Ethyl-eicosapentaenoate (E-EPA) attenuates motor impairments and inflammation in the MPTP-probenecid mouse model of Parkinson's disease

Behavioural Brain Research 226 (2012) 386–396

Contents lists available at SciVerse ScienceDirect

Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr

Research report

Ethyl-eicosapentaenoate (E-EPA) attenuates motor impairments and inflammation in the MPTP-probenecid mouse model of Parkinson’s disease D.W. Luchtman a,1 , Q. Meng a , C. Song a,b,∗ a Department of Biomedical Sciences, University of Prince Edward Island & National Institute for Nutrisciences and Health, 550 University Avenue, Charlottetown, PEI, C1A 4P3, Canada b National Engineering Institute for the Development of Endangered Medicinal Resources in Southwest China, Guangxi Botanic Garden of Medicinal Plants, 189 Changgang Road, 530023, Nanning China

a r t i c l e

i n f o

Article history: Received 11 August 2011 Received in revised form 16 September 2011 Accepted 20 September 2011 Available online 29 September 2011 Keywords: Parkinson’s disease 1-Methyl-4-phenyl-1,2,3,6tetrahydropyridine Eicosapentaenoic acid Behavior Neurotransmitters Inflammation

a b s t r a c t Parkinson’s disease (PD) is a neurodegenerative disorder, characterized by hypokinesia, but also mood and cognitive disorders. Neuropathologically, PD involves loss of nigrostriatal dopamine (DA) and secondary non-dopaminergic abnormalities. Inflammation may contribute to PD pathogenesis, evident by increased production of pro-inflammatory cytokines. PD onset has been positively associated with dietary intake of omega-(n)-6 polyunsaturated fatty acids (PUFA). On the other hand, omega-(n)-3 PUFA may benefit PD. One of these n-3 PUFA, eicosapentaenoic acid (EPA), is a neuroprotective lipid with anti-inflammatory properties, but its neuroprotective effects in PD are unknown. Thus, we presently tested the hypothesis that EPA can protect against behavioral impairments, neurodegeneration and inflammation in the 1methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-probenecid (MPTP-P) mouse model of PD. MPTP-P injections caused hypokinesia in the rotorod and pole test, hyperactivity in the open field, and impaired mice on the cued version (procedural memory) of the Morris water maze. MPTP-P caused a loss of nigrostriatal DA and altered neurochemistry in the frontal cortex and hippocampus. Furthermore, striatal levels of pro-inflammatory cytokines were increased, while the brain n-3/n-6 lipid profile remained unaltered. Feeding mice a 0.8% ethyl-eicosapentaenoate (E-EPA) diet prior to MPTP-P injections increased brain EPA and docosapentaenoic acid (DPA) but not docosahexaenoic acid (DHA) or n-6 PUFA. The diet attenuated the hypokinesia induced by MPTP-P and ameliorated the procedural memory deficit. E-EPA also suppressed the production of pro-inflammatory cytokines. However, E-EPA did not prevent nigrostriatal DA loss. Based on this partial protective effect of E-EPA, further testing may be warranted. © 2011 Published by Elsevier B.V.

1. Introduction Parkinson’s disease (PD) is a neurodegenerative disorder, characterized by motor impairments and neuropsychiatric

Abbreviations: ␣-LA, alpha-linolenic acid; 5-HT, serotonin; 5-HIAA, 5hydroxyindoleacetic acid; AA, arachidonic acid; COX-2, cyclo-oxygenase-2; cPLA2, cytosolic phospholipase A2; DA, dopamine; DHA, docosahexaenoic acid; DOPAC, 3,4-dihydroxyphenylacetic acid; DPA, docosapentaenoic acid; E-EPA, ethyl-eicosapentaenoate; EPA, eicosapentaenoic acid; GLA, gamma-linolenic acid; HVA, homovanillic acid; IFN-␥, interferon-gamma; IL-1␤, interleukin-1␤; IL-10, interleukin-10; LA, linoleic acid; MPP+, 1-methyl-4-phenylpyridinium; MPTP, 1-methyl 4-phenyl-1,2,3,6-tetrahydropyridine; MPTP-P, MPTP-probenecid; n-3, omega-3; n-6, omega-6; NA, noradrenaline; PD, Parkinson’s disease; PPAR, peroxisome proliferator-activated receptor; PUFA, polyunsaturated fatty acid; SNpc, substantia nigra pars compacta; TNF-␣, tumor necrosis factor-␣. ∗ Corresponding author at: Department of Psychology, Life Science Centre, Dalhousie University, 1355 Oxford Street, Halifax B3H 4R2, Canada. Tel.: +1 902 494 2036 (Canada)/+86 771 244 3036 (China). E-mail address: [email protected] (C. Song). 1 Present address: Department of Pharmacology, 9-36 Medical Sciences Building, University of Alberta, Edmonton, AB, T6G 2H7, Canada. 0166-4328/$ – see front matter © 2011 Published by Elsevier B.V. doi:10.1016/j.bbr.2011.09.033

symptoms, including anxiety and depression. Progressive degeneration of nigrostriatal dopaminergic neurons leads to the loss of dopaminergic terminals and dopamine (DA) in the striatum. Neurodegeneration may also extend to non-dopaminergic and extra-nigrostriatal systems, which may underlie the complex symptomatology of PD. Indeed, cognitive impairment and dementia may occur in PD as well [1]. Although Parkinsonian cognitive impairment is typically characterized by fronto-striatal executive dysfunction, atrophy in the hippocampus, an important brain region in memory processing, has been demonstrated [2]. Among various pathogenesis, neuroinflammation may be important in PD [3]. Activated microglia have been reported in PD and the nigrostriatal system has the brain’s highest density of microglia [3,4]. Activated microglia can produce pro-inflammatory cytokines, which may be directly neurotoxic [3]. Furthermore, enzymes involved in the arachidonic acid (AA) cascade are activated in PD [5,6] and these enzymes can be triggered by cytokines [7]. Cytosolic phospholipase A2 (cPLA2) catalyzes the release of omega-(n)-6 polyunsaturated fatty acid (PUFA) AA from neuronal membranes, and further modification of AA by cyclooxygenase 2 (COX-2) leads to the production of oxidants and

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396

potentially pro-inflammatory eicosanoids [8]. Increased intake of n-6 in proportion to omega-(n)-3 PUFA, as in Western diets, increases the availability of n-6 substrate for COX-2, which may be linked to the increased prevalence of neurodegenerative disorders such as PD [9]. Furthermore, administration of Parkinsonian toxin 1-methyl-4-phenylpyridium (MPP+), the active metabolite of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), to mouse brain slices increased AA content and cPLA2 expression [10], further suggesting involvement of the AA metabolic pathway and inflammation in PD pathogenesis. On the other hand, therapeutic benefit of omega-(n)-3 PUFA in neurodegenerative disorders has been reported [11]. By influencing biophysical properties of membranes and metabolism to eicosanoids, n-3 PUFA modulate neurochemistry, signal transduction and gene expression [11,12]. Well documented are their anti-inflammatory effects. Eicosapentaenoic acid (EPA), can suppress the formation of inflammatory eicosanoids, by competing with AA for COX-2 [12]. Furthermore, by activating peroxisome proliferator-activated receptors (PPARs), n-3 PUFA can suppress the transcription of inflammatory genes, including COX-2 [13,14]. These pleiotropic effects of n-3 PUFA may explain the behavioral (motor, mood and cognition) changes observed with modulation of brain n-3 content by dietary restriction or supplementation [15]. As current therapies are limited to symptom reduction [16], treatment for PD is urgent. Recent studies [17–19] demonstrating neuroprotective effects of n-3 PUFA in experimental PD used a fish-oil compound largely containing docosahexaenoic acid (DHA). While DHA is the major n-3 PUFA of neuronal membranes and EPA is only a trace component, the latter is known to be a neuroactive and protective lipid [20,21]. EPA is extensively utilized to modulate cellular functions, which may explain its lower quantity in neuronal membranes [22]. EPA also prevented interleukin-1␤ (IL-1␤)-induced inflammation, anxiety, depression and memory-impairment in rodents [23–26]. Furthermore, we previously demonstrated that EPA has some modest neuroprotective action against experimental PD induced by MPP+ in mouse brain slices [10]. To obtain more elaborate in vivo evidence, we presently tested the hypothesis that a diet enriched in EPA can attenuate behavioral impairments, neurodegeneration and inflammation induced by MPTP in mice. We used the MPTP-probenecid (MPTP-P) model, since it causes the most profound striatal and extra-striatal neuropathology as well as inflammation [27,28]. 2. Material and methods 2.1. Animals All experimental procedures were carried out in agreement with the Canadian Council on Animal Care (CCAC), and approved by the Animal Care Committee of the University of Prince Edward Island. Male C57BL/6 mice (Charles River, Canada), 6 weeks of age, were habituated for 1 week to the animal colony and had free access to food (regular chow, PMI® Richmond, Indiana, US) and water at all times. Animals were housed individually in a 12–12 light–dark cycle with room temperature at 22 ± 1 ◦ C.

2.2. Experimental procedures and methods For an overview of the experimental design, see Fig. 1.

2.2.1. Behavioral pre-training One week following animal habituation, the mice were pre-trained on the rotorod and pole-test until ceiling performance was reached. As evident in Fig. 1, the time between pre-training and experimental sessions was substantial, at 79 days. While some may argue that mice will forget the acquired tasked, we found this not to be the case, as saline treated (control) mice still performed near maximum level (rotorod distance moved and pole test total time) on the first day of post-MPTP experimentation. Since we were interested in the effect of experimental treatment on motor performance “per se” and not motor learning, we considered it important to pre-train mice prior to any experimental manipulation.

387

2.2.2. Experimental diets Immediately following pre-training, the regular rat chow was replaced with a diet consisting of regular chow supplemented with 0.8% ethyl-eicosapentaenoate (E-EPA) (Amarin Neuroscience Ltd, UK) (N = 20) or palm oil (Sigma, Canada) (N = 20). The latter served as control diet, as it ensured comparable caloric values, contains very low amounts of linoleic acid (C18:2, n-6) and no n-3 fatty acids [24]. The nutritional composition and procedures of the diets are outlined in Table 1. The feeding period prior to MPTP-P injections was 6 weeks, but the diet continued until the animals were sacrificed. It was previous determined in C57Bl/6 mice that 6 weeks is the sufficient feeding time to raise brain EPA levels significantly [10]. 2.2.3. MPTP-P administration 100 mg MPTP hydrochloride (Sigma, Canada) was dissolved in 20 mL sterile saline, rendering a 5 mg/mL solution that was injected. Probenecid (Sigma, Canada), is an adjuvant drug that reduces the renal excretion of MPTP and thus its excretion from the body, thereby prolonging the neurotoxic effect of MPTP during the 3.5 days inter-injection interval. It was dissolved in 16 mL 0.1 N NaOH, which was stabilized with 1 M Tris–HCl (pH 6.8) to a pH of 7.4 and final volume adjusted to 32 mL with ddH2 O [27]. Probenecid solutions were prepared freshly for each injection. The chronic MPTP-P dosing regimen was adapted from previous protocols [28] and conducted as described in Ref. [27]. In brief, mice were administered 10 injections of MPTP 25 mg/kg s.c. combined with probenecid 250 mg/kg i.p. (n = 20), or saline s.c. combined with probenecid 250 mg/kg i.p. (n = 20), 3.5 days apart. Thus, there were four experimental groups: (1) Palm and MPTP-P (MP); (2) E-EPA and MPTP-P (ME); (3) Palm and saline/P (P); (4) E-EPA and saline/P (E). Since probenecid by itself does not affect neurochemistry or behavior [28], an extra experimental group to investigate the effect of probenecid was not carried out in the present study. 2.2.4. Behavior testing A test battery similar to Ref. [27] was used to assess the hypokinetic, mood and cognitive aspects of Parkinsonism. All tests were done between 9 am and 2 pm under normal animal room lighting. The rotarod (hypokinesia) and open field (hypokinesia and anxiety) were tested on the second and third day after the last injection respectively. The pole test (hypokinesia) was performed on the fourth day following injection. For spatial learning and memory, the Morris water maze was carried out one week after injections. 2.2.4.1. Rotorod. The protocol was modified from previous procedures [29]. The animals were positioned on the rotorod (IITC Life Science,CA, USA), which was programmed to rotate with linearly increasing speed from 1 rpm to 30 rpm in 300 s. Automatic sensors captured when animals fell off the rod and the total distance moved (m) automatically calculated. During pre-training, mice performed 4 trials a day and rested 1 min between trials. On the post-MPTP-P testing day, the average of total distance moved on 5 successive trials was measured. Between each trial, animals rested for 1 min. 2.2.4.2. Open field. A white plastic rectangular box (70 cm × 30 cm × 20 cm) with 5 cm2 squares drawn on the bottom was used for assessment of mouse locomotor and anxiety-related behavior in a bright, open environment, as previously done [27]. The mouse was placed in the center of the open field. Grid crosses (total distance moved), rearing and grooming were manually counted for 5 min, in normal room lighting (no extra bright light source was placed immediately above the apparatus). Video recordings of each session were used to measure the percentage of time spent in the center of the arena, which is a measure of anxiety [30]. The equipment was cleaned with 70% alcohol and water between trials. Each animal performed only 1 trial. 2.2.4.3. Pole test. The method was adapted from Ogawa et al. [31]. The pole test consisted of a 50 cm high wooden pole, 0.5 cm in diameter, wrapped in gauze to prevent slipping and the base position in the home cage. A rubber ball was glued on top of the pole to prevent animals from sitting on the top and to help position the animals on the pole (by sliding the forepaws over the ball and holding the animal by the tail). The time that animals required to climb down the pole was measured. If the mouse did not descend in 60 s, it was guided. During pre-training as well as post-MPTP-P sessions, each animal performed 3 successive trials, with a 5 min inter-trial interval. The average of the three trials was taken for statistical analyses. 2.2.4.4. Morris water maze. The method used for the Morris water maze was adapted from a more recent version [32]. Briefly, the water tank (1 m in diameter, with 40 cm high walls) was divided into 4 equally sized quadrants and designated as north (N), west (W), south (S), and east (E). Water was made opaque by the addition of water-soluble and non-toxic tempera paint (Funstuff) and temperature maintained at 23 ± 1 ◦ C. A security camera was mounted above the center of the tank, and swimming trials were recorded. One day prior to testing, animals were habituated to the water by swimming for 1 min without the escape platform. During the testing days, mice started each trial facing the rim of the pool and were given 60 s

388

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396

Fig. 1. Experimental design. This diagram presents an overview of the experimental design and group division in the experiment.

Table 1 Fatty acid composition of diets. Standard chow fat contents (% of total mass)a Crude fat Fatty acid contents in crude fat (% of crude fat) C18:1 Oleic C18:2, n-6 Linoleic C18:3, n-6 Gamma Linolenic C18:3, n-3 Linolenic C20:4, n-6 Arachidonic C20:5, n-3 Eicosapentaenoic C22:5, n-3 Docosapentaenoic C22:6, n-3 Docosahexaenoic Total n-6 Total n-3

4.5 27.6 28.9 <0.1 2.31 0.30 0.85 0.19 0.96 29.7 4.54

Experimental diet PUFA content (% of total mass) Total n-6 content palm oil diet Total n-3 content palm oil diet Total EPA content palm oil diet Total n-6 content E-EPA diet Total n-3 content E-EPA diet Total EPA content E-EPA diet

1.4 0.2 0.04 1.34 1 0.84

N-6:n-3 ratio palm oil diet N-6:n-3 ratio E-EPA diet

7:1 1.3:1

For preparation of the diets, the master bottle of E-EPA (oil) was taken from −20 ◦ C freezer and oil aliquoted in several 15 mL Falcon tubes that were flushed with nitrogen gas to slow oxidation. The appropriate volume of oil (0.8% in predetermined amount of chow) was then added to the chow and carefully mixed. For palm oil preparation, the palm oil had to be heated (50 ◦ C) first to liquidize the fat. The prepared diets were fed immediately to the mice (10 g per mouse per day). The diets were fed daily and prepared each time for three days. The EPA diet was stored in −20 ◦ C between feeding to further slow down oxidation. The palm oil diet was stored in 4 ◦ C between feeding. a Courtesy of LabDiet, PMI Nutrition International, LLC.

to complete a trial. On the first testing day, mice were trained to locate and escape on a Perspex escape platform (diameter 9 cm) that was made visible (cued version of the test) by elevating it slightly above the water and placing a brightly colored object on top of it (flag). Mice were given 4 trials and with each trial, the location of the platform changed to a different quadrant in a pseudorandom order. The purpose of this cued test was to assess procedural memory (habit learning) as well as the motivation of the mice to escape [33,34]. It also offered mice practice to escape on the platform. During the subsequent three testing days (spatial version of the test), the platform was submerged 0.5 cm below the water surface and the animals trained to locate the hidden platform. Again, mice were given 4 trials, with inter-trial times of approximately 5 min. The submerged platform remained in the same position (in the middle of one of the four quadrants), but the starting quadrant of the mice was pseudorandomly changed with each of the four trials. Those mice that failed to reach the platform within 1 min were guided towards it and then permitted to stay on for 15 s. The main dependent measure during cued and spatial training was escape latency (s). On the fourth day, the platform was removed after the fourth trial. During this probe trial, the swimming behavior was recorded over 60 s, while the mouse searched for the removed platform. The percentage of search time in the correct quadrant (the one that contained the platform) and the total number of crossings of the exact position where the platform was located (annulus) were measured. Furthermore, swimming speed and the total number of quadrant entries (each time an animal passed the boundaries of a quadrant, an entry was counted) were measured. In post hoc analyses from videorecorded material, swimming speed was measured as pixels/s with Mousetracker version 1.0 by Adriano Tort.

2.2.5.1. Neurotransmitters and their metabolites in the striatum, frontal cortex and hippocampus. Concentrations of DA, noradrenaline (NA), serotonin (5-HT), DA metabolites 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) and 5-HT metabolite 5-hydroxyindoleacetic acid (5-HIAA), were measured in the supernatants by HPLC with electrochemical detector. HPLC analyses were conducted as described in detail in Ref. [27]. Data was expressed as pg per milligram of protein content.

2.2.5. Sacrifice and biochemical analyses Following the Morris water maze, the animals rested for 2 days. On the next day, animals were sacrificed by cervical dislocation. Brains were rapidly removed and the striatum, frontal cortex, hippocampus and midbrain dissected on ice. For analyses of neurochemistry and brain cytokines, brain samples of one hemisphere were sonicated (Misonix Inc., XL2000, NY, USA) in ice cold lysis buffer (1 L contained 8.8 mg l-ascorbic acid, 8.8 mg; 9.3 mL 70% HClO4 , and 100 mg EDTA), centrifuged at 10,000 × g for 25 min at 4 ◦ C and stored in −20 ◦ C. These samples were also used for BCA protein assay (Sigma, Canada, #QPBCA) to measure total protein content in the samples. For mRNA and PUFA analyses, brain samples of the other hemisphere were flash frozen in liquid nitrogen and stored at −80 ◦ C.

2.2.5.4. Real-time PCR of cPLA2 and COX-2. Brain tissue was lysed in TRI-reagent (Sigma, Canada) (1 mL per 50–100 mg tissue) by sonication (Misonix Inc., XL2000, NY, USA) at Power level 2. The RNA extraction method was followed according to manufacturer’s instructions. Complementary DNA (cDNA) was synthesized using the Quantiscript reverse transcription kit (Qiagen, Canada) according to manufacturer’s instructions. Primer sequences for the target genes (Table 2) were obtained from NCBI’s Nucleotide database and designed in PRIMER3 (Biology Workbench 3.2). Synthesis of these primers was performed by Sigma–Aldrich (Canada). Primers were tested for specificity in NCBI-nucleotide-BLAST. Polymerase chain reactions were carried out with Quantitect SYBR Green (Qiagen, Canada, \) master mix kits, according to manufacturer’s instructions.

2.2.5.2. Cytokine measurements in the striatum and midbrain. A customized 4-plex mouse cytokine panel (Biorad, CA, USA) consisting of fluorescent beads for proinflammatory cytokines interleukin-(IL)-1␤, tumor necrosis factor-(TNF)-␣ and interferon-(IFN)-␥ and anti-inflammatory cytokine interleukin-(IL)-10 was used with a Luminex protein suspension array system (Bioplex 200, Biorad, CA, USA), according to manufacturer’s instructions. This assay is highly sensitive and has been demonstrated to work well with detecting cytokines from brain tissues [27,35]. Data was expressed as pg per milligram of protein content. 2.2.5.3. Lipid extraction and gas chromatography. Frontal cortex tissue was analyzed with gas chromatography (GC). Lipids were extracted according to Folch’s method, as described in detail [10]. The fatty acid profiles were expressed as a percentage of the total microgram of fatty acid (weight percentage). Frontal cortex tissue was used as significant changes in PUFA content are observed in this tissue following dietary supplementation [17].

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396 Table 2 Primer names, sequences and product sizes (mouse source). Primer name

Sequence

Actin F Actin R cPLA2 F cPLA2 R COX 2 F COX 2 R

TGTTACCAACTGGGACGACA CTGGGTCATCTTTTCACGGT GACAGCTCCGACAGTGATGA GACCCAACTTGCTTGGTTGT CCAGATGCTATCTTTGGGGA GCTCGGCTTCCAGTATTGAG

389

Nonetheless, MPTP-P significantly decreased the distance mice walked on the rotorod only in the palm oil group (P-MP: t = 3.9, P < 0.05, Bonferroni post hoc after two-way ANOVA) (Fig. 3A), suggesting that E-EPA partially attenuated the MPTP-P induced rotorod impairment. 3.2.2. Pole test A significant main effect of MPTP-P was found, which significantly interacted with DIET (MPTP-P × DIET: F[1,34] = 4.2, P < 0.05). Bonferroni tests showed that MPTP-P significantly increased the total time which the mice needed to climb down the pole (P-MP: t = 3.48, P < 0.01) and E-EPA attenuated this effect (MP-ME: t = 2.72, P < 0.05) (Fig. 3B). 3.2.3. Open field On line crosses, a significant main effect of MPTP-P was found, but not of DIET. MPTP-P significantly increased the line crosses (distance moved), but again only in the palm oil group (P-MP: t = 2.77, P < 0.05, Bonferroni after two-way ANOVA) (Fig. 3C). It should be noted that E-EPA fed mice also increased line crosses and this effect bordered significance. In a former pilot study, we found that E-EPA significantly increased line crosses and rearings compared to palm oil. To determine whether this increase in the activity had a possible anxiety component, the percentage of time spent in the center area was measured. While MPTP-P did not have an effect on center exploration, there was a main effect of DIET. The percentage of time spent in the center area was increased by E-EPA feeding compared to palm oil in saline treated mice (P-E: t = 2.61, P < 0.05) (Fig. 3D). There were no significant main or interaction effects on grooming and rearing (data not shown).

Fig. 2. Body weight. Progression of body weight (g) over the course of the experiment in the four treatment groups. Data are presented as mean ± SEM.

2.3. Statistics Statistics were performed on GraphPad Prism 4.03 and SPSS 17.0 for Windows. Data were analyzed with two-way ANOVA, with main effects for factors MPTP-P, DIET and their interaction (MPTP-P × DIET). For body weight measurements and Morris water maze performance, a three-way repeated-measures ANOVA was performed, with between-subject factors DIET and MPTP-P; within-subject factors TIME (the weeks that body weight was sampled) or DAY (testing days in the Morris water maze), and their interactions. Bonferroni post hoc analysis was used to compare significance among the four groups in case of significant main effects. Bivariate correlations were calculated with the Pearson correlation coefficient r.

3. Results 3.1. Body weight Body weight was significantly increased over time in all four groups (TIME: F[16,518] = 88.9, P < 0.0001) as the mice were growing, but with progression of MPTP-P treatment (from day 8 of MPTP-P treatment and onwards), MPTP-P started to have a suppressing effect on body weight (MPTP-P × TIME (F[16,518] = 4.23, P < 0.0001), as evident in Fig. 2. Although MPTP-P only suppressed body weight in the palm oil treated mice, the E-EPA diet did not fully attenuate this effect, as the difference between MP and ME groups was not significant. The E-EPA diet by itself also did not affect body weight. 3.2. Behavior 3.2.1. Rotorod A significant main effect of MPTP-P was found, but not of DIET and there was no interaction between MPTP-P and DIET.

3.2.4. Morris water maze No significant main effect of MPTP-P or DIET was found on latency to escape on the visual platform. However, there was an interaction between MPTP-P and DIET that approached significance (F[1,31] = 3.12, P < 0.05), justifying post hoc tests. Bonferroni tests revealed that compared to saline treated mice, MPTP-P treated mice that were fed palm oil but not E-EPA had significantly longer escape latencies (P-MP: t = 2.38, P < 0.05) (Fig. 4A) without a change in swimming speed (data not shown), suggesting that E-EPA partially attenuated the effect of MPTP-P. In the hidden platform version of the test, a learning curve over the three testing days can be seen in Fig. 3B, which was confirmed by the significant within-subject factor (DAY: F[2,62] = 27.7, P < 0.0001) of the repeated measures analyses. Importantly, there was no interaction between DAY and the treatment factors MPTP-P or DIET, suggesting that all groups had a similar rate of learning over the three testing days (Fig. 4B). However, there was a significant between-subject effect of MPTP-P (F[1,33] = 29.3, P < 0.0001), suggesting that the toxin did affect the average escape latency compared to saline. Multivariate ANOVA for each day showed that MPTP-P significantly increased escape latency on day 1 in palm-oil fed mice (P-MP: t = 4.5, P < 0.01, Bonferroni test) (Fig. 4C), but not significantly on day 2 and 3 (data not shown). No effect of MPTP-P or E-EPA was found on velocity (data not shown). Probe trial performance is an important measure of acquired spatial memory. On total number of annulus crossings (Fig. 4D) or percentage of time spent in the correct quadrant, which indicate whether the subjects memorized, respectively, the exact and coarse location of the platform, there were no significant effects of MPTP-P or DIET. On the total number of quadrant entries however, there was a main effect of MPTP-P and it was found that MPTP-P significantly increased the total number of quadrant entries (PMP: t = 2.8, P < 0.05, Bonferroni test after two-way ANOVA), which was reversed by E-EPA treatment (MP-ME: t = 2.5, P < 0.05) (Fig. 4E).

390

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396

Fig. 3. Hypokinesia and anxiety-related behavior. (A) Rotorod performance in the four groups, measured as the total distance moved (m), averaged over 5 consecutive trials. (B) Pole test total time (s) in the four groups, measured as the time the mouse needed to complete the test (inverse and climb down the pool), averaged over three consecutive trials. (C) Line crosses in the open field, a measure of locomotor activity, measured during the 5-min trial. (D) Anxiety in the open field, measured as the percentage of time spent in the center area, during the 5 min session. Data are presented as mean ±SEM. *P < 0.05 (P-MP); **P < 0.01 (P-MP); # P < 0.05 (MP-ME); ˆ P < 0.05 (P-E). P (palm oil + saline-probenecid); E (E-EPA + saline-probenecid); MP (palm oil + MPTP-probenecid); ME (E-EPA + MPTP-probenecid).

Fig. 4. Morris water maze: Performance in the Morris water maze in the four groups. (A) Escape latency (s) in the “cued version” (visual platform) of the test, averaged over four trials. (B) Learning curve of escape latency (s) over the three consecutive days of testing in the “spatial version” (hidden platform) of the test. Presented is the daily average of four trials. (C) Group comparison on day 1 of the “spatial version” of the test. (D) and (E) Performance on the “probe trial”, in which the platform was removed from the pool. (D) Group comparison on annulus crossings; the mean number times each mouse swam through the “annulus”, the exact area where the platform was located during the preceding spatial learning (hidden platform) trials. (E) The mean number of quadrant entries during the probe trial. Results are presented as mean ±SEM. *P < 0.05 (P-MP); **P < 0.01 (P-MP); # P < 0.05 (MP-ME). P (palm oil + saline-probenecid); MP (palm oil + MPTP-probenecid); ME (E-EPA + MPTP-probenecid).

2.7, P < 0.05

3.1, P < 0.05 2.6, P < 0.05

6.5, P < 0.001 3.9, P < 0.01

E-ME P-MP P-E

14.5, P < 0.001 6.2, P < 0.001 4.1, P < 0.01

MP-ME

14.95, P < 0.001 6.7, P < 0.001 4.1, P < 0.01 4.3, P < 0.01 5.3, P < 0.001 3.8, P < 0.01

430 (40.4) 35.5 (15.05) 17.5 (6.5) 0.071 (0.004) 0.24 (0.01) 34.9 (6.5) 7.1 (1.8)a 5.9 (0.9) 1.05 (0.42)

Presented are the neurotransmitter concentrations (pg/mg protein) in the striatum, for each treatment group. Results of Bonferroni t-tests are shown as well, in case significant omnibus tests were found. Alpha level was set at 0.05. For the Bonferroni tests, t and P-values are shown. P (palm oil + saline-probenecid); E (E-EPA + saline-probenecid); MP (palm oil + MPTP-probenecid); ME (E-EPA + MPTP-probenecid). DA, dopamine; DOPAC, 3,4dihydroxyphenylacetic acid; HVA, homovanillic acid; DOPAC/DA, ratio of DOPAC to DA (indicator of DA turnover); HVA/DA, ratio of HVA to DA (indicator of DA turnover); NA, noradrenaline/norepinephrine; 5HT, serotonin; 5-HIAA, 5-hydroxyindoleacetic acid; 5-HIAA/5-HT, 5-HIAA to 5-HT ratio (indicator of 5-HT turnover).

MP Mean (SEM) N = 10 E Mean (SEM) N=9 P Mean (SEM) N=9 Neurotransmitter

3.3.3. Hippocampus In this brain region, MPTP-P, but not DIET caused a main effect on DA. MPTP-P depleted DA by 34% in the palm oil (P-MP: P < 0.05, Bonferroni test after two-way ANOVA) and 49% in E-EPA fed mice (E-ME: P < 0.01). On metabolite DOPAC, there was a significant interaction between MPTP-P and DIET (F[1,32] = 14.92, P < 0.001), justifying a post hoc test. MPTP-P increased DOPAC levels in the hippocampus by 116% in palm-oil fed mice (P-MP: P < 0.01). Although E-EPA by itself also increased DOPAC levels by 116% (P-E: P < 0.05), when combined with MPTP-P, E-EPA reduced DOPAC levels to those of palm oil control (saline) levels (MP-ME: P < 0.05). Due to depletion of DA in the hippocampus and increased levels of metabolite

Table 3 Effect of MPTP-P and E-EPA treatment on neurochemistry in the striatum.

3.3.2. Frontal cortex Both MPTP-P and DIET caused main effects on DA levels, but there was no interaction. Post hoc Bonferroni testing indicated that MPTP-P decreased DA levels by 35% in the palm (P-MP: P < 0.05) and 40% in E-EPA fed mice (E-ME: P < 0.01). Interestingly, E-EPA treatment by itself increased frontal cortex DA levels by 43% compared to palm oil (P-E: P < 0.05). MPTP-P increased the DOPAC/DA ratio by 107% in E-EPA fed mice (E-ME: P < 0.05), compared to 33% in palm oil fed mice (not-significant). There was also a main effect of MPTPP on NA and post hoc tests revealed that NA was depleted by 13% in E-EPA (E-ME: P < 0.001) versus 7% (not significant) in palm oil fed mice. Interestingly, the cortical 5-HIAA/5-HT ratio was 73% higher in MPTP-P treated E-EPA fed mice compared to MPTP-P treated palm oil fed mice (MP-ME: P < 0.05).

ME Mean (SEM) N = 8

3.3.1. Striatum MPTP-P caused significant main effects on catecholamine DA and its metabolites DOPAC and HVA, but no interaction with DIET was found on all these measures. Indeed, the decrease of DA by MPTP-P was remarkably similar between diets; 79% in the palm oil (P-MP: P < 0.001, Bonferroni test after two-way ANOVA) and 81% in E-EPA group (E-ME: P < 0.001). Similarly, DOPAC was depleted 63% in the palm (P-MP: P < 0.001) and 62% in the E-EPA group (E-ME: P < 0.001). HVA was depleted 50% in the palm oil (P-MP: P < 0.01) and 52% in the E-EPA group (E-ME: P < 0.01). While HVA and DOPAC were decreased, DA was depleted more, so the ratios of DOPAC and HVA to DA were significantly increased. For both ratios, there was a significant main effect of MPTP-P, but again no interaction with DIET. Nonetheless, the DOPAC to DA was increased by 125% in the palm oil group (P-MP: P < 0.01), but only 40% in the E-EPA group (not significant), suggesting that the E-EPA diet did cause some modulation of the MPTP-P effect on DA turnover. However, the HVA/DA ratio was not different between diet groups, as it was increased 250% in the palm (P-MP: P < 0.001) and 260% in the E-EPA group (E-ME: P < 0.001). Quite differently, however, on NA, there was a significant interaction between MPTP-P and the DIET factor (F[1,30] = 11,6, P = 0.0021). MPTP-P significantly decreased NA level by 58% in the palm oil group (P-MP: P < 0.01). Although E-EPA feeding by itself (no MPTP-P treatment) also decreased NA level (PE: P < 0.01), it significantly reversed the MPTP-P-induced depletion of NA (MP-ME: P < 0.05). There was also an interaction between MPTP-P and DIET factors on 5-HT metabolite 5-HIAA (F[1,29] = 3.9, P = 0.052). Post hoc tests indicated that MPTP-P increased 5-HIAA levels by 19% (P-MP: P < 0.05) and this effect was attenuated by E-EPA (MP-ME: P < 0.05).

455 (80.6) 35.7 (6.3) 18.9 (9.5) 0.086 (0.01) 0.20 (0.03) 18.5 (1.6) 7.5 (1.7) 8.7 (1.1) 1.62 (0.42)

Results for neurochemistry in the striatum, frontal cortex and hippocampus are respectively shown in Tables 3–5 .

2156 (95.6) 93.5 (4.3) 36.03 (5.5) 0.05 (0.002) 0.09 (0.008) 17.2 (2.5) 5.9 (0.14) 6.3 (0.7) 1.52 (0.39)

3.3. Neurochemistry

2200 (105.5) 94.5 (11.5) 37.8 (8.9) 0.04 (0.008) 0.08 (0.01) 46.2 (8.3) 5.03 (0.11) 7.3 (0.75)1 1.85 (0.39)

On velocity in the probe trial, there were no significant effects of MPTP-P or DIET.

391

DA DOPAC HVA DOPAC/DA HVA/DA NA 5-HT 5-HIAA 5-HIAA/5-HT

Bonferroni t-test (only significant mean differences) t, P < 0.05; P < 0.01; P < 0.001

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396

392

Table 4 Effect of MPTP/P and E-EPA treatment on neurotransmitter content in the frontal cortex. P Mean (SEM) N = 9

E Mean (SEM) N = 9

MP Mean (SEM) N = 10

ME Mean (SEM) N = 8

DA DOPAC HVA DOPAC/DA HVA/DA NA 5-HT 5-HIAA 5-HIAA/5-HT

8.6 (0.9) 1.5 (0.2) 15.6 (0.9) 0.18 (0.04) 1.83 (0.26) 181.2 (4.0) 1.1 (0.2) 6.1 (0.5) 5.5 (0.66)

12.3 (1.3) 1.4 (0.2) 14.1 (1.7) 0.13 (0.04) 1.44 (0.13) 190.2 (3.2) 0.9 (0.2) 6.9 (0.6) 7.6 (0.87)

5.6 (0.6) 1.5 (0.1) 14.7 (1.6) 0.24 (0.04) 2.37 (0.34) 169.2 (4.1) 0.8 (0.1) 7.2 (0.8) 9.1 (1.5)

7.4 (1.9) 1.4 (0.1) 11.1 (1.4) 0.27 (0.05) 2.40 (0.40) 167.1 (3.8) 0.5 (0.07) 7.3 (0.3) 13.2 (1.3)

Bonferroni t-test (only significant mean differences) t, P < 0.05, 0.01, 0.001 P-E

P-MP

E-ME

2.6, P < 0.05

2.5, P < 0.05

3.1, P < 0.01

MP-ME

2.9, P < 0.05 4.3, P < 0.001

3.11, P < 0.05

Presented are the neurotransmitter concentrations (pg/mg protein) in the frontal cortex, for each treatment group. Results of Bonferroni t-tests are shown as well, in case significant omnibus tests were found. Alpha level was set at 0.05. For the Bonferroni tests, t and P-values are shown. CP (palm oil + saline-probenecid); CE (E-EPA + saline-probenecid); MP (palm oil + MPTP-probenecid); ME (E-EPA + MPTP-probenecid). DA, dopamine; DOPAC, 3,4-dihydroxyphenylacetic acid; HVA, homovanillic acid; DOPAC/DA, ratio of DOPAC to DA (indicator of DA turnover); HVA/DA, ratio of HVA to DA (indicator of DA turnover); NA, noradrenaline/norepinephrine; 5HT, serotonin; 5-HIAA, 5-hydroxyindoleacetic acid; 5-HIAA/5-HT, 5-HIAA to 5-HT ratio (indicator of 5-HT turnover).

Table 5 Effect of MPTP-P and E-EPA treatment on neurotransmitter content in the hippocampus. Neuro-transmitter

DA DOPAC HVA DOPAC/DA HVA/DA NA 5-HT 5-HIAA 5-HIAA/5-HT

P Mean (SEM) N = 9

2.4 (0.3) 0.6 (0.05) 9.1 (1.5) 0.30 (0.05) 4.07 (1.01) 132 (6.6) 3.0 (0.2) 5.4 (0.7) 2.5 (0.08)

E Mean (SEM) N = 9

2.9 (0.3) 1.3 (0.2) 10.3 (1.4) 0.46 (0.09) 3.6 (0.63) 129 (4.4) 2.5 (0.2) 4.2 (0.6) 2.6 (0.21)

MP Mean (SEM) N = 10

1.6 (0.2) 1.3 (0.2) 10.0 (1.5) 0.82 (0.11) 6.2 (0.91) 135 (5.2) 2.1 (0.2)a 4.8 (0.5) 3.2 (0.31)

ME Mean (SEM) N = 8

1.5 (0.2) 0.6 (0.2) 8.7 (1.8) 0.30 (0.04) 4.8 (0.75) 132 (5.2) 2.8 (0.5) 5.1 (0.7) 2.6 (0.09)

Bonferroni t-test (only significant mean differences) t, P < 0.05, 0.01, 0.001 P-E

P-MP

E-ME

2.5, P < 0.05 3.6, P < 0.01

4.1, P < 0.01

2.8, P < 0.05

4.3, P < 0.01

MP-ME 2.8, P < 0.05 3.8, P < 0.01

2.8, P < 0.05

Presented are the neurotransmitter concentrations (pg/mg protein) in the hippocampus, for each treatment group. Results of Bonferroni t-tests are shown as well, in case significant omnibus tests were found. Alpha level was set at 0.05. For the Bonferroni tests, t and P-values are shown. P (palm oil + saline-probenecid); E (E-EPA + saline-probenecid); MP (palm oil + MPTP-probenecid); ME (E-EPA + MPTP-probenecid). DA, dopamine; DOPAC, 3,4-dihydroxyphenylacetic acid; HVA, homovanillic acid; DOPAC/DA, ratio of DOPAC to DA (indicator of DA turnover); HVA/DA, ratio of HVA to DA (indicator of DA turnover); NA, noradrenaline/norepinephrine; 5HT, serotonin; 5-HIAA, 5-hydroxyindoleacetic acid; 5-HIAA/5-HT: 5-HIAA to 5-HT ratio (indicator of 5-HT turnover).

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396

Neuro-transmitter

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396

DOPAC, there was a main effect of MPTP-P on DOPAC/DA levels and this effect interacted with DIET (MPTP-P × DIET: F[1,32] = 14.02, P < 0.001). MPTP-P significantly increased the DOPAC/DA ratio by 166% (P < 0.01) and E-EPA treatment attenuated this effect (MPME: P < 0.01). On 5-HT, an interaction effect between MPTP-P and DIET factors (F[1,28] = 5.07, P < 0.05) was found. Post hoc analyses indicated that MPTP-P significantly decreased hippocampal 5-HT by 30% (P-MP: P < 0.05), but only in the palm oil group.

393

MPTP-P, and prevented an increase in striatal pro-inflammatory cytokines. While E-EPA did attenuate some alterations in striatal and hippocampal neurochemistry, there was no clear pattern of protection at neurochemical level. Furthermore, the E-EPA diet did not attenuate the loss of nigrostriatal DA, which is a major aspect of PD neuropathology. With this mixed data, combined with the protective effects of n-3 PUFA reported in other studies, we suggest that E-EPA may have potential as a protectant in PD, but further research is warranted.

3.4. Cortex n-3 and n-6 fatty acid contents 4.1. Behavior and neurochemistry MPTP-P did not affect the concentration of EPA, docosapentaenoic acid (DPA) and DHA in cortical tissue, but E-EPA feeding by itself had significant main effects on EPA and DPA. The post hoc tests showed that the E-EPA diet significantly increased the concentrations of EPA in both saline (P-E: t = 8.593, P < 0.001) and MPTP-P treated mice (MP-ME: t = 9.691, P < 0.001) (Fig. 5A). Likewise, DPA was significantly increased by E-EPA feeding in both saline (P-E: t = 8.607, P < 0.001) and MPTP-P treated mice (MP-ME: t = 7.289, P < 0.001) (Fig. 5B). On the other hand, the E-EPA diet did not affect DHA concentration (Fig. 5C) or n-3 fatty acid ␣-LA (data not shown). With respect to n-6 fatty acids, neither MPTP-P nor the DIET factor had any effect on the tissue content of the major n-6 fatty acids linoleic acid (LA), gamma-linolenic acid (GLA) and AA (data not shown). 3.5. Cytokine concentrations in the striatal and midbrain In the striatum, there were significant main effects of MPTPP on TNF-␣ and IFN-␥ concentrations. The effect of MPTP-P on TNF-␣ levels interacted significantly with DIET [MPTP-P × DIET: F(1,32) = 3.92, P < 0.05]. Indeed, post hoc tests showed that MPTPP increased TNF-␣ expression compared to saline in the palm oil group (P-MP: t = 2.93, P < 0.05), and this effect was attenuated by E-EPA feeding (MP-ME: t = 2.4, P < 0.05) (Fig. 6A). Post hoc analyses also indicated that MPTP-P significantly increased IFN-␥ levels compared to saline, and this effect was only found significant in the palm oil group (P-MP: t = 2.47, P < 0.05) (Fig. 6B), suggesting that E-EPA partially attenuated the effect of MPTP-P. No significant effects were detected on IL-1␤ and anti-inflammatory cytokine IL10 (data not shown). In the midbrain, MPTP-P as well as DIET caused a significant main effect on anti-inflammatory cytokine IL-10. Post hoc analyses indicated that MPTP-P increased IL-10 in the palm oil group (P-MP: t = 2.4, P < 0.05) and this effect was attenuated by E-EPA feeding (MP-ME: t = 2.7, P < 0.05) (Fig. 5C). 3.6. Striatal and midbrain mRNA expression of COX-2 and cPLA2 In the striatum, there was a significant interaction between MPTP-P and DIET on COX-2 mRNA expression (MPTP-P × DIET: F[1,15] = 11.34, P < 0.01). The post hoc analyses showed that COX2 mRNA expression was significantly downregulated by MPTP-P compared to saline in the palm oil group (P-MP: t = 3.15, P < 0.05) (Fig. 7A). E-EPA feeding by itself also significantly downregulated COX-2 mRNA expression compared to palm oil in saline treated mice (P-E: t = 2.6, P < 0.05) (Fig. 7A). There were no significant effects of MPTP-P or DIET on cPLA2 mRNA expression (Fig. 7B). In the midbrain, no significant effects of either MPTP-P or E-EPA treatment on cPLA2 and COX-2 mRNA expression were found (data not shown). 4. Discussion For the first time, we tested whether E-EPA can protect against experimental PD induced by MPTP-P. The E-EPA diet significantly increased brain levels of EPA and DPA, attenuated or improved the hypokinesia and procedural memory-impairment induced by

Consistent with PD hypokinesia, MPTP-P presently caused an impairment of pole and rotorod performance, but also open field hyperactivity. This MPTP-induced hyper- rather than hypo-activity in spontaneous movement paradigms, such as the open field, has been described frequently, especially when more chronic and high accumulative dose regimens are used [27,36,37]. This may be a form of hyperactive disinhibition, caused by a fronto-striatal hypodopaminergic state [38]. It was probably not anxiety related, as open field centre square exploration was not different from controls. Other studies also did not report MPTP effects on mood [43]. The increase in quadrant entries, indicating an increase in swimming activity in MPTP-P treated mice, as observed in the water maze, may be similar disinhibition. There was presently little evidence that MPTP-P treated mice had a spatial memory impairment. Only a few studies have assessed spatial memory in MPTP treated mice and while a spatial memory impairment was observed in one (which used another MPTP-regimen) [39], these results concur with others studies demonstrating only an impairment in the cued or working memory version of the test [33,40]. The visual (cued) platform test relies on basal ganglia rather than hippocampal integrity [33,34]. It is unclear why MPTP-P treated mice had longer escape latencies on the first day of hidden platform testing. This may be a working memory impairment, as mice were trying to learn the new task, while attempting to memorize the location of the platform. Swimming was not impaired in MPTP-P treated mice and there was no change in velocity. Consistent with PD neuropathology [1], the most robust effects of MPTP-P on neurochemistry were found in the striatum. Presently, MPTP-P caused a massive depletion of DA, while the ratios of DOPAC and HVA to DA increased, as more DA was released and metabolized from remaining nigrostriatal termininals, a compensatory mechanism observed in PD 51] [41]. As in PD [1,41], MPTP-P also caused changes in striatal non-dopaminergic neurochemistry. Similar to our previous findings with this model [27], 5-HT metabolite levels were increased. Since MPP+ has less affinity to the 5-HT transporter, this effect may be an adaptation or compensation in striatal serotonergic neurochemistry in response to MPTP-P rather than a direct effect of the neurotoxin [1,41,42]. These potentially adaptive neurochemical changes may be behaviorally relevant, as pole test and rotorod performance in the MPTP-P palm oil group were correlated to striatal DA (R = .638, P < 0.05) and 5-HT metabolism (−R = .669, P < 0.05) respectively. Further consistent with PD [1,41], MPTP-P caused significant extra-striatal neuropathology. While a depletion of cortical DA is a common finding with MPTP-P [27] and other MPTP regimens [37], a new finding was the abnormalities in hippocampal neurochemistry. While the hippocampus is generally ignored in MPTP research, these findings may be consistent with hippocampal neuropathology observed in some PD patients [5]. Yet the behavioral relevance of this effect is unclear. Long-term potentiation (LTP), a neurophysiological process during learning and memory, has been linked to dopaminerginc neurochemistry in the hippocampus [44], however, memory performance was presently not affected by MPTP-P. Furthermore, we recently (data submitted) found that

394

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396

Fig. 5. Brain n-3 PUFA concentrations: Levels of brain fatty acids in the four treatment groups, measured from frontal cortex extracts with GC. The fatty acid profiles were expressed as a percentage of the total microgram of fatty acid (weight percentage). (A) Eicosapentaenoic acid (C20:5, n-3); (B) Docosapentaenoic acid (C22:5, n-3); (C) Docosahexaenoic acid (C22:6, n-3). Data are presented as mean ± SEM. ***P < 0.001 (E-P); ### P < 0.001 (ME-MP). P (palm oil + saline-probenecid); E (E-EPA + saline-probenecid); MP (palm oil + MPTP-probenecid); ME (E-EPA + MPTP-probenecid).

Fig. 6. Cytokine levels: Striatal and midbrain cytokines in the four treatment groups, measured with the Bioplex protein suspension array system from striatal and midbrain extracts. Data was expressed as pg per milligram of protein content. (A) Striatal pro-inflammatory cytokine TNF-␣. (B) Striatal pro-inflammatory and Th-1 cytokine IFN-␥. (C) Midbrain anti-inflammatory cytokine IL-10. Data are presented as mean ± SEM. *P < 0.05 (P-MP); # P < 0.05 (ME-MP). P (palm oil + saline-probenecid); MP (palm oil + MPTPprobenecid); ME (E-EPA + MPTP-probenecid).

MPP+ administration to hippocampal brain slices does not affect LTP, despite a significant decrease in DA content. 4.2. Brain PUFA, and inflammation Neuroinflammation is suggested to be an important aspect of PD pathogenesis [6]. Consistent with cytokine production and

microglia activation in PD and the MPTP-P model [27,45], MPTPP presently increased TNF-␣ and IFN-␥ in the striatum. Knock-out of both these major pro-inflammatory cytokines or receptors in mice causes resistance to MPTP neurotoxicity [46,47]. The present increase in anti-inflammatory cytokine IL-10 in the midbrain, which was documented before (in the striatum) in another MPTP regimen [48], may be a compensatory mechanism against the nigral

Fig. 7. cPlA2 and COX-2 mRNA expression. Striatal COX-2 and cPLA2 mRNA expression, measured with real-time (q)-PCR from striatal mRNA extracts. Presented are the fold changes of target gene expression relative to reference gene (beta-actin) expression, calculated with the Ct correction. (A) Cyclo-oxygenase-2 (COX-2); (B) Cytosolic phospholipase A2 (cPLA2). Data are presented as mean ± SEM. *P < 0.05 (P-MP); # P < 0.05 (P-E). P (palm oil + saline-probenecid); E (E-EPA + saline-probenecid); MP (palm oil + MPTP-probenecid).

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396

degeneration. Another source of inflammation and oxidative stress in PD may be an increase in the conversion of AA to potentially pro-inflammatory eicosanoids by cPLA2 and COX-2. Inhibition of cPLA2 and COX-2 resulted in protection against MPTP in mice [5,6]. Furthermore, short-term administration of MPP+, the active metabolite of MPTP, to striatal brain slices increases AA content and cPLA2 mRNA expression [10]. However, in the present MPTP-P regimen, which causes more severe and prolonged neurotoxicity, brain n-6 or n-3 PUFAs and cPLA2 mRNA expression were not altered, and COX-2 was down- rather than upregulated. A number of studies also reported that PUFA profiles remained unaltered in PD patients, MPTP treated monkeys [49] and mice [17]. Nonetheless, since a link between dietary intake of n-6 PUFA and PD has been suggested [9], brain PUFA metabolism may be altered at some stage in PD, possibly an early stage, or acutely following exposure to Parkinsonian neurotoxin, as demonstrated previously [10]. While the downregulation of COX-2 mRNA was unexpected, gene expression results should be interpreted with caution. For instance, an upregulation followed by a downregulation of genes related to the dopaminergic system was found before in another MPTP study [50]. 4.3. E-EPA a functional, neuroactive lipid Omega-3 PUFA, mainly DHA, may be beneficial in PD [17–19]. However, little is known about the neuroprotective effects of EPA in PD. We previously observed that an E-EPA diet equivalent to the present diet can attenuate some neurodegenerative changes induced by acute MPP+ administration to striatal brain slices [10]. Interestingly, similar to the present study, the E-EPA diet only increased tissue EPA and DPA levels and not DHA. The fact that E-EPA did not increase DHA content may be due to insufficient conversion of DPA to DHA. Nonetheless, E-EPA by itself can evidently affect behavior and neurochemistry. For instance, E-EPA presently caused an anxiolytic effect in the open field, which is consistent with the mood-enhancing effects of n3 PUFA [15]. Omega-3 PUFAs are also well-known to modulate cognition [15]. Presently, E-EPA by itself did not affect memory performance, but the diet did alter brain physiology, as DA and metabolite levels were significantly increased in the frontal cortex and hippocampus. This is the first study to investigate the effects of E-EPA in an in vivo PD model, as all other studies used fish oil (containing both EPA and DHA). Our initial findings were encouraging, as E-EPA prevented hypokinesia, hyperactivity and a procedural memory deficit in the water maze, and loss of body weight induced by MPTP-P, one of the most severe and valid regimens of the MPTP model [27]. While these findings are remarkable, there was no protection against the hallmark PD neuropathology, a nigrostriatal DA depletion. This finding is consistent with a study demonstrating that fish oil can attenuate motor abnormalities induced by 6-hydroxydopamine (6-OHDA), without attenuating the nigrostriatal dopaminergic neuropathology [19]. Yet in another study, a fish oil compound concentrated in DHA partially prevented an MPTPinduced striatal DA loss [17], but the DHA diet lasted 10 months and motor behavior was not measured. Furthermore, another MPTP regimen was used. No conclusive causal data could be obtained to explain how E-EPA prevented the MPTP-P induced motor deficits. E-EPA did attenuate the increase in striatal DA turnover and 5-HT metabolite levels, which were respectively correlated to pole test and rotorod performance. However, in the frontal cortex and hippocampus, the effects of EPA in MPTP-P treated mice followed a highly complicated and unpredictable pattern, from which no clear conclusions about neuroprotection could be made. Perhaps a more reasonable mechanism of neuroprotection by E-EPA is suppression of inflammation. Consistent with the anti-inflammatory effects of n-3 PUFA, the E-EPA diet presently

395

prevented increases of striatal TNF-␣ and partially attenuated the increase in IFN-␥, a potent microglia activator. These effects of E-EPA may have neuroprotective relevance, since TNF-␣ and IFN␥ knock-out mice are resistant to MPTP neurotoxicity [46,47]. Anti-inflammatory treatments have protective effects in PD and experimental PD [51], but unlike E-EPA, they have side effects [52]. The anti-inflammatory effects of E-EPA may also explain that it prevented MPTP-P induced weight loss, as these cytokines can have anorexigenic effects [53]. Altogether, while the present evidence in support of neuroprotective effects of E-EPA in experimental PD is limited by the lack of nigrostriatal protection, it does demonstrate that E-EPA, despite its low content in the brain, is a neuroactive lipid that may have some beneficial effect in PD. Further research with different concentrations of E-EPA, different diet times and different dosing regimens of the MPTP model may provide further insights. Role of the funding source This study was supported by a grant to Cai Song from Laxdale/Amarin Neruoscience Ltd., UK, Canadian Institutes for Health Research (CIHR) and Atlantic Innovation Fund, Canada. Acknowledgement We thank Mr. Di Shao for his expert technical support during the study. References [1] Dauer W, Przedborski S. Parkinson’s disease: mechanisms and models. Neuron 2003;39:889–909. [2] Farlow MR, Cummings J. A modern hypothesis: the distinct pathologies of dementia associated with Parkinson’s disease versus Alzheimer’s disease. Dement Geriatr Cogn Disord 2008;25:301–8. [3] Teismann P, Schulz JB. Cellular pathology of Parkinson’s disease: astrocytes, microglia and inflammation. Cell Tissue Res 2004;318:149–61. [4] Lawson LJ, Perry H, Dri P, Gordon S. Heterogeneity in the distribution and morphology of microglia. Neuroscience 1990;39:151–70. [5] Klivenyi P, Beal MF, Ferrante RJ, Andreassen OA, Wermer M, Chin MR, et al. Mice deficient in group IV cytosolic phospholipase A2 are resistant to MPTP neurotoxicity. J Neurochem 1998;71:2634–7. [6] Teismann P, Vila M, Choi DK, Tieu K, Wu DC, Jackson-Lewis V, et al. COX-2 and neurodegeneration in Parkinson’s disease. Ann N Y Acad Sci 2003;991:272–7. [7] Okuno T, Nakatsuji Y, Kumanogoh A, Moriya M, Ichinose H, Sumi H, et al. Loss of dopaminergic neurons by the induction of inducible nitric oxide synthase and cyclooxygenase-2 via CD 40: relevance to Parkinson’s disease. J Neurosci Res 2005;81:874–82. [8] Calder PC. Polyunsaturated fatty acids and inflammation. Prostaglandins Leukot Essent Fatty Acids 2006;75:197–202. [9] Miyake Y, Sasaki S, Tanaka K, Fukushima W, Kiyohara C, Tsuboi Y, et al. Dietary fat intake and risk of Parkinson’s disease: a case-control study in Japan. J Neurol Sci 2010;288:117–22. [10] Meng Q, Luchtman DW, El Bahh B, Zidichouski JA, Song C. Ethyleicosapentaenoate modulates changes in neurochemistry and brain lipids induced by Parkinsonian neurotoxin 1-methyl-4-phenylpyridinium in mouse brain slices. Eur J Pharmacol 2010;649:127–34. [11] Dyall MT, Michael-Titus AT. Neurological benefits of omega-3 fatty acids. Neuromol Med 2008;10:219–35. [12] Youdim KA, Martin A, Joseph JA. Essential fatty acids and the brain: possible health implications. Int J Dev Neurosci 2000;18:383–99. [13] Kitajka K, Sinclair AJ, Weisinger RS, Weisinger HS, Mathai M, Jayasooriya AP, et al. Effects of dietary omega-3 polyunsaturated fatty acids on brain gene expression. Proc Natl Acad Sci USA 2004;101:10931–6. [14] Bernardo A, Minghetti L. PPAR-gamma agonists as regulators of microglial activation and brain inflammation. Curr Pharm Des 2006;12:93–109. [15] Fedorova I, Salem Jr N. Omega-3 fatty acids and rodent behavior. Prostaglandins Leukot Essent Fatty Acids 2006;75:271–89. [16] Rezak M. Current pharmacotherapeutic treatment options in Parkinson’s disease. Dis Mon 2007;53:214–22. [17] Bousquet M, Saint-Pierre M, Julien C, Salem Jr N, Cicchetti F, Calon F. Beneficial effects of dietary omega-3 polyunsaturated fatty acid on toxin-induced neuronal degeneration in an animal model of Parkinson’s disease. FASEB J 2008;22:1213–25. [18] Bousquet M, Gibrat C, Saint-Pierre M, Julien C, Calon F, Cicchetti F. Modulation of brain-derived neurotrophic factor as a potential neuroprotective

396

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30] [31]

[32]

[33]

[34]

[35]

D.W. Luchtman et al. / Behavioural Brain Research 226 (2012) 386–396 mechanism of action of omega-3 fatty acids in a Parkinsonian animal model. Prog Neuropsychopharmacol Biol Psychiatry 2009;33:1401–8. Delattre AM, Kiss A, Szawka RE, Anselmo-Franci JA, Bagatini PB, Xavier LL, et al. Evaluation of chronic omega-3 fatty acids supplementation on behavioral and neurochemical alterations in 6-hydroxydopamine-lesion model of Parkinson’s disease. Neurosci Res 2010;66:256–64. Kawashima A, Harada T, Kami H, Yano T, Imada K, Mizuguchi K. Effects of eicosapentaenoic acid on synaptic plasticity, fatty acid profile and phosphoinositide 3-kinase signaling in rat hippocampus and differentiated PC12 cells. J Nutr Biochem 2010;21:268–77. Peet M, Brind J, Ramchand CN, Shah S, Vankar GK. Two double-blind placebo-controlled pilot studies of eicosapentaenoic acid in the treatment of schizophrenia. Schizophr Res 2010;49:243–51. Chen CT, Liu Z, Ouellet M, Calon F, Bazinet RP. Rapid beta-oxidation of eicosapentaenoic acid in mouse brain: an in situ study. Prostaglandins Leukot Essent Fatty Acids 2009;80:157–63. Song C, Horrobin D. Omega-3 fatty acid ethyl-eicosapentaenoate, but not soybean oil, attenuates memory impairment induced by central IL-1beta administration. J Lipid Res 2004;45:1112–21. Song C, Li X, Leonard BE, Horrobin DF. Effects of dietary n-3 or n-6 fatty acids on interleukin-1beta-induced anxiety, stress, and inflammatory responses in rats. J Lipid Res 2003;44:1984–91. Song C, Manku MS, Horrobin DF. Long-chain polyunsaturated fatty acids modulate interleukin-1beta-induced changes in behavior, monoaminergic neurotransmitters, and brain inflammation in rats. J Nutr 2008;138: 954–63. Song C, Phillips AG, Leonard BE, Horrobin DF. Ethyl-eicosapentaenoic acid ingestion prevents corticosterone-mediated memory impairment induced by central administration of interleukin-1beta in rats. Mol Psychiatry 2004;9:630–8. Luchtman DW, Shao D, Song C. Behavior, neurotransmitters and inflammation in three regimens of the MPTP mouse model of Parkinson’s disease. Physiol Behav 2009;98:130–8. Petroske E, Meredith GE, Callen S, Totterdell S, Lau YS. Mouse model of Parkinsonism: a comparison between subacute MPTP and chronic MPTP/probenecid treatment. Neuroscience 2001;106:589–601. Rozas G, Labandeira Garcia JL. Drug-free evaluation of rat models of Parkinsonism and nigral grafts using a new automated rotarod test. Brain Res 1997;749:188–99. Kong WX, Chen SW, Li YL, Zhang YJ, Wang R, Min L, et al. Effects of taurine on rat behaviors in three anxiety models. Pharmacol Biochem Behav 2006;83:271–6. Ogawa N, Hirose Y, Ohara S, Ono T, Watanabe Y. A simple quantitative bradykinesia test in MPTP-treated mice. Res Commun Chem Pathol Pharmacol 1985;50:435–41. Zhu XR, Maskri L, Herold C, Bader V, Stichel CC, Güntürkün O, et al. Nonmotor behavioural impairments in parkin-deficient mice. Eur J Neurosci 2007;26:1902–11. Da Cunha C, Wietzikoski S, Wietzikoski EC, Miyoshi E, Ferro MM, AnselmoFranci JA, et al. Evidence for the substantia nigra pars compacta as an essential component of a memory system independent of the hippocampal memory system. Neurobiol Learn Mem 2003;79:236–42. Carlson ES, Fretham SJ, Unger E, O’Connor M, Petryk A, Schallert T, et al. Hippocampus specific iron deficiency alters competition and cooperation between developing memory systems. J Neurodev Disord 2010 Sep 1;2(3): 133–43. Datta SC, Opp MR. Lipopolysaccharide-induced increases in cytokines in discrete mouse brain regions are detectable using Luminex xMAP technology. J Neurosci Methods 2008;175:119–24.

[36] Chia LG, Ni DR, Cheng LJ, Kuo JS, Cheng FC, Dryhurst G. Effects of 1methyl-4-phenyl-1,2,3,6-tetrahydropyridine and 5,7-dihydroxytryptamine on the locomotor activity and striatal amines in C57BL/6 mice. Neurosci Lett 1996;218:67–71. [37] Rousselet E, Joubert C, Callebert J, Parain K, Tremblay L, Orieux G, et al. Behavioral changes are not directly related to striatal monoamine levels, number of nigral neurons, or dose of Parkinsonian toxin MPTP in mice. Neurobiol Dis 2003;14:218–28. ˜ J. Adrenergic hyperactivity and metanephrine excess in the [38] Espejo EF, Minano nucleus accumbens after prefrontocortical dopamine depletion. J Neurophysiol 2001;85:1270–4. [39] Deguil J, Chavant F, Lafay-Chebassier C, Pérault-Pochat MC, Fauconneau B, Pain S. Neuroprotective effect of PACAP on translational control alteration and cognitive decline in MPTP Parkinsonian mice. Neurotox Res 2010;17: 142–55. [40] Pothakos K, Kurz MJ, Lau YS. Restorative effect of endurance exercise on behavioral deficits in the chronic mouse model of Parkinson’s disease with severe neurodegeneration. BMC Neurosci 2009;20:10–6. [41] Rabey JM, Burns RS. Neurochemistry. In: Weiner WJ, Factor SA, editors. Parkinson’s disease: diagnosis and clinical management. New York, NY: Demos Medical Publishing; 2002. p. 195–211. [42] Rozas G, Liste I, Guerra MJ, Labandeira-Garcia JL. Sprouting of the serotonergic afferents into striatum after selective lesion of the dopaminergic system by MPTP in adult mice. Neurosci Lett 1998;245:151–4. [43] Vuˇckovic´ MG, Wood RI, Holschneider DP, Abernathy A, Togasaki DM, Smith A, et al. Memory, mood, dopamine, and serotonin in the 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine-lesioned mouse model of basal ganglia injury. Neurobiol Dis 2008;32:319–27. ˜ V, Solís JM, et al. D1 but not D5 [44] Granado N, Ortiz O, Suárez LM, Martín ED, Cena dopamine receptors are critical for LTP, spatial learning, and LTP-Induced arc and zif268 expression in the hippocampus. Cereb Cortex 2008;18:1–12. [45] Schintu N, Frau L, Ibba M, Caboni P, Garau A, Carboni E, et al. PPARgammamediated neuroprotection in a chronic mouse model of Parkinson’s disease. Eur J Neurosci 2009;29:954–63. [46] Ferger B, Leng A, Mura A, Hengerer B, Feldon J. Genetic ablation of tumor necrosis factor-alpha (TNF-alpha) and pharmacological inhibition of TNFsynthesis attenuates MPTP toxicity in mouse striatum. J Neurochem 2004;89: 822–33. [47] Mount MP, Lira A, Grimes D, Smith PD, Faucher S, Slack R, et al. Involvement of interferon-gamma in microglial-mediated loss of dopaminergic neurons. J Neurosci 2007;27:3328–37. [48] Ciesielska A, Joniec I, Przybyłkowski A, Gromadzka G, Kurkowska-Jastrzebska I, Członkowska A. Dynamics of expression of the mRNA for cytokines and inducible nitric synthase in a murine model of the Parkinson’s disease. Acta Neurobiol Exp 2003;63:117–26. [49] Julien C, Berthiaume L, Hadj-Tahar A, Rajput AH, Bédard PJ, Di Paolo T, et al. Postmortem brain fatty acid profile of levodopa-treated Parkinson disease patients and Parkinsonian monkeys. Neurochem Int 2006;48:404–14. [50] Kühn K, Wellen J, Link N, Maskri L, Lübbert H, Stichel CC. The mouse MPTP model: gene expression changes in dopaminergic neurons. Eur J Neurosci 2003;317:1–12. [51] Asanuma M, Miyazaki I. Nonsteroidal anti-inflammatory drugs in experimental Parkinsonian models and Parkinson’s disease. Curr Pharm Des 2008;14:1428–34. [52] Peura DA, Goldkind L. Balancing the gastrointestinal benefits and risks of nonselective NSAIDs. Arthritis Res Ther 2005;7:S7–13. [53] Langhans W. Signals generating anorexia during acute illness. Proc Nutr Soc 2007;66:321–30.