Neurobiology of Learning and Memory 116 (2014) 59–68

Contents lists available at ScienceDirect

Neurobiology of Learning and Memory journal homepage: www.elsevier.com/locate/ynlme

Dissociable effects of dorsal and ventral hippocampal DHA content on spatial learning and anxiety-like behavior Eldin Jašarevic´ a,b,c,⇑, Patrick M. Hecht a,b, Kevin L. Fritsche d,e, David Q. Beversdorf a,b,c,f,g, David C. Geary a,c,⇑ a

Interdisciplinary Neuroscience Program, University of Missouri, Columbia, MO 65211, United States Thompson Center for Autism and Neurodevelopmental Disorders, University of Missouri, Columbia, MO 65211, United States c Department of Psychological Sciences, University of Missouri, Columbia, MO 65211, United States d Division of Animal Sciences, University of Missouri, Columbia, MO 65211, United States e Department of Nutrition and Exercise Physiology, University of Missouri, Columbia, MO 65211, United States f Department of Radiology, University of Missouri, Columbia, MO 65211, United States g Department of Neurology, University of Missouri, Columbia, MO 65211, United States b

a r t i c l e

i n f o

Article history: Received 9 May 2014 Revised 19 August 2014 Accepted 21 August 2014 Available online 30 August 2014 Keywords: Docosahexaenoic acid Arachidonic acid Spatial learning Anxiety-like behavior Hippocampus

a b s t r a c t Chronic deficiency of dietary docosahexaenoic acid (DHA) during critical developmental windows results in severe deficits in spatial learning, anxiety and hippocampal neuroplasticity that parallel a variety of neuropsychiatric disorders. However, little is known regarding the influence of long-term, multigenerational exposure to dietary DHA enrichment on these same traits. To characterize the potential benefits of multigenerational DHA enrichment, mice were fed a purified 10:1 omega-6/omega-3 diet supplemented with either 0.1% preformed DHA/kg feed weight or 1.0% preformed DHA/kg feed weight through three generations. General locomotor activity, spatial learning, and anxiety-like behavior were assessed in adult male offspring of the third generation. Following behavioral assessments, ventral and dorsal hippocampus was collected for DHA and arachidonic acid (AA) analysis. Animals consuming the 0.1% and 1.0% DHA diet did not differ from control animals for locomotor activity or on performance during acquisition learning, but made fewer errors and showed more stable across-day performance during reversal learning on the spatial task and showed less anxiety-like behavior. Consumption of the DHA-enriched diets increased DHA content in the ventral and dorsal hippocampus in a region-specific manner. DHA content in the dorsal hippocampus predicted performance on the reversal training task. DHA content in the ventral hippocampus was correlated with anxiety-like behavior, but AA content in the dorsal hippocampus was a stronger predictor of this behavior. These results suggest that long-term, multigenerational DHA administration improves performance on some aspects of complex spatial learning, decreases anxietylike behavior, and that modulation of DHA content in sub-regions of the hippocampus predicts which behaviors are likely to be affected. Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction Docosahexaenoic acid (DHA [22:6(n-3)]) is the most abundant omega-3 (n-3) polyunsaturated fatty acid in the mammalian brain and provides structural integrity to neuronal membranes,

⇑ Corresponding authors. Address: Interdisciplinary Neuroscience Program, Center for Translational Neuroscience, University Hospital, M741 One Hospital Drive, DC069.10, Columbia, MO 65212, United States (E. Jašarevic´). Address: Interdisciplinary Neuroscience Program, Department of Psychological Sciences, University of Missouri, 210 McAlester Hall, Columbia, MO 65211, United States (D.C. Geary). Fax: +1 (573) 882 7710. E-mail addresses: [email protected] (E. Jašarevic´), [email protected] (D.C. Geary). http://dx.doi.org/10.1016/j.nlm.2014.08.009 1074-7427/Ó 2014 Elsevier Inc. All rights reserved.

functions as a precursor for a variety of signaling molecules, including prostaglandins and neuroprotectins, and regulates gene expression through the activation of transcription factors (Bazan, 2006; Innis, 2007; Mitchell, Niu, & Litman, 2003; Xiao, Huang, & Chen, 2005). The window of DHA accumulation in the brain is tightly conserved in mammals, whereby a sharp DHA accretion spurt begins during the last trimester of pregnancy and remains elevated until 24 months in humans and from gestation day 7 to postnatal day 21 in rodents (Green & Yavin, 1993, 1998; Yavin, Glozman, & Green, 2001). DHA preferentially accumulates in the brain in a region-specific manner, primarily in the frontal cortex, hippocampus, hypothalamus and striatum, and dietary depletion of neural DHA results in region-specific alterations in neuroplasticity and in behavioral deficits (Chung, Chen, & Su, 2008; Su, 2010).

60

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

Given the importance of DHA in normal neurodevelopment and function, suboptimal acquisition of DHA during critical developmental windows may be associated with increased risk for neuropsychiatric disorders. Indeed, low levels of plasma DHA have been observed in a variety of neuropsychiatric conditions, including bipolar disorders, depression, suicide, and Alzheimer’s disease (Akbar, Calderon, Wen, & Kim, 2005; Calon et al., 2004; Cole & Frautschy, 2006; McNamara et al., 2007; Su et al., 2008; Vaz, Kac, Nardi, & Hibbeln, 2014). Animal models of chronic dietary DHA depletion, primarily attained through a multigenerational dietary deficiency of a-linolenic acid (ALA [18:3(n-3)]), reduces neural DHA levels to 3–5% of total fatty acids in the lipid fraction and mimics many clinical manifestations of DHA depletion, including poor performance on spatial learning tasks, deficits in olfactory discrimination, and increased anxiety- and depressive-like behaviors (Fedorova & Salem, 2006; Moriguchi, Greiner, & Salem, 2000). Based on these studies, it is possible that these behavioral deficits following chronic depletion of DHA are likely related to impaired neurotransmission of serotonergic neurons and decreased dopamine release in the dopamine mesolimbic pathway. Supplementation of ALA-enriched diet recovers brain DHA levels to the more typical 8–12% of total fatty acids, reverses behavioral deficits, and increases proteins related to synaptic plasticity, including brain derived neurotrophic factor and phosphorylated CREB in the hippocampus (Bhatia et al., 2011; Fedorova & Salem, 2006; Zimmer et al., 2002). The recovery of behavior and neuroplasticity is reported among individuals and in rodent models with chronically low levels of DHA, but the beneficial role of dietary enrichment with preformed DHA for individuals consuming a diet with a balanced fatty acid composition remains relatively unknown. Preferential accumulation during development suggests that DHA is critical for normal hippocampal-dependent behavior (Su, 2010). Chronic ALA depletion decreases hippocampal DHA levels, and results in severe deficits in hippocampal-dependent spatial learning and underlying neural mechanisms, including decreases in adult neurogenesis, suppression of long-term potentiation, and decreased activation of synaptic proteins (Su, 2010). However, the overwhelming focus in the literature on the contribution of DHA levels to hippocampal-dependent learning overshadows the function of the hippocampus in behaviors that are not specifically related to spatial navigation. Based on differences in patterns of gene expression and anatomical projections along the dorsal–ventral axis of the hippocampus, it has been suggested that these subregions may be involved in different functional behaviors (Dong, Swanson, Chen, Fanselow, & Toga, 2009; Fanselow & Dong, 2010). Lesions of the dorsal hippocampus (DH) significantly impaired acquisition and retention on a spatial navigation task, whereas lesions of the ventral hippocampus (VH) had no effect on spatial performance (Klur et al., 2009; Moser, Moser, Forrest, Andersen, & Morris, 1995). Conversely, lesions of the VH increased open arm entries and time spent in the open arms of the elevated plus maze and dampened stress-induced rise in corticosterone, both of which are consistent with a reduction in fear- or anxietyrelated behavior. Lesions to the DH did not result in these behavioral changes (Fanselow & Dong, 2010; Kjelstrup et al., 2002). Taken together, these studies suggest that DH contributes to acquisition and retention of spatial memories whereas the VH is necessary for fear-related behavior. While n-3 dietary deficiency studies using a multigenerational deprivation paradigm have revealed severe deficits in cognition and affect (Fedorova & Salem, 2006), little is known about the effects of a multigenerational exposure to a balanced n-6/n-3 ratio diet and supplementation of preformed DHA on cognition and affect and the underlying neural correlates. As brain DHA status modulates both spatial learning and fear-related behavior, the present multigenerational study was designed to (i) determine whether dietary enrichment

of preformed DHA improves acquisition and reversal learning on a spatial navigation task and reduces anxiety-like behavior, (ii) measure DHA levels in the dorsal and ventral hippocampus, (iii) and assess whether DHA levels in the DH and VH differentially predict behavioral outcomes. 2. Materials and methods 2.1. Animals Sixty (60) 6–8 week old C57BL/6J females (P0 females) were purchased from Jackson Laboratories (Bar Harbor, Maine) and fed the control diet for at least two weeks during habituation to the vivarium. Following habituation, animals were randomly placed on one of three diets: (i) the CTL diet with no preformed DHA (CTL, n = 20), (ii) the CTL diet supplemented with 0.1% by weight DHA (0.1% DHA, n = 20), (iii) or the CTL diet supplemented with 1.0% by weight DHA (1.0% DHA, n = 20). Diet production and composition is discussed in detail below. Internal routine screens of standard rodent chow preparations revealed differences in the composition of fatty acids that may result in physiologically relevant differences in omega-3 polyunsaturated fatty acid (PUFA) status and brain DHA levels. A multigenerational breeding scheme was used to control for this potential confound. P0 females were mated with animals consuming the CTL diet (detailed below) and females were single-housed upon detecting a mating plug. Females remained on the assigned diet through gestation and lactation, and F1 offspring remained on the same maternal diet. Similar to the P0 breeding scheme, F1 females were mated with CTL males, remained on the assigned diet during gestation and lactation, and F2 offspring were maintained the same diet as their mothers. Based on the disproportionately high rates of first litter mortality in C57BL/6J mice (Brown, Mathieson, Stapleton, & Neumann, 1999) and the interaction of diet and parity on DHA levels (Ozias, Carlson, & Levant, 2007), only second litter males were used for this experiment. Only male offspring were used in these experiments based on previous reports that suboptimal dietary fatty acid composition leads to greater disruption in spatial learning in males than females (Lindqvist et al., 2006). Animals were housed in clear polycarbonate cages (32 cm  18 cm  24 cm) provided with aspen bedding and a nestlet. All animals were maintained under standard conditions (25 ± 2 °C and 50% ± 10% humidity), with ad libitum access to water provided in glass bottles and a diet specific to each treatment group, and on 12:12 h light cycle with lights on at 0600 CST. All experiments were approved by University of Missouri Animal Care and Use Committee and performed in accordance with National Institutes of Health Animal Care and Use Guidelines. 2.2. Diet composition The rodent diets in this study started with the AIN-93G purified-diet profile (Dyets Inc., #110700) as the base. The AIN93G diet uses solely soybean oil as the source of fat, resulting in a 7:1 n-6:n3 ratio. Early studies measuring n-6:n-3 ratios in Western populations have reported ratios of up to 50:1 n-6:n-3 (Simopoulos, 2002); however, more recent efforts have shown that the average n-6:n-3 ratio of individuals consuming a Western-type diet is closer to a 10:1 n-6:n-3 ratio (Blasbalg, Hibbeln, Ramsden, Majchrzak, & Rawlings, 2011; Daniel et al., 2009; Sun, Ma, Campos, Hankinson, & Hu, 2007). In order to model the average fatty acid composition of a typical Western diet, we altered the AIN93G base profile by adding corn and soy oil at a 2:1 ratio to achieve a 10:1 n-6:n-3 ratio AIN93G diet (Dyets Inc., #103619). This modified diet served as the control (CTL) and as the base for the experimental DHA diets; the

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

CTL diet contained no preformed DHA but all of the essential fatty acids necessary for biosynthesis of DHA. The experimental diets were identical to the CTL, except for the addition of 0.1% by weight DHA (Dyets Inc., #103597) or 1.0% by weight DHA (Dyets Inc., #103598) (a generous gift from OmegaProtein Inc., Houston, TX). Both DHA diets were stabilized against auto-oxidation by the addition of a synthetic antioxidant (i.e., 0.02 g tertiary-butylhydroquinone/100 g fat) (Irons & Fritsche, 2005). The DHA doses in the diets were chosen based on a cross-study meta-regression dose– response analyses that showed plasma phospholipid DHA concentrations increase in a dose-dependent, saturable manner in response to dietary DHA, and identified the highest sensitivity to dietary intake to be between 0.1 and 2.0 g/day (Arterburn, Hall, & Oken, 2006). A ten-fold dose range of 0.1% and 1.0% DHA was chosen to capture the entire range of maximal sensitivity to dietary DHA and to keep the amount of linoleic acid in each DHA diet matched to that found in the CTL diet. Table 1 provides detailed fatty acid composition of the experimental diets. 2.3. Open field test On PND 60, general locomotor and exploratory behavior was assessed using the open field test. The apparatus was constructed out of clear Plexiglas, measured 45 cm  45 cm  22 cm, overlaid on an 8  8 grid, and the outer walls were covered with dark poster board to ensure mice could not see beyond the apparatus. Mice were tested for 10 min during the light portion of the cycle and activity was recorded using the automated video tracking software AnyMaze (Stoelting AnyMaze Software, Wood Dale, IL, USA). No observers were present in the room during testing. The field was wiped down with 70% ethanol alcohol (EtOH) between each testing subject. Total distance travelled, time spent in center quadrant (4  4 central grid), time spent in the periphery, and number of entries into each zone were recorded using the software. 2.4. Barnes maze acquisition and reversal learning 2.4.1. Apparatus Spatial learning was assessed in the Barnes maze as previously reported (Jasarevic, Williams, Roberts, Geary, & Rosenfeld, 2012; Jasarevic et al., 2011, 2013), with some modifications. The maze consisted of a white circular platform 75 cm in diameter that is brightly lit from above. The platform was elevated 50 cm above the floor by a stand. Twenty holes measuring 5 cm in diameter were evenly spaced around the perimeter. A triangular black escape box (17.8 cm length  5.1–10.2 cm wide  7 cm high) containing a small ramp could be slid into place beneath any hole. The floor beneath the stand was covered with black fabric, and curtains (28 cm in height) of the same color were hung above the maze

Table 1 Fatty acid composition of diets. Standard diet 0.1% DHA diet (% by weight in diet)

1% DHA diet

Fatty acid 16:0 18:0 18:1 18:2n6 18:3n3 20:5n3 22:5n3 22:6n3

0.77 0.24 1.76 3.86 0.37 nd nd nd

0.77 0.24 1.60 3.86 0.38 nd 0.02 0.10

0.39 0.16 1.00 4.01 0.02 0.11 0.26 1.01

Ratio n-6/n-3

10.4

7.7

2.9

Note: nd = not detectable.

61

150 cm high from floor level. These curtains surrounded the apparatus to ensure that the mice were only using the visual cues provided in the maze, and not distal cues within the testing room. Four shapes (triangle, square, circle, and star) were placed at evenly spaced intervals on the inside of the maze wall. Three 150-W lights were hung above the platform to create a potentially averse environment and motivate the mice to escape from the brightly lit, open space. 2.4.2. Procedure At the beginning of each testing day, animals were transferred from the vivarium to the testing room 30 min before behavioral assessments to reduce any stress related to transport. Before testing, mice were randomly assigned to one of the 20 exit holes, which determined the location of the target exit during the remainder of acquisition testing. Mice completed three phases of testing: habituation (1 d, 1 trial), acquisition (5 d, 2 trials/day), and reversal training (5 d, 2 trials/day). 2.4.3. Habituation Mice completed one habituation trial for 300 s to become familiar with the maze environment and to practice descending into their target exit hole. 2.4.4. Acquisition Mice completed 5 days of acquisition training with 2 trials per day and a 20–30 min inter-trial interval. For each trial, mice were placed in an opaque starting box in the center of the maze. To prevent use of a fixed motor response or side bias to locate the escape hole, the direction mice were facing when placed into the starting box was randomized across trials. After 30 s, the starting box was raised and the trial was started. If a mouse failed to enter the target exit in 300 s it was gently guided to the escape hole using a small glass cup. After entering the escape hole, the mouse remained there for 30 s before being returned to the home cage. To reduce odor cues, the maze surface and escape box were cleaned with 70% EtOH between each trial. At the end of each testing day the maze was rotated 90° to the left or right, although the 4 intramaze visual cues remained in the same positions relative to the escape hole. Rotating the maze ensured that the hole at the spatial location where escape could occur varied across training trials. For each trial, latency to enter the escape hole and distance travelled was recorded by the AnyMaze tracking software. Search errors were counted when the mouse dipped its head into a blind hole and repeated successive dips into the same hole were considered a single error. The strategies used to locate the escape hole were classified as one of three search strategies. The random strategy was operationally defined as localized searches of holes separated by maze center crosses. The serial strategy was defined as a systematic search of consecutive holes in a clockwise or counterclockwise direction. The direct strategy was defined as navigating directly to the target exit hole with no center crosses and 63 errors. 2.4.5. Reversal training During reversal training (d 6–10) the location of the escape hole was moved 180° from its location during acquisition training and the maze was visually separated into 4 quadrants using the AnyMaze software. As with acquisition training, mice completed 5 days of reversal training with 2 trials per day and an inter-trial interval of 20–30 min. Latency, distance travelled, error, and search strategies were recorded. Perseveration, or the time spent in the quadrant of the target exit assigned during acquisition training, was calculated as the proportion of time spent in the target quadrant relative to time to reach the reversal learning exit hole (e.g., [(time in acquisition target/latency) ⁄ 100]).

62

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

2.5. Elevated plus maze

2.8. Statistical analysis

Following testing in the Barnes maze, exploratory and fearrelated behaviors were measured in the elevated plus maze (EPM). The EPM was constructed of black polypropylene in a plus configuration with two opposite open arms (30 cm), a middle platform (5  5 cm), and two opposing closed arms (30 cm). The maze was supported 100 cm above the floor by a stand constructed of polypropylene. Each animal was placed on the center of the platform and allowed to explore the maze for 300 s. The direction the animal faced when placed in the center of the maze was alternated to control for potential side biases. After each test, the apparatus was cleaned with 70% EtOH. Each trial was recorded with AnyMaze software, which automatically scores total time spent in open and closed arms and number of closed and open arm entries and center entries. Arm entry was defined as both front paws and shoulders placed into the area. On the occasion an animal jumped off the maze, it was gently placed back in the center, and the trial was continued.

All data are presented as mean ± SEM and analyzed in R using lattice and linear and nonlinear mixed effects models packages (Deepayan, 2008; Pinheiro, Bates, DebRoy, & Sarkar, 2013; Team, 2014). All analyses were conducted at the level of the individual. In the open field test, the relations between diet and distance travelled, time spent in the center and outside quadrant, and number of entries into center and quadrant were analyzed with a oneway ANOVA. For the Barnes maze, as different neurobiological processes may be involved in acquisition and reversal learning (Shuai et al., 2010), associated data were analyzed separately. Mean scores for acquisition and reversal training path length, latency, and errors were analyzed with 3 (diet)  5 (day) repeated-measures ANOVAs; means across the 2 trials/day were used. The discrete Barnes search strategy was analyzed with a 3 (diet)  5 (day) binomial logistic analysis separately for acquisition and reversal training. The outcome of interest was the probability of use of the direct strategy and thus the random and serial strategies were combined and contrasted with the direct strategy. For the EPM, distance travelled, time spent in open arms, open arm entries, closed arm entries, proportion of total time spent in open arms were submitted to a one-way ANOVA. Percent DHA and AA of total fatty acids in the dorsal and ventral hippocampus was analyzed with a 2 (region)  3 (diet) ANOVA. Tukey’s HSD post hoc comparison was used for all group-level contrasts. Stepwise regressions were then used to assess the relations between dorsal and hippocampal DHA and AA levels and behavioral measures on the Barnes maze and EPM.

2.6. Hippocampus collection Twenty-four hours following completion of behavioral testing, animals were anesthetized using CO2, rapidly decapitated and brains were excised and placed into a Zivic adult mouse brain matrix (Zivic Instruments, Pittsburgh PA). Bilateral 1.0 mm micropunches of dorsal (Bregma 1.00 to 2.00 mm) and ventral (Bregma 2 mm to 3 mm) hippocampus were obtained from two serial 1 mm thick sections spanning the hippocampus. The disappearance of the corpus callosum was used as a landmark for the dorsal hippocampus and a bilateral punch that was lateral and posterior to the third ventricle was made to collect the ventral hippocampus. Samples were quickly frozen in liquid nitrogen and stored in 80 °C until fatty acid analysis.

2.7. Brain fatty acid analysis Total lipids were extracted from brain homogenates following the method of Folch, Lees, and Sloane Stanley (1957). Ten volumes of 0.32 M sucrose solution to brain wet weight was added to brain tissue and homogenized. 100 lL of brain homogenate was pipetted into a screwcap glass tube containing 0.1 mg of internal standard (23:0 dissolved in methanol). One mL of 0.5 N methanolic base was added and the sample was boiled at 100 °C for 15 min. The samples were cooled to room temperature, upon which 2 mL of 14% BF3-methanol was added and the sample was boiled at 100 °C for 20 s. Once cooled to room temperature, 1 mL of isooctane was added and the mixture was gently inverted. Five mL of saturated NaCl in ddH2O was added and the mixture was agitated. After the phases separated, the upper isooctane layer was separated from the aqueous phase with a glass Pasteur pipette into a pre-rinsed conical glass tube through another Pasteur pipette filled with anhydrous sodium sulfate granules. Isooctane was evaporated under N2 and resuspended in 100 lL of heptane. The sample was injected into the GLC system. Fatty acid methyl esters were analyzed using a gas chromatograph (Agilent 7809A) equipped with a 60 m, 0.25 mm I.D., and 0.15 lm film DB-23 column. Gas chromatograph conditions were a helium flow rate of 1 mL/min with an initial temperature of 140 °C held for 5 min. The column temperature was then increased to 250 °C at a rate of 2 °C/min and held at 250 °C for 15 min. Fatty acid peaks were identified based on relative retention times using pure methyl ester standards (Nu-Chek Prep, Elysian, MN).

3. Results 3.1. General activity patterns During the open field test, there were no significant main effects of diet on locomotor activity or exploration, including total distance travelled, time spent in outside quadrant, time spent in center, or distance travelled in the outside quadrant or center (Fig. 1, Ps > 0.05). 3.2. Spatial learning During the acquisition phase of the Barnes maze learning, latencies to reach the target exit hole decreased across days (Fig. 2A; F4,280 = 65.36, p < 0.001); however, there was no main effect of diet on latency (p = 0.13), or a day  diet interaction (p = 0.75). Similarly, number of training errors (Fig. 2B; F4,280 = 67.68, P < 0.001) and path length (Fig. 2C; F4,280 = 41.68, p < 0.001) decreased across acquisition training, but neither the diet (ps = 0.48 and 0.23, respectively) nor the day  diet interactions (ps = 0.54 and 0.74, respectively) was significant. As shown in Fig. 2C, animals in all diet groups predominantly used a random search strategy on the first day of acquisition training. Across subsequent days there was a significant increase in the use of the direct search strategy (F1,325 = 100.06, P < 0.001), but the day  diet interaction (P = 0.61) was not significant. There was, however, a main effect of diet (p = 0.053), but none of the pairwise group contrasts were significant (ps > .25). A significant day  diet interaction for latency to reach the new target hole was observed during reversal training (Fig. 2a; F8,285 = 4.73 p < 0.001). Animals consuming either 0.1% DHA or the 1.0% DHA diet exhibited shorter latency than CTL males (ps < 0.001) but there were no differences between the two DHA diet groups (p = 0.64). The main effect of diet (F2,30 = 13.71, p < 0.001), day (F4,285 = 21.78, p < 0.001), and the day  diet inter-

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

63

Fig. 1. General locomotor activity in mice consuming a diet with no DHA (black), 0.1% DHA (gray) or 1.0% DHA (white). Mean (± SEM) time spent in outside and center quadrants (A and C), and distance travelled in outside and center zone (B and D). n = 9–12 males per group.

action (F8,285 = 4.53 p < 0.001) were significant for reversal training errors (Fig. 2a). The interaction emerged because mice consuming either the 0.1% DHA or 1.0% DHA diet committed fewer errors across reversal training than CTL males (ps < 0.01), with no difference between animals in the DHA diet groups (p = 0.93). Path length decreased across days of reversal training (F4,285 = 8.63, p < 0.001) but this was qualified by a day  diet interaction (F8,285 = 3.21, p = 0.04); mice consuming either the 0.1% DHA or 1.0% DHA diet exhibited shorter path lengths than CTL males (ps = 0.007 and 0.03, respectively), with no difference between mice consuming either DHA diet (p = 0.95). Overall, animals decreased perseveration across reversal training (F1,338 = 23.28, p < 0.001) but there was no main effect of diet (p = 0.10) or day  diet interaction (p = 0.48). There was also a main effect of day (F1,328 = 17.99, p < 0.001), and diet (F1,327 = 12.1647, p < 0.001) on probability of using the direct search strategy during reversal training (Fig. 2); the day  diet interaction was not significant (p = 0.1467). Compared to CTL mice, animals consuming the 1.0% DHA diet used the direct strategy more often (p = 0.02) whereas the contrast of the 0.1% DHA mice with CTL mice was trending (p = 0.071); mice in the two DHA diet groups did not differ (p = 0.78). 3.3. Anxiety-like behavior There was a significant main effect of diet on time spent in open arms (F2,30 = 12.093, p < 0.001), number of open arm entries (F1,30 = 8.975, p = 0.001), distance travelled in open arms (F1,30 = 7.614, p < 0.001), time spent in closed arms (F1,30 = 13.888, p = 0.002), and proportion of time spent in open arms relative to time spent in closed arms (F1,30 = 12.09, p < 0.001) (Fig. 3). 0.1% DHA animals made more entries into the open arms ðX ¼ 19:50  4:796Þ than CTL animals ðX ¼ 9:92  6:022, p = 0.001). 0.1% DHA animals

ðX ¼ 132:54  45:08 sÞ and 1.0% DHA animals ðX ¼ 134:87 32:30 sÞ spent more time in the open arms than CTL animals ðX ¼ 59:14  44:48 s, ps < 0.01). 0.1% DHA animals ðX ¼ 581:08 156:79 cmÞ and 1.0% DHA animals ðX ¼ 488:74  230:58 cmÞ covered more distance in the open arms than CTL animals ðX ¼ 251  245 cm, ps < 0.05). CTL animals ðX ¼ 202:69  47:37 sÞ spent more time in the closed arms than 0.1% DHA animals (X ¼ 124:65  41:21 s) and 1.0% DHA animals (X ¼ 132:18 19:75 s, ps < 0.01). No other comparisons were significant. 3.4. Hippocampal DHA and AA composition Data for DHA and AA levels in the dorsal and ventral hippocampus was first collapsed to assess whether there was a differences in DHA and AA levels in total hippocampus. There was a main effect of diet on hippocampal DHA content (F2,57 = 15.941, p < 0.001), and AA content (F2.57 = 6.794, p < 0.01). Both 0.1% DHA and 1.0% DHA diet animals had higher DHA content in the hippocampus than CTL animals (ps < 0.05), whereas only the 1.0% DHA diet animals exhibited lower AA content than CTL animals (p = 0.001). Fig. 4 shows that when the dorsal and ventral hippocampus were treated as distinct regions, there were main effects of diet (F2.54 = 17.193, p < 0.01) and region (F1.54 = 4.165, p < 0.05) on DHA content; the diet  region interaction was not significant (p = .391). Overall, there was more DHA in the dorsal than ventral hippocampus. Post hoc comparisons revealed that DHA content in the DH of animals consuming the 1.0% DHA diet was greater than that of CTL diet animals (p = 0.001) and 0.1% DHA animals (p = 0.031); no other contrasts were significant (ps > 0.26). Animals consuming the 1.0% DHA diet had higher DHA content in the VH than CTL animals (p = 0.02), as did animals consuming the 0.1% DHA diet (p = 0.033); the contrast of the two diet groups was not significant (p 0.70).

64

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

Fig. 2. Spatial learning in mice exposed to diet a containing no DHA, 0.1% DHA or 1.0% DHA. Mean (±SEM) latency, error and distance travelled (A) across acquisition (A1–A5) and reversal (R1–R5) training. Mean (±SEM) perseveration (B) across reversal training. Perseveration is operationalized as proportion of time spent in the acquisition target quadrant during reversal training. Strategy use [random (black), serial (gray), and direct (white)] during acquisition and reversal training (C). n = 9–12 males per group.

Fig. 3. Anxiety-like behavior in mice consuming a diet a containing no preformed DHA (black), 0.1% DHA (gray) or 1.0% DHA (white). Mean (±SEM) for time spent in open arms, distance travelled in open arms, open arm entries, and proportion of time spent in open arms. ⁄Difference between CTL and DHA diet groups (see Section 3 for details). n = 9–12 males per group.

There was a main effect of diet on proportion of AA (F2.54 = 7.241, p < 0.01), but neither hippocampal region (P = 0.478) nor its interaction with diet (P = 0.129) drove this difference (Fig. 4). Pairwise comparisons revealed that 1.0% DHA diet animals exhibited lower AA content than CTL animals (p = 0.001); no other comparisons were significant (ps > 0.16).

3.5. Relations between hippocampal DHA levels and behavior The role of the dorsal and ventral hippocampus on complex behavior has been previously reported to be region-specific, with the DH primarily responsible for information processing and spatial learning, whereas anxiety-like behaviors are under the partial

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

65

Fig. 4. Mean (±SEM) for DHA (A) and Arachidonic acid (B) of total fatty acid content in the dorsal and ventral hippocampus in mice exposed to diet a containing no preformed DHA, 0.1% DHA or 1.0% DHA. Same letters indicate no difference within region, whereas different letters indicate significant differences within region (see Section 3 for detailed comparisons). n = 9–12 males per group.

control of the VH. Accordingly, we examined the relation between DHA content in the DH and VH and performance parameters from the Barnes maze and elevated plus maze, focusing on behaviors with significant diet or diet  day effects. Correlations revealed a significant inverse relation between DH DHA and reversal learning errors rates (r = 0.501, p = 0.004) and error variance (i.e., acrossday changes in error rates) (r = 0.386, p = 0.036), consistent with previous deficiency studies; a correlation matrix of hippocampal DHA and AA content and reversal training error rates and error variance is provided in Table S1. The stepwise multiple regression was conducted including AA and DHA content in the DH and VH as predictors of reversal training error and error variance. At step 1 DHA content in DH entered into the equation and was significantly related to reversal training errors (b = 0.62, t = 3.49, p = 0.002). DHA content in VH, or AA content in the DH or VH did not enter into the equation at step 2 (ts = 1.03, 0.36, 1.13, respectively, ps > 0.27). The same procedure indicated that error variance was simultaneously predicted by DHA content in both the DH (b = 0.46, t = 2.796, p = 0.01) and VH (b = 0.48, t = 2.792, p = 0.01). Although time spent in the open arms of the EPM was positively correlated with DHA content in the VH (r = 0.358, p = 0.052, n = 30), consistent with deficiency studies, this relation did not emerge once DHA in DH and AA in DH and VH were included in the stepwise regression; a correlation matrix of hippocampal DHA and AA content and EPM variables is provided in Table S2. At step 1 AA content in DH entered into the equation and was significantly related to time spent in the open arms (b = 0.48, t = 3.14, p = 0.004). However, DHA content in the VH did not emerge as a significant predictor, nor did DHA content in DH or AA content in the VH (ts = 0.90, 1.08, 0.11, respectively, ps > 0.29). 4. Discussion Chronic dietary deficiency of n-3 PUFAs during development and adulthood results in significant disruption of spatial learning and increased anxiety-like behavior and repletion of neural DHA is only partially successful in rescuing these deficits (Calon et al., 2004; Federova, Nahed, Baumann, DiMartino, & Norman, 2009; Fedorova & Salem, 2006; Gamoh et al., 1999; Moriguchi et al., 2000). The latter suggests there may be a critical developmental

window during which DHA accumulation is necessary for the normal expression of these behaviors in adulthood (Fedorova & Salem, 2006). Although numerous studies have highlighted the importance of DHA accretion during the prenatal and early postnatal period, the predominant focus has been within the context of maternal dietary deficiency of n-3 PUFAs (Al et al., 1995; Connor, Neuringer, & Lin, 1990; Crawford, Costeloe, Ghebremeskel, & Phylactos, 1998; Destaillats et al., 2010; Garcia-Calatayud et al., 2005; Green & Yavin, 1998; Harauma, Salem, & Moriguchi, 2010; Mozurkewich et al., 2011, 2013; Mulder, King, & Innis, 2014; Rioux, Belanger-Plourde, Leblanc, & Vigneau, 2011; Sable, Kale, & Joshi, 2013; Salem et al., 2001; Uauy & Dangour, 2006; Yavin, 2006; Yavin et al., 2001). Across numerous animal models involving manipulation of diet composition, route of administration, and sources of n-3 PUFAs, a specific vulnerability of spatial learning and memory and anxiety-like behavior and their neural correlates has been observed in adults (Fedorova & Salem, 2006; Ross, Seguin, & Sieswerda, 2007; Su, 2010). Yet, little is known about the effects of a multigenerational exposure to a balanced n-6/n-3 ratio diet and supplementation of preformed DHA on spatial learning and anxiety-like behavior and the underlying hippocampal correlates. The present study tested the prediction that a 3 generation exposure to preformed DHA will confer a benefit on spatial learning and anxiety-like behavior and that these effects will be related to DHA levels in the dorsal and ventral subregions of the hippocampus, respectively. The finding that supplementation of DHA had no effect on spatial performance during acquisition training suggests the amount of omega-3 fatty acid precursors in the standard diet was sufficient for normal development and maintenance of these behaviors. However, supplementation of preformed DHA improved performance on spatial reversal training. Mice consuming both doses of DHA exhibited shorter latencies and fewer errors across reversal training. It is unlikely that these diet effects are related to differences in general activity and locomotor activity because no differences emerged in the open field task. The clear difference in the role of DHA enrichment on acquisition and reversal training may be related to the distinct molecular mechanisms and neuroanatomical structures required during these different learning processes (Shuai et al., 2010; Thompson, Kao, & Yang, 1981). During reversal training, functional plasticity of the hippocampal encoding

66

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

network is required because the encoding rule acquired during acquisition training has to be inhibited in favor of a new one representing the modified cue-exit configurations (Garthe, Behr, & Kempermann, 2009; Wiskott, Rasch, & Kempermann, 2006). When task conditions change, such as in the transition period between day 5 of acquisition training and day 1 of reversal training, interference between previously encoded contingencies and newly appearing ones are predicted to affect maze performance, which was revealed by increased latency and errors and reversal from direct to random search strategies. The increased latency and errors across reversal training in CTL mice may reflect the time course of inhibiting the old encoding rule and adopting the new one. If so, DHA enrichment may facilitate this process by decreasing interference between retrieval of the old and new cues during the reversal condition. Similarly, rate of perseveration across reversal training may reflect a preference for the encoding rule acquired during acquisition training until the mouse learns and reliably retrieves the new encoding rule (Garthe et al., 2009; Wiskott et al., 2006). These results compliment previous findings showing that chronic deficiency of n-3 PUFAs results in robust deficits in reversal training performance, characterized by increased latency, error, and perseveration (Federova et al., 2009). The beneficial role of DHA enrichment on this performance may be related to adult hippocampal neurogenesis. Maternal DHA deficiency appears to delay the onset and rate of neurogenesis during prenatal development, whereas DHA enrichment has been shown to promote neurogenesis both in vitro and in vivo (Brand, Crawford, & Yavin, 2010; Coti Bertrand, O’Kusky, & Innis, 2006; Innis, 2007; Katakura et al., 2009; Kawakita, Hashimoto, & Shido, 2006). In addition, enhanced performance on spatial learning tasks has been found to be associated with enhanced cell proliferation and increased dendritic spine density of CA1 pyramidal neurons in the hippocampus following DHA supplementation (He, Qu, Cui, Wang, & Kang, 2009). The present results suggest that the association between DHA supplementation and neurogenesis may be specifically related to reversal learning. Pharmacological suppression of hippocampal neurogenesis during acquisition and reversal training has shown a significant increase in perseveration and utilization of less efficient search strategies during reversal training but no effect on acquisition training performance (Garthe et al., 2009). Proportion of DHA in the dorsal hippocampus predicted number of errors and error variance across reversal training, although DHA in the VH was also independently related to the latter effect. A positive correlation between dietary omega-3 fatty acid intake and behavioral accuracy on a hippocampal dependent relational memory task was recently reported in a cohort of children (Baym et al., 2014), highlighting the potential translational importance of our findings. Despite its translational implications, the present results must be interpreted with some caution based on the observation that AA content in the dorsal hippocampus was reduced in mice consuming the 1.0% DHA diet. Preparation of the 1.0% DHA diet resulted in physiologically relevant levels of eicosapentaenoic acid (Table 1). The decreased AA content in the dorsal hippocampus of mice consuming the 1.0% DHA diet is consistent with previous studies that have shown that administration of DHA in the presence of eicosapentaenoic acid results in increased DHA content with a concomitant decrease in AA content (Green & Yavin, 1995). Arachidonic acid is a key regulator of normal neural functioning: Activation of the perforant pathway and induction of long-term potentiation in the hippocampus is, at least in part, under the control of arachidonic acid (Williams, Errington, Lynch, & Bliss, 1989), and suppression of AA content may affect these neurophysiological properties as related to learning and memory. Nevertheless, animals consuming the 0.1% DHA diet exhibited comparable performance in the Barnes maze without altering AA content, highlighting

a potential dose-dependent property of dietary DHA supplementation; a relatively small dose of DHA supplementation maximizes the benefits and higher doses may result in alteration of neural lipid membranes, potentially producing unintended costs. Moreover, the stepwise regression indicated that hippocampal AA content did not influence reversal errors or error variance above and beyond the influence of DHA content. It is well established that the hippocampus contributes to fear and anxiety memory encoding associated with exploration of a novel environment. More recent studies have suggested that the ventral hippocampus controls fear and anxiety independently of learning (Dong et al., 2009; Fanselow & Dong, 2010; Kjelstrup et al., 2002). In addition, chronic dietary deficiency of DHA increases anxiety-like and fear-related behaviors (Bhatia et al., 2011; Carrie, Clement, de Javel, Frances, & Bourre, 2000), but the relation between DHA content in the ventral hippocampus and anxiety-like behavior has not been evaluated. Multigenerational exposure to dietary DHA enrichment increased time spent and distance travelled in the open arms of the EPM, and ventral hippocampus DHA content was correlated with time spent in the open arms, in keeping with results from deficiency studies. However, ventral hippocampus DHA content was no longer predictive of time in open arms, once the effects of AA content in the dorsal hippocampus were controlled. Whether this result reflects the tradeoffs associated with DHA supplementation and AA content noted above, or a contribution of dorsal hippocampus AA to suppression of anxiety-related behaviors – as contrasted with enhancement of them with dietary deficiency – remains to be determined. Future studies would also benefit from exploring the consequence of DHA deficiency on behavior and subregion specific gene networks and to what extent DHA enrichment is capable of reversing behavioral deficits and the accompanying dysregulation of gene expression patterns.

5. Conclusion The present work characterized subregion differences in DHA content within the mouse hippocampus following multigenerational exposure to preformed dietary DHA. DHA enrichment did not affect general locomotor activity, but improved performance on a spatial learning task that requires the animals to inhibit a previously learned encoding rule and learn and use a new rule, and performance on this task was predicted by dorsal hippocampus DHA content. DHA enrichment also suppressed anxiety-like behavior relative to controls, but this effect may be more strongly related to AA in the dorsal hippocampus than to DHA in ventral hippocampus as is found in diet deficiency studies.

Acknowledgments We thank OmegaProtein for their generous donation of pure DHA oil used in our experimental diets; OmegaProtein had no influence on the design, implementation, or interpretation of the study. We also thank Jim Browning and Connor Fraser for technical assistance, and Dr. Fumihiro Matsui for critical feedback to previous versions of this manuscript. Research Council of the University of Missouri (DQB & DCG) and the University of Missouri Research Investment Fund (DQB) supported this work.

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.nlm.2014.08.009.

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

References Akbar, M., Calderon, F., Wen, Z., & Kim, H. Y. (2005). Docosahexaenoic acid: A positive modulator of Akt signaling in neuronal survival. Proceedings of the National Academy of Sciences of the United States of America, 102, 10858–10863. Al, M. D., van Houwelingen, A. C., Kester, A. D., Hasaart, T. H., de Jong, A. E., & Hornstra, G. (1995). Maternal essential fatty acid patterns during normal pregnancy and their relationship to the neonatal essential fatty acid status. The British Journal of Nutrition, 74, 55–68. Arterburn, L. M., Hall, E. B., & Oken, H. (2006). Distribution, interconversion, and dose response of n-3 fatty acids in humans. The American Journal of Clinical Nutrition, 83, 1467S–1476S. Baym, C. L., Khan, N. A., Monti, J. M., Raine, L. B., Drollette, E. S., Moore, R. D., et al. (2014). Dietary lipids are differentially associated with hippocampal-dependent relational memory in prepubescent children. The American Journal of Clinical Nutrition, 99, 1026–1032. Bazan, N. G. (2006). Cell survival matters: Docosahexaenoic acid signaling, neuroprotection and photoreceptors. Trends in Neurosciences, 29, 263–271. Bhatia, H. S., Agrawal, R., Sharma, S., Huo, Y. X., Ying, Z., & Gomez-Pinilla, F. (2011). Omega-3 fatty acid deficiency during brain maturation reduces neuronal and behavioral plasticity in adulthood. PLoS One, 6, e28451. Blasbalg, T. L., Hibbeln, J. R., Ramsden, C. E., Majchrzak, S. F., & Rawlings, R. R. (2011). Changes in consumption of omega-3 and omega-6 fatty acids in the United States during the 20th century. The American Journal of Clinical Nutrition, 93, 950–962. Brand, A., Crawford, M. A., & Yavin, E. (2010). Retailoring docosahexaenoic acidcontaining phospholipid species during impaired neurogenesis following omega-3 alpha-linolenic acid deprivation. Journal of Neurochemistry, 114, 1393–1404. Brown, R. E., Mathieson, W. B., Stapleton, J., & Neumann, P. E. (1999). Maternal behavior in female C57BL/6J and DBA/2J inbred mice. Physiology & Behavior, 67, 599–605. Calon, F., Lim, G. P., Yang, F., Morihara, T., Teter, B., Ubeda, O., et al. (2004). Docosahexaenoic acid protects from dendritic pathology in an Alzheimer’s disease mouse model. Neuron, 43, 633–645. Carrie, I., Clement, M., de Javel, D., Frances, H., & Bourre, J. M. (2000). Phospholipid supplementation reverses behavioral and biochemical alterations induced by n3 polyunsaturated fatty acid deficiency in mice. Journal of Lipid Research, 41, 473–480. Chung, W. L., Chen, J. J., & Su, H. M. (2008). Fish oil supplementation of control and (n-3) fatty acid-deficient male rats enhances reference and working memory performance and increases brain regional docosahexaenoic acid levels. The Journal of Nutrition, 138, 1165–1171. Cole, G. M., & Frautschy, S. A. (2006). Docosahexaenoic acid protects from amyloid and dendritic pathology in an Alzheimer’s disease mouse model. Nutrition and Health, 18, 249–259. Connor, W. E., Neuringer, M., & Lin, D. S. (1990). Dietary effects on brain fatty acid composition: The reversibility of n-3 fatty acid deficiency and turnover of docosahexaenoic acid in the brain, erythrocytes, and plasma of rhesus monkeys. Journal of Lipid Research, 31, 237–247. Coti Bertrand, P., O’Kusky, J. R., & Innis, S. M. (2006). Maternal dietary (n-3) fatty acid deficiency alters neurogenesis in the embryonic rat brain. The Journal of Nutrition, 136, 1570–1575. Crawford, M. A., Costeloe, K., Ghebremeskel, K., & Phylactos, A. (1998). The inadequacy of the essential fatty acid content of present preterm feeds. European Journal of Pediatrics, 157(Suppl. 1), S23–S27. Daniel, C. R., McCullough, M. L., Patel, R. C., Jacobs, E. J., Flanders, W. D., Thun, M. J., et al. (2009). Dietary intake of omega-6 and omega-3 fatty acids and risk of colorectal cancer in a prospective cohort of U.S. men and women. Cancer Epidemiology, Biomarkers & Prevention: A Publication of the American Association for Cancer Research, Cosponsored by the American Society of Preventive Oncology, 18, 516–525. Deepayan, S. (2008). Lattice: Multivariate data visualization with R. New York: Springer. Destaillats, F., Joffre, C., Acar, N., Joffre, F., Bezelgues, J. B., Pasquis, B., et al. (2010). Differential effect of maternal diet supplementation with alpha-Linolenic acid or n-3 long-chain polyunsaturated fatty acids on glial cell phosphatidylethanolamine and phosphatidylserine fatty acid profile in neonate rat brains. Nutrition & Metabolism, 7, 2. Dong, H. W., Swanson, L. W., Chen, L., Fanselow, M. S., & Toga, A. W. (2009). Genomic–anatomic evidence for distinct functional domains in hippocampal field CA1. Proceedings of the National Academy of Sciences of the United States of America, 106, 11794–11799. Fanselow, M. S., & Dong, H. W. (2010). Are the dorsal and ventral hippocampus functionally distinct structures? Neuron, 65, 7–19. Federova, I., Nahed, H., Baumann, M. H., DiMartino, C., & Norman, S. J. (2009). An n-3 fatty acid deficiency impairs rat spatial learning in the Barnes maze. Behavioral Neuroscience, 123, 196–205. Fedorova, I., & Salem, N. Jr., (2006). Omega-3 fatty acids and rodent behavior. Prostaglandins, Leukotrienes, and Essential Fatty Acids, 75, 271–289. Folch, J., Lees, M., & Sloane Stanley, G. H. (1957). A simple method for the isolation and purification of total lipides from animal tissues. The Journal of Biological Chemistry, 226, 497–509. Gamoh, S., Hashimoto, M., Sugioka, K., Shahdat Hossain, M., Hata, N., Misawa, Y., et al. (1999). Chronic administration of docosahexaenoic acid improves

67

reference memory-related learning ability in young rats. Neuroscience, 93, 237–241. Garcia-Calatayud, S., Redondo, C., Martin, E., Ruiz, J. I., Garcia-Fuentes, M., & Sanjurjo, P. (2005). Brain docosahexaenoic acid status and learning in young rats submitted to dietary long-chain polyunsaturated fatty acid deficiency and supplementation limited to lactation. Pediatric Research, 57, 719–723. Garthe, A., Behr, J., & Kempermann, G. (2009). Adult-generated hippocampal neurons allow the flexible use of spatially precise learning strategies. PLoS One, 4, e5464. Green, P., & Yavin, E. (1993). Elongation, desaturation, and esterification of essential fatty acids by fetal rat brain in vivo. Journal of Lipid Research, 34, 2099– 2107. Green, P., & Yavin, E. (1995). Modulation of fetal rat brain and liver phospholipid content by intraamniotic ethyl docosahexaenoate administration. Journal of Neurochemistry, 65, 2555–2560. Green, P., & Yavin, E. (1998). Mechanisms of docosahexaenoic acid accretion in the fetal brain. Journal of Neuroscience Research, 52, 129–136. Harauma, A., Salem, N., Jr., & Moriguchi, T. (2010). Repletion of n-3 fatty acid deficient dams with alpha-linolenic acid: Effects on fetal brain and liver fatty acid composition. Lipids, 45, 659–668. He, C., Qu, X., Cui, L., Wang, J., & Kang, J. X. (2009). Improved spatial learning performance of fat-1 mice is associated with enhanced neurogenesis and neuritogenesis by docosahexaenoic acid. Proceedings of the National Academy of Sciences of the United States of America, 106, 11370–11375. Innis, S. M. (2007). Dietary (n-3) fatty acids and brain development. The Journal of Nutrition, 137, 855–859. Irons, R., & Fritsche, K. L. (2005). Omega-3 polyunsaturated fatty acids impair in vivo interferon-gamma responsiveness via diminished receptor signaling. The Journal of Infectious Diseases, 191, 481–486. Jasarevic, E., Sieli, P. T., Twellman, E. E., Welsh, T. H., Jr., Schachtman, T. R., Roberts, R. M., et al. (2011). Disruption of adult expression of sexually selected traits by developmental exposure to bisphenol A. Proceedings of the National Academy of Sciences of the United States of America, 108, 11715–11720. Jasarevic, E., Williams, S. A., Roberts, R. M., Geary, D. C., & Rosenfeld, C. S. (2012). Spatial navigation strategies in Peromyscus: A comparative study. Animal Behaviour, 84, 1141–1149. Jasarevic, E., Williams, S. A., Vandas, G. M., Ellersieck, M. R., Liao, C., Kannan, K., et al. (2013). Sex and dose-dependent effects of developmental exposure to bisphenol A on anxiety and spatial learning in deer mice (Peromyscus maniculatus bairdii) offspring. Hormones and Behavior, 63, 180–189. Katakura, M., Hashimoto, M., Shahdat, H. M., Gamoh, S., Okui, T., Matsuzaki, K., et al. (2009). Docosahexaenoic acid promotes neuronal differentiation by regulating basic helix-loop-helix transcription factors and cell cycle in neural stem cells. Neuroscience, 160, 651–660. Kawakita, E., Hashimoto, M., & Shido, O. (2006). Docosahexaenoic acid promotes neurogenesis in vitro and in vivo. Neuroscience, 139, 991–997. Kjelstrup, K. G., Tuvnes, F. A., Steffenach, H. A., Murison, R., Moser, E. I., & Moser, M. B. (2002). Reduced fear expression after lesions of the ventral hippocampus. Proceedings of the National Academy of Sciences of the United States of America, 99, 10825–10830. Klur, S., Muller, C., Pereira de Vasconcelos, A., Ballard, T., Lopez, J., Galani, R., et al. (2009). Hippocampal-dependent spatial memory functions might be lateralized in rats: An approach combining gene expression profiling and reversible inactivation. Hippocampus, 19, 800–816. Lindqvist, A., Mohapel, P., Bouter, B., Frielingsdorf, H., Pizzo, D., Brundin, P., et al. (2006). High-fat diet impairs hippocampal neurogenesis in male rats. European Journal of Neurology: The Official Journal of the European Federation of Neurological Societies, 13, 1385–1388. McNamara, R. K., Hahn, C. G., Jandacek, R., Rider, T., Tso, P., Stanford, K. E., et al. (2007). Selective deficits in the omega-3 fatty acid docosahexaenoic acid in the postmortem orbitofrontal cortex of patients with major depressive disorder. Biological Psychiatry, 62, 17–24. Mitchell, D. C., Niu, S. L., & Litman, B. J. (2003). Enhancement of G protein-coupled signaling by DHA phospholipids. Lipids, 38, 437–443. Moriguchi, T., Greiner, R. S., & Salem, N. Jr., (2000). Behavioral deficits associated with dietary induction of decreased brain docosahexaenoic acid concentration. Journal of Neurochemistry, 75, 2563–2573. Moser, M. B., Moser, E. I., Forrest, E., Andersen, P., & Morris, R. G. (1995). Spatial learning with a minislab in the dorsal hippocampus. Proceedings of the National Academy of Sciences of the United States of America, 92, 9697–9701. Mozurkewich, E., Chilimigras, J., Klemens, C., Keeton, K., Allbaugh, L., Hamilton, S., et al. (2011). The mothers, Omega-3 and mental health study. BMC Pregnancy and Childbirth, 11, 46. Mozurkewich, E. L., Clinton, C. M., Chilimigras, J. L., Hamilton, S. E., Allbaugh, L. J., Berman, D. R., et al. (2013). The Mothers, Omega-3, and Mental Health Study: A double-blind, randomized controlled trial. American Journal of Obstetrics and Gynecology, 208(313), e1–e9. Mulder, K. A., King, D. J., & Innis, S. M. (2014). Omega-3 fatty acid deficiency in infants before birth identified using a randomized trial of maternal DHA supplementation in pregnancy. PLoS One, 9, e83764. Ozias, M. K., Carlson, S. E., & Levant, B. (2007). Maternal parity and diet (n-3) polyunsaturated fatty acid concentration influence accretion of brain phospholipid docosahexaenoic acid in developing rats. The Journal of Nutrition, 137, 125–129. Pinheiro, J., Bates, D., DebRoy, S., & Sarkar, D. (2013). nlme: Linear and nonlinear mixed effects models. R package version 3.1-113.

68

E. Jašarevic´ et al. / Neurobiology of Learning and Memory 116 (2014) 59–68

Rioux, F. M., Belanger-Plourde, J., Leblanc, C. P., & Vigneau, F. (2011). Relationship between maternal DHA and iron status and infants’ cognitive performance. Canadian Journal of Dietetic Practice and Research: A Publication of Dietitians of Canada = Revue canadienne de la pratique et de la recherche en dietetique: une publication des Dietetistes du Canada, 72, 76. Ross, B. M., Seguin, J., & Sieswerda, L. E. (2007). Omega-3 fatty acids as treatments for mental illness: Which disorder and which fatty acid? Lipids in Health and Disease, 6, 21. Sable, P. S., Kale, A. A., & Joshi, S. R. (2013). Prenatal omega 3 fatty acid supplementation to a micronutrient imbalanced diet protects brain neurotrophins in both the cortex and hippocampus in the adult rat offspring. Metabolism: Clinical and Experimental, 62, 1607–1622. Salem, N., Jr., Moriguchi, T., Greiner, R. S., McBride, K., Ahmad, A., Catalan, J. N., et al. (2001). Alterations in brain function after loss of docosahexaenoate due to dietary restriction of n-3 fatty acids. Journal of Molecular Neuroscience: MN, 16, 299–307 (discussion 17-21). Shuai, Y., Lu, B., Hu, Y., Wang, L., Sun, K., & Zhong, Y. (2010). Forgetting is regulated through Rac activity in Drosophila. Cell, 140, 579–589. Simopoulos, A. P. (2002). The importance of the ratio of omega-6/omega-3 essential fatty acids. Biomedicine & Pharmacotherapy = Biomedecine & Pharmacotherapie, 56, 365–379. Su, H. M. (2010). Mechanisms of n-3 fatty acid-mediated development and maintenance of learning memory performance. The Journal of Nutritional Biochemistry, 21, 364–373. Su, K. P., Huang, S. Y., Chiu, T. H., Huang, K. C., Huang, C. L., Chang, H. C., et al. (2008). Omega-3 fatty acids for major depressive disorder during pregnancy: Results from a randomized, double-blind, placebo-controlled trial. The Journal of Clinical Psychiatry, 69, 644–651. Sun, Q., Ma, J., Campos, H., Hankinson, S. E., & Hu, F. B. (2007). Comparison between plasma and erythrocyte fatty acid content as biomarkers of fatty acid intake in US women. The American Journal of Clinical Nutrition, 86, 74–81.

Team, R. C. (2014). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Thompson, R., Kao, L., & Yang, S. (1981). Rapid forgetting of individual spatial reversal problems in rats with parafascicular lesions. Behavioral and Neural Biology, 33, 1–16. Uauy, R., & Dangour, A. D. (2006). Nutrition in brain development and aging: Role of essential fatty acids. Nutrition Reviews, 64, S24–S33 (discussion S72-91). Vaz, J. S., Kac, G., Nardi, A. E., & Hibbeln, J. R. (2014). Omega-6 fatty acids and greater likelihood of suicide risk and major depression in early pregnancy. Journal of Affective Disorders, 152–154, 76–82. Williams, J. H., Errington, M. L., Lynch, M. A., & Bliss, T. V. (1989). Arachidonic acid induces a long-term activity-dependent enhancement of synaptic transmission in the hippocampus. Nature, 341, 739–742. Wiskott, L., Rasch, M. J., & Kempermann, G. (2006). A functional hypothesis for adult hippocampal neurogenesis: Avoidance of catastrophic interference in the dentate gyrus. Hippocampus, 16, 329–343. Xiao, Y., Huang, Y., & Chen, Z. Y. (2005). Distribution, depletion and recovery of docosahexaenoic acid are region-specific in rat brain. The British Journal of Nutrition, 94, 544–550. Yavin, E. (2006). Versatile roles of docosahexaenoic acid in the prenatal brain: From pro- and anti-oxidant features to regulation of gene expression. Prostaglandins, Leukotrienes, and Essential Fatty Acids, 75, 203–211. Yavin, E., Glozman, S., & Green, P. (2001). Docosahexaenoic acid accumulation in the prenatal brain: Prooxidant and antioxidant features. Journal of Molecular Neuroscience: MN, 16, 229–235 (discussion 79-84). Zimmer, L., Vancassel, S., Cantagrel, S., Breton, P., Delamanche, S., Guilloteau, D., et al. (2002). The dopamine mesocorticolimbic pathway is affected by deficiency in n-3 polyunsaturated fatty acids. The American Journal of Clinical Nutrition, 75, 662–667.