Neurobiology of Disease 45 (2012) 529–538

Contents lists available at SciVerse ScienceDirect

Neurobiology of Disease

j ourna l homepage: www.e lsev ie r .com/ locate /ynbd i

High-fat diet exacerbates MPTP-induced dopaminergic degeneration in mice

M. Bousquet a,b, I. St-Amour a,b,c, M. Vandal a,b, P. Julien d, F. Cicchetti a,e,⁎,1, F. Calon a,b,f,⁎⁎,1

a Centre de Recherche du CHUL (CHUQ), Axe Neurosciences, Québec, QC, Canadab Faculté de Pharmacie, Université Laval, Québec, QC, Canadac Département de Recherche et Développement, Héma-Québec, Québec, Canadad Centre de Recherche sur les Maladies Lipidiques, CHUL (CHUQ), Québec, QC, Canadae Département de Psychiatrie & Neurosciences, Faculté de Médecine, Université Laval, Québec, QC, Canadaf Institut des nutraceutiques et des aliments fonctionnels (INAF), Université Laval, Québec, QC, Canada

Abbreviations: ARA, arachidonic acid; BDNF, brain-decontrol diet; DA, dopamine; DAT, dopamine transporteDOPAC, 3,4-dihydroxyphenylacetic acid; DPA, docosapetraenoic acid; GDNF, glial cell line-derived neurotrophacidic protein; GM-CSF, granulocyte macrophage colhigh-fat diet; IFN, interferon; IL, interleukin; LA, linoleictactic protein; MIP, macrophage inflammatory protei1,2,3,6-tetrahydropyridine; n−3/n−6 PUFA, omegafatty acid; PD, Parkinson's disease; PDS-95, post-synaptnigra pars compacta; TH, tyrosine hydroxylase; TNF, tum⁎ Correspondence to: F. Cicchetti, Centre de Rech

Neurosciences, T2-05, 2705, boulevard Laurier, QuFax: +1 418 654 2753.⁎⁎ Correspondence to: F. Calon, CentredeRechercheduC

T2-05, 2705, boulevard Laurier, Québec, QC, Canada G1V 4GE-mail addresses: Francesca.Cicchetti@crchul.ulaval.c

Frederic.Calon@crchul.ulaval.ca (F. Calon).1 Co-senior authors.

Available online on ScienceDirect (www.scienced

0969-9961/$ – see front matter © 2011 Elsevier Inc. Alldoi:10.1016/j.nbd.2011.09.009

a b s t r a c t

a r t i c l e i n f o

Article history:Received 26 June 2011Revised 31 August 2011Accepted 13 September 2011Available online 21 September 2011

Keywords:Parkinson's diseaseObesityInsulin resistanceDopamineFlow cytometryCytokinesLeukocytesDopaminergic neuronsNutrition

The identification of modifiable nutritional risk factors is highly relevant to the development of preventive strate-gies for neurodegenerative disorders including Parkinson's disease (PD). In this study, adult C57BL/6micewere fedeither a control (CD — 12%kcal) or a high-fat diet (HFD — 60%kcal) for 8 weeks prior to MPTP exposure, a toxinwhich recreates a number of pathological features of PD. HFD-fedmice significantly gainedweight (+41%), devel-oped insulin resistance and a systemic immune response characterizedby an increase in circulating leukocytes andplasmatic cytokines/chemokines (interleukin-1α, MCP-1, MIP-1α). As expected, the MPTP treatment producednigral dopaminergic degeneration as evidenced by the loss of striatal dopamine and the decreased number ofnigral tyrosine hydroxylase (TH)- and dopamine transporter-expressing neurons (23% and 25%, respectively).However, exposure to HFD exacerbated the effects of MPTP on striatal TH (23%) and dopamine levels (32%), indi-cating that diet-induced obesity is associated with a reduced capacity of nigral dopaminergic terminals to copewith MPTP-induced neurotoxicity. Since high-fat consumption is commonplace in our modern society, dietaryfat intake may represent an important modifiable risk factor for PD.

rived neurotrophic factor; CD,r; DHA, docosahexaenoic acid;ntaenoic acid; DTA, docosate-ic factor; GFAP, glial fibrillaryony stimulating factor; HFD,acid; MCP, monocyte chemo-

n; MPTP, 1-methyl-4-phenyl--3/omega-6 polyunsaturatedic protein-95; SNpc, substantiaor necrosis factor.

erche du CHUL (CHUQ), Axeébec, QC, Canada G1V 4G2.

HUL (CHUQ), AxeNeurosciences,2. Fax: +1 418 654 2761.a (F. Cicchetti),

irect.com).

rights reserved.

© 2011 Elsevier Inc. All rights reserved.

Introduction

Parkinson's disease (PD) is a neurodegenerative disorder affectingover 1% of the population over 60 years of age (de Lau and Breteler,2006). A progressive loss of dopaminergic neurons within the substantianigra pars compacta (SNpc) accompaniedwith a drastic decrease in stria-tal dopamine (DA) are part of the pathological features discernible in the

brain of PD patients (Calon et al., 2003; Damier et al., 1999; Lees et al.,2009). The dopaminergic neurodegeneration seemingly leads to impor-tant motor perturbations such as resting tremors, bradykinesia, akinesiaand muscular rigidity. PD is also characterized by non-motor manifesta-tions such as autonomic system failure, depression and dementia, forwhich the pathophysiology is still unclear (Chaudhuri and Schapira,2009; Lees et al., 2009). For the vast majority of PD patients, the originof neurodegeneration remains unknown and less than 15% of PD casescan be attributed to specific genetic mutations as identified in α-synu-clein (SNCA), LRRK2, PINK1, parkin and DJ1 genes (Meissner et al., 2011).A number of environmental risk factors have been suggested to play akey role in the development of PD but because they are readily modifi-able, nutritional factors have received considerable attention over theyears. Several preclinical and epidemiological studies have indeed under-lined the impact of nutrients such as dairy products, numerous vitamins,caffeine and iron on PD pathogenesis despite the fact that some of theseassociations are still being debated (Chen et al., 2007; Costa et al., 2010;Etminan et al., 2005; Powers et al., 2009). In line with the present study,we have previously reported the neuroprotective effects of a long-termhigh intake of omega-3 polyunsaturated fatty acid (n−3 PUFA) in amouse model of PD (Bousquet et al., 2008; Calon and Cicchetti, 2008).

Increasing evidence suggest that diet-induced weight gain is at theroot of a plethora of health problems (Colditz et al., 1995; Felson,1996;Willett et al., 1995, 1999).While obesity andmetabolic disorders,

Table 1Diet composition

Content Diets

CD HFD

Protein (% w/w) 20.3 27.4Carbonhydrate (% w/w) 66.0 25.3Fat (% w/w) 5.0 35.1Calories (kcal/g) 3.9 5.3

Ingredients (g/kg)Casein 200.0 200.0DL-methionine 3.0 3.0Corn starch 150.0 125.0Sucrose 500.0 52.5Cellulose, BW200 50.0 50.0Corn oil 30.0 0.0Safflower oil 0.0 125.0Soybean oil 10.0 0.0Lard 0.0 135.0Canola oil 10.0 0.0Cholesterol, USP 0.6 3.1

Fatty acid content as determined by gas chromatography (g/kg)n−3 PUFAs 2.3 1.5n−6 PUFAs 25.5 111.9n−6/n−3 PUFA ratio 11.1 74.0

530 M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

such as diabetes, are well known outcomes of unhealthy high-fat diet(HFD), more recent studies have added neurodegenerative diseases tothe list of possible deleterious consequences of high-fat consumption(Bayer-Carter et al., 2011; Gao et al., 2007; Hiltunen et al., 2011; Julienet al., 2010; Mody et al., 2011). For example, retrospective and prospec-tive epidemiological studies suggest an associationwith ahigh consump-tion of saturated fat from animalmeat and a higher risk of developing PD(Gao et al., 2007). Obesity, a predictable effect of high-fat intake, has alsobeen studied in relation to PD in prospective epidemiological studies.While a significant weight loss is observed 2 to 4 years prior to diagnosis(Chen et al., 2003), a greater bodymass index (BMI),midlife triceps skin-fold thickness, waist circumference and waist-to-hip ratio have beenlinked to increased risk of developing PD in never smokers (Abbottet al., 2002; Chen et al., 2004; Hu et al., 2006). Other studies, however,have not reported a significant link between obesity and PD (Logroscinoet al., 2007). With a prevalence increasing with age (Wild et al., 2004),diet-induced obesity often leads to metabolic disorders, including typeII diabetes, and chronic peripheral inflammation (Das, 2001). Interest-ingly, a recently published prospective study evaluating the relation-ship between diabetes and the development of PD found an overall41% increased risk (odd ratio) among individuals who have sufferedfrom diabetes for 10 years or more (Xu et al., 2011).

High-fat intake in the general population and resultingmetabolic pe-ripheral perturbations arewidespread. Since high-fat consumption couldbe a modifiable risk factor, it is critical to further investigate their effectson the development of neurodegenerative diseases. We evaluated theimpact of a 2-month high intake in calories from fat on insulin resistanceand immune systemic response as well as in the brain of a mousemodelof PD. Our main a priori hypothesis was that a HFD exacerbates the ef-fects of the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine(MPTP) in mice compared to mice fed a control diet (CD).

Materials and methods

Experimental design, diet and MPTP treatment

Mice were housed in groups of 3 to 4 per cage and were kept understandard conditions throughout the experiments, with free access tofood and water. All procedures were approved by the Laval UniversityAnimal Welfare Committee (#09-098). Forty-nine C57BL/6 male micewere either fed a HFD (60% of calories) or a CD (12% of calories) from2months of age (corresponding to adulthood) until the time of sacrifice(see Table 1 for complete description of the diets) (Julien et al., 2010).The MPTP treatment was performed at 4 months of age and consistedof 7 i.p. injections of a MPTP·HCl solution (20 mg/kg, free base; Sigma,St. Louis, MO) freshly dissolved in 0.9% saline. MPTP was administeredtwice on thefirst two days of the experimental protocol at 12-h intervals,and once a day on the three subsequent days, as previously described(Bousquet et al., 2008; Gibrat et al., 2009). Remaining animals received0.9% saline i.p. in replacement of the MPTP injections. To ensure equiva-lent dosage of the neurotoxin to eachmouse and to prevent any lethal in-toxication of obese mice, MPTP doses were calculated for each mousewith respect to the body surface area (BSA) instead of body weight(Cheung et al., 2009). Briefly, the BSA is described as: k*m0.667, m beingtheweight (g) of themouse and k (Meeh constant) is determined empir-ically as 9.822 for the C57B/L6 mouse strain (Cheung et al., 2009). Themean weight of the mice on the control diet was 31 g for a calculatedBSA of 97.04 (BSAmean); these mice received 0.62 mg MPTP (on a20 mg/kg basis). The BSA of all mice was determined prior to calculationof MPTP doses (mg) using this formula: BSA×0.62/BSAmean.

Insulin sensitivity assessment and plasmatic insulin concentration

Insulin sensitivity was evaluated after 8 weeks of diet. Mice wereinjected with human insulin (1 g/kg) after 6 h of fasting. Blood glucosewasmonitored (OneTouch UltraMini, LifeScan, Milpitas, CA, USA) at 15,

30, 45, 60 and 90 min post-insulin injection. Plasma insulin was mea-sured using an Ultrasensitive Insulin ELISA assay (Mercodia, Uppsala,Sweden) according to the manufacturer's protocol.

Flow cytometry analyses and Q-plex cytokine assay

Blood was collected via the saphenous vein using EDTA-coatedcapillary tubes. Whole blood cell populations were analyzed usingPE conjugated-anti-CD45, AF647 conjugated anti-CD3, AF647conjugated-anti-CD4 and PE conjugated anti-CD8 for T cells, PE-Cy7conjugated anti-B220 for B cells and AF488 conjugated anti-Gr1 forgranulocytes detection (all antibodies are from eBioscience, SanDiego, CA). Briefly, 5 μl of whole blood cells was stained for 15 minat room temperature in 800 μl PBS with 1% fetal bovine serum (Ther-mo Scientific Hyclone, Logan, UT) and immediately analyzed by flowcytometry using a CyFlow ML (Partec, Swedesboro, NJ). Serum cyto-kine concentrations (1:2 plasma dilution) were evaluated with anELISA-based fully quantitative Q-Plex™Mouse Cytokine-Screen (16-plex — assay detection limit: interleukin (IL)-1α — 14 pg/ml, IL-1β — 50 pg/ml, IL-2 — 16 pg/ml, IL-3 — 14 pg/mg, IL-4 — 14 pg/ml,IL-5 — 12 pg/ml, IL-6 — 12 pg/ml, IL-10 — 8 pg/mg, IL-12 —

28 pg/ml, IL-17— 32 pg/ml, interferon (IFNγ)— 32 pg/ml, tumor ne-crosis factor (TNFα) — 22 pg/ml, macrophage inflammatory protein(MIP-1α) — 12 pg/ml, monocyte chemotactic protein (MCP-1) —

12 pg/ml, granulocyte macrophage colony stimulating factor (GM-CSF) — 14 pg/ml and RANTES— 30 pg/ml) from Quansys Biosciences(Logan, UT) according to the manufacturer's instructions.

Tissue preparation for post-mortem analyses

Fourteen days following the last MPTP injection, animals were sacri-ficed under deep anesthesiawith ketamine/xylazine and perfused via in-tracardiac infusion with 1X PBS containing protease inhibitors (Sigma)and phosphatase inhibitors (1 mM tetrasodium pyrophosphate and50 mM sodium fluoride). Brains were collected, the frontal cortex wasdissected for fatty acid analyses and the two hemispheres were separat-ed. The left hemisectionwas snap-frozen in 2-methyl-butane and storedat −80 °C for cryostat coronal brain sections of 20 μm for high perfor-mance liquid chromatography (HPLC) analysis. The right hemisectionwas separated at the level of bregma −1.70 mm, the caudal sectionwas post-fixed in 4% paraformaldehyde pH 7.4, while the striatum con-tained in the rostral section was dissected for Western immunoblotting.

531M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

Lipid extraction and gas chromatography

Approximately 20 mg of frozen frontal cortex tissue from eachmousewas used for fatty acid profile analyses. The frontal cortex was selectedbecause it isminimally affected byMPTP treatment and allowedus to iso-late the effect of the diet on brain fatty acid profiles (Bousquet et al., 2008,2011). Weighed brain tissues were homogenized successively with 0.9%NaCl, butylhydroxytoluene-methanol (Sigma, St. Louis, MO) and chloro-form (J.T. Baker, Phillipsburg, NJ) with 22:3n-3 methyl ester as internalstandard (Nu-Chek Prep, Elysian, MN) at a concentration of 500 μg/g oftissue. After centrifugation at 2400 g for 7 min, the lower layer was col-lected (Folch et al., 1957). This procedure was repeated and the two ex-tracts were pooled and dried under a stream of N2. Lipid extracts weretransmethylated with methanol:benzene (4:1) and acetyl chloride at98 °C for 90 min. After cooling down, 6% K2CO3was added. A 15-min cen-trifugation at 514 g allowed phase separation and the upper layer wascollected in a gas chromatography autosampler vial and capped underN2. Fatty acid methyl ester profiles in brain tissue were obtained by cap-illary gas chromatography using a temperature gradient on a HP5890 gaschromatograph (Hewlett Packard, Toronto, Canada) equippedwith aHP-88 capillary column (100 m×0.25 mm i.d.×0.20 μm film thickness; Agi-lent Technologies) coupled with a flame ionization detector. Heliumwasused as a carrier gas (split ratio 1:80). Fatty acids were identified accord-ing to their retention time, using the following standard mixtures as abasis for comparison: the FAME 37 mix (Supelco Inc., Bellefonte, PA)and the GLC-411 fatty acid mix (Nu-Chek Prep Inc, Elysian, MN), aswell as the following methylated fatty acids C22:5 n−6 (Larodan AB,Malmö, Sweden) and C22:5 n−3 (Supelco Inc., Bellefonte, PA). Amixtureof trans fatty acids containing C18:2 n−6 cis/trans (Supelco IncBellefonte, PA), a mixture of cis/trans C18:3 n−3 (Supelco Inc Bellefonte,PA), the following Nu-Chek fatty acids: C14:1 trans-9, C16:1 trans-9, andfinally isoforms of C18:1 (cis-6, trans-6, cis-11, trans-11: Nu-Chek, andcis-12, cis-13, Supelco Inc.) were also used as a standard mixture.

Visceral fat dissection and adipocyte quantification

Epididymal fat was dissected at sacrifice, weighted and immediatelypost-fixed in 4% paraformaldehyde pH 7.4 for 24 h. Paraffin sections of10 μmwere cut using a cryostat. Sections were first hydrated in ethanoland stained in successive baths of hematoxylin (10 min),water (3 min),70% alcohol and 1% HCl solution and washed with running water. Lith-ium carbonate dips, running water and eosin staining (30 s) were thenperformed prior to alcohol dehydration and coverslipping. Adipocytecircumferences were measured under bright-field illumination usingStereo Investigator software (MicroBrightfield, Colchester, VT) integrat-edwith an E800Nikonmicroscope (Nikon Canada Inc., Mississauga, ON,Canada). Approximately 80 to 100 cell perimeters per animal werequantified with Neurolucida modeling software (MicroBrightfield).

Immunohistochemistry

Paraformaldehyde post-fixed sections were processed using stan-dard immunohistochemical procedures as previously described(Bousquet et al., 2008). Briefly, sections were incubated overnight at4 °C with rabbit anti-tyrosine hydroxylase (TH) (1:5000; Pel-Freez, Rog-ers, AR;) in 0.1% Triton X-100, and 5% normal goat serum in PBS(137 mM NaCl, 2.7 mM KCl, 10 mM sodium phosphate dibasic, 2 mMpotassium phosphate monobasic, pH 7.4). The overnight incubationwas followed by 1 h incubation at room temperature in a PBS solutioncontaining 0.1% Triton X-100, 5% normal goat serum and biotinylatedgoat anti-rabbit IgG (Vector Laboratories, Burlington, ON, Canada;1:1500). An avidin–biotin peroxidase complex (ABC; Vector Laborato-ries, Burlington, ON, Canada), combined with a 3,3′-diaminobenzidinetetrahydrochloride (Sigma) immunoreaction, was used to visualizebound antibodies. Following a reaction of 3,3′-diaminobenzidine

tetrahydrochloridewith TH, sectionswere counterstainedwith cresyl vi-olet, dehydrated, and coverslipped.

In situ hybridization

The DA transporter (DAT) probe, a 2238-bp fragment, was clonedinto the pBluescript II SK+ plasmid. Linearization was made with Notlenzyme. The antisense probe was synthesized with [35S]UTP and T7RNA polymerase. Brain sections were mounted onto Snowcoat X-traslides (Surgipath, Winnipeg, MB, Canada) and air-dried overnight atroom temperature. Brain sections were prepared for overnight hybrid-ization as previously described (Bousquet et al., 2008; Gibrat et al.,2010). The [35S]UTP-radiolabeled complementary RNA probes wasadded to an hybridization mix (1X Denhart's solution, 10% dextran sul-fate, 50% deionized formamide and 35S coupled 2×106 cpm/μl probe)and heated at 80 °C for 5 min. Each slide was covered with 100 μl ofthe hybridization solution and coverslipped. The hybridization was car-ried out overnight on a slidewarmer at 58 °C. After hybridization, slideswere rinsed in successive baths of standard salt sodium citrate andRNase A solution before being dehydrated in increasing concentrationsof ethanol. Tissue sections were then exposed to Biomax MR autoradi-ography films (Kodak, New Haven, CT) for 5 h.

Deffating was performed with four sequential baths of ethanol, twobaths of xylene and three baths of ethanol. Following these steps, slideswere dipped in NTP emulsion (Kodak, New Haven, CT, USA) and pre-pared at 42 °C, air-dried for 2 h and stored in the dark for 7 days at 4 °C.The emulsion was then developed in D-19 developer (3.5 min) (Kodak,New Haven, CT, USA), rinsed in deionised water and fixed (5 min) inRapid Fixer solution (Kodak, New Haven, CT, USA). Slides were rinsedin deionised water for 1 h and then colored. Coloration was performedusing thionine (1 min), followed by water, ethanol dips, an additionalthree ethanol (1 min) and three xylene baths (3 min). Slides were cover-slipped with DPX mounting media.

Stereological quantification

The loss of dopaminergic neurons was determined by stereologi-cal counts of TH- or DAT-positive cells under bright-field illumination,as previously described (Bousquet et al., 2008; Gibrat et al., 2010). Toinsure accurate cell count of the area more specifically affected by theMPTP administration, corresponding to anterior and medial sections(Jackson-Lewis et al., 1995), every fourth section between antero-posterior levels of −2.75 to −3.25 was analyzed using the Stereo In-vestigator software (MicroBrightfield) integrated with an E800 Nikonmicroscope (Nikon Canada Inc.). After delineation of the SNpc at lowmagnification (4× objective), a point gridwas overlaid onto each section.Stained cells were countedwith the optical fractionatormethod at highermagnification (20× objective). The counting variables were as follow:distance between counting frames (150 μm×150 μm), counting framesize (100 μm) and guard zone thickness (2 μm). Cells were countedonly if they did not intersect forbidden lines. The optical fractionatormethod was used to count TH-positive and TH-negative (cresyl violet-positive only) cellular profiles (Glaser and Glaser, 2000). Stereologicalcounts were performed blindly by two independent investigators.

High performance liquid chromatography

DA and 3,4-dihydroxyphenylacetic acid (DOPAC) were measuredby HPLC with electrochemical detection, according to a previouslypublished protocol (Bousquet et al., 2011). Extracts of anterior striata(stereotaxic coordinates, A–P levels, +1.34 mm to +0.94 mm) werecollected from mouse brain, and 200 μl of perchloric acid (0.1 N; Mal-linckrodt Baker, Phillipsburg, NJ) was added to generate a superna-tant. Fifty microliters of supernatant from striatal tissues wasdirectly injected into the HPLC consisting of a Waters 717 plus Auto-sampler automatic injector, a Waters 1525 Binary Pump equippedwith

0.0

0.1

0.2

0.3

0.4

CD CDHFD HFD

Vehicle MPTP

###

0.0

0.5

1.0

1.5

2.0

2.5

CD CDHFD HFD

Vehicle MPTP

###Wei

gh

t (g

)

Ave

rag

e fo

od

inta

ke

(kca

l per

wee

k)

Vis

cera

l fat

(g

)

Ad

ipo

cyte

siz

esP

erim

eter

s (m

m)

AU

C

Gly

cem

ia (

mM

)

CD HFD0

25

50

75

100

125

CD HFD0

100

200

300

400

500

600

700

800

900***

***

0 15 30 45 60 900

3.5

7.0

10.5

14.0

CD

HFD

Time after insulin injection (min)

CD HFD0.0

9.0

18.0

27.0

36.0

45.0***

12 11 11 14

5 5 5 5

24 25 24 25

20 23

0

0.5

1.0

1.5

2.0

2.5

CD HFD11 11

Insu

lin (

μg/l)

**

A B C

D E F G

H

Fig. 1. HFD induces weight gain and insulin resistance. The high intake in calories from fat (A) led to a significant weight gain in mice fed the HFD (B). Dissection and weighing of theepididymal fat revealed an increase of this specific area of the visceral white adipose tissue (C). This fat accumulation was visualized and quantified by hematoxylin and eosin stain-ing (D). Photomicrographs of representative animal from vehicle-treated fed the CD (E-upper panel) and the HFD (E-lower panel). The assessment of glucose response following asingle insulin injection showed higher blood glucose levels in obese mice, which persisted 90 min post-injection (F). The AUC quantification corroborated the significant increase ofblood glucose and confirmed the insulin resistance in mice fed the HFD (G). Insulin levels were increased in HFD-fed mice (H). **=pb0.01; ***=pb0.001 vs CD; ###=pb0.001 vsvehicle on HFD; ♦♦♦=pb0.001 vs MPTP-treated on CD. Scale bar in E=100 μm. The number of animals per group is indicated in each bar graph.

CD HFD0

2.5

5.0

7.5

10.0

12.5

CD HFD0

0.25

0.50

0.75

1.00

1.25

CD HFD0

0.5

1.0

1.5

2.0

CD HFD0

1

2

3

4

5

6

7

CD HFD0

0.25

0.50

0.75

1.00

1.25

CD HFD0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Leu

kocy

tes

(x10

6 ce

lls/m

l)

Gra

nu

locy

tes

(x10

6 ce

lls/m

l)

T L

ymp

ho

cyte

s(x

106

cells

/ml)

B

Lym

ph

ocy

tes

(x10

6 ce

lls/m

l)

CD

8+(x

106

cells

/ml)

C

D4+

(x10

6 ce

lls/m

l)

A

D

EC

FB

**

**

*

*

*

*

15 18

15 18 15 18 15 18

15 16 15 18

Fig. 2. HFD triggers peripheral immunologic alterations. The two-month dietary treatment induced a significant increase in blood cells as measured by flow cytometry. Leukocytes(A), granulocytes (B), B and T lymphocytes (C and D) as well as CD4 and CD8 positives cells (E and F) were increased as compared to CD-fed mice. *=pb0.05; **=pb0.01 vs CD.The number of animals per group is indicated in each bar graph.

532 M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

0

10

20

30

40

50

60

70

CD HFD0

25

50

75

IL-1

α (p

g/m

l)

CD HFD0

10

20

30

40

50

60

70

80

90

MC

P-1

(p

g/m

l)

BA

17 18 18 20

CD HFD11 13

C

MIP

-1α

(pg

/ml)

Fig. 3. HFD induces changes in serum cytokine profiles. Interleukin-1α (IL-1α) (A), the monocyte chemotactic protein-1 (MCP-1) (B) and the macrophage inflammatory protein-1α(MIP-1α) (C) were quantified using a Multiplex ELISA assay and showed increased levels in HFD-fed mice. *=pb0.05; **=pb0.01 vs CD. The number of animals per group is in-dicated in each bar graph.

Table 2Plasmatic levels of cytokines and chemokines 2 months post-HFD

Cytokines andchemokines

CD HFD pValue

Mean±S.E.M. N Mean±S.E.M. N

IL-1β 117.1±28.0 12 77.2±21.6 13 0.2674IL-2 Below detection limitIL-3 Below detection limitIL-4 Below detection limitIL-5 76.7±19.7 13 84.8±12.9 14 0.7336IL-6 66.6±9.2 14 53.39±8.07 15 0.2907IL-10 Below detection limitIL-12 65.9±21.7 11 15.8±10.3 12 0.0559IL-17 Below detection limitIFNγ 38.2±7.5 14 46.2±4.5 15 0.3552TNFα 80.5±14.3 12 Below detection limitGM-CSF 51.2±12.2 12 57.4±7.5 14 0.6574RANTES 141.7±12.1 20 142.4±9.2 20 0.9674

Values of IL1-α, MCP-1 and MIP-1α are presented in Fig. 3.p Value: unpaired Student t-test with Welch's correction.Abbreviations: CD: control diet, GM-CSF: granulocyte macrophage colony stimulatingfactor, HFD: high-fat diet, IFNγ: interferon, IL: interleukin, TNFα: tumor necrosisfactor.

533M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

anAtlantis dC18 (3 μm)column, aWaters 2465Electrochemical Detector,and a glassy carbon electrode (Waters, Lachine, QC, Canada). The electro-chemical detector was set at 10 nA. The mobile phase consisted of47.8 mM NaH2PO4, 0.9 mM sodium octyl sulfate (Mallinckrodt Baker),0.4 mM EDTA, 2 mM NaCl and 8% MeOH (Mallinckrodt Baker) at pH 2.9and was delivered at 0.8 ml/min. Peaks were identified using the Breezesoftware (Waters). HPLC data were normalized to protein concentra-tions, as determined with a bicinchoninic acid protein assay kit usingBSA as standard (Pierce, Rockford, IL) and according to the manufac-turer's protocol.

Sample preparation and Western immunoblotting

Striatal samples were homogenized in 8 volumes of tris-bufferedsaline (TBS) containing a cocktail of protease inhibitors (Roche, Mis-sissauga, ON, Canada) and phosphatase inhibitors (1 mM tetrasodiumpyrophosphate and 50 mM sodium fluoride), sonicated (3×10 s) andthen centrifuged at 100,000×g for 20 min at 4 °C. The supernatantwas collected, stored at −80 °C and constituted the TBS-soluble in-tracellular and extracellular fraction. The residual pellet was homog-enized in lysis buffer (150 mM NaCl, 10 mM NaH2PO4, 1% (v/v)Triton X-100, 0.5% SDS, and 0.5% sodium deoxycholate) includingthe same protease and phosphatase inhibitors. Following a 20-mincentrifugation (100,000×g), supernatants were stored at −80 °C togenerate the detergent-soluble fraction which comprises membrane-bound and nuclear proteins (Julien et al., 2010). A bicinchoninic acidprotein assay was utilized to quantify protein concentrations in eachfraction. Twenty micrograms of total protein per sample was added tolaemmli loading buffer and heated to 95 °C for 5 min. Samples werethen subjected to SDS-polyacrylamide (8%) gel electrophoresis. Pro-teins were electroblotted onto 0.45 μm Immobilon PVDF membranes(Millipore, Billerica, MA) and blocked in 5% nonfat dry milk and 1%BSA in 1X PBS-tween for 1 h. Membranes were immunoblotted with aprimary antibody rabbit anti-TH (Pel-Freez; 1:5000) to identify dopa-minergic elements, a mouse anti-glial fibrillary acidic protein (GFAP)(Sigma; 1:10,000) for astrogliosis and a rabbit anti-NFκB (Santa CruzBiotechnology, Santa Cruz, CA; 1:1000). For neurotrophic factor assess-ment, a rabbit anti-BDNF (Santa Cruz; 1:1000) and a rabbit anti-GDNF(BioVision San Francisco. CA; 1:1000) were used while a mouse anti-synaptophysin (Chemicon-Millipore; 1:10,000) and a mouse anti-PSD-95 (Neuromab, Davis, CA; 1:5000) were employed to evaluatecomponents of the synapse. Finally, a mouse anti-actin (ABM Inc,Richmond, BC, Canada; 1:10,000) was utilized as a loading control. De-tection with appropriate secondary antibodies, goat anti-rabbit oranti-mouse (Jackson Immunoresearch, West Grove, PA; 1:100,000),followed by the addition of chemiluminescence reagents (KPL, Man-del Scientific, Guelph, ON, Canada), were then performed. For

oxidative stress assessment, carbonylated group within proteinswere derivatized to 2,4-dinitrophenylhydrazone by reaction with2,4-dinitrophenylhydrazine prior to Western immunoblot accordingto the manufacturer's protocol (OxyBlot Protein Oxidation DetectionKit, Millipore, Billerica, MA). Band intensities were quantified using aKODAK Image Station 4000 Digital Imaging System (Molecular Im-aging Software version 4.0.5f7, Eastman Kodak, New Haven, CT).

Statistical analyses and image preparation

Statistical analyses were performed using JMP software version 8.0.2(SAS Institute Inc., Cary, IL) and Prism 4 (GraphPad software, CA). Stu-dent's t-testswere used to analyze food intake,mouseweight, insulin re-sistance test, inflammatory cells, cytokines and fatty acid profiles. Two-way ANOVAs were performed for all the other experimentations andno significant interaction between diet and treatment (vehicle orMPTP) were found. Therefore, comparisons were based on one-wayANOVAs andNewman–Keulsmultiple comparison testswere performedfor post-hoc analyses. Log adjustmentswere used to normalize varianceswhenneeded. Photomicrographswere takenwith aMicrofire 1.0 camera(Optronics, Goleta, CA) connected to an E800 Nikon microscope (NikonInc., Québec, QC, Canada) using the Picture Frame imaging software

A

Total n-3 PUFAs

CD HFD0

10

20

DHA

CD HFD0

10

20

LA

CD HFD0

0.5

1.0

1.5 ***

DTA

CD HFD0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

***

n-3:n-6 PUFA ratio

CD HFD0

0.25

0.50

0.75

1.00

1.25

***

n-6 DPA

CD HFD0

0.1

0.2

0.3

0.4

0.5

0.6

*

ARA

CD HFD0

2.5

5.0

7.5

10.0

12.5**

Total n-6 PUFAs

CD HFD0

10

20

***

B

C

D

E

F

G

H

% o

f to

tal f

atty

aci

ds

% o

f to

tal f

atty

aci

ds

% o

f to

tal f

atty

aci

ds

% o

f to

tal f

atty

aci

ds

Rat

io%

of

tota

l fat

ty a

cid

s

% o

f to

tal f

atty

aci

ds

% o

f to

tal f

atty

aci

ds

22 24

23 25

23 25

23 25

23 25

23 25

22 25

23 25

Fig. 4. HFD alters brain fatty acid profiles. Frontal cortex fatty acid profiles were per-formed by gas chromatography and revealed increased levels in HFD-fed mice inn−6 PUFA linoleic acid (LA) (A), docosatetraenoic acid (DTA) (B), docosapentaenoicacid (DPA) (C), arachidonic acid (ARA) (D) as well as in total n−6 PUFAs (E). n−3PUFA docosahexaenoic acid (DHA) (F) and total n−3 PUFAs (G) levels remained sim-ilar between dietary treatments. These cerebral fatty acid changes led to a significantdecrease in the n−3:n−6 PUFA ratio (H). *=pb0.05; **=pb0.01; ***=pb0.001 vsCD. The number of animals per group is indicated in each bar graph.

534 M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

(Microbrightfield, Colchester, VT). All images were prepared for illustra-tion with Adobe Photoshop 7.0.

Results

Impact of the HFD on the peripheral system

As a result of the HFD and increased calorie intake (Fig. 1A; Student's ttest; F(groups)1,30=96.71; pb0.001),mice significantly gainedweight asevidenced by a 41% increase in the HFD group compared to CD-fed mice(Fig. 1B; Student's t test; F(groups)1,47=151.32; pb0.001). Consequentlyto the weight gain, post-mortem dissection of the epididymal fat padrevealed an increased accumulation of white adipose tissue (Fig. 1C;ANOVA; F(groups)3,52=37.68; pb0.001) combined to an increase insize of adipocytes (larger circumference) (Fig. 1D–E; ANOVA; F(groups)3,16=49.17; pb0.001). The high intake in fat also triggered insu-lin resistance (Fig. 1F and G; Student's t test; F(groups)1,42=44.48;pb0.001), which was further accompanied by increased levels of plas-matic insulin (Fig. 1H; Student's t test; F(groups)1,20=17.92; pb0.01).No significant differences between MPTP-treated mice on CD or HFDwere found for insulin quantification 14 days post-MPTP (data notshown).

Prior toMPTP administration, several components relating to system-ic inflammation were assessed to evaluate the effects of the HFD on theimmune system. Flow cytometry analyses revealed an increased numberof overall leukocytes (Fig. 2A; Student's t test; F(groups)1,31=10.65;pb0.01). Granulocytes (Fig. 2B; Student's t test; F(groups)1,31=7.69;pb0.01), as well as B and T lymphocytes (Fig. 2C–D; Student's t test; F(groups)1,31–29=5.45, 7.17; pb0.05), including CD4 (T helper) andCD8 (T cytotoxic) positive cells (Fig. 2E–F; Student's t test; F(groups)1,31=7.24, 6.78; pb0.05), were significantly increased inmice fed the HFD. Two months following the beginning of the diet, 16cytokines were examined using a Multiplex ELISA assay and the high-fat intake led to an increase in IL-1α (Fig. 3A; Student's t test; F(groups)1,33=4.95; pb0.05), MCP-1 (Fig. 3B; Student's t test; F(groups)1,22=7.88; pb0.01) and MIP-1α (Fig. 3C; Student's t test; F(groups)1,22=5.13; pb0.05). IL-1β, -2, -3, -4, -5, -6, -10, -12, -17,IFNγ, TNFα, GM-CSF and RANTES were also quantified but remainedsimilar between groups or under the detection limit (Table 2). Plasmasamples were also analyzed 1 h following the last MPTP injection aswell as 5 and 14 days post-MPTP administration. Although no signifi-cant differences were observed for the 2 last time points, mice fed theHFD displayed significantly higher plasma levels of IL-1α than micefed the CD, 1 h post-MPTP (data not shown).

Brain fatty acid profiles following high-fat intake

Cerebral fatty acids were evaluated in the frontal cortex. The HFDcontained elevated n−6 PUFAs and saturated fat compared to the CDand its consumption rapidly led to insulin resistance in C57BL/6 mice(Lucas et al., 2011; Winzell et al., 2011). As previously shown by ourgroup (Julien et al., 2010), the same HFD formula over a period of9 months induces alterations of brain fatty acid profiles, such as a de-crease in the n−3:n−6 PUFA ratio (Julien et al., 2010). Here, the 2-month dietary treatment induced an increase in levels of total (Fig. 4E;Student's t test; F(groups)1,46=23.24; pb0.001) and individual n−6PUFAs, such as linoleic (LA; C18:2 n−6, Fig. 4A; Student's t test;F(groups)1,45=21.33; pb0.001), docosatetraenoic (DTA; C22:4 n−6,Fig. 4B; Student's t test; F(groups)1,46=32.59; pb0.001), docosapen-taenoic (DPA; C22:5 n−6, Fig. 4C; Student's t test; F(groups)1,46=5.11;pb0.05) and arachidonic acids (ARA; C20:4 n−6, Fig. 4D; Student'st test; F(groups)1,46=11.19; pb0.01) in the frontal cortex of animalssubjected to the HFD. Levels of docosahexaenoic acid (DHA; C22:6 n−3) as well as total n−3 PUFAs (Fig. 4F–G) remained similar be-tween mice on either diet. These modifications in brain fatty acids

535M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

led to an overall 12% decrease in the n−3:n−6 PUFA ratio (Fig. 4H;Student's t test; F(groups)1,44=115.46; pb0.001).

Impact of HFD on MPTP-induced dopaminergic degeneration

The subacuteMPTP regimen employed here consistently produces asignificant nigral dopaminergic neuronal degeneration of nearly 30%,which is accompanied by a 75% decrease in striatal DA levels (Bousquetet al., 2008; Gibrat et al., 2009). Nigral dopaminergic neuronswere eval-uated using TH- and DAT labeling. As quantified by stereologicalcell counts, MPTP treated-mice depicted a reduction in cell densityof TH- (Fig. 5A–E; ANOVA; F(groups)3,37=4.43; pb0.05 in MPTPmice on HFD vs vehicle groups) and DAT-positive cells (Fig. 5F–J;ANOVA; F(groups)3,39=4.38; pb0.05 in MPTP mice on CD and HFD vsvehicle groups). Striatal DA and DOPAC contents, as assessed by HPLC,were decreased in MPTP-treated mice as compared to mice treatedwith saline (Fig. 6A–B; ANOVA F(groupes)3,43=23.52; pb0.001). Thetwo-month dietary treatment potentiated the DA depletion as themice fed the HFD were characterized by a 32% lower level of striatalDA after the MPTP insult compared to mice fed the CD (Student t test;F(groups)1,22=4.37; pb0.05 MPTP CD vs MPTP HFD). Interestingly,Western blot quantification of TH protein levels in the striatum, asses-sing dopaminergic terminals, also revealed an exacerbated effect ofthe neurotoxinMPTP inmice fed the HFD (Fig. 6C–D). Indeed, TH levelswere decreased by 43% and 58% in MPTP-treated mice on CD and HFD,respectively (ANOVA; F(groups)3,44=52.23; pb0.001 in MPTPmice on

Fig. 5. MPTP treatment leads to nigral dopaminergic degeneration. The state of the dopaminhydroxylase (TH) (A to E) and in situ hybridization for the dopamine transporter (DAT) (F tpositive cells as pointed by the black arrows. Stereological counts of either marker revealedhicle-treated animals on either diet. *=pb0.05. Scale bar in B applies for C to E=300 μm anper group is indicated in each bar graph.

CD and HFD vs vehicle groups; pb0.05 MPTP CD vs MPTP HFD). Two-way ANOVA analyses confirmed the effect of MPTP on dopaminergicmarkers utilized in this study but could not detect the effect of the die-tary treatment.

Underlying mechanisms for the increased vulnerability of striataldopaminergic terminals

In order to identify possible molecular mechanisms associated withthe exacerbation of MPTP-induced striatal dopaminergic depletion inHFDmice,Western blot analyseswere performed for neuroinflammatorymolecules, neurotrophic factors, oxidized proteins and synaptic compo-nents (Table 3). These mechanisms were specifically selected becauseof their contribution to PD pathophysiology and the probable capacityof HFD to act via these pathways (see Discussion for details). However,no significant differences were observed for any of these parameters inresponse toHFD orMPTP administration,with the exception of the astro-glial response, where an increase in GFAP protein levels was observed inMPTP-treated mice which was, however, independent of the dietarytreatment (Table 3).

Discussion

Based on our previouswork demonstrating the beneficial effects of an−3 PUFA-enriched diet in the MPTP mouse model (Bousquet et al.,2008), we sought to investigate the consequences of a converse

ergic system within the SNpc was analyzed using immunohistochemistry for tyrosineo J). The higher magnification of the photomicrograph in the inset (G) illustrates DAT-a decreased in the total number of nigral cells in MPTP-treated mice compared to ve-d scale bar in G applies for H to J=400 μm (inset in G=25 μm). The number of animals

Fig. 6. HFD exacerbates the detrimental effects of MPTP in the striatum. MPTP-treated mice displayed a significant decrease in striatal DA (A) and in one of its metabolite DOPAC (B)as quantified by HPLC and to a greater extent in mice fed the HFD. Western blot analyses of striatal TH levels showed a greater loss of TH proteins in MPTP-treated mice fed the HFDas compared to MPTP-treated mice fed CD (C and D). *=pb0.05; ***=pb0.001. The number of animals per group is indicated in each bar graph.

536 M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

nutritional behavior, the HFD, in a mouse model of parkinsonism. Thecurrent experiment explored both the cerebral consequences followingthe HFD andMPTP treatment as well as the peripheral corollaries of theHFD including metabolic disturbances and the inflammatory response.Findings from this study revealed a more notable DA reduction andTH-positive fiber degeneration in the striatum of MPTP-treated micefed a HFD for 2 months, despite comparable nigral neuronal cellcount. The high intake in calories from fat caused a weight gain, whichwas accompanied by insulin resistance, increased circulating immunecells and elevated levels of cytokines as well as chemokines. Our datathus suggest that the peripheral consequences of diet-induced obesityare associated with an enhanced vulnerability of dopaminergic termi-nals to MPTP in mice exposed to HFD.

The present observations are in accordance with results from twostudies performed in different animal models of parkinsonism (Choiet al., 2005; Morris et al., 2010). The study conducted by Choi et al. in-cluded two distinct MPTP regimens, 4 subcutaneous (s.c.) injections at2 h interval of a low(5 mg/kg) or a high (15 mg/kg) dose. An exacerbatedeffect of the 8-week high-fat intake was only found in mice exposed tolow doses of MPTP (Choi et al., 2005). HFD was also investigated in the6-hydroxydopamine-lesioned (6-OHDA) rat model of PD (Morris et al.,2010). As observed with s.c. or i.p. administration of MPTP, 6-OHDA-induced DA depletions in the SNpc and striatum were increased in ratsfed the HFD, which correlated with the homeostatic model assessmentof insulin resistance (HOMA-IR) as well as with the adiposity (Morris

et al., 2010). Here we observed discrepant outcomes between the nigraland striatal compartments; the dopaminergic terminals of HFD-fedmice were more profoundly affected by the MPTP treatment than thenigral cell bodies, suggesting that terminals are more susceptible tocombined effects of MPTP and HFD. If this finding can be extrapolatedto humans, the prevalent high consumption of fat may increase the vul-nerability of the dopaminergic system to exogenous insult and precipi-tates the onset of the pathology.

High-fat intake, which can lead to excess visceral adiposity, is wellknown to result in metabolic disorders such as insulin and glucose per-turbations, increased systemic inflammation, hypertension and dyslipi-demia (Scaglione et al., 2010). Low-grade chronic inflammatory state,caused by HFD, is also thought to play a critical role in the pathogenesisof metabolic syndrome. Here, we observed an increase of inflammatorycells such as leukocytes as well as cytokines and chemokines, in accor-dance with other studies performed in rodents fed a HFD and also inthe blood of diet-induced obese individuals (Nieman et al., 1999; Strisselet al., 2010;Wu et al., 2007). Interestingly, total leukocyte count stronglycorrelates with BMI in humans (Nieman et al., 1999). Furthermore, thechemokine MCP-1, which plays an important role in recruiting macro-phages to adipose tissue, is also increased in the blood and visceral fatof genetically-induced obese mice (ob/ob) (Sartipy and Loskutoff,2003). Here, an increased in plasmatic IL-1α was observed inMPTP-treated mice fed the HFD 1 h following the last MPTP injec-tion. This increase of IL-1α could have, in part, contributed to generate

Table 3Mechanisms underlying the striatal MPTP-induced degeneration exacerbated by theHFD

Markers Vehicle MPTP

CD HFD CD HFD

Neuroinflammatory responseGFAP 1660±111 1545±178 2812±148⁎⁎⁎ 2673±119⁎⁎⁎

NFκB 3679±171 3567±172 3588±142 3508±208

Oxidative damageCarbonylatedproteins

2080±140 2040±167 2019±198 2235±151

Neurotrophic factorspro-BDNF 267.8±43.7 296.9±37.7 277.1±23.5 283.6±18.3BDNF 15.9±2.7 17.9±3.2 14.8±1.3 18.1±1.6GDNF 375.9±35.0 359.7±28.6 360.7±20.6 383.9±25.9

Synaptic proteinsSynaptophysin 4622±216 4741±233 4367±148 4435±152PSD-95 4354±312 3871±337 4083±259 4145±240

Levels are expressed as mean±S.E.M.Abbreviations: BDNF: brain-derived neurotrophic factor, CD: control diet, GDNF: glialcell line-derived neurotrophic factor, GFAP: glial fibrillary acidic protein, HFD: high-fat diet, NFκB: nuclear factor kappa B, PSD-95: post-synaptic density.⁎⁎⁎ =pb0.001 Vs Vehicle on CD and HFD.

537M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

a pro-inflammatory environment within the brain, thereby increasingthe vulnerability of the dopaminergic terminals. Although the involve-ment of inflammation to PD pathogenesis remains unclear, its systemicand cerebral occurrence has been confirmed in several PD animalmodelsas well as in patients (Hunot et al., 1999; Kurkowska-Jastrzebska et al.,1999). For instance, increased levels of several cytokines and chemokinesincluding IL-6, IL-1α and MCP-1 are found in the blood, cerebrospinalfluid and/or brain parenchyma samples from PD patients (Reale et al.,2009).While an overall depletion in circulating lymphocytes is observed,an elevated number of activated lymphocytes (CD4+CD25+) has alsobeen found in the blood of PD patients (Bas et al., 2001; Hisanaga et al.,2001). In addition, infiltrating lymphocytes have been observed withinthe brain parenchyma of PD patients (Hunot et al., 1999). Taken togeth-er, the intensification of the peripheral inflammatory response caused byhigh intake in calories from fat, could in part explain the increased vul-nerability to MPTP of mice fed the HFD compared to MPTP-treatedmice on CD. However, the type of MPTP regimen employed here createsa modest inflammatory response, with limited microglial activationpeaks which decrease to control levels 24 h after the last MPTP injection(Drouin-Ouellet et al., in press). Nevertheless, using GFAP immunoblot-ting, we detected a strong astroglial response persisting up to 2 weekspost-MPTP treatment as previously observed (Bousquet et al., 2011)but the dietary intake of fat did not alter GFAP immunoreactivity(Table 3). Therefore, although the HFD clearly induced an inflammatoryresponse in the periphery, it remains unclear whether a coinciding neu-roinflammation played a role in the amplified vulnerability toMPTP seenhere. It could therefore be postulated that an increase in circulating im-mune cells and cytokine expression could have increased the blood–brain barrier permeability, thereby generating a temporary infiltrationof inflammatory cells in MPTP-treated mice fed the HFD (Nerurkar etal., 2011). The increased levels in circulating cytokines and chemokinescould also have directly affected neuronal signaling cascades (Banks,2005; Laflamme and Rivest, 1999), thus exacerbating the dopaminergicterminal depletion.

Previous studies have identified several other cerebral effects of aHFD in rodents that may underlie the exacerbated effects of the MPTPtreatment observed here. For example, the effect of low MPTP doseadministered subcutaneously to mice fed a HFD (sacrificed 7 dayspost-MPTP) was in part explained via increased nitrogen speciesand phosphorylated neuronal nitric oxide synthase in the striatumand SNpc of mice (Choi et al., 2005). A number of evidence has also

unveiled the effect of PUFAs on neurotrophic factors (Bousquet et al.,2009; Rao et al., 2007). Increased fat intake has been associatedwith de-creased levels of BDNF with impaired hippocampal neurogenesis inadult male rats (Park et al., 2010). Finally, increased protein carbonylswere reported in the hippocampus of 20-month old mice fed a HFDfor 4 months (Bruce-Keller et al., 2010;Morrison et al., 2010). However,as summarized in Table 3, we did not detect changes in striatal neuro-trophic factors or oxidative damage by quantification of protein carbonylsthroughWestern blot assessment. In addition, no differences in pre- andpost-synaptic components were noted as previously reported in an ani-malmodel of Alzheimer's disease fed aHFD for 9 monthswhere an aggra-vation of amyloid-β and tau pathologies were observed (Table 3) (Julienet al., 2010).

One the other hand, the increase in n−6 PUFAs, especially ARA, result-ing from theHFDmayhave also contributed to the results presented here.Indeed, ARA, via cyclooxygenase and lipoxygenase action, is the precursorof several bioactive eicosanoids such as prostaglandins, leukotrienes andthromboxanes leading to a proinflammatory environment (Funk, 2001).Cascades leading to eicosanoid production originating from PUFAs arepresent in neurological disorders including PD (Teismann et al., 2003).In patients, although no overall differences in fatty acid profiles werenoted, patients taking levodopa and experiencing motor complicationshad higher levels of cerebral ARA compared to individuals devoid ofthese medication-induced side effects and to age-matched controls(Julien et al., 2006). In agreement with this, dietary n−3 PUFAs, whichalso lead to a decrease of brain ARA, is neuroprotective in the sameMPTP model. This supports a mechanistic relationship between cerebralPUFAs and their metabolic products with PD pathogenesis.

Conclusion

The peripheral perturbations caused by a HFD are well established,whereas the outcomes of these unhealthy diets on the brain are less un-derstood. Taken together, the present data suggest that the detrimentalconsequences of high intake in calories from fat translate into a reducedcapacity of the brain to cope with a neurodegenerative stress. Exposureto HFD, and consequently obesity, could confer a greater susceptibilityto environmental toxins and accelerate PD pathogenesis. Further investi-gations are needed to ascertain the direct contribution of systemic im-pairments, notably insulin resistance and increased immune response,to PD and other neurodegenerative diseases and to ultimately determinewhether HFD is a modifiable risk factor for PD.

Acknowledgments

This studywas supported by the Parkinson Society Canada and by theInstitute of Nutrition, Metabolism and Diabetes (INMD) of the CanadianInstitutes of Health Research (CIHR) and the Canada Foundation for Inno-vation (FC and FC). MB and MV were supported by a CIHR scholarshipand IS was supported by an Industrial Innovation PhD Scholarship fromCRSNG/FQRNT.

References

Abbott, R.D., Ross, G.W., White, L.R., Nelson, J.S., Masaki, K.H., Tanner, C.M., Curb, J.D.,Blanchette, P.L., Popper, J.S., Petrovitch, H., 2002. Midlife adiposity and the futurerisk of Parkinson's disease. Neurology 59, 1051–1057.

Banks, W.A., 2005. Blood–brain barrier transport of cytokines: a mechanism for neuro-pathology. Curr. Pharm. Des. 11, 973–984.

Bas, J., Calopa, M., Mestre, M., Mollevi, D.G., Cutillas, B., Ambrosio, S., Buendia, E., 2001.Lymphocyte populations in Parkinson's disease and in rat models of parkinsonism.J. Neuroimmunol. 113, 146–152.

Bayer-Carter, J.L., Green, P.S., Montine, T.J., VanFossen, B., Baker, L.D., Watson, G.S.,Bonner, L.M., Callaghan, M., Leverenz, J.B., Walter, B.K., Tsai, E., Plymate, S.R.,Postupna, N., Wilkinson, C.W., Zhang, J., Lampe, J., Kahn, S.E., Craft, S., 2011.Diet intervention and cerebrospinal fluid biomarkers in amnestic mild cognitiveimpairment. Arch. Neurol. 68, 743–752.

Bousquet, M., Saint-Pierre, M., Julien, C., Salem Jr., N., Cicchetti, F., Calon, F., 2008. Ben-eficial effects of dietary omega-3 polyunsaturated fatty acid on toxin-induced

538 M. Bousquet et al. / Neurobiology of Disease 45 (2012) 529–538

neuronal degeneration in an animal model of Parkinson's disease. FASEB J. 22,1213–1225.

Bousquet, M., Gibrat, C., Saint-Pierre, M., Julien, C., Calon, F., Cicchetti, F., 2009. Modu-lation of brain-derived neurotrophic factor as a potential neuroprotective mecha-nism of action of omega-3 fatty acids in a parkinsonian animal model. Prog.Neuropsychopharmacol. Biol. Psychiatry 33, 1401–1408.

Bousquet, M., Gue, K., Emond, V., Julien, P., Kang, J.X., Cicchetti, F., Calon, F., 2011. Trans-genic conversion of omega-6 into omega-3 fatty acids in a mouse model of Parkin-son's disease. J. Lipid Res. 52 (2), 263–271.

Bruce-Keller, A.J., White, C.L., Gupta, S., Knight, A.G., Pistell, P.J., Ingram, D.K., Morrison, C.D.,Keller, J.N., 2010. NOX activity in brain aging: exacerbation by high fat diet. Free Radic.Biol. Med. 49, 22–30.

Calon, F., Cicchetti, F., 2008. Canwe prevent Parkinson's diseasewith n−3 polyunsaturatedfatty acids? Futur. Lipidol. 3, 133–137.

Calon, F., Morissette, M., Rajput, A.H., Hornykiewicz, O., Bedard, P.J., Di Paolo, T., 2003.Changes of GABA receptors and dopamine turnover in the postmortem brains ofparkinsonians with levodopa-induced motor complications. Mov. Disord. 18,241–253.

Chaudhuri, K.R., Schapira, A.H., 2009. Non-motor symptoms of Parkinson's disease: do-paminergic pathophysiology and treatment. Lancet Neurol. 8, 464–474.

Chen, H., Zhang, S.M., Hernan, M.A., Willett, W.C., Ascherio, A., 2003. Weight loss in Par-kinson's disease. Ann. Neurol. 53, 676–679.

Chen, H., Zhang, S.M., Schwarzschild, M.A., Hernan, M.A., Willett, W.C., Ascherio, A.,2004. Obesity and the risk of Parkinson's disease. Am. J. Epidemiol. 159, 547–555.

Chen, H., O'Reilly, E., McCullough, M.L., Rodriguez, C., Schwarzschild, M.A., Calle, E.E.,Thun, M.J., Ascherio, A., 2007. Consumption of dairy products and risk of Parkin-son's disease. Am. J. Epidemiol. 165, 998–1006.

Cheung, M.C., Spalding, P.B., Gutierrez, J.C., Balkan, W., Namias, N., Koniaris, L.G.,Zimmers, T.A., 2009. Body surface area prediction in normal, hypermuscular,and obese mice. J. Surg. Res. 153, 326–331.

Choi, J.Y., Jang, E.H., Park, C.S., Kang, J.H., 2005. Enhanced susceptibility to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine neurotoxicity in high-fat diet-induced obesity.Free Radic. Biol. Med. 38, 806–816.

Colditz, G.A., Willett, W.C., Rotnitzky, A., Manson, J.E., 1995. Weight gain as a risk factorfor clinical diabetes mellitus in women. Ann. Intern. Med. 122, 481–486.

Costa, J., Lunet, N., Santos, C., Santos, J., Vaz-Carneiro, A., 2010. Caffeine exposure andthe risk of Parkinson's disease: a systematic review and meta-analysis of observa-tional studies. J. Alzheimers Dis. 20 (Suppl. 1), S221–S238.

Damier, P., Hirsch, E.C., Agid, Y., Graybiel, A.M., 1999. The substantia nigra of the humanbrain. II. Patterns of loss of dopamine-containing neurons in Parkinson's disease.Brain 122 (Pt 8), 1437–1448.

Das, U.N., 2001. Is obesity an inflammatory condition? Nutrition 17, 953–966.de Lau, L.M., Breteler, M.M., 2006. Epidemiology of Parkinson's disease. Lancet Neurol.

5, 525–535.Drouin-Ouellet, J., Gibrat, C., Bousquet, M., Calon, F., Kriz, J., Cicchetti, F., in press. The

Role of the MYD88-dependent Pathway in MPTP-induced Brain Dopaminergic De-generation. Journal of Neuroinflammation.

Etminan, M., Gill, S.S., Samii, A., 2005. Intake of vitamin E, vitamin C, and carotenoidsand the risk of Parkinson's disease: a meta-analysis. Lancet Neurol. 4, 362–365.

Felson, D.T., 1996. Weight and osteoarthritis. Am. J. Clin. Nutr. 63, 430S–432S.Folch, J., Lees, M., Sloane Stanley, G.H., 1957. A simple method for the isolation and pu-

rification of total lipides from animal tissues. J. Biol. Chem. 226, 497–509.Funk, C.D., 2001. Prostaglandins and leukotrienes: advances in eicosanoid biology. Sci-

ence 294, 1871–1875.Gao, X., Chen, H., Fung, T.T., Logroscino, G., Schwarzschild, M.A., Hu, F.B., Ascherio, A.,

2007. Prospective study of dietary pattern and risk of Parkinson disease. Am. J.Clin. Nutr. 86, 1486–1494.

Gibrat, C., Saint-Pierre, M., Bousquet, M., Levesque, D., Rouillard, C., Cicchetti, F., 2009.Differences between subacute and chronic MPTP mice models: investigation of do-paminergic neuronal degeneration and alpha-synuclein inclusions. J. Neurochem.109, 1469–1482.

Gibrat, C., Bousquet, M., Saint-Pierre, M., Levesque, D., Calon, F., Rouillard, C., Cicchetti, F.,2010. Cystamine prevents MPTP-induced toxicity in young adult mice via the up-reg-ulation of the brain-derived neurotrophic factor. Prog. Neuropsychopharmacol. Biol.Psychiatry 34, 193–203.

Glaser, J.R., Glaser, E.M., 2000. Stereology, morphometry, and mapping: the whole isgreater than the sum of its parts. J. Chem. Neuroanat. 20, 115–126.

Hiltunen, M., Khandelwal, V.K., Yaluri, N., Tiilikainen, T., Tusa, M., Koivisto, H., Krzisch, M.,Vepsalainen, S., Makinen, P., Kemppainen, S., Miettinen, P., Haapasalo, A., Soininen, H.,Laakso, M., Tanila, H., 2011. Contribution of genetic and dietary insulin resistance toAlzheimer phenotype in APP/PS1 transgenic mice. J. Cell Mol. Med. doi:10.1111/j.1582-4934.2011.01384.x (Electronic publication ahead of print).

Hisanaga, K., Asagi, M., Itoyama, Y., Iwasaki, Y., 2001. Increase in peripheral CD4 bright+ CD8 dull+ T cells in Parkinson disease. Arch. Neurol. 58, 1580–1583.

Hu, G., Jousilahti, P., Nissinen, A., Antikainen, R., Kivipelto, M., Tuomilehto, J., 2006.Body mass index and the risk of Parkinson disease. Neurology 67, 1955–1959.

Hunot, S., Dugas, N., Faucheux, B., Hartmann, A., Tardieu, M., Debre, P., Agid, Y., Dugas,B., Hirsch, E.C., 1999. FcepsilonRII/CD23 is expressed in Parkinson's disease and in-duces, in vitro, production of nitric oxide and tumor necrosis factor-alpha in glialcells. J. Neurosci. 19, 3440–3447.

Jackson-Lewis, V., Jakowec, M., Burke, R.E., Przedborski, S., 1995. Time course and mor-phology of dopaminergic neuronal death caused by the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine. Neurodegeneration 4, 257–269.

Julien, C., Berthiaume, L., Hadj-Tahar, A., Rajput, A.H., Bedard, P.J., Di Paolo, T., Julien, P.,Calon, F., 2006. Postmortem brain fatty acid profile of levodopa-treated Parkinsondisease patients and parkinsonian monkeys. Neurochem. Int. 48, 404–414.

Julien, C., Tremblay, C., Phivilay, A., Berthiaume, L., Emond, V., Julien, P., Calon, F., 2010.High-fat diet aggravates amyloid-beta and tau pathologies in the 3xTg-AD mousemodel. Neurobiol. Aging 31, 1516–1531.

Kurkowska-Jastrzebska, I., Wronska, A., Kohutnicka, M., Czlonkowski, A., Czlonkowska, A.,1999. The inflammatory reaction following 1-methyl-4-phenyl-1,2,3, 6-tetrahydropyridine intoxication in mouse. Exp. Neurol. 156, 50–61.

Laflamme, N., Rivest, S., 1999. Effects of systemic immunogenic insults and circulatingproinflammatory cytokines on the transcription of the inhibitory factor kappaBalpha within specific cellular populations of the rat brain. J. Neurochem. 73, 309–321.

Lees, A.J., Hardy, J., Revesz, T., 2009. Parkinson's disease. Lancet 373, 2055–2066.Logroscino, G., Sesso, H.D., Paffenbarger Jr., R.S., Lee, I.M., 2007. Bodymass index and risk of

Parkinson's disease: a prospective cohort study. Am. J. Epidemiol. 166, 1186–1190.Lucas, E.A., Li, W., Peterson, S.K., Brown, A., Kuvibidila, S., Perkins-Veazie, P., Clarke, S.L.,

Smith, B.J., 2011. Mango modulates body fat and plasma glucose and lipids in micefed a high-fat diet. Br. J. Nutr. 1–11.

Meissner,W.G., Frasier,M., Gasser, T., Goetz, C.G., Lozano, A., Piccini, P., Obeso, J.A., Rascol, O.,Schapira, A., Voon, V., Weiner, D.M., Tison, F., Bezard, E., 2011. Priorities in Parkinson'sdisease research. Nat. Rev. Drug Discov. 10, 377–393.

Mody, N., Agouni, A., McIlroy, G.D., Platt, B., Delibegovic, M., 2011. Susceptibility todiet-induced obesity and glucose intolerance in the APP (SWE)/PSEN1 (A246E)mouse model of Alzheimer's disease is associated with increased brain levels ofprotein tyrosine phosphatase 1B (PTP1B) and retinol-binding protein 4 (RBP4),and basal phosphorylation of S6 ribosomal protein. Diabetologia 54, 2143–2151.

Morris, J.K., Bomhoff, G.L., Stanford, J.A., Geiger, P.C., 2010. Neurodegeneration in an an-imal model of Parkinson's disease is exacerbated by a high-fat diet. Am. J. Physiol.Regul. Integr. Comp. Physiol. 299, R1082–R1090.

Morrison, C.D., Pistell, P.J., Ingram, D.K., Johnson, W.D., Liu, Y., Fernandez-Kim, S.O.,White, C.L., Purpera, M.N., Uranga, R.M., Bruce-Keller, A.J., Keller, J.N., 2010. Highfat diet increases hippocampal oxidative stress and cognitive impairment in agedmice: implications for decreased Nrf2 signaling. J. Neurochem. 114, 1581–1589.

Nerurkar, P.V., Johns, L.M., Buesa, L.M., Kipyakwai, G., Volper, E., Sato, R., Shah, P., Feher, D.,Williams, P.G., Nerurkar, V.R., 2011. Momordica charantia (bitter melon) attenuateshigh-fat diet-associated oxidative stress and neuroinflammation. J. Neuroinflamma-tion. 8, 64.

Nieman,D.C., Henson, D.A., Nehlsen-Cannarella, S.L., Ekkens,M., Utter, A.C., Butterworth,D.E.,Fagoaga, O.R., 1999. Influence of obesity on immune function. J. Am. Diet. Assoc. 99,294–299.

Park, H.R., Park, M., Choi, J., Park, K.Y., Chung, H.Y., Lee, J., 2010. A high-fat diet impairsneurogenesis: involvement of lipid peroxidation and brain-derived neurotrophicfactor. Neurosci. Lett. 482, 235–239.

Powers, K.M., Smith-Weller, T., Franklin, G.M., Longstreth Jr., W.T., Swanson, P.D.,Checkoway, H., 2009. Dietary fats, cholesterol and iron as risk factors for Parkin-son's disease. Parkinsonism Relat. Disord. 15, 47–52.

Rao, J.S., Ertley, R.N., Lee, H.J., DeMar Jr., J.C., Arnold, J.T., Rapoport, S.I., Bazinet, R.P.,2007. n−3 polyunsaturated fatty acid deprivation in rats decreases frontal cortexBDNF via a p38 MAPK-dependent mechanism. Mol. Psychiatry. 12, 36–46.

Reale, M., Iarlori, C., Thomas, A., Gambi, D., Perfetti, B., Di Nicola, M., Onofrj, M., 2009.Peripheral cytokines profile in Parkinson's disease. Brain Behav. Immun. 23, 55–63.

Sartipy, P., Loskutoff, D.J., 2003. Monocyte chemoattractant protein 1 in obesity and in-sulin resistance. Proc. Natl. Acad. Sci. U. S. A. 100, 7265–7270.

Scaglione, R., Di Chiara, T., Cariello, T., Licata, G., 2010. Visceral obesity and metabolicsyndrome: two faces of the same medal? Intern. Emerg. Med. 5, 111–119.

Strissel, K.J., DeFuria, J., Shaul, M.E., Bennett, G., Greenberg, A.S., Obin, M.S., 2010. T-cellrecruitment and Th1 polarization in adipose tissue during diet-induced obesity inC57BL/6 mice. Obesity (Silver Spring). 18, 1918–1925.

Teismann, P., Tieu, K., Choi, D.K., Wu, D.C., Naini, A., Hunot, S., Vila, M., Jackson-Lewis, V.,Przedborski, S., 2003. Cyclooxygenase-2 is instrumental in Parkinson's disease neuro-degeneration. Proc. Natl. Acad. Sci. U. S. A. 100, 5473–5478.

Wild, S., Roglic, G., Green, A., Sicree, R., King, H., 2004. Global prevalence of diabetes:estimates for the year 2000 and projections for 2030. Diabetes Care 27, 1047–1053.

Willett, W.C., Manson, J.E., Stampfer, M.J., Colditz, G.A., Rosner, B., Speizer, F.E.,Hennekens, C.H., 1995. Weight, weight change, and coronary heart disease inwomen. Risk within the ‘normal’ weight range. JAMA. 273, 461–465.

Willett, W.C., Dietz, W.H., Colditz, G.A., 1999. Guidelines for healthy weight. N. Engl. J.Med. 341, 427–434.

Winzell, M.S., Coghlan, M., Leighton, B., Frangioudakis, G., Smith, D.M., Storlien, L.H.,Ahren, B., 2011. Chronic glucokinase activation reduces glycaemia and improvesglucose tolerance in high-fat diet fed mice. Eur. J. Pharmacol. 663, 80–86.

Wu, H., Ghosh, S., Perrard, X.D., Feng, L., Garcia, G.E., Perrard, J.L., Sweeney, J.F., Peterson, L.E.,Chan, L., Smith, C.W., Ballantyne, C.M., 2007. T-cell accumulation and regulated on acti-vation, normal T cell expressed and secreted upregulation in adipose tissue in obesity.Circulation 115, 1029–1038.

Xu, Q., Park, Y., Huang, X., Hollenbeck, A., Blair, A., Schatzkin, A., Chen, H., 2011. Diabe-tes and risk of Parkinson's disease. Diabetes Care 34, 910–915.