1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

NP5316_proof ■ 27 November 2013 ■ 1/11

Neuropharmacology xxx (2013) 1e11

Contents lists available at ScienceDirect

Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm

Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depressionq Q6

Yuliya E. Borre a, b, *, Theodora Panagaki a, Pim J. Koelink a, Mary. E. Morgan a, Hendrikus Hendriksen a, b, Johan Garssen a, c, Aletta D. Kraneveld a, Berend Olivier a, b, Ronald S. Oosting a, b a Division of Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Faculty of Science, Utrecht University, PO Box 80082, 3508 TB Utrecht, The Netherlands b Rudolf Magnus Institute of Neuroscience, Utrecht University, PO Box 80082, 3508 TB Utrecht, The Netherlands c Danone Research, Center for Specialized Nutrition, Wageningen, The Netherlands

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 April 2013 Received in revised form 13 November 2013 Accepted 16 November 2013

Rising neurodegenerative and depressive disease prevalence combined with the lack of effective pharmaceutical treatments and dangerous side effects, has created an urgent need for the development of effective therapies. Considering that these disorders are multifactorial in origin, treatments designed to interfere at different mechanistic levels may be more effective than the traditional single-targeted pharmacological concepts. To that end, an experimental diet composed of zinc, melatonin, curcumin, piperine, eicosapentaenoic acid (EPA, 20:5, n-3), docosahexaenoic acid (DHA, 22:6, n-3), uridine, and choline was formulated. This diet was tested on the olfactory bulbectomized rat (OBX), an established animal model of depression and cognitive decline. The ingredients of the diet have been individually shown to attenuate glutamate excitoxicity, exert potent anti-oxidant/anti-inflammatory properties, and improve synaptogenesis; processes that all have been implicated in neurodegenerative diseases and in the cognitive deficits following OBX in rodents. Dietary treatment started 2 weeks before OBX surgery, continuing for 6 weeks in total. The diet attenuated OBX-induced cognitive and behavioral deficits, except long-term spatial memory. Ameliorating effects of the diet extended to the control animals. Furthermore, the experimental diet reduced hippocampal atrophy and decreased the peripheral immune activation in the OBX rats. The ameliorating effects of the diet on the OBX-induced changes were comparable to those of the NMDA receptor antagonist, memantine, a drug used for the management of Alzheimer’s disease. This proof-of-concept study suggests that a diet, which simultaneously targets multiple disease etiologies, can prevent/impede the development of a neurodegenerative and depressive disorders and the concomitant cognitive deficits. Ó 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

Keywords: Neurodegeneration Depression Neuroprotection Olfactory bulbectomy Dietary intervention Memantine Anosmia Hippocampus Inflammation Cognition

1. Introduction Rising neurodegenerative and depressive disease prevalence combined with a lack of effective pharmaceutical treatments has created an urgent need for novel therapeutic approaches.

Q1

q This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited. * Corresponding author. Rudolf Magnus Institute of Neuroscience, Utrecht University, PO Box 80082, 3508 TB Utrecht, The Netherlands. E-mail address: [email protected] (Y.E. Borre).

Neurodegenerative disorders are multifactorial in origin with a complex set of pathological pathways (Dauer and Przedborski, 2003; Anand et al., 2012; Chopra et al., 2011). Therefore, simultaneous manipulation of these pathways may exert higher or better therapeutic efficacy than a single approach alone (Wollen, 2010). To prevent a disease onset, most people will probably prefer the daily intake of natural food supplements above the daily intake of drugs (Rozin et al., 2004). Hence, dietary components have emerged as potential preventatives and/or treatments for neurodegenerative and depressive disorders. Alzheimer’s disease, in particular, is being targeted with dietary treatments due to the limited adverse side effects as compared to pharmacological interventions (Wollen, 2010; Kamphuis and Wurtman, 2009; Scheltens et al., 2013).

0028-3908/$ e see front matter Ó 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

Please cite this article in press as: Borre, Y.E., et al., Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depression, Neuropharmacology (2013), http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Q2 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

NP5316_proof ■ 27 November 2013 ■ 2/11

2

Y.E. Borre et al. / Neuropharmacology xxx (2013) 1e11

Clinical studies show higher efficacy of the dietary interventions especially at the early phases of the disease, whereas at later stages treatments are rather disappointing (Kamphuis and Scheltens, 2010) The aim of this study was to investigate the neuroprotective effects of a dietary intervention in olfactory bulbectomized (OBX) rats, an animal model of neurodegeneration and depression. Removal of the olfactory bulbs leads to neuronal degeneration in several different brain areas and cognitive decline (Wrynn et al., 2000; Song and Leonard, 2005; Borre et al., 2012a). Although the initial mechanism of neurodegeneration induction in the OBX rats differs from that in neurodegenerative diseases in humans, the secondary changes, such as NMDA-receptor-mediated excitotoxicity, impaired structural plasticity, and neuroinflammation, have much in common with human neurodegenerative diseases. In this proof-of-concept study it was hypothesized that a combination diet of unique nutritional ingredients, individually known to be (partly) beneficial in the OBX model, would effectively impede cognitive decline and neurodegeneration when chronically administered starting two weeks before the surgery. The diet was composed of zinc, melatonin, curcumin, piperine, eicosapentaenoic acid (EPA 20:5, n-3), docosahexaenoic acid (DHA, 22:6, n-3), uridine, and choline (Table 1). These ingredients have been individually shown to attenuate glutamate excitoxicity, exert potent antioxidant/anti-inflammatory properties, or improve synaptogenesis; processes that all have been implicated in neurodegenerative diseases and in the cognitive deficits following OBX in rodents. Some of these ingredients have already been tested in OBX rats (Nowak et al., 2003; Tasset et al., 2010; Song et al., 2009), but in the studies so far, dietary interventions started several weeks after surgery-a time point at which the OBX-induced secondary neurodegenerative processes have been completed (Song and Leonard, 2005). Clinical data combined with our previous studies (Borre et al., 2012a,b,c) suggest a higher efficacy in early intervention rather than late treatment. Therefore, current study employed an alternative approach of commencing treatment regiment prior to the OBX and exploring prophylactic/preventive therapies. Zinc is a potent antagonist of the NMDA receptor (Dingledine et al., 1999; Smart et al., 1994) and has been shown to modulate glutamate neurotransmission (Morris and Levenson, 2012; Watt et al., 2010) and to be neuroprotective (Bancila et al., 2004). Under normal conditions, zinc plays a critical role in learning and memory (Mocchegiani et al., 2005; Bitanihirwe and Cunningham, 2009). Curcumin has antioxidant and anti-inflammatory properties (Cole et al., 2007; Braidy et al., 2010; Ishrat et al., 2009). Since curcumin has a poor absorption rate, which undermines its bioavailability, piperine has been added. Piperine is a major alkaloidal constituent of black pepper. It is a powerful inhibitor of hepatic and intestinal glucuronidation, and increases the bioavailability of many drugs including curcumin (Atal et al., 1985; Shoba et al., 1997). In addition, piperine possesses potential

Table 1 Experimental and control diet compositions. Rats were given 20 g of the diet per day. The delivered dose is on a mg/kg of food. Active ingredient

Control (AIN-93) g/kg diet

Experimental diet

Zinc Curcumin Piperine Melatonin Choline Uridine Soy oil

0.03 0 0 0 1.09 0 7%

1.63 0.25 0.06 0.03 9.5 15.48 3% soyaþ4% tuna oil (25% DHA/6% EPA)

antidepressant effects by inhibiting monoamine oxidase A and B, improves cognitive performance, and has anti-inflammatory and anti-oxidative properties (Chonpathompikunlert et al., 2010; Wattanathorn et al., 2008; Selvendiran et al., 2003). Melatonin is a hormone involved in sleep regulation and a free radical scavenger. It is able to pass the bloodebrain barrier, making this compound a potential neuroprotective agent (Reiter et al., 2000; Olcese et al., 2009; Ramirez-Rodriguez et al., 2011). UMP, choline and the n-3 fatty acids EPA and DHA were added as well. These compounds stimulate neurite outgrowth, dendritic spine formation and are necessary for brain phosphatide synthesis (Wurtman, 2011). Recently, it was shown that feeding mice for 3 month with DHA alone is already sufficient to increase n-3 fatty acids in brain (Broersen et al., 2013). Importantly, a diet containing, in addition to DHA, UMP, choline, some vitamins and selenium reduced the amyloid plaque burden in Alzheimer’s disease transgenic, indicating Q3 that increased amounts of n-3 fatty acids in the brain lipids by itself are not sufficient? It is the combination of relevant nutrients that has the relevant biological effect. On the other hand, animal studies have shown that n-3 poly-unsaturated fatty acid supplementation alone is already sufficient to suppress inflammation (James et al., 2000; Calder, 2001), attenuate the stress response and improve cognition (Song et al., 2003, 2009; Ikemoto et al., 2001). Importantly, it has been shown in recent clinical trials that SouvenaidÒ, a medicinal food mixture that contains n-3 fatty acids, choline, and UMP as main ingredients, improves memory performance in drugnaïve patients with mild Alzheimer’s disease (Kamphuis et al., 2011; Scheltens et al., 2013). Concentrations of various supplements included in the experimental diet were derived from the literature where these concentrations were effective in attenuating cognitive and behavioral deficits in either OBX rodents, other animal models of neurodegeneration or clinical setting (SouvenaidÒ). For example, zinc showed antidepressant and cognition enhancing effects in the OBX (Nowak et al., 2003) Melatonin normalized OBXinduced aberrant changes in the open field (Tasset et al., 2010). Curcumin administration reduced OBX-induced hyperactivity and fear memory deficits (Xu et al., 2005). EPA treatment attenuated OBX-induced behavioral and cognitive deficits (Song et al., 2009). Piperine significantly improved spatial memory and neurodegeneration in AD animal model (Chonpathompikunlert et al., 2010). SouvenaidÒ improves memory performance in patients with mild Alzheimer’s disease (Kamphuis et al., 2011; Scheltens et al., 2013). Q4 The above-described diet was given to the rats starting 2 weeks before surgery and continued for 6 weeks. Animals were assessed in behavioral and cognitive paradigms at several time points during the treatment regimen. To compare the efficacy of the experimental diet to the existing pharmacological treatment for Alzheimer’s disease, a separate group of animals was treated with memantine, an NMDA receptor antagonist. To ensure OBX-induced behavioral and cognitive deficits were independent of anosmia, a separate group of animals were made anosmic by destroying the nose epithelium with a ZnSO4 infusion. After the completion of the behavioral screening, the modulating effects of experimental diet on molecular, immune and cellular parameters in the brain and spleen were evaluated. Rats were tested at several time points during the treatment regimen in the open field (assessing locomotor activity of the animals), passive avoidance (a task assessing fear memory), T-maze (a task assessing short-term spatial memory) and hole board (a task assessing spatial and reference memory) paradigms. Because hippocampus is involved in various cognitive functions and is affected in neurodegenerative disorders and OBX animals, we analyzed hippocampi for atrophy and cell loss using nissl staining in order to assess OBX-induced cell in hippocampus. To assess if experimental diet attenuated OBX-induced

Please cite this article in press as: Borre, Y.E., et al., Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depression, Neuropharmacology (2013), http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

NP5316_proof ■ 27 November 2013 ■ 3/11

Y.E. Borre et al. / Neuropharmacology xxx (2013) 1e11

abnormal inflammatory response, a key feature in various neurodegenerative disorders, we performed histological examination of the spleen. 2. Methods and materials 2.1. Animals and surgery Sprague-Dawley male rats (Harlan, Zeist, The Netherlands) weighing between 240 g and 270 g upon arrival were used. Animals were on a 12-h light/dark cycle, with lights going off at 7.00 pm. Olfactory bulbectomy surgery was performed as previously described (Breuer et al., 2007). Briefly, animals were anesthetized with isofluorane gas anesthetic (3%e4%), mixed with oxygen. Two burr holes (8 mm anterior to bregma and 2 mm from the midline) were drilled and the olfactory bulbs were aspirated through a blunt hypodermic needle. Animals receiving sham surgery went through a similar procedure as the bulbectomized rats but without the removal of their olfactory bulbs. Animals were housed in groups of four in macrolon IV cages separated by group (OBX separated from the Sham animals). All experiments were performed in accordance with the Dutch guidelines for care and use of laboratory animals and were approved by the Ethical Committee for Animal Research of Utrecht University. 2.2. Induction of anosmia with zinc sulfate Infusion of a ZnSO4 solution (Sigma-Aldrich) into the nasal cavity causes necrosis of the olfactory epithelium (Alberts, 1974) and consequently e loss of smell (Mar et al., 2000). Animals were anesthetized with isofluorane gas anesthetic (3%e 4%), mixed with oxygen. The animals were then held on their back with their hind limbs elevated to prevent ZnSO4 entry into the trachea or esophagus. A flexible Tygon tube connected to a 1 ml syringe was inserted 14 mm into the nostrils. Per nostril, 250 ml of the ZnSO4 solution (0.3 M) or saline was instilled slowly. Animals were given 5 days to recover before behavioral testing. Animals treated with ZnSO4 remained anosmic throughout the duration of the experiment, suggesting that one instillation of zinc sulfate is sufficient to damage the nasal epithelium, resulting in loss of olfactory function. These results are in accordance with previous reports (Alberts, 1974; Mar et al., 2000) Because animals treated with ZnSO4 remained anosmic throughout the duration of the experiment, histological verification was not necessary. Moreover, van Denderen et al., 2001 provided histological evidence that ZnSO4 infusion leads to damage of the olfactory fiber density and consequently anosmia.

3

and 4 weeks thereafter). Experiment II was conducted to compare the efficacy of our experimental dietary intervention with a known pharmacological treatment -memantine. Forty animals were divided into four groups (n ¼ 10 per group): (1) sham operated þ vehicle (Sham-Veh), (2) sham-operated þ memantine (ShamMem), (3) OBX þ vehicle (OBX-Veh), and (4) OBX þ memantine (OBX-Mem). In Experiment III, the effects of ZnSO4-induced anosmia on the cognitive, behavioral, immunological and cell count changes were investigated. Twenty-four animals were divided into two groups (n ¼ 12 per group): (1) saline, and (2) ZnSO4. Animals were assessed in a battery of behavioral and cognitive tests, starting 3 days post bulbectomy and 5 days after ZnSO4 infusion (Fig. 1). After behavioral tests, the animals were sacrificed, and the brains, serum and spleens were obtained and processed for further analysis. At the end of the experiment, the animals were sacrificed by decapitation, and olfactory bulb ablation was verified for the bulbectomized rats. Sham surgeries were verified by ensuring that the olfactory bulbs remained intact. All OBX surgeries were performed correctly, that is, there were no animals with partial bulbectomies or damaged prefrontal cortices. 2.4. Dietary treatment Prior to treatment intervention and surgery, animals were randomly divided into four groups (described in detail above) of twelve rats each (Sham-C; Sham-Exp; OBX-C; OBX-Exp). The control diet was a standard animal laboratory purified diet (AIN-93, Research Diets Inc, http://www.researchdiets.com) (Reeves, 1993,1997). The experimental diet consisted of the control diet supplemented as described in Table 1. The relative calorific content of the two diets is approximately 4000 kcal/kg. The experimental diet was prepared by Research Diet Services (Wijk Bij Duurstede, The Netherlands). The dietary treatment started 14 days prior to the surgery and continued for 28 days post surgery. The tolerability of the diets was assessed by measuring the amount of food eaten per day and general well being of the animals (i.e.: weight, fur quality). To ensure the weight loss observed in the OBX animals was not due to the decreased food consumption, we provided 20 g per rat each morning and weighed the remaining food pellets 24-h later. Animals were fed 20 g each morning and weighed the remaining food pellets 24-h later. On average the rats consumed 17 g/day. Water was provided ad libitum. 2.5. Drug treatment Memantine (20 mg/2 ml/kg) (Lundbeck, Denmark) or water was orally administered (via gavage) daily starting 2 days prior to the OBX surgery and continued for 28 days. The administration period and dose were chosen on the basis of our published data (Borre et al., 2012a).

2.3. Experimental design The current study consisted of three independent experiments (Fig. 1). In Experiment I, the effects of the experimental diet on the OBX syndrome were examined. Therefore, 48 animals were divided into four groups (n ¼ 12 per group): (1) sham operated þ control diet (Sham-C), (2) sham operated þ experimental diet (Sham-Exp), (3) OBX þ control diet (OBX-C), and (4) OBX þ experimental diet (OBXExp). Both diets were fed to the animals for 6 weeks (2 weeks prior to the surgery

Fig. 1. Experimental design. A. Experiment I: Dietary treatment scheme. OBX surgery was performed on day 0. Each animal was tested in all paradigms. B. Experiment II: Memantine treatment scheme. OBX surgery was performed day 0. Each animal was tested in all paradigms. C. Experiment III: Anosmia induced by intranasal ZnSO4 infusion in rats. ZnSO4 infusion was performed on day 0. Each animal was tested in all paradigms. Abbreviations: FR: Food restriction; HBA: Holeboard acquisition; HLB: Holeboard test. OF: open field; PAT: Passive avoidance training; PA: Passive avoidance test; OT: olfactory test.

2.6. Cognitive and behavioral assessment Olfactory function. Olfactory function was tested in bulbectomized, sham, anosmic and intact animals prior to, 5 days after, and after completion of the cognitive tests following the surgery and ZnSO4 infusion. The olfactory function assessment was adapted from Mar et al. (2000). Briefly, a small cloth soaked in vanilla or almond extract was presented on the top, back corner of the home cage and the latency to sniff in the corner containing the cloth was recorded with a maximum duration of 60 s. Total activity in the open field was assessed in a 92  92  40 cm arena as described before (Borre et al., 2012a). The light intensity was 20 lux at floor level. Locomotion of the animals was automatically tracked (TSE system, Germany). Each animal was placed in the center of the open field and allowed to explore for 5 min, after which they were returned to their home cage. Step-through Passive avoidance (PA). The PA procedure used in this study is described in detail elsewhere (Borre et al., 2012a). In short, during acquisition, animals were placed in the light compartment and the latency to enter the dark compartment with all four feet was measured in seconds. Once the animal entered the dark compartment, the door was closed and a mild footshock (0.6 mA/3 s) was delivered. All the animals were trained prior to the OBX surgery or ZnSO4 treatment and tested after these treatments. During the retention trials (no shock presented), the rat was placed in the light compartment and the latency to cross to the dark compartment was measured. T-maze spontaneous alternation (T-maze). Spatial working memory was assessed using the T-maze paradigm. This procedure was adapted from Deacon and Rawlins (2006) and is described in detail elsewhere (Borre et al., 2012c). Briefly, a trial consisted of two runs, with an inter-trial interval of 2 min. After the rat had been placed in the start arm, the animal was free to choose between both goal arms. The percentage of spontaneous alternation was calculated from the alternation ratio multiplied by 100%. Total of 6 trials were conducted over 2 days (3 trials per day, 1 h between trials). Holeboard (HLB). The holeboard procedure used in this study is described in detail elsewhere (Borre et al., 2012a). In short, prior to being trained in this task, rats were placed on a food restriction schedule for 7 days in which their body weights were lowered and maintained at 80e85% of their free-feeding body weights. In the training trials, 4 holes (holes number 2, 7, 9, and 16 with the top left hole being #1 and bottom right being #16) were baited with a 45 mg sugar pellet (Research Diets, Hampton, UK). Training days consisted of 4 consecutive trials per day. Trials

Please cite this article in press as: Borre, Y.E., et al., Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depression, Neuropharmacology (2013), http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

NP5316_proof ■ 27 November 2013 ■ 4/11

4

Y.E. Borre et al. / Neuropharmacology xxx (2013) 1e11

continued until all 4 baits were consumed or 3 min had expired. The apparatus was cleaned with 70% ethanol between subjects. Rats were assessed in the holeboard task for 7 consecutive days. Reference memory was calculated from the total number of visits to baited holes divided by the total number of visits to all holes multiplied by 100%. Four successive trials a day were performed in the training test periods, and the mean values from the four trials were used for the statistical calculation.

comparison tests. Total activity in the open field, histological and immunohistochemical data were analyzed using a two-way ANOVA followed by one-way ANOVA and subsequent post hoc analysis with Bonferonni correction. Reference memory, as measured in the holeboard test, was analyzed by repeated measures ANOVAs followed by two-way ANOVA followed by post hoc analysis at each separate time point. Grubb’s test was performed to identify outliers. Statistical calculations were carried out using SPSS 18 or GraphPad Prism version 5 for Windows (GraphPad Software, San Diego, CA).

2.7. Tissue preparation After the last test day, animals were sacrificed and their brains were removed. The hippocampus was dissected from the left hemisphere and weighed. Right hemispheres were fixed in 10% formalin solution for two days at room temperature and then transferred into a 30% sucrose solution at 4  C. Decreased body weight of the OBX rats has been reported in the literature (Song and Leonard, 2005) and was also observed in the current experiments (data not shown). The hippocampus weight was normalized to the body weight. Because only left hippocampi was weighted and because weight between left and right hippocampi of the rats do not differ (unpublished data), the weights of the left hippocampus were multiplied by 2 and divided by the animal’s weight in kg. Spleens were taken out, fixed in 10% formalin for 24 h, then processed and embedded in paraffin. 2.8. Histology 2.8.1. Cell count in the hippocampus 25-ìm-thick coronal slices of the dorsal and ventral hippocampus were collected and Nissl-stained. Sections were then examined using an Olympus BX50 optical microscope equipped with a Leica DFC digital camera. The total number of cells visible using a 100 total magnification was counted in the CA3, CA1, and the dentate gyrus (DG) areas of the hippocampus. CA1, CA3, and DG areas were calculated automatically (Image JÒ software) and the number of the nissl-positive cells were normalized by areas of CA1, CA3, and DG. Every fourth hippocampal section was counted ( 50 slices per rat; 8e10 rats per group). Cell counts were obtained “blindly”, averaged and are reported as absolute number of cells per area counted. 2.8.2. T-cell count in the spleen The 5-mm-thick spleen sections were deparrafinized, rehydrated and endogenous peroxidase was blocked with 0.03% H2O2 in methanol for 30 min. Next, the antigen was retrieved by boiling the slides for 15 min in a 10mMTris/1 mM EDTA pH 9.0 buffer. After cooling down to room temperature (RT) and rinsing with PBS for three times, the slides were blocked with 5% goat serum (Dakocytomation, Glustrup, Denmark) in 1% BSA in PBS for 20 min at RT. Afterward the sections were incubated overnight at 4  C with anti-CD3 antibodies (1:1000, Dakocytomation) in 1% BSA/PBS for 45 min. After 5 thorough washings with PBS, the slides were incubated with biotinylated goat-anti-rabbit antibodies (1:200, Dakocytomation) and streptavidine avidinebiotin complex/HRP (Vectastain Elite ABC, Vector Laboratories, Burlingame, CA USA)) for 45 min at RT. The signal was developed with 0.015% H2O2 (Merck, Darmstadt, Germany) in 0.05% diaminobenzidine-tetrahydrochloride (SigmaAldrich, St. Louis, MO USA) in 0.01 M Tris-HCL pH 7.6 for 10 min resulting in a brown staining product. Sections were counterstained with Mayers’ haematoxylin (Merck), then dehydrated and mounted. Slides without primary antibody incubations were included as negative controls. The stained tissue sections were examined under the microscope and T-cell accumulation was detected by applying a 10 magnification. Areas outside the white pulps were defined as the regions of interest in which counting process was performed automatically (Image JÒ software). 2.8.3. Determination of zinc concentration in serum Trunk blood was collected in tubes and centrifuged. Serum was separated, frozen and stored at 20  C until further analysis. Each sample was diluted 1:10 with water. Determination of zinc was carried out by flame atomic absorption spectrometry. The equipment used was a Pye Unicam SP-9 800 AA Spectrophotometer with deuterium background correction (air flow: 4.5 l/min, acetylene flow: 1.1 l/min, analytical wavelength: 213.9 nm) (Smith et al., 1979). 2.8.4. Determination of phospholipid fatty acids in blood cells Blood was centrifuged and cell pellets were stored at 80  C until analysis. Phospholipids were separated from total cellular lipids using Bond-ElutÒ solidphase extraction columns and the Vac-Elut SPS 24Ô system. Phospholipid extracts were converted into methyl esters by using 10% BF3 in methanol at 100  C for 60 min. After hexane extraction, derivatized phospholipids were dissolved in iso-octane, and the fatty acid composition was analyzed by gas chromatography using a capillary column (50 m  0.25 mm, CP-SIL88-fame). Peaks were identified by commercial reference standards (Faber et al., 2011). 2.8.5. Statistics All results were expressed as means  standard error (SEM) and differences among means were considered significant if p  0.05. Nonparametric statistics (Kruskale Wallis test) were applied to analyze passive avoidance latency times and T-maze spontaneous alteration followed by post-hoc analysis using Dunn’s multiple

3. Results 3.1. Validation of anosmia in OBX and ZnSO4 treated animals As expected, both OBX and ZnSO4 infusion resulted in a complete loss of smell. Anosmia was present throughout the experiment. The severity of olfactory impairment in all animals was assessed by measuring the latency to sniff a novel odor (vanilla or almond extract) presented in the home cage. OBX and ZnSO4 rats did not sniff the odor within the allotted 60 s, whereas all sham and intact animals sniffed the novel odor within the 60 s sessions (intact: 7.94 s  1.29; sham: 8.2 s  0.92). 4. Effects of the dietary and pharmacological interventions on the OBX-induced cognitive and behavioral deficits 4.1. Both the experimental diet and memantine attenuated OBXinduced short-term spatial memory deficit assessed in T- maze The OBX-induced spatial memory deficit was reversed by the experimental diet as shown in Figure 2A (H(4) ¼ 22.62, p < 0.001 Memantine intervention (Fig. 2B) also attenuated the OBXmediated spatial memory deficit: H(4) ¼ 15.80, p < 0.01. 4.2. Both the experimental diet and memantine rescued OBXinduced fear memory impairment as measured in passive avoidance All animals were trained in passive avoidance task prior to the OBX surgeries. The experimental diet (Fig. 2C) and memantine (Fig. 2D) partly rescued OBX-induced fear memory loss in the passive avoidance retention task. Dietary intervention: (H(4) ¼ 26.83, p < 0.001; Memantine intervention: (H(4) ¼ 23.53, p < 0.001). 4.3. The experimental diet failed to rescue the OBX-induced spatial memory deficit as assessed in the holeboard We have previously demonstrated that memantine fails to rescue bulbectomy-induced spatial memory deficit in the holeboard (Borre et al., 2012a). The animals were trained in the holeboard prior to the surgery for 7 consecutive days until their reference memory index was above 80%. All animals reached that level within 8 days. Retention trials were conducted for 3 successive days. Repeated measures ANOVA revealed effects attributable to the surgery (F (1, 40) ¼ 540.15, p < 0.001) and treatment (F (1, 40) ¼ 17.70, p < 0.001) without a significant interaction (F(2,80) ¼ 1.13, p > 0.05). Further analysis on each test day separately revealed that bulbectomized rats showed a robust reference memory deficit when compared to the sham animals as demonstrated by the main effect of surgery (Day 1: F (1, 40) ¼ 116.98, p < 0.001; Day 2: F (1, 40) ¼ 239.99, p < 0.001; Day 3: F (1, 40) ¼ 309.97, p < 0.001)) and no main effect of experimental diet on Day 1: F(1,40) ¼ 1.65, p > 0.05); Day 2: F(1,40) ¼ 3.63, p > 0.05). In contrast, on day 3, a main effect of diet was found. Post hoc analysis revealed that sham animals on experimental diet performed significantly better in comparison to sham animals fed the control diet (Day 3: F(1,40) ¼ 7.16, p < 0.05), whereas the reference memory of the bulbectomized rats both on control and experimental diet remained unchanged (Fig. 2E).

Please cite this article in press as: Borre, Y.E., et al., Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depression, Neuropharmacology (2013), http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

NP5316_proof ■ 27 November 2013 ■ 5/11

Y.E. Borre et al. / Neuropharmacology xxx (2013) 1e11

5

Fig. 2. Both dietary (A) and memantine (B) treatments attenuated OBX-mediated spatial memory deficit as assessed in the T-maze. Alteration ratio is shown. Bulbectomized animals treated with the diet (C) and memantine (D) demonstrated an improved fear memory (animals were trained prior to OBX surgery). The latency time to enter the dark compartment is shown. (E) The experimental diet failed to rescue OBX-induced spatial memory deficit in the holeboard. However, the experimental diet improved spatial memory performance in the sham animals. Reference memory is shown. Each bar represents mean  SEM, *p < 0.05, **p < 0.01 and ***p < 0.001. Data from A, B, C and D were analyzed using Kruskall Wallis followed by Dunn’s post hoc tests. Data from E were analyzed using the repeated measures ANOVA followed by two-way ANOVA post hoc tests. Group sizes n ¼ 10e12.

4.4. Both the experimental diet and memantine normalized OBXinduced hyperactivity Two-way ANOVA analysis revealed a significant interaction between the surgery and the experimental diet (F (1, 40) ¼ 17.71, p < 0.01), an overall effect of the treatment F (1, 40) ¼ 9.65, p < 0.01

and no significant effect of the surgery (F (1, 40) ¼ 0.53, ns) (Fig. 3A). For the memantine treatment, a significant interaction between the surgery and the treatment was noted (F (1, 38) ¼ 3.93, p < 0.01), as well as an overall effect of the treatment F (1, 38) ¼ 10.80, p < 0.01 and surgery (F (1, 38) ¼ 10.57, p < 0.01) (Fig. 3B). OBX animals fed the control diet or treated with the memantine vehicle were

Please cite this article in press as: Borre, Y.E., et al., Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depression, Neuropharmacology (2013), http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

Y.E. Borre et al. / Neuropharmacology xxx (2013) 1e11

significantly more hyperactive when compared with their controls and experimental diet and memantine attenuated the OBX-induced hyperactivity. 4.5. OBX-induced changes in the hippocampal weights are attenuated by both the experimental diet and memantine treatment The hippocampus is involved in various cognitive functioning (Fanselow and Dong, 2010) and its function is impaired in the OBX animals (Song and Leonard, 2005). To examine the integrity of the hippocampus, we removed and weighed the left hippocampus of the animals from each group (the right side of the brain was used for staining). Hippocampal weight was normalized against bodyweight (average of the final body weight for Sham-C: 490  5.7; Sham-Exp: 477  3.7; OBX-C: 442  6.5; OBX-Exp: 467  6.9). Twoway ANOVA analysis revealed a significant interaction between the surgery and the diet (F (1, 40) ¼ 12.38, p < 0.01), an overall effect of the diet F (1, 40) ¼ 17.37, p < 0.001 and a significant effect of the surgery (F (1, 40) ¼ 17.15, p < 0.001) (Fig. 4A). For the memantine treatment, a significant interaction between the surgery and the drug was found (F (1, 36) ¼ 6.43, p < 0.05), as well as an overall effect of the surgery F (1, 36) ¼ 18.22, p < 0.001, but not of the drug (F (1, 36) ¼ 2.06, ns) (Fig. 4B). Post-hoc analyses revealed that the hippocampi of the OBX animals fed the control diet and treated with the vehicle weighed significantly less compared to the sham controls. Both the experimental diet and memantine attenuated the OBX-induced hippocampal atrophy. 4.6. Dietary and memantine treatment partially rescued the OBXinduced cell loss in the hippocampus Since the hippocampus is a heterogeneous structure with functional separation of the ventral and dorsal parts (Moser and Moser, 1998), we performed the cell count accordingly following Nissl staining. OBX resulted in a significant reduction of the cell count in the ventral and dorsal hippocampus (Fig. 5) with both treatments partly rescuing bulbectomy-mediated cell loss (Statistics are presented in Table 2). Bulbectomy resulted in a significantly lower cell count in CA3, CA1 and DG areas of the dorsal hippocampus (5A-F). Experimental diet and memantine failed to rescue OBX-induced cell loss in the CA3 and CA1, whereas both interventions attenuated cell loss in the DG areas. In ventral hippocampus (5G-L), bulbectomy resulted in a significantly lower cell count in CA3, CA1 and DG areas. Experimental diet and memantine rescued OBX-induced cell loss in the CA3, CA1 and DG.

*

4.8. Effects of ZnSO4-induced anosmia To investigate the role of anosmia in the OBX syndrome, the nose epithelium was damaged by ZnSO4 infusion in a separate group of animals. ZnSO4 treated animals did not differ from salinetreated animals in any of the tested parameters (Table 3). Importantly, using the ZnSO4 treated animals, we have further validated our holeboard paradigm and have proven that the lack of an effect of memantine or the diet on the spatial memory of the OBX rats in this paradigm is not an experimental artifact. 4.9. Bioavailability validation of the experimental diet in serum The 6 week experimental diet increased the percentage of the (n-3) fatty acids EPA and DHA, whereas it decreased the (n-6) fatty acid arachidonic acid (AA) in the total phospholipid fatty acids in red blood cells of the sham and bulbectomized animals 4 weeks following the bulbectomy. Two-way ANOVA demonstrated no significant treatment  surgery interaction for the levels of EPA (F(1,28) ¼ 0.91, ns; DHA (F(1,28) ¼ 0.88, ns; or AA (F(1,28) ¼ 2.56, ns). There was no main effect of surgery for either fatty acid: EPA (F(1, 28) ¼ 1.59, ns; DHA (1,28) ¼ 0.66, ns; AA (F(1,28) ¼ 0.79, ns; with a main treatment effect: EPA (F(1, 28) ¼ 324.9, p < 0.001; DHA (1,28) ¼ 578.5, p < 0.001; AA (F(1,28) ¼ 41.71, p < 0.001. These data confirm the biochemical bioavailability of the experimental diet

B.

**

em

h

M XB O

O

B

X-

Ve

em Sh

am

-M

-V

B

X-

eh

Ex

C O

B

X-

O

Sh

am

-E

-C am Sh

**

130 120 110 100 90 80 70 60 50 40 30 20 10 0

p

Total Activity (m)

***

The presence of increased numbers T-cells in lymphoid organs signify the induction of an immune response. Since OBX induces systemic inflammation (Song et al., 2009), we investigated whether experimental diet reduces OBX-mediated immune response in the spleen. For the dietary intervention (Fig. 6A), two-way ANOVA analysis revealed a significant interaction between the surgery and treatment (F(1,24) ¼ 8.05, p < 0.01), with significant effects of the diet (F (1, 24) ¼ 5.04, p < 0.05), and surgery (F (1, 24) ¼ 11.77, p < 0.01). For the memantine intervention (Fig. 6B), two-way ANOVA analysis revealed a significant interaction between the surgery and treatment (F(1,24) ¼ 5.14, p < 0.05), a significant effect of the surgery (F (1, 24) ¼ 7.63, p < 0.05) and no drug treatment effect (F (1, 24) ¼ 3.13, ns). Further analysis demonstrated that OBX increased the numbers of T- cells, suggesting the presence of systemic immune activation and possibly active inflammation. Both the experimental diet and memantine significantly reduced splenic T-cells in the bulbectomized rats.

am

130 120 110 100 90 80 70 60 50 40 30 20 10 0

xp

A.

4.7. Dietary and memantine treatment decreased the OBX-induced immune response

Sh

6

Total activity (m)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

NP5316_proof ■ 27 November 2013 ■ 6/11

Fig. 3. Experimental diet (A) and memantine (B) decreased OBX-induced hyperactivity. Data are expressed as mean  SEM of total activity during 5 min of the open field test.*p < 0.05, **p < 0.01 and ***p < 0.001 as determined by Bonferroni multiple comparisons following a two-way ANOVA. Group sizes: n ¼ 10e12.

Please cite this article in press as: Borre, Y.E., et al., Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depression, Neuropharmacology (2013), http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

Y.E. Borre et al. / Neuropharmacology xxx (2013) 1e11

50

em O

B

X-

M

Ve X-

-V am

B

p XB O

h

0

Ex

C XB O

-E am Sh

Sh

am

-C

xp

0

O

50

100

em

100

150

-M

150

**

200

am

200

***

250

eh

250

300

Sh

***

Sh

***

300

7

B. Hippocampal weight (mg) per g BW

A. Hippocampal weight (mg) per g BW

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

NP5316_proof ■ 27 November 2013 ■ 7/11

Fig. 4. Experimental diet (A) and memantine (B) attenuated OBX-induced hippocampal atrophy. Hippocampal weights were normalized by body weight (BW). Data are expressed as mean  SEM. **p < 0.01 and ***p < 0.001 as determined by Bonferroni multiple comparisons following a two-way ANOVA. Group sizes: n ¼ 10e12.

(Fig. 7). Percentages of the rest of the fatty acids are shown in Table 4. Moreover, the experimental diet increased zinc concentration in serum in both bulbectomized and sham animals 4 weeks following the bulbectomy and after 6 weeks of experimental diet regime (Concentration (mg/L) Sham-C: 1.34  0.009; Sham-Exp: 3.99  0.27; OBX-C: 1.49  0.12; OBX-Exp: 4.02  0.40). Two-way ANOVA revealed only a main treatment effect (F(1,40) ¼ 0.11, p < 0.001). 5. Discussion In the present study, we demonstrated that our multi-targeted experimental diet attenuated many of the OBX-induced cognitive, cellular, and immune changes in a manner that was largely comparable to memantine. The beneficial effects of the diet intervention, unlike memantine treatment, also extended to the control animals. The treatment with experimental diet started 2 weeks before surgery and continued for 6 weeks. It is important to note that the current study was a proof-of-concept study. Since the experimental design incorporated both preventative and neurorestorative strategies, it is not possible to attribute the ameliorating effects of the diet to either approach. The diet may have prevented OBX-induced neurodegeneration via NMDA-receptor mediated cell death (zinc), neuroinflammation (DHA, EPA, curcumin and piperine) and oxidative stress (melatonin). On the other hand, the neurorestorative properties of the diet (EPA, DHA, choline and UMP) may also have played a role after the OBX-surgery. The idea behind our study was to show the efficacy of the experimental diet at the behavioral and at several mechanistic levels, rather than establishing the contribution of each individual food component. This approach was chosen because human neurodegenerative and depressive disorders are multifarious in origin and thus a single drug or targeting a single mechanism will probably not be sufficient to halt neurodegenerative processes; secondly, the available animal models have only limited predictive validity. For example, memantine (a single-target drug) is ineffective in halting mild Alzheimer’s disease (Schneider et al., 2011), but in animal models this drug shows clear neuroprotective properties (Thomas and Grossberg, 2009; Borre et al., 2012a). Our current study provides evidence that the multi-targeted food concept is effective in an animal model; the next step should be testing this experimental diet in human patients rather than optimizing the mixture in order to alleviate symptoms in the animal model. Although dietary intervention is a promising therapeutic strategy, developing a human diet is a challenging task due to inter-individual variation,

differences in tolerability for certain food components between individuals, and the occurrence of adverse interactions when multiple dietary components are given. The right dose can also be affected by exposure to other drugs as well as factors such as circadian rhythms. One of the main challenges remains translating animal studies to human patients. Olfactory bulbectomy (OBX) in rats has been widely used as an experimental model of depression (Song and Leonard, 2005), a mood disorder associated with cognitive impairments (Murrough et al., 2011) and long chain n-3 fatty acid deficits (McNamara and Strawn, 2013). Despite the fact that the initial induction of neurodegeneration in the OBX rat differs from that in neurodegenerative disorders in humans, the bulbectomy-mediated pathologies such as immune activation, neuronal death, and cognitive abnormalities are comparable to those seen in the human patient. Imaging studies confirm excessive trans-neuronal degeneration following bilateral bulbectomy, which takes about 4 weeks to develop (Wrynn et al., 2000; Skelin et al., 2008). We have previously demonstrated that treatment starting prior to the onset of the OBX rather than well after produced a higher therapeutic efficacy (Borre et al., 2012a). Similarly, in the human situation, early intervention seems to be more beneficial as by the time the patient is diagnosed, extensive neuronal damage has usually already occurred. This creates a great need for developing early intervention and/or preventive strategies as current pharmacotherapy focuses on the neurorestorative strategies with limited efficacy. Although we did not measure oxidative status among experimental groups, other studies have provided evidence of an increased oxidative stress in the OBX animals (Rinwa et al., 2013; Tasset et al., 2010) 4 weeks following the bulbectomy (Tasset et al., 2010; Rinwa et al., 2013). Ablation of olfactory bulbs is reported to be associated with production of oxygen reactive species and saturation of antioxidant enzymes (Tasset et al., 2010). It is likely that in our experiment OBX resulted in oxidative stress, which was attenuated by experimental diet. It has been demonstrated that treatment with melatonin (Tasset et al., 2010) decreased OBXinduced oxidative stress documented by reduced lipid peroxidation, and increases in GSH and the activity of the studied antioxidant enzymes. Moreover, beneficial effects of the combination of curcumin and piperine against OBX-induced oxidative stress have been also reported (Rinwa et al., 2013). Experimental diet ingredients such as melatonin combined with curcumine and piperine, either individually or in synergy, may have attenuated the oxidative stress in the OBX animals. However, because the experimental diet consisted of several components with potent antioxidant properties, such as melatonin, curcumin, DHA and EPA, it is

Please cite this article in press as: Borre, Y.E., et al., Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depression, Neuropharmacology (2013), http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

Y.E. Borre et al. / Neuropharmacology xxx (2013) 1e11

p Ex XB

xp -E am

O B XC

**

60 50 40

h

M O

O

B

B

X-

X-

-M am Sh

Sh

Ve

eh -V am

X-

em

30 em

Number of nissl-positive cells/mm2

***

70

*

60 50 40

Sh

p Ex XO

B

xp -E

am Sh

DG Ventral Hippocampus

***

70

***

60 50 40

em M XB O

O

B

X-

Ve

h

em am

am Sh

-M

-V

eh

em

30

Sh

Number of nissl-positive cells/mm2

am

p

O B XC

30 -C

Number of nissl-positive cells/mm2

***

70

M XB O

O

O

DG Dorsal Hippocampus 80

Ex XB

h Ve XB O

Sh

Sh

X-

am

Ex

p

-C

Number of nissl-positive cells/mm2

30

DG Ventral Hippocampus

O

O B XC

xp -E am Sh em

eh -V

40

M

-M am Sh

am Sh

30 -M

M XB O

40

am

em

h Ve XB O

Sh

am

-M

-V

eh

em

30

50

Sh

40

**

60

***

L.

CA1 Ventral Hippocampus Number of nissl-positive cells/mm2

*

50

am

-V

p Ex XB

***

Sh

30

K.

CA3 Ventral Hippocampus 60

40

O

O B XC

xp -E am Sh

Sh

am

-C

30

**

50

am

40

**

Sh

*

50

60

*

50

I.

CA1 Ventral Hippocampus

-C

Number of nissl-positive cells/mm2

**

J.

am Sh

B

O

O

H.

CA3 Ventral Hippocampus 60

em

30

em X-

XB

am Sh

M

Ve

h

em -M

-V am Sh

G.

***

60

B

-E

40

eh

30

DG Dorsal Hippocampus 70

B

40

50

B

50

*

O

60

F.

CA1 Dorsal Hippocampus 60

em

Number of nissl-positive cells/mm2

70

am Sh

Sh

XB

O

Sh

*

80

xp

-C am

p Ex

XB

-E am

am Sh

30

O

E.

CA3 Dorsal Hippocampus

eh

Number of nissl-positive cells/mm2

C

xp

30

O

40

40

O B XC

50

**

50

h

60

**

60

Ve

70

C.

CA1 Dorsal Hippocampus Number of nissl-positive cells/mm2

*

80

D.

Number of nissl-positive cells/mm2

B.

CA3 Dorsal Hippocampus

-C

Number of nissl-positive cells/mm2

A.

X-

8

Number of nissl-positive cells/mm2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

NP5316_proof ■ 27 November 2013 ■ 8/11

Fig. 5. Dietary and memantine treatment partially rescued the OBX-induced cell loss in the ventral and dorsal hippocampi. A, B, C represent CA3, CA1, DG areas of dorsal hippocampus under the dietary treatment; D, E, F represent CA3, CA1, DG areas of the dorsal hippocampus under the memantine treatment. G, H, I represent CA3, CA1, DG areas of ventral hippocampus under the dietary treatment; J, K, L represent CA3, CA1, DG areas of the ventral hippocampus under the memantine treatment. Data are expressed as mean  SEM of nissl-positive cell counts per mm2 *p < 0.05 and **p < 0.01, ***p < 0.001 as determined by Bonferroni multiple comparisons following a two-way ANOVA. Group sizes: n ¼ 8-10.

Please cite this article in press as: Borre, Y.E., et al., Neuroprotective and cognitive enhancing effects of a multi-targeted food intervention in an animal model of neurodegeneration and depression, Neuropharmacology (2013), http://dx.doi.org/10.1016/j.neuropharm.2013.11.009

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

Y.E. Borre et al. / Neuropharmacology xxx (2013) 1e11 Table 2 Two -way ANOVA analysis of the nissl-positive cells counted in hippocampal subfields CA1, CA3 and DG in the dorsal and ventral parts of hippocampus. Group sizes: n ¼ 8e10. Memantine

ns ns