory in object recognition test (C). *p < 0.05, **p < 0.01.

hoc analysis showed that the mice which receivedunenriched EVOO or 3×- or 5×-enriched EVOO hadsignificantly greater GR activity than the mice thatreceived regular diet (see Fig. 4B).

The one-way ANOVA for SOD produced a signif-icant effect for treatment F(3,36) = 4.101, p < 0.01.Tukey’s post hoc analysis indicated that the mice thatreceived unenriched EVOO or 3×- or 5×-enrichedEVOO had significantly greater SOD activity than themice that received regular diet (see Fig. 4C).

The one-way ANOVA for HNE levels produceda significant effect for treatment F(4,35) = 3.312,p < 0.05. Tukey’s post hoc analysis indicated that themice that received 3×- or 5×-enriched EVOO had sig-nificantly reduced HNE levels compared to the micethat received regular diet (see Fig. 4D).

The one-way ANOVA for 3-NT produced a signif-icant effect for treatment F(4,32) = 2.831, p < 0.05.Tukey’s post hoc analysis indicated that the mice thatreceived 3×- or 5×-enriched EVOO had significantly

greater 3-NT levels than the mice that received regulardiet (see Fig. 4E).

DISCUSSION

The current set of studies shows that EVOO admin-istration is associated with improved learning andmemory in the SAMP8, a mouse model of Alzheimer’sdisease (16–18; 23, 24), when compared to mice fedbutter, a source of animal fat not high in polypheno-lics. Learning was improved in the T-maze whereasmemory was improved in both the T-maze and theobject-place recognition test in mice fed EVOO. Incomparison, SAMP8 mice fed coconut oil, a sourceof vegetable oils that are not high in polyphenolics,produced a significant effect on learning but in nei-ther test of memory. The higher GSH levels suggestedthat the possible mechanism of action was reduction ofoxidative stress. In the second study, mice fed EVOO

S.A. Farr et al. / EVOO Improves Learning and Memory 89

A B

C D

E

Fig. 4. SAMP8 mice administered regular EVOO or any of the enriched EVOOs had significantly greater GSH levels than mice fed regularrodent chow (A). SAMP8 mice administered regular EVOO or a 3×- or 5×-enriched EVOO had significantly greater GR and SOD activity thanthe SAMP8 mice fed regular rodent chow (B and C). SAMP8 fed the 3×- or 5×-enriched EVOO had significantly reduced HNE and 3NT levelscompared to SAMP8 mice fed regular rodent chow (D and E). *p < 0.01, **p < 0.01, ***p < 0.001. These results indicate that EVOO enriched3 to 5 times the polyphenols of regular EVOO improves all indicators of oxidative stress.

that contained 3 or 5 times the amount of polyphenolsfound in unenriched EVOO had improved acquisitionand retention in the T-maze and improved retentionin the novel object-place recognition test. The SAMP8mouse has been shown to have oxidative damage in thebrain [21–23] and cognitive defects have been reversedby antioxidants [24]. On measures of oxidative stressthe mice had increased GSH levels, GR activity, andSOD activity and reduced HNE and 3-NT levels inthe brain indicating that reversal of oxidative dam-age was involved in the improvement of cognition.SAMP8 mice are a spontaneous model for amyloid-�

protein precursor overproduction and have excess A�,learning and memory defects, decreased production ofacetylcholine and defective transport at the blood-brainbarrier limiting egress of A� from the brain [16, 19,36–38].

The Mediterranean diet has been associated withincreased lifespan and decreased incidences of hyper-tension, cardiovascular disease, and cancer [39]. Somestudies have suggested that adhering to a strict Mediter-ranean diet decreases mortality by 50–65% [40]. Inaddition, Mediterranean diets appear to improve cogni-tion (8–10). EVOO is a staple of the Mediterranean diet

90 S.A. Farr et al. / EVOO Improves Learning and Memory

and is rich in antioxidants. Antioxidants are consideredto be the primary reason for these benefits.

EVOO is a natural product particularly rich inmonounsaturated fatty acids (MUFA), mainly oleicacid, a constituent of the biological membranes thatmay progressively substitute polyunsaturated fattyacids (PUFA). Cellular membranes rich in MUFA haveoptimal fluidity and are less prone to lipid peroxi-dation, a process that requires two or more doublelinks to take place [41]. Other elements containedin EVOO are antioxidant molecules, such as alpha-tocopherol, phenolic compounds, and coenzyme Q.These are able to counter the toxics effects of oxy-gen metabolism such as free radical formation and soprotect cells against oxidative damage [42]. Severalstudies demonstrate the ability of olive oil to mod-ify cellular membrane structure and to reduce theiroxidative modifications [43]. Polyphenols in EVOOinclude over 230 chemical substances belonging todifferent classes, such as aliphatic and triterpenic alco-hols, sterols, hydrocarbons and volatile compounds.The antioxidants in EVOO include carotenes, alpha-tocopherol, and hydrophilic and lipophilic phenolicsubstances. It is worth noting that while tocopherolsand carotenes can be bound in other plant and animalfats, the simple polyphenols, such as phenyl acids andphenyl alcohols (e.g., several oleuropein derivatives)are compounds found exclusively in olives and the oilsextracted mechanically from them [25, 44].

Antioxidants are found in many types of foods suchas berries, chocolate, EVOO, green tea, and wine. Stud-ies involving the consumption of these foods haveimproved cognition, reversed oxidative damage, andimproved heart function [7, 11, 45, 46]. Polyphenolsare considered to be potent antioxidants contained inthese foods. Polyphenols also have antiamyloidogeniceffects [15, 47, 48]. A small study suggested that blue-berry supplementation may improve memory in olderhumans [49].

Interestingly, two recent reports showed effects ofone of the phenols found in high concentrations in ourEVOO, the dialdehydic form of deacetoxy-ligstrosideaglycone, on key mediators of Alzheimer’s disease.This phenol was shown in vitro to abrogate fibrilliza-tion of tau by locking tau into the naturally unfoldedstate [50]. Another study showed that this phenolwas able to alter the oligomerization state of sol-uble oligomers of A�1-42 peptide while protectingneurons from the synaptopathological effects of theseoligomers [51]. This is an interesting result which mayhelp to explain mechanistically the positive effects weobserved; however, it would be inaccurate to attribute

our results to a single compound since we used anEVOO with high concentrations of many phenolsand other types of anti-oxidant and anti-inflammatorymolecules.

EVOO is a food that naturally contains polyphe-nols. Studies involving rats administered EVOO withhigh (H-EVOO) and low (L-EVOO) concentrations ofpolyphenols found a protective effect of the H-EVOOon blood pressure in rats fed a high-calorie diet from12 months of age to senescence [14, 52]. In mice, thesenescence-accelerated P1 mouse fed olive oil for 26weeks starting at 6 weeks of age had significantly fewerApoA-II amyloid fibril deposits and anti-single-strandDNA antibodies in the peripheral tissue and blood com-pared to mice fed fish oil or perilla oil [53]. In SAMP8mice, tea was shown to reverse cognitive deficits andmorphological changes in the brain [6]. Here, weshowed that SAMP8 mice fed EVOO had improvedmemory and reversed oxidative damage in the braincompared to mice fed butter, or a regular chow diet.Two polyphenols, ferulic acid and p-coumaric acid,have been identified in coconut oil, perhaps account-ing for the effect of coconut oil on acquisition in theSAMP8 mice [54].

The current studies indicate that chronic con-sumption of EVOO improves cognitive function andoxidative brain damage in aged SAMP8 mice. TheEVOO with enhanced polyphenols improved cogni-tive function and oxidative damage to a greater extentthan regular EVOO. The current studies suggest thatthe polyphenols in EVOO are involved in the cognitiveenhancing ability as well as in the reversal of oxida-tive damage in the SAMP8 mouse model of dementiaassociated with the overproduction of A�. In addition,these studies suggest that adding additional polyphe-nols to EVOO can have even greater benefits on healththan regular EVOO.

ACKNOWLEDGMENTS

This work was supported by VA Medical Center, St.Louis, MO.

Authors’ disclosures available online (http://www.j-alz.com/disclosures/view.php?id=974).

REFERENCES

[1] Trichopoulou A, Critselis E (2004) Mediterranean diet andlongevity. Eur J Cancer Prev 13, 453-456.

[2] Perez-Lopez FR, Chedraui P, Haya J, Cuadros JL (2009)Effects of the Mediterranean diet on longevity and age-relatedmorbid conditions. Maturitas 64, 67-79.

S.A. Farr et al. / EVOO Improves Learning and Memory 91

[3] Guallar-Castillon P, Rodriguez-Artalejo F, Tormo MJ,Sanchez MJ, Rodriguez L, Quiros JR, Navarro C, Molina E,Martinez C, Marin P, Lopez-Garcia E, Larranaga N, HuertaJM, Dorronsoro M, Chirlaque MD, Buckland G, Barricarte A,Banegas JR, Arriola L, Ardanaz E, Gonzalez CA, Moreno-Iribas C (2010) Major dietary patterns and risk of coronaryheart disease in middle-aged persons from a Mediterraneancountry: The EPIC-Spain cohort study. Nutr Metab Cardio-vasc Dis, doi:10.1016/j.numecd.2010.06.004.

[4] Lopez-Miranda J, Perez-Jimenez F, Ros E, De Caterina R,Badimon L, Covas MI, Escrich E, Ordovas JM, Soriguer F,Abia R, de la Lastra CA, Battino M, Corella D, Chamorro-Quiros J, Delgado-Lista J, Giugliano D, Esposito K, EstruchR, Fernandez-Real JM, Gaforio JJ, La Vecchia C, Lairon D,Lopez-Segura F, Mata P, Menendez JA, Muriana FJ, OsadaJ, Panagiotakos DB, Paniagua JA, Perez-Martinez P, Per-ona J, Peinado MA, Pineda-Priego M, Poulsen JE, QuilesJL, Pamirez-Tortosa MC, Puano J, Serra-Majem L, Sola R,Solanas M, Solfrizzi V, de la Torre-Fornell R, TrichopoulouA, Uceda M, Villalba-Montoro JM, Villar-Ortiz JR, VisioliR, Yiannakouris N (2010) Olive oil and health: Summary ofthe II international conference on olive oil and health con-sensus report, Jaen and Cordoba (Spain) 2008. Nutr MetabCardiovasc Dis 20, 284-294.

[5] Calabrese V, Cornelius C, Trovato A, Cavallaro M, MancusoC, Di Rienzo L, Condorelli D, De Lorenzo A, CalabreseEJ (2010) The hormetic role of dietary antioxidants in freeradical-related diseases. Curr Pharm Des 16, 877-883.

[6] Chan YC, Hosoda K, Tsai CJ, Yamamoto S, Wang MF (2006)Favorable effects of tea on reducing the cognitive deficitsand brain morphological changes in senescence-acceleratedmice. J Nutr Sci Vitaminol 52, 266-273.

[7] Papandreou MA, Dimakopoulou A, Linardaki ZI, CordopatisP, Klimis-Zacas D, Margarity M, Lamari FN (2009) Effectof a polyphenol-rich wild blueberry extract on cognitiveperformance of mice, brain antioxidant markers and acetyl-cholinesterase activity. Behav Brain Res 198, 352-358.

[8] Scarmeas N, Stern Y, Mayeux R, Manly JJ, Schupf N,Luchsinger JA (2009) Mediterranean diet and mild cognitiveimpairment. Arch Neurol 66, 216-225.

[9] Feart C, Samieri C, Rondeau V, Amieva H, Portet F, DartiguesJF, Scarmeas N, Barberger-Gateau P (2009) Adherence to aMediterranean diet, cognitive decline, and risk of dementia.JAMA 302, 638-648.

[10] Scarmeas N, Stern Y, Tang MX, Mayeux R, Luchsinger JA(2006) Mediterranean diet and risk for Alzheimer’s disease.Ann Neurol 59, 912-921.

[11] Shukitt-Hale B, Cheng V, Joseph JA (2009) Effects of black-berries on motor and cognitive function in aged rats. NutrNeurosci 12, 135-140.

[12] Shih PH, Chan YC, Liao JW, Wang MF, Yen GC(2010) Antioxidant and cognitive promotion effects ofanthocyanin-rich mulberry (Morus atropurpurea L) onsenescence-accelerated mice and prevention of Alzheimer’sdisease. J Nutr Biochem 21, 598-605.

[13] Assuncao M, Santos-Marques MJ, Carvalho F, LukoyanovNV, Andrade JP (2011) Chronic green tea consumptionprevents age-related changes in rat hippocampal formation.Neurobiol Aging 32, 707-717.

[14] Pitozzi V, Jacomelli M, Zaid M, Luceri C, Bigagli E, LodoviciM, Ghelardini C, Vivoli E, Norcini M, Gianfriddo M, EspostoS, Servili M, Morozzi G, Baldi E, Bucherelli C, Dolara P,Giovannelli L (2010) Effects of dietary extra-virgin olive oilon behaviour and brain biochemical parameters in ageing rats.Br J Nutr 103, 1674-1683.

[15] Ho L, Chen LH, Wang J, Zhao W, Talcott ST, Ono K, TeplowD, Humala N, Cheng A, Percival SS, Ferruzzi M, Janle E,Dickstein Dl (2009) Heterogeneity in red wine polypheno-lic contents differentially influences Alzheimer’s disease-typeneuropathology and cognitive deterioration. J Alzheimers Dis16, 59-72.

[16] Kumar VB, Farr SA, Flood JF, Kamlesh V, Franko M,Banks WA, Morley JE (2000) Site-directed antisense oligonu-cleotide decreases the expression of amyloid precursor proteinand reverses deficits in learning and memory in aged SAMP8mice. Peptides 21, 1769-1775.

[17] Morley JE, Kumar VB, Bernardo AE, Farr SA, Uezu K,Tumosa N, Flood JF (2002) Beta-amyloid precursor polypep-tide in SAMP8 mice affects learning and memory. Peptides21, 1761-1767.

[18] Morley JE, Farr SA, Flood JF (2002) Antibody to amyloid betaprotein alleviates impaired acquisition, retention, and memoryprocessing in SAMP8 mice. Neurobiol Learn Mem 78, 125-138.

[19] Takeda T (2009) Senescence-accelerated mouse (SAM) withspecial references to neurodegeneration models, SAMP8 andSAMP10 mice. Neurochem Res 34, 639-659.

[20] Pallas M, Camins A, Smith MA, Perry G, Lee HG, CasadesusG (2008) From aging to Alzheimer’s disease: Unveiling“the switch” with the senescence-accelerated mouse model(SAMP8). J Alzheimers Dis 15, 615-624.

[21] Butterfield DA, Koppal T, Howard B, Subramaniam R, Hall N,Hensley K, Yatin S, Allen K, Aksenov M, Aksenova M et al.(1998) Structural and functional changes in proteins inducedby free radical-mediated oxidative stress and protective actionof the antioxidants N-tert-butyl-alpha-phenylnitrone and vita-min E. Ann N Y Acad Sci 854, 448-462.

[22] Poon HF, Castegna A, Farr SA, Thongboonkerd V, Lynn BC,Banks WA, Morley JE, Klein JB, Butterfield DA (2004) Quan-titative proteomics analysis of specific protein expression andoxidative modification in aged senescence-accelerated-prone8 mice brain. Neuroscience 126, 915-926.

[23] Poon HF, Farr SA, Thongboonkerd V, Lynn BC, Banks WA,Morley JE, Klein JB, Butterfield DA (2005) Proteomic anal-ysis of specific brain proteins in aged SAMP8 mice treatedwith alpha-lipoic acid: Implications for aging and age-relatedneurodegenerative disorders. Neurochem Int 46, 159-168.

[24] Farr SA, Poon HF, Dogrukol-Ak D, Drake J, Banks WA,Eyerman E, Butterfield DA, Morley JE (2003) The antioxi-dants alpha-lipoic acid and N-acetylcysteine reverse memoryimpairment and brain oxidative stress in aged SAMP8 mice.J Neurochem 84, 1173-1183.

[25] Montedoro GF, Servili M, Baldioli M, Selvaggini R, MiniatiE, Macchioni A (1993) Simple and hydrolyzable compoundsin virgin olive oil. Spectroscopic characterizations of the sec-oiridoid derivatives. J Agric Food Chem 41, 2228-2234.

[26] Beauchamp GK, Keast RS, Morel D, Lin J, Pika J, Han Q, LeeCH, Smith AB, Breslin PA (2005) Phytochemistry: Ibuprofen-like activity in extra-virgin olive oil. Nature 437, 45-46.

[27] Andrewes P, Busch JL, de Joode T, Groenewegen A,Alexandre H (2003) Sensory properties of virgin olive oilpolyphenols: Identification of deacetoxy-ligstroside aglyconas a key contributor to pungency. J Agric Food Chem 51,1415-1420.

[28] Morello JR, Motilva MJ, Tovar MJ, Romero MP (2004)Changes in commercial virgin olive oil (cv Arbequina) duringstorage, with special emphasis on the phenolic fraction. FoodChem 85, 357-364.

[29] Singleton VL, Orthofer R, Lamuela-Raventos RM (1999)Analysis of total phenols and other oxidation substrates and

92 S.A. Farr et al. / EVOO Improves Learning and Memory

antioxidants by means of Folin-Ciocalteu reagent. Meth Enzy-mol 299, 152-178.

[30] Hammond RS, Tull LE, Stackman RW (2004) On thedelay-dependent involvement of the hippocampus in objectrecognition memory. Neurobiol Learn Mem 82, 26-34.

[31] Ridnour LA, Winters RA, Ercal N, Spitz DR (1999) Measure-ment of glutathione, glutathione disulfide, and other thiols inmammalian cell tissue homogenates using high-performanceliquid chromatography separation of N-(1-pyrenyl)maleimidederivatives. Methods Enzymol 299, 258-267.

[32] Winters RA, Zukowski J, Ercal N, Matthews RH, SpitzDR (1995) Analysis of glutathione, glutathione disulfide,cysteine, homocysteine, and other biological thiols by high-performance liquid chromatography following derivatizationby n-(1-pyrenyl)maleimide. Anal Biochem 227, 14-21.

[33] Draper HH, Squires EJ, Mahmoodi H, Wu J, Agarwal S,Hadley M (1993) A comparative evaluation of thiobarbituricacid methods for the determination of malondialdehyde inbiological materials. Free Radic Biol Med 15, 353-363.

[34] McCord JM, Keele BB Jr, Fridovich I (1971) An enzyme-based theory of obligate anaerobiosis: The physiologicalfunction of superoxide dismutase. Proc Natl Acad Sci U S A68, 1024-1027.

[35] Bradford MA (1976) Rapid and sensitive method for the quan-titation of microgram quantities of protein utilizing of theprinciple of protein-dye binding. Anal Biochem 72, 248-256.

[36] Banks WA, Kumar VB, Farr SA, Nakaoke R, Robinson SM,Morley JE (2011) Impairments in brain-to-blood transportof amyloid-� and reabsorption of cerebrospinal fluid in ananimal model of Alzheimer’s disease are reversed by antisensedirected against amyloid-� protein precursor. J AlzheimersDis 23, 599-605.

[37] Farr SA, Banks WA, Uezu K, Sano A, Gaskin FS, MorleyJE (2003) Antibody to beta-amyloid protein increases acetyl-choline in the hippocampus of 12 month SAMP8 male mice.Life Sci 73, 555-562.

[38] Banks WA, Robinson SM, Verma S, Morley JE. Efflux ofhuman and mouse amyloid beta proteins 1-40 and 1-42 frombrain: Impairment in a mouse model of Alzheimer’s disease.Neuroscience 121, 487-492.

[39] Martinez-Gonzalez MA, Bes-Rastrollo M, Serra-Majem L,Lairon D, Estruch R, Trichopoulou A. Mediterranean foodpattern and the primary prevention of chronic disease: recentdevelopments. Nutr Rev 67 (Suppl 1), S111-S116.

[40] Imoberdorf R, Ruhlin M, Ballmer PE (2005) (Eating anddrinking during aging) (2005) Ther Umsch 62, 847-851.

[41] Scislowski V, Bauchart D, Gruffat D, Laplaud PM, DurandD (2005) Effect of dietary n-6 and n-3 polyunsaturated fattyacids on peroxidizability of lipoproteins in steers. Lipids 40,1245-1256.

[42] Visioli F, Galli C (2002) Biological properties of olive oilphytochemicals. Crit Rev Food Sci Nutr 42, 209-221.

[43] Frankel EN (2011) Nutritional and biological properties ofextra virgin olive oil. J Agric Food Chem 59, 785-792.

[44] Servili M, Selvaggini R, Esposto S, Taticchi A, MontedoroG, Morozzi G (2004) Health and sensory properties of virginolive oil hydrophilic phenols: Agronomic and technologicalaspects of production that affect their occurrence in the oil. JChromatogr 1054, 113-127.

[45] Assuncao M, Santos-Marques MJ, Carvalho F, Andrade JP(2010) Green tea averts age-dependent decline of hippocam-pal signaling systems related to antioxidant defenses andsurvival. Free Radic Biol Med 48, 831-838.

[46] Huertas JR, Martinez-Velasco E, Ibanez S, Lopez-Frias M,Ochoa JJ, Quiles J, Parenti Castelli G, Mataix J, Lenaz G(1999) Virgin olive oil and coenzyme Q10 protect heart mito-chondria from peroxidative damage during aging. Biofactors9, 337-343.

[47] Ono K, Yoshiike Y, Takashima A, Hasegawa K, Naiki H,Yamada M (2003) Potent anti-amyloidogenic and fibril-destabilizing effects of polyphenols in vitro: Implications forthe prevention and therapeutics of Alzheimer’s disease. J Neu-rochem 87, 172-181.

[48] Wang YJ, Thomas P, Zhong JH, Bi FF, Kosaraju S, Pol-lard A, Fenech M, Zhou XF (2009) Consumption of grapeseed extract prevents amyloid-beta deposition and attenuatesinflammation in brain of an Alzheimer’s disease mouse. Neu-rotox Res 15, 3-14.

[49] Krikorian R, Shidler MD, Nash TA, Kalt W, Vinqvist-Tymchuk MR, Shukitt-Hale B, Joseph JA (2010) Blueberrysupplementation improves memory in older adults. J AgricFood Chem 58, 3996-4000.

[50] Li W, Sperry JB, Crowe A, Trojanowski JQ, Smith AB, 3rd,Lee VM (2009) Inhibition of tau fibrillization by oleocanthalvia reaction with the amino groups of tau. J Neurochem 110,1339-1351.

[51] Pitt J, Roth W, Lacor P, Smith AB, 3rd, BlankenshipM, Velasco P, De Felice F, Breslin P, Klein WL (2009)Alzheimer’s-associated Abeta oligomers show altered struc-ture, immunoreactivity and synaptotoxicity with low doses ofoleocanthal. Toxicol Appl Pharmacol 240, 189-197.

[52] Jacomelli M, Pitozzi V, Zaid M, Larrosa M, Tonini G, Mar-tini A, Urbani S, Taticchi A, Servili M, Dolara P, Jacomelli M(2010) Dietary extra-virgin olive oil rich in phenolic antiox-idants and the aging process: Long-term effects in the rat. JNutr Biochem 21, 290-296.

[53] Umezawa M, Higuchi K, Mori M, Matushita T, Hosokawa M(2009) Effect of dietary unsaturated fatty acids on senile amy-loidosis in senescence-accelerated mice. J Gerontol A Biol SciMed Sci 64, 646-652.

[54] Marina AM, Man YB, Nazimah SA, Amin I (2009) Antiox-idant capacity and phenolic acids of virgin coconut oil. Int JFood Sci Nutr 60 (Suppl 2), 114-123.