Prenatal, but not early postnatal, exposure to a Western diet improves ...

The FASEB Journal article fj.201500208R. Published online March 16, 2016. THE

JOURNAL

• RESEARCH •

www.fasebj.org

Prenatal, but not early postnatal, exposure to a Western diet improves spatial memory of pigs later in life and is paired with changes in maternal prepartum blood lipid levels Caroline Clouard,*,1 Bas Kemp,* David Val-Laillet,† Walter J. J. Gerrits,‡ Andrea C. Bartels,* and J. Elizabeth Bolhuis*

*Department of Animal Sciences, Adaptation Physiology Group, and ‡Department of Animal Sciences, Animal Nutrition Group, Wageningen University, Wageningen, The Netherlands; and †Unité de Recherche 1341 Alimentation et Adaptations Digestives, Nerveuses et Comportementales, Institut National de la Recherche Agronomique, Saint Gilles, France

Maternal obesity and perinatal high-fat diets are known to affect cognitive development. We examined the effects of late prenatal and/or early postnatal exposure to a Western-type diet, high in both fat and refined sugar, on the cognition of pigs (Sus scrofa) in the absence of obesity. Thirty-six sows and their offspring were assigned to 1 of 4 treatments in a 2 3 2 factorial arrangement, with 8 wk prenatal and 8 wk postnatal exposure to a Western diet (enriched in fat, sucrose, and cholesterol) or control diets as factors. Compared to controls, piglets exposed to the prenatal Western diet showed enhanced working and reference memory during the acquisition and reversal phases of a spatial hole-board task. Mothers fed the prenatal Western diet had higher prepartum blood cholesterol and free fatty acid levels. Postnatal exposure to the Western diet did not affect piglet cognitive performance, but it did increase postpartum maternal and postweaning piglet cholesterol levels. The Western diet had no effect on maternal or offspring insulin sensitivity or leptin levels. In conclusion, a prenatal Western diet improved memory function in pigs, which was paired with changes in prepartum maternal blood cholesterol levels. These findings highlight the key role of late fetal nutrition for long-term programming of cognition.—Clouard, C., Kemp, B., Val-Laillet, D., Gerrits, W. J. J., Bartels, A. C., Bolhuis, J. E. Prenatal, but not early postnatal, exposure to a Western diet improves spatial memory of pigs later in life and is paired with changes in maternal prepartum blood lipid levels. FASEB J. 30, 000–000 (2016). www.fasebj.org

ABSTRACT:

KEY WORDS:

cholesterol

•

cognition

•

fat

•

prenatal programming

In recent decades, modern human societies have been subjected to a fundamental and unprecedented rise in the availability of highly palatable and low-cost foods that are high in saturated fat, refined sugars, and cholesterol. The widespread availability of these Western-type foods has led to a drastic increase in the exposure to dietary fat, sugars, and cholesterol during the fetal and early-life period, which is a highly sensitive developmental window. It is now widely accepted that perinatal nutrition can influence offspring development and may result in long-term ABBREVIATIONS: CC, control–control diet; CV, coefficient of variability;

CW, control–Western diet; HDL-C, high-density lipoprotein cholesterol; NEFA, nonesterified fatty acid; QUICKI, quantitative insulin sensitivity check index; RM, reference memory; TC, total cholesterol; TG, triglyceride; WC, Western–control diet; WM, working memory; WW, Western–Western diet 1

Correspondence: Adaptation Physiology Group, Wageningen University, P.O. Box 338, 6700 AH Wageningen, The Netherlands. E-mail: [email protected]

doi: 10.1096/fj.201500208R This article includes supplemental data. Please visit http://www.fasebj.org to obtain this information.

0892-6638/16/0030-0001 © FASEB

•

refined sugar

adaptations of metabolic, physiologic, behavioral, and brain functions in later life, a phenomenon known as early life programming (1, 2). Accordingly, exposure to high levels of dietary fat or refined sugars during the perinatal period in humans and rodents has been found to predispose the offspring to metabolic diseases (3–5). Furthermore, rodent studies indicate that fetal and/or early-life exposure to high-fat diets, often associated with maternal obesity, may also alter brain integrity and cognitive functions in the offspring (6–10). While the effects of perinatal dietary fat on cognition have been studied extensively, data about the impact of a perinatal Western diet, which is high in both saturated fat and refined sugars, are scarce and equivocal (11, 12). Yet in our modern Western societies, excessive consumption of fat is frequently associated with the consumption of excessive amounts of sugars (13). Additionally, it remains unknown which period, late prenatal and/or early postnatal, is the most susceptible to dietary influences for cognitive development and function in later life. Further research in a pertinent animal model is thus needed to 1

determine whether exposure to a Western diet before or after birth can affect cognitive function in later life, even in the absence of severe metabolic alterations or excessive weight gain in mothers and/or offspring. The pig (Sus scrofa), like the human, has a perinatal brain growth spurt that extends from late prenatal to early postnatal life, unlike the brain growth spurt of rats, which is largely postnatal (14, 15). The porcine brain also shows strong anatomic and functional similarities to that of humans; both species have a gyrencephalic brain (16, 17), and the fatty acid composition of the piglet brain is similar to that of human infants (18). Finally, both pigs and humans are monogastric omnivores, with comparable gastrointestinal tract anatomy, physiology, and development, and with relatively similar nutrient requirements (for reviews, see 19, 20). These similarities with the human, as well as its high cognitive abilities during behavioral tasks (17, 21), make the pig a relevant animal model to examine the relationship between perinatal nutrition and cognitive development. Therefore, we examined the effects of 8 wk prenatal and/or 8 wk postnatal exposure to a Western diet enriched in saturated fat, sucrose, and cholesterol on the cognitive performance of piglets in later life. We hypothesized that both prenatal and postnatal exposure to the Western diet would impair spatial learning, working memory (WM), and reference memory (RM) of piglets tested 3 to 8 wk after the end of the dietary intervention. We assessed cognitive performance data in relation to blood parameters reflecting lipid and glucose metabolism. MATERIALS AND METHODS The Animal Care and Use Committee of Wageningen University approved this experiment. Animals and housing A total of 36 multiparous sows (Topigs 20) and their litters (Tempo 3 Topigs 20) from the Swine Innovation Centre (Wageningen University and Research Centre, Sterksel, The Netherlands) were used. The experiment was carried out in 2 successive replicates, with 18 sows per replicate. During gestation, the sows were housed in groups of 4 or 5 per pen (12 m2) and were moved to individual farrowing pens (240 cm 3 180 cm) in 2 rooms 1 wk before the expected farrowing date, 115 d after insemination. A jute bag was available in the pen until after parturition to be used as nesting material. Rooms had a natural light–dark cycle, and temperature was maintained at a minimum of 23°C. After birth, piglets remained in the sow’s farrowing pen with littermates until weaning. Within 3 d after birth, if needed, piglets were cross-fostered within dietary treatment groups to balance litter sizes. At weaning (27.8 6 0.26 d after parturition, i.e., about 4 wk of age), 3 female piglets (1.47 6 0.04 kg) with a birth weight the closest to the average birth weight of the females of the litter were selected per sow. Three sows were excluded from the experiment for health reasons. The piglets were selected from 32 of the remaining 33 litters. At weaning, the 96 selected female piglets (8.42 6 0.13 kg) were transported (i.e., about 100 km) to the experimental farm (CARUS, Wageningen University and Research Centre, The Netherlands) and housed in groups of 3 littermates per pen (280 3 180 cm) in 2 identical and adjacent rooms. Pens were enriched with wood shavings and equipped with a chain 2

Vol. 30

July 2016

with screws attached to it as a toy. Starting from the end of the dietary intervention (i.e., 8 wk of age), straw was added daily to the pens. Room temperature was maintained at a minimum of 25°C for the first 2 postweaning weeks, then was gradually decreased to 20°C. Lights were on from 7 AM to 7 PM. Distribution of piglets over and within the 2 rooms was balanced for treatment. Dietary treatments The sows (n = 9 sows per treatment) and their piglets (n = 8 pens of 3 littermates selected from 8 sows per treatment) were allocated to 1 of 4 dietary treatments—Western–Western (WW), control–Western (CW), Western–control (WC), and control– control (CC)—in a 2 3 2 factorial design with 8 wk prenatal and 8 wk postnatal exposure to the Western diet or control diets as factors. Four experimental pelleted diets were formulated, along with 3 control diets: for the sows, a standard gestation diet and a standard lactation diet; for the piglets, a standard starter diet; and for both sows and piglets, an all-in-one Western diet (18.5% saturated fat, 12% sucrose, and 1% cholesterol). Ingredient and nutrient compositions of the 4 diets are listed in Supplemental Table 1. The gestating sows and piglets were fed standard commercial pig diets before the start of the experiment (i.e., before d 59 of gestation) and after the end of the dietary intervention (i.e., 4 wk after weaning), respectively. The control standard diets were formulated to meet or exceed the recommendations for the animals’ requirements by the Centraal Veevoeder Bureau (Central Feedstuff Bureau, Lelystad, The Netherlands) (22). During the last 8 wk of gestation, half of the sows were fed the Western diet and the other half were fed the control gestation diet (i.e., the prenatal diet). Starting the day of parturition, half of the sows of each group were fed the Western diet and the other half were fed the control lactation diet during the 4 wk of lactation. Starting at weaning, piglets remained on the same dietary treatment (Western diet or control starter diet) as their mothers for 4 extra weeks (i.e., postnatal diet). Hence, 4 groups of piglets were used: WW, WC, CW, and CC. Allocation to the treatments was balanced for parity, and treatments were evenly distributed between replicates, as well as within and between the rooms during gestation and lactation, and after weaning. During gestation and lactation, the sows were fed twice a day according to the normal net energy recommendations for pregnant and lactating sows. Because the organoleptic properties of the Western diet greatly differed from that of the gestation and lactation diets, 4-d feed transitions (i.e., 25, 50, 75, and 100% of the Western diet in the total feed ration) were systematically done to prevent neophobic response of the sows when first exposed to the Western diet, and vice versa. Additionally, 16 d before weaning, piglets also had ad libitum access to the Western or starter diet as a creep feed to facilitate the transition to solid feed after weaning. From weaning until the end of the experiment, piglets were allowed ad libitum access to the experiment diets. Sows and piglets had ad libitum access to water throughout the experiment. Spatial learning and memory task From 3 to 8 wk after the end of the dietary intervention (i.e., about 11 to 15 wk of age), 2 piglets (with weight closest to the average weight of the 3 pen mates) per pen were individually subjected to a spatial learning and memory test known as the hole-board task (23). This test allows for the assessment of learning, WM, and RM, as well as for the assessment of cognitive flexibility in responding to a new spatial configuration, which is sometimes referred to as reversal (23). The test arena (5.3 3 5.3 m) had wooden black walls (80 cm high) and contained 4 entrances that could be opened with

The FASEB Journal x www.fasebj.org

CLOUARD ET AL.

Figure 1. A–D) Map of test arena and configurations of baited buckets in hole-board task.

guillotine doors from the surrounding corridor that also served as a waiting area for the piglets (Fig. 1). The corridor contained a jute bag and some toys to keep the waiting piglets busy during trials. Sixteen gray metallic buckets 19 cm in diameter were attached to the floor of the arena in a 4 3 4 matrix with a mutual distance of 1.1 m. During the test, 4 of the buckets were baited with a chocolate raisin, and 4 different configurations of baited buckets were used (Fig. 1). These configurations were obtained by using a single configuration in its 0, 90, 180, and 270° orientations. Each piglet was individually tested on a fixed configuration of baited buckets. The baited configurations differed between the 2 piglets within each pen, and were balanced between treatment groups and rooms. Four different starting entrances were used for the piglets to prevent them from developing a fixed pattern of visits that would reduce their WM load (21, 24). Two entrances were used per day of test with 1 entrance per trial. To prevent the use of odor cues to locate the baited buckets, all buckets also contained a perforated darkbrown bottom under which fresh chocolate raisins were placed before test trials. Habituation During the 2 wk before testing (i.e., 9 working days), the piglets were habituated to the experimenters, the buckets, and the testing arena. On the first day of habituation, the piglets were habituated to the experimenter, the bucket, and chocolate raisins and apples in their pens. During the second and third days, the PRENATAL WESTERN DIET IMPROVES COGNITION

piglets were allowed to explore with pen mates the waiting area and testing arena with 16 baited buckets. During the fourth and fifth days, the piglets were allowed to explore the waiting area with the pen mates but entered the test arena individually while all the 16 buckets were baited. Finally, during the 4 last days of habituation, the piglets were individually habituated to the testing area with 8 baited buckets, with different configurations across days and trials. Acquisition trials After the habituation period, the piglets were individually exposed to 2 massed trials per day on 14 consecutive workdays, starting on a Wednesday and excluding weekends, for 28 total acquisition trials. Before the start of a trial, all 3 pen mates were led into the waiting area. The trial started when the piglet had 4 legs in the testing arena and was terminated when the piglet found all 4 rewards or after 180 s. Every time the piglet visited a baited bucket for the first time, a clicker sound was produced. If the piglet completed the task (i.e., found the 4 rewards in less than 180 s), the experimenter opened the exit guillotine door and orally congratulated the piglet in cheerful high-pitched voice (“good job!” and “well done!”), and the piglet received a piece of apple. If the piglet did not complete the task within the 180 s, a sirenlike sound was produced to indicate the end of the trial to the pig, which could thereafter rejoin its pen mates; the experimenter silently opened the door, and the piglet received no piece of apple. After each trial, the piglet was led back into the waiting area 3

with its pen mates, and the other piglet of the pen was tested. Then the same 2 piglets were tested a second time (i.e., massed trials), and the 3 pen mates were brought back into their home pen. After each trial, feces were removed from the testing and waiting areas. The order of testing between and within groups of pen mates was alternated across days and was balanced over treatments.

analyzed using a porcine insulin radioimmunoassay kit with an interassay CV of 3 and an intraassay CV of 10% (EMD Millipore, Billerica, MA, USA). Plasma leptin concentrations (mU/ml) were analyzed using a Multi-Species Leptin radioimmunoassay kit with an interassay and intraassay CV of 10% (EMD Millipore). Plasma glucose concentrations (mg/dl) were analyzed using enzymatic colorimetric assay with an interassay CV of 2 and an intraassay CV of 1% [glucose GOD-PAP method using Cobas of Roche (Basel, Switzerland) or Hitachi (Yokohama, Japan) analyzers].

Reversal trials After the acquisition phase, piglets were exposed to a reversal phase, with 16 trials during 8 d following the same procedure as for the acquisition phase. In the reversal phase, piglets were assigned to a different configuration of baited buckets to assess their cognitive flexibility (A→C, B→D, C→A, and D→B; Fig. 1). Measurements The following parameters were scored live: latency to first bucket visit, total number of bucket visits, total number of collected rewards, time needed to finish the trial (i.e., latency to fourth reward or the maximum time of 180 s), WM errors (i.e., all revisits to baited buckets), RM errors (i.e., all (re)visits to nonbaited buckets). After the trials, the general WM score (later referred as WM score), which reflects the ability of the animals to avoid revisits to buckets already visited during a trial (i.e., short-term memory), was calculated as the ratio between the number of different buckets visited and the total number of visits. The RM score, which reflects the ability of the animals to discriminate between baited and unbaited buckets (i.e., long-term memory), was calculated as the ratio between the number of visits to the baited set of buckets and the number of visits to all buckets. The intervisit interval (i.e., average time between 2 successive bucket visits) was also calculated as follows: (time to finish the trial 2 latency to first bucket visit)/(number of bucket visits 2 1) (23).

Blood sampling and analyses Blood was collected from sows before the start of the dietary treatment (i.e., d 45 of gestation), before entering the farrowing pens (i.e., d 108 of gestation, about 1 wk before parturition), and at the end of lactation (i.e., about 4 wk after parturition, weaning day). Blood was collected from all the piglets at the end of the dietary intervention (i.e., about 8 wk of age) and 8 wk after (i.e., after the hole-board test, about 16 wk of age). After an overnight fasting, blood samples were collected from the jugular vein of the sows and the experimental piglets in 2 BD Vacutainer tubes (Becton Dickinson, San Diego, CA, USA), one with EDTA and the other with heparin, and stored on ice immediately after sampling. During blood sampling, 8-wk-old piglets were manually restrained by 2 experimenters in a supine position, while 16-wk-old piglets and sows were fixed using a snout string. The tubes were centrifuged for 10 min at 1300 g at 4°C, and plasma was stored at 280°C until further analyses. Tubes containing EDTA were used for the analysis of insulin, leptin, nonesterified fatty acids (NEFA), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and triglycerides (TG). Tubes containing heparin were used for the analysis of glucose. Plasma TG, TC, and HDL-C concentrations (mM) were analyzed in the Cooperating Aid Agencies [(SHO) Laboratory, SHO Centra voor Medische Diagnostiek, Velp, The Netherlands]. Plasma NEFA concentrations (mM) were analyzed using the enzymatic method with acyl-CoA oxidase and the Wako NEFAHR (2) reagent kit with an interassay coefficient of variability (CV) of 2 and an intraassay CV of 4% (Wako Pure Chemicals, Tokyo, Japan). Plasma insulin concentrations (mU/ml) were 4

Vol. 30

July 2016

Statistical analyses Statistical analyses were conducted by SAS 9.1.3 (SAS Institute, Cary, NC, USA). Three sows were excluded from the experiment for health reasons, resulting in 8 sows for CC treatment, 7 sows for WC treatment, and 9 sows for CW and WW treatments. For all the analyses on sow performance and physiologic data, sow nested within prenatal diet, postnatal diet, and batch was the experimental unit. Two pens of piglets were excluded from the experiment because of a methodologic error in replicate 2, resulting in 8 pens for CW and WC treatments, and 7 pens for WW and CC treatments. For all the analyses on piglet data, pen nested within prenatal diet, postnatal diet, and batch was the experimental unit. If the model residuals were not normally distributed, analyses were performed on (arc sine) square root or logarithmically transformed data. In the spatial task, block mean values of 4 consecutive trials were calculated (25), resulting in 7 and 4 blocks for acquisition and reversal phases, respectively. Data were analyzed using a repeated mixed model with observations on the same pen in time taken as repeated measures. Prenatal diet, postnatal diet, block, their interactions, and batch were included as fixed effects. A first-order autoregressive covariance structure (ar(1) gave the best fit and was used to account for within-pen variation. The effects of the prenatal diet, postnatal diet, and their interaction on the cognitive flexibility of piglets during the first day of reversal testing was assessed using a mixed model with prenatal diet, postnatal diet, their interactions, and batch as fixed effects. Fasting levels of blood metabolites of sows and piglets were analyzed using a repeated mixed model with prenatal diet, postnatal diet, time, their interactions, and batch as fixed effects. A QUICKI (quantitative insulin sensitivity check index), a measure of insulin sensitivity used in clinical studies (26), was calculated for each piglet and sow, as 1/[log (fasting insulin, mU/ml) + log(fasting glucose, mg/dl)], and was analyzed with the same model. Data on sows’ glucose and piglets’ leptin concentrations both contained one outlier, which was characterized using the Dixon test (27) and omitted from analysis. Significant interaction effects were further analyzed using the difference of the least square means with Tukey’s adjustment. Data are presented as (untransformed) means 6 SEM.

RESULTS Spatial task Acquisition trials During the hole-board acquisition trials, all variables were affected by trial blocks, with increased memory scores and fewer memory errors over blocks, indicating that piglets learned the task and improved their performance over time (Supplemental Fig. 1A). Performance of piglets during the acquisition trials of the hole-board task is presented in Fig. 2. Piglets exposed to the prenatal Western diet had higher WM scores [F(1,25) = 4.31, P = 0.05; Fig. 2A] and made fewer WM errors


CLOUARD ET AL.

Figure 2. Performance of piglets in hole-board acquisition trials. (A) RM and (B) WM scores, (C ) RM and (D) WM errors, (E ) latency to fourth reward, (F ) intervisit interval, (G) total number of visits, and (H ) number of rewards consumed. Piglets were subjected to 8 wk prenatal and 8 wk postnatal exposure to Western (W) and/or control (C) diet. Data are expressed as means 6 † SEM and were analyzed with mixed model: 0.05 , P # 0.10; *P # 0.05.

[F(1,25) = 5.62, P = 0.03; Fig. 2C] than piglets exposed to the prenatal control diet. Piglets exposed to the prenatal Western diet also made fewer RM errors [F(1,25) = 4.80, P = 0.04; Fig. 2D], although RM scores were not significantly affected by the prenatal diet [F(1,25) = 2.32, P = 0.14; Fig. 2B]. The latency to first bucket visit (Western: 2.0 6 0.33 s, control: 1.6 6 0.13 s, F(1,25) = 1.11, P = 0.30) and to fourth reward [F(1,25) = 0.45, P = 0.51; Fig. 2E] were not influenced by the prenatal diet, but the intervisit interval tended to be longer for piglets exposed to the prenatal Western diet than for piglets exposed to the control diet [F(1,25) = 3.60, P = 0.07; Fig. 2F]. Piglets exposed to the prenatal Western diet made fewer visits in total [F(1,25) = 5.96, P = 0.02; Fig. 2G], and they consumed fewer rewards during the first trial block [Western: 3.2 6 0.16, control: 3.6 6 0.10; prenatal diet 3 trial block interaction: F(6,156) = 2.43, P = 0.03; Fig. 2H] than piglets exposed to the prenatal control diet. No effects of the postnatal diet or the postnatal diet 3 block interaction were found on any of the variables (P . 0.10 for all). Reversal trials The performance of the piglets during the first day of reversal testing was not affected by prenatal diet, postnatal diet, or their interactions (data not shown, P . 0.10 for all). During the hole-board reversal trials, all variables were affected by trial blocks, with increased memory scores and fewer memory errors over trial blocks, indicating that piglets learned the new configuration and improved their performance over time (Supplemental Fig. 1B). PRENATAL WESTERN DIET IMPROVES COGNITION

Performance of piglets during the reversal trials of the hole-board task are presented in Fig. 3. Piglets exposed to the prenatal Western diet tended to have higher WM [F(1,25) = 3.29, P = 0.08; Fig. 3A] and RM scores [F(1,25) = 3.74, P = 0.06; Fig. 3B] than piglets exposed to the prenatal control diet. They also made fewer RM errors [F(1,25) = 4.66, P = 0.04; Fig. 3D]. Compared to the prenatal control diet, the prenatal Western diet increased the intervisit interval duration [F(1,25) = 4.23, P = 0.05; Fig. 3F] and reduced the total number of visits [F(1,25) = 4.62, P = 0.04; Fig. 3G]. Piglets exposed to the prenatal Western diet also tended to have a longer latency to visit the first bucket than piglets exposed to the prenatal control diet [Western: 2.0 6 0.18 s, control: 1.5 6 0.07 s, F(1,25) = 3.86, P = 0.06]. WM errors (Fig. 3C), the latency to fourth reward (Fig. 3E), and the number of rewards consumed (Fig. 3H) were not influenced by the prenatal diet (P . 0.10 for all). No effects of the postnatal diet or the postnatal diet 3 block interactions were found on any of the variables (P . 0.10 for all). Fasting blood levels of glucose, insulin, and leptin Fasting blood concentrations of glucose, insulin, and leptin of sows and piglets are presented in Fig. 4. Sows Fasting plasma glucose concentrations tended to be lower in sows fed the prenatal Western diet than in sows fed the 5

Figure 3. Performance of piglets in hole-board reversal trials. (A) RM and (B) WM scores, (C ) RM and (D) WM errors, (E ) latency to fourth reward, (F ) intervisit interval, (G) total number of visits, and (H ) number of rewards consumed. Piglets were subjected to 8 wk prenatal and 8 wk postnatal exposure to Western (W) and/or control (C) diet. Data are expressed as means 6 SEM and were analyzed with mixed model: †0.05 , P # 0.10; *P # 0.05.

control gestation diet [F(1,31) = 3.54, P = 0.07; Fig. 4A]. Fasting plasma concentrations of leptin (Fig. 4C) and insulin (Fig. 4B) and the QUICKI of sows (CC 0.38 6 0.005, CW 0.39 6 0.01, WC 0.39 6 0.006, WW 0.39 6 0.01) were not affected by prenatal diet, postnatal diet, or their interactions P . 0.10 for all). All variables were influenced by time (P , 0.001 for all; Fig. 4A–C). Piglets Fasting plasma glucose concentrations at the end of the dietary intervention were higher in piglets fed the prenatal Western diet than in piglets fed the prenatal control diet (time 3 prenatal diet interaction, [F(1,26) = 4.86, P = 0.04; Fig. 4D]. Piglets fed the postnatal Western diet had reduced fasting plasma insulin concentrations [F(1,26) = 4.12, P = 0.05; Fig. 4E]. Fasting plasma leptin concentrations (Fig. 4F) and the QUICKI of piglets (CC 0.45 6 0.02, CW 0.45 6 0.01, WC 0.44 6 0.008, WW 0.44 6 0.01) were not influenced by prenatal diet, postnatal diet, or their interactions (P . 0.10 for all). The QUICKI, however, was influenced by a time 3 prenatal diet interaction, with a larger decrease over time in the piglets fed the prenatal control diet [8 wk: 0.48 6 0.01, 16 wk: 0.42 6 0.01, F(1,26) = 4.31, P = 0.05]. All variables were influenced by time (P , 0.05 for all; Fig. 4D–F). Blood lipid levels Fasting blood levels of TG, TC, HDL-C, and NEFA of sows and piglets are presented in Fig. 5. No effects of prenatal 6

Vol. 30

July 2016

diet 3 postnatal diet interactions were found on any variables (P . 0.10 for all). Sows Fasting plasma TC (Fig. 5A), HDL-C (Fig. 5B), and NEFA (Fig. 5D) concentrations just before parturition were higher in sows exposed to the prenatal Western diet than in sows fed the control diet [prenatal diet 3 time interaction, F(2,61) = 140, P , 0.001, F(2,61) = 30.1, P , 0.001 and F(2,62) = 4.41, P = 0.02, respectively]. Fasting plasma TC and HDL-C concentrations at the end of lactation were higher in sows fed the postnatal Western diet than in sows fed the control diets [postnatal diet 3 time interaction, F(2,61) = 77.1, P , 0.001 and F(2,61) = 7.76, P = 0.001, respectively]. Fasting plasma TG concentrations were not affected by prenatal diet, postnatal diet, or their interactions (P . 0.10 for all; Fig. 5C). All variables were influenced by time (P , 0.05 for all; Fig. 5A–D). Piglets At the end of the dietary intervention, but not 8 wk later, fasting blood concentrations of TC [time 3 postnatal diet interaction, F(1,26) = 49.9, P , 0.001; Fig. 5E] and HDL-C ] time 3 postnatal diet interaction, F(1,26) = 99.8, P , 0.001; Fig. 5F] were higher, while fasting blood NEFA concentrations were lower [time 3 postnatal diet interaction, F(1,26) = 7.74, P = 0.01; Fig. 5H] in piglets exposed to the postnatal Western diet than in control piglets. Fasting plasma TG concentrations were reduced by the prenatal


CLOUARD ET AL.

Figure 4. Plasma levels of glucose (A, D), insulin (B, E), and leptin of sows (C, F) (top) and piglets (bottom). Piglets were subjected to 8 wk prenatal and 8 wk postnatal exposure to Western (W) and/or control (C) diet. Data are expressed as means 6 SEM and were analyzed with mixed model. Start treatment, d 45 of gestation; parturition, d 108 of gestation; end lactation, 4 wk after parturition (i.e., weaning). End treatment, 8 wk of age; 8 wk later, 16 wk of age.

Western diet [F(1,25) = 5.70, P = 0.03; Fig. 5G]. All variables were influenced by time (P , 0.10 for all; Fig. 5E–G). DISCUSSION Our study reveals that 8 wk fetal exposure to a Western diet enriched in fat, sucrose, and cholesterol enhanced spatial memory of piglets in later life. Piglets exposed before birth to the Western diet showed improved long-term RM and short-term WM in a spatial cognitive task. They also tended to show longer intervisit intervals and, in the reversal, a longer latency to first bucket visit than control piglets, suggesting that, if anything, they were slightly less motivated for the task (23), thus eliminating enhanced motivation as a potential explanation for their improved cognitive performance. We found no effect of the prenatal Western diet on responses of the pigs in a novelty test (Clouard et al., personal communication), so we also discarded differences in anxiety as an explanation for its impact on cognition. While some early life environmental factors, such as stress or enrichment, have been found to affect only 1 of the 2 memory constructs (14, 28), which rely on different brain structures (29), the prenatal Western diet enhanced both WM and RM, indicating broader effects on brain development. To our knowledge, our study is the first to disentangle the influence of late fetal vs. early postnatal exposure to a Western diet on cognitive abilities in the pig, an animal model that has, like the human, a perinatal brain growth spurt extending from late gestation to early postnatal life (15, 28), and to demonstrate the susceptibility of the developing brain to a Western diet in the late fetal period. As the effects of the diet were found PRENATAL WESTERN DIET IMPROVES COGNITION

even 11 wk after the prenatal dietary intervention, our study highlights major long-term programming effects of late fetal nutrition on cognitive functioning. Our findings contradict a number of rodent studies reporting impaired spatial cognition after fetal and/or early life exposure to high-fat diets (7–9). Apart from the differences in the dynamics of brain growth spurt between pigs and rodents, the type of cognitive task used may partly explain the discrepancies between studies. The holeboard is, unlike some tasks used in rodent studies (6–9), a nonaversive task that does not elicit stress or fear. In line with this, Bilbo and Tsang (10) found enhanced spatial cognitive performance, but no increase in anxiety levels, in female rat offspring exposed to a maternal high-fat diet. Discrepancies between our study and others might alternatively reside in differences in diet composition, notably with regard to sugar content. While a plethora of studies have addressed the effects of diets enriched exclusively in fat, to our knowledge, only 2 studies have examined the effects of perinatal exposure to a Western diet on cognition, with inconsistent results reported (11, 12), indicating that variation in ingredient composition is likely not the only factor underlying the effect of the diet on cognition. Intriguingly, in most prior studies in rodents reporting impaired cognitive performance in the offspring, the impact of the diet was confounded with maternal obesity (6–8, 11). In our study, neither the sows nor the piglets developed overweight, which indicates that factors related to maternal obesity (e.g., inflammation, adiposity), rather than the composition of the diet per se, may explain the adverse effects found on offspring’s cognition in previous studies, as suggested in humans (30). Rodent studies also suggested a link between insulin resistance and the 7

Figure 5. Plasma levels of TC (A, E), HDL-C (B, F), TG (C, G), and NEFA of sows (D, H) (top) and piglets (bottom). Piglets were subjected to 8 wk prenatal and/or 8 wk postnatal exposure to Western (W) and/or control (C) diet. Data are expressed as means 6 SEM and were analyzed with mixed model. Start treatment, d 45 of gestation; parturition, d 108 of gestation; end lactation, 4 wk after parturition (i.e., weaning). End treatment, 8 wk of age; 8 wk later, 16 wk of age.

cognitive impairments induced by high-fat/sugar diets (31, 32). In our study, however, exposure to the Western diet did not induce insulin resistance in sows or piglets, thus indicating that the effects of the diet on cognition were likely not related to changes in insulin sensitivity. In our study, the prenatal Western diet increased the sows’ blood cholesterol and NEFA levels just before parturition, in line with prior studies in mice dams fed a highfat diet (8). It is possible that changes in maternal lipid availability during the late fetal period mediated the effects of the Western diet on offspring’s cognition. In mammals, cholesterol has a key role in fetal (brain) development, as it is an essential component of brain cell membranes, where it contributes to synapse formation or synaptic structural plasticity (33). Although transport of lipids across the pig placenta is rather poor, limited amounts of fatty acids can cross the placental barrier (34, 35). Furthermore, although the ability of cholesterol to cross the blood–brain barrier is still debated, dietary cholesterol has been found to increase free cholesterol in the cerebrum of piglets (36, 37) and rats (33). We therefore speculate that the elevated levels of circulating cholesterol in the sows during late gestation were positively associated with increased cholesterol levels in the fetal brain in our study. Increased cholesterol levels in the fetal brain may have altered synaptic plasticity in memory-related brain structures (e.g., hippocampus (38)), thus resulting in enhanced memory function, as previously reported in young rats (33). In mammals, during the last third of gestation (i.e., during the growth spurt), the fetus requires an increased amount of energy for optimal growth and development. 8

Vol. 30

July 2016

Glucose is the main source of energy for fetal (brain) development (39), and it is the major nutrient crossing the placental barrier in mammals (40). To fulfill the energy requirement for fetal growth, transfer of glucose from the maternal blood to the fetus is enhanced at the end of gestation, which results in maternal hypoglycemia (41). At the end of gestation, mammalian mothers usually change to a catabolic state, which favors an enhanced breakdown of lipid stores to provide energy to the fetus via placental transfer of nutrients, and compensate for low glucose stores. Maternal NEFA may thus be used for the synthesis of ketone bodies, which easily cross the placenta to reach the fetus and which are used as an alternative energetic substrate for brain development (40, 41). Our study revealed that sows fed the prenatal Western diet had higher NEFA levels before parturition than control sows. The higher levels of maternal NEFA may have resulted in a larger synthesis of ketone bodies, thus enabling a better compensation for low glucose levels and a higher energy supply for fetal brain development, which may have contributed to the enhancement of the offspring’s cognitive performance. It is worth noting that cholesterol and NEFA levels were also affected by the postnatal Western diet in our study. Indeed, exposure to the Western diet after birth increased blood TC and HDL-C levels of sows at the end of lactation, which was paired with increased circulating cholesterol levels and decreased NEFA levels in piglets at the end of the dietary intervention. These metabolic alterations, however, were not paired with changes in piglet cognitive performance, suggesting that diet-induced metabolic changes


CLOUARD ET AL.

occurring after birth (i.e., after fetal brain development), do not impact offspring’s cognitive development in later life. These findings further support the important role of gestational, rather than postpartum, maternal lipid metabolism for fetal brain development and long-term cognitive function in pigs. Piglets fed the Western diet before birth had lower TG levels than control piglets at the time of testing, indicating that prenatal nutrition also programmed long-lasting changes in blood lipid levels of the offspring. Increased TG levels after the intake of fat or refined sugars have been associated with impaired spatial memory in young male rats (42) and minipigs (43). These animal studies support findings in elderly people showing negative associations between TG levels and cognitive abilities (44, 45). It is thus possible that the memory improvements found in our study were mediated by reduced blood TG levels during testing. In conclusion, we reported long-lasting enhancing effects of late fetal, but not early postnatal, exposure to a Western diet on spatial memory of piglets in later life, in the absence of maternal/offspring obesity or insulin resistance. Our data suggest that the beneficial effects of the diet on cognition may be mediated by changes in maternal blood cholesterol and NEFA levels before parturition. Our findings highlight the key role of late prenatal nutrition, and its associated effects on maternal prepartum blood lipid levels, for fetal (brain) development and long-term programming of cognitive functions in pigs. Considering the widespread availability of foods high in fat, sugars, and cholesterol in Western societies, these findings in the porcine model bring new insight for further research on the programming effects of prenatal Western diets on cognition in the human infant. This work was supported in part by a postdoctoral study grant from the Fyssen Foundation (to C.C.). The authors thank M. Ooms, J. Wijnen, M. van der Vosse, and M. Brouwers for their help with the experiment, and R. Koopmans (Adaptation Physiology group of Wageningen University) for his assistance with laboratory analyses. The authors gratefully acknowledge S. J. Koopmans [Wageningen University and Research Center (UR) Livestock Research] for his advice on the experimental setup. The authors also thank P. Peters from the experimental farm of Sterksel [Swine Innovation Centre (VIC) Wageningen UC], and R. Ernste, B. van den Top, and A. Jansen from the animal research facilities in Wageningen (CARUS) for taking care of the animals and for their technical assistance. Scientific interactions with the Institut National de la Recherche Agronomique Diet Impacts and Determinants: Interactions and Transitions Metaprogramme made it possible to share insights on the methodology and interpretation of result.

4. 5. 6.

7.

8.

9.

10. 11.

12.

13. 14. 15. 16. 17. 18.

19. 20. 21.

REFERENCES 1. Bateson, P., Gluckman, P., and Hanson, M. (2014) The biology of developmental plasticity and the predictive adaptive response hypothesis. J. Physiol. 592, 2357–2368 2. Barker, D. J. P. (1992) Fetal and Infant Origins of Adult Disease, BMJ Publishing Group, London 3. Férézou-Viala, J., Roy, A. F., Sérougne, C., Gripois, D., Parquet, M., Bailleux, V., Gertler, A., Delplanque, B., Djiane, J., Riottot, M., and Taouis, M. (2007) Long-term consequences of maternal high-fat feeding on hypothalamic leptin sensitivity and diet-induced obesity in PRENATAL WESTERN DIET IMPROVES COGNITION

22. 23.

24. 25.

the offspring. Am. J. Physiol. Regul. Integr. Comp. Physiol. 293, R1056–R1062 Tamashiro, K. L., and Moran, T. H. (2010) Perinatal environment and its influences on metabolic programming of offspring. Physiol. Behav. 100, 560–566 O’Reilly, J. R., and Reynolds, R. M. (2013) The risk of maternal obesity to the long-term health of the offspring. Clin. Endocrinol. (Oxf.) 78, 9–16 White, C. L., Pistell, P. J., Purpera, M. N., Gupta, S., Fernandez-Kim, S. O., Hise, T. L., Keller, J. N., Ingram, D. K., Morrison, C. D., and Bruce-Keller, A. J. (2009) Effects of high fat diet on Morris maze performance, oxidative stress, and inflammation in rats: contributions of maternal diet. Neurobiol. Dis. 35, 3–13 Yu, H., Bi, Y., Ma, W., He, L., Yuan, L., Feng, J., and Xiao, R. (2010) Long-term effects of high lipid and high energy diet on serum lipid, brain fatty acid composition, and memory and learning ability in mice. Int. J. Dev. Neurosci. 28, 271–276 Tozuka, Y., Kumon, M., Wada, E., Onodera, M., Mochizuki, H., and Wada, K. (2010) Maternal obesity impairs hippocampal BDNF production and spatial learning performance in young mouse offspring. Neurochem. Int. 57, 235–247 Page, K. C., Jones, E. K., and Anday, E. K. (2014) Maternal and postweaning high-fat diets disturb hippocampal gene expression, learning, and memory function. Am. J. Physiol. Regul. Integr. Comp. Physiol. 306, R527–R537 Bilbo, S. D., and Tsang, V. (2010) Enduring consequences of maternal obesity for brain inflammation and behavior of offspring. FASEB J. 24, 2104–2115 Rodriguez, J. S., Rodr´ıguez-González, G. L., Reyes-Castro, L. A., Ibáñez, C., Ram´ırez, A., Chavira, R., Larrea, F., Nathanielsz, P. W., and Zambrano, E. (2012) Maternal obesity in the rat programs male offspring exploratory, learning and motivation behavior: prevention by dietary intervention pre-gestation or in gestation. Int. J. Dev. Neurosci. 30, 75–81 Wright, T. M., King, M. V., Davey, W. G., Langley-Evans, S. C., and Voigt, J. P. (2014) Impact of cafeteria feeding during lactation in the rat on novel object discrimination in the offspring. Br. J. Nutr. 112, 1933–1937 Horgan, G. W., and Whybrow, S. (2012) Associations between fat, sugar and other macronutrient intakes in the National Diet and Nutrition Survey. Nutr. Bull. 37, 213–223 Conrad, C. D. (2010) A critical review of chronic stress effects on spatial learning and memory. Prog. Neuropsychopharmacol. Biol. Psychiatry 34, 742–755 Dobbing, J., and Sands, J. (1979) Comparative aspects of the brain growth spurt. Early Hum. Dev. 3, 79–83 Sauleau, P., Lapouble, E., Val-Laillet, D., and Malbert, C. H. (2009) The pig model in brain imaging and neurosurgery. Animal 3, 1138–1151 Lind, N. M., Moustgaard, A., Jelsing, J., Vajta, G., Cumming, P., and Hansen, A. K. (2007) The use of pigs in neuroscience: modeling brain disorders. Neurosci. Biobehav. Rev. 31, 728–751 Rooke, J. A., Shanks, M., and Edwards, S. A. (2000) Effect of offering maize, linseed or tuna oils throughout pregnancy and lactation on sow and piglet tissue composition and piglet performance. Anim. Sci. 71, 289–299 Spurlock, M. E., and Gabler, N. K. (2008) The development of porcine models of obesity and the metabolic syndrome. J. Nutr. 138, 397–402 Clouard, C., Meunier-Salaün, M. C., and Val-Laillet, D. (2012) Food preferences and aversions in human health and nutrition: how can pigs help the biomedical research? Animal 6, 118–136 Kornum, B. R., and Knudsen, G. M. (2011) Cognitive testing of pigs (Sus scrofa) in translational biobehavioral research. Neurosci. Biobehav. Rev. 35, 437–451 Convention Bureau Netherlands (2007) Chemical Composition and Nutritional Value of Feedstuffs and Feeding Standards, CVB Series No. 36, CVB, The Hague, The Netherlands Van der Staay, F. J., Gieling, E. T., Pinzón, N. E., Nordquist, R. E., and Ohl, F. (2012) The appetitively motivated “cognitive” holeboard: a family of complex spatial discrimination tasks for assessing learning and memory. Neurosci. Biobehav. Rev. 36, 379–403 Arts, J. W. M., van der Staay, F. J., and Ekkel, E. D. (2009) Working and reference memory of pigs in the spatial holeboard discrimination task. Behav. Brain Res. 205, 303–306 Antonides, A., Schoonderwoerd, A. C., Nordquist, R. E., and van der Staay, F. J. (2015) Very low birth weight piglets show 9

26.

27. 28.

29. 30. 31. 32.

33. 34. 35. 36.

10

improved cognitive performance in the spatial cognitive holeboard task. Front. Behav. Neurosci. 9, 43 Muniyappa, R., Lee, S., Chen, H., and Quon, M. J. (2008) Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage. Am. J. Physiol. Endocrinol. Metab. 294, E15–E26 Sokal, R., and Rohlf, F. J. (1995) Biometry, 3rd ed., New York, W. H. Freeman Elizabeth Bolhuis, J., Oostindjer, M., Hoeks, C. W. F., de Haas, E. N., Bartels, A. C., Ooms, M., and Kemp, B. (2013) Working and reference memory of pigs (Sus scrofa domesticus) in a holeboard spatial discrimination task: the influence of environmental enrichment. Anim. Cogn. 16, 845–850 Yoon, T., Okada, J., Jung, M. W., and Kim, J. J. (2008) Prefrontal cortex and hippocampus subserve different components of working memory in rats. Learn. Mem. 15, 97–105 Van Lieshout, R. J. (2013) Role of maternal adiposity prior to and during pregnancy in cognitive and psychiatric problems in offspring. Nutr. Rev. 71(Suppl 1), S95–S101 Winocur, G., and Greenwood, C. E. (2005) Studies of the effects of high fat diets on cognitive function in a rat model. Neurobiol. Aging 26 (Suppl 1), 46–49 Stranahan, A. M., Norman, E. D., Lee, K., Cutler, R. G., Telljohann, R. S., Egan, J. M., and Mattson, M. P. (2008) Diet-induced insulin resistance impairs hippocampal synaptic plasticity and cognition in middle-aged rats. Hippocampus 18, 1085–1088 Ya, B. L., Liu, W. Y., Ge, F., Zhang, Y. X., Zhu, B. L., and Bai, B. (2013) Dietary cholesterol alters memory and synaptic structural plasticity in young rat brain. Neurol. Sci. 34, 1355–1365 Thulin, A. J., Allee, G. L., Harmon, D. L., and Davis, D. L. (1989) Utero-placental transfer of octanoic, palmitic and linoleic acids during late gestation in gilts. J. Anim. Sci. 67, 738–745 Père, M.-C. (2003) Materno-foetal exchanges and utilisation of nutrients by the foetus: comparison between species. Reprod. Nutr. Dev. 43, 1–15 Schoknecht, P. A., Ebner, S., Pond, W. G., Zhang, S., McWhinney, V., Wong, W. W., Klein, P. D., Dudley, M., Goddard-Finegold, J.,

Vol. 30 July 2016

37.

38.

39. 40. 41.

42.

43.

44.

45.

and Mersmann, H. J. (1994) Dietary cholesterol supplementation improves growth and behavioral response of pigs selected for genetically high and low serum cholesterol. J. Nutr. 124, 305–314 Boleman, S. L., Graf, T. L., Mersmann, H. J., Su, D. R., Krook, L. P., Savell, J. W., Park, Y. W., and Pond, W. G. (1998) Pigs fed cholesterol neonatally have increased cerebrum cholesterol as young adults. J. Nutr. 128, 2498–2504 Wang, D., and Zheng, W. (2015) Dietary cholesterol concentration affects synaptic plasticity and dendrite spine morphology of rabbit hippocampal neurons. Brain Res. 1622, 350–360 Vannucci, R. C., and Vannucci, S. J. (2000) Glucose metabolism in the developing brain. Semin. Perinatol. 24, 107–115 Herrera, E. (2002) Lipid metabolism in pregnancy and its consequences in the fetus and newborn. Endocrine 19, 43–55 Herrera, E. (2000) Metabolic adaptations in pregnancy and their implications for the availability of substrates to the fetus. Eur. J. Clin. Nutr. 54(Suppl 1), S47–S51 Jurdak, N., Lichtenstein, A. H., and Kanarek, R. B. (2008) Dietinduced obesity and spatial cognition in young male rats. Nutr. Neurosci. 11, 48–54 Haagensen, A. M. J., Klein, A. B., Ettrup, A., Matthews, L. R., and Sørensen, D. B. (2013) Cognitive performance of Göttingen minipigs is affected by diet in a spatial hole-board discrimination test. PLoS One 8, e79429 Goh, V. H. H., and Hart, W. G. (2014) The association of metabolic syndrome and aging with cognition in Asian men. Aging Male 17, 216–222 Reynolds, C. A., Gatz, M., Prince, J. A., Berg, S., and Pedersen, N. L. (2010) Serum lipid levels and cognitive change in late life. J. Am. Geriatr. Soc. 58, 501–509


Received for publication January 7, 2016. Accepted for publication March 2, 2016.

CLOUARD ET AL.