| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
* Department of Degenerative Neurological Diseases, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Tokyo, Japan;
Japan Association for the Advancement of Medical Equipment, Tokyo, Japan; and
Core Research for Evolutional Science and Technology, Japan Science and Technology Agency, Saitama, Japan
1Correspondence: Department of Degenerative Neurological Diseases, National Institute of Neuroscience, National Center of Neurology and Psychiatry, 4-1-1 Ogawahigasi, Kodaira, Tokyo 187-8502, Japan. E-mail: wada{at}ncnp.go.jp
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: lactation • neural progenitor cells • hight-fat diet • metabolic abnormalities • oxidative stress • brain development
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. An increase in the prevalence of obese women of childbearing age is also a great public health concern; for example, in the United States, as over 50% of women (aged 20 to 39 yr) are reported to be overweight or obese (3)
. Female obesity is associated with increased risk of almost every common complication of pregnancy (4)
. Moreover, obesity in the mother has been documented to play a direct role in the transmission of obesogenic and diabetogenic traits to the next generation. Studies using animal models demonstrated that offspring born from dams complicated by either high-fat diet (HFD)-induced obesity or related diabetes are at increased risk of obesity, impaired glucose tolerance, hypertension, and other facets of metabolic disorders in their adult lives (5
6
7
8)
.
In addition to the effect of maternal obesity on the metabolic functions and endocrine systems of offspring, maternal obesity and overnutrition have also been shown to influence the nervous system of offspring [e.g., hypothalamic leptin sensitivity (9)
and appetite-regulating networks (8
, 10)
]. These findings suggest that maternal obesity can affect brain development of offspring, but the development of brain regions besides the endocrine-related areas has not been extensively studied. Early brain development has a tremendous effect on the success of people throughout their lives. In adult rodents, obesity and its related metabolic disorders also appear to negatively affect cognitive function and neural plasticity (11
, 12)
. In obese dams suffering from metabolic disorders, excess lipids and other nutritious components can be transferred to developing offspring via the placenta and/or breast milk (13
14
15
16)
. Consequently, HFD-induced obesity in dams may have an effect on the development and neural plasticity of the brains of offspring.
The hippocampus is one of the most morphologically and synaptically plastic areas in the brain. Recent studies have shown that hippocampal activities are remarkably affected by maternal care, stress, and physical activity (17
18
19)
. Furthermore, unlike in most brain regions, in the hippocampal dentate gyrus, neural progenitor cells are located, and produce functional new neurons throughout postnatal life (20
21
22
23)
. Recent works have demonstrated that newborn neurons could contribute to the hippocampus-dependent cognitive functions, such as learning and memory (24
25
26
27
28
29
30
31)
. Neurogenesis is composed of at least two components: cell proliferation and cell survival. Cell proliferation refers to the production of new cells, whereas cell survival refers to the number of newborn cells that survive to maturity. Very interestingly, it has been reported that hippocampal neurogenesis in offspring can also be modulated by maternal conditions, such as nutrient deficits, stress, and physical activities, during the early development of the offspring (26
, 28
, 32
33
34
35)
. It has also been demonstrated that long-term consumption of an HFD and diabetes leads to the reduction of hippocampal function and neurogenesis in adults as well (36
37
38)
. From these findings, we can postulate the possibility that maternal HFD-induced obesity would have an effect on hippocampal formation during development of offspring.
In the present study, we used a nutritional model that is closer to the modern human lifestyle, characterized by a high-calorie and high-fat diet, to investigate its potential effect on hippocampal development of offspring. Here, we found that HFD offspring developed hyperlipidemia and showed peroxidized lipid accumulations in serum and hippocampal dentate gyrus during early development. Furthermore, impaired hippocampal neurogenesis was observed during postnatal development of HFD offspring, suggesting that lipid-mediated oxidative stress reduces the process of neuronal production in this area. This study is the first demonstration that maternal metabolic conditions can affect oxidative status and progenitor cell division in offspring brain.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
for 6 wk until the mating (6
, 9)
(
Fig. 3A
). The macronutrient composition and caloric distribution in the CE-2 and HFD-32 are detailed in Table 1
. Mice continued on the same diet during pregnancy and lactation (until lactational day 16). On lactational day 16, both groups were fed the ND. We adopted these schedules to eliminate the possibility that offspring could eat the dropped HFD just before weaning. In some experiments, dams were fed the ND or HFD until delivery as described above. After delivery, they were fed the opposite diet until lactational day 16 (
Fig. 3A
). From lactational day 16 on, both groups were fed the ND. The day of parturition was set as d 0, and litters were culled on d 0 to 6–8 offspring per mother. Offspring were kept with their dams until weaning on d 22. Only male offspring were used for all analyses, and a maximum of two male offspring was taken from each litter at each developmental stage to remove any litter effects as reported previously (40
, 41)
. Thereafter, weaned offspring were housed 4–5 per cage, and all of them were fed an ND until the end of the study. To avoid autooxidation of the HFD, diets were stored at –80°C until use, and diets were exchanged every 2 d. All animal experiments were performed in strict accordance with the guidelines of the National Institute of Neuroscience, National Center of Neurology and Psychiatry (NCNP), and were approved by the Animal Investigation Committee of the Institute.
|
|
|
|
Measurement of body weight and food intake
The body weights of 5- to 11-wk-old female mice (dams) were measured on a weekly basis. As for the fetal body weights, after dams (n=6/group) were sacrificed at gestational day 18, fetuses were separated from the placenta, and then they were cleaned and weighed. Postnatal offspring were weighed at postnatal day 2 (P2), and weekly from the day of birth until P70.
The measurement of food consumption was adopted from Ohki-Hamazaki et al. (42)
. Briefly, mice were housed individually. Consumption of mouse chow as well as body weight was measured for 4 consecutive days at the indicated ages. Feeding efficiency [the daily diet consumption (g) per body weight (kg)] was determined in ND-fed females and HFD-fed females at 11 wk of age (n=8/group), and the lactational periods (d 5-9 during lactation) (n=8/group). ND offspring and HFD offspring were also measured at 5 and 10 wk of age (n=6-9/group).
Serum parameters determination
Mice were killed between 10:00 and 12:00 AM in all experiments. For the collection of postnatal blood, the trunk blood was collected from age-matched male offspring (n=6-7/group), allowed to stand for 50 min at room temperature, and then centrifuged at 3000 rpm for 10 min. Serum aliquots were collected and stored at –80°C until use. For the collection of fetal ones, pregnant dams (n=6/group) at gestational day 18 were used. The trunk blood of decapitated fetus was rapidly collected by using heparinized glass capillaries (Terumo, Tokyo, Japan). Blood from the 2-3 fetuses in the same litter were pooled, and their serums were stored as indicated above. Twenty-five male fetuses in ND group and 22 male fetuses in HFD group were used for the collection of their serums. A serum from pooled fetal blood was measured as one sample (i.e., n=1).
Serum total cholesterol and triglyceride were determined enzymatically with an autoanalyzer (Fuji Dry-Chem 5500; Fujifilm, Tokyo, Japan). The serum level of free fatty acids (FFAs) was measured using a commercial enzymatic method (nonesterified fatty acid C-Test Wako; Wako, Wako, Japan). Serum levels of lipid peroxidation were estimated by measuring with thiobarbituric acid reaction substances (TBARS; Cayman Chemical, Ann Arbor, MI, USA). Protein oxidation was evaluated by measuring carbonyl group content in serum proteins, using a commercial enzyme immunoassay kit (protein carbonyl enzyme immunoassay kit (BioCell Corp., Papatoetoe, New Zealand). Serum corticosterone and cortisol levels were assayed using commercial enzyme immunoassay kits (corticosterone and cortisol; Cayman Chemical). All experimental procedures were according to the manufacturers’ protocols.
Glucose-tolerance test
A glucose-tolerance test was carried out in 11-wk-old ND-fed and HD-fed dams (n=6/group). Their 10-wk-old male progeny (both ND- and HFD-fed) were also tested (n=6/group). Mice were fasted for 24 h, and then glucose (1 g/kg body weight) was injected intraperitoneally (i.p.). Blood glucose levels were monitored in tail-blood samples using an Accu-Check Compact Plus Blood Glucose Meter (Accu-Check, Roche Diagnostics, Tokyo, Japan) before (0) and 15, 30, 60, and 120 min after glucose injection as reported previously (42)
.
Histology
Mice were deeply anesthetized with diethyl ether and perfused with phosphate buffered saline (PBS), followed by 4% paraformaldehyde (PFA) in PBS. Tissues were postfixed with 4% PFA for 24 h. As for embryos at E18, brains were fixed in 4% PFA by immersion for 24 h.
White adipose tissues were obtained from the abdomen. The tissues were postfixed with 4% PFA and embedded in paraffin wax. Tissue sections were cut at a thickness of 5 µm and stained with hematoxylin and eosin. Sections were examined under an Olympus BX51 light microscope (Olympus, Tokyo, Japan).
Immunohistochemistry on free-floating 40-µm-thick coronal sections was performed as described previously (23
, 43)
. Sections were preincubated with blocking solution containing 3% bovine serum albumin in 0.1% Triton/PBS for 60 min. Incubation with primary antibodies was performed at 4°C for 1–2 d. Secondary antibodies were applied to sections for 2 h at room temperature.
The following primary antibodies and dilutions were used: malondialdehyde (MDA, 1:200; mouse-IgG; NOF Corporation, Tokyo, Japan), 5-bromodeoxyuridine (BrdU, 1:200; rat-IgG; Harlan, Loughborough, UK), Pax6 (1:500; rabbit-IgG; Covance, Berkeley, CA, USA), calbindin-D-28K (1:500; rabbit-IgG; Chemicon, Temecula, CA, USA), phospho-Histone-H3 (pHH3) (1:200; rabbit-IgG; Upstate, Temecula, CA, USA), and doublecortin (DCX) (1:100; goat-IgG; Santa Cruz Biotechnology, Santa Cruz, CA, USA). The secondary antibodies consisted of Alexa 488-conjugated goat anti-rabbit IgG (1:500; Molecular Probes, Eugene, OR, USA), Alexa 488-conjugated rabbit anti-goat IgG (1:500; Molecular Probes), Alexa 568-conjugated donkey anti-rat IgG (1:500; Molecular Probes), and Cy5-conjugated donkey anti-mouse IgG (1:500; Jackson Immuno Research, West Grove, PA, USA). In some experiments for MDA staining, the sections were incubated with secondary antibody of HRP-labeled anti-mouse antibody (Dako, Kyoto, Japan) for 1 h. The colored reaction product was developed with 3,3'-diaminobenzidine tetrahydrochloride solution (Dako). Nuclei were stained with 4', 6'-diamino-2-phenylindole (DAPI) (1:1000; Sigma, St. Louis, MO, USA). Images were taken with a confocal microscope (TCS SP2; Leica, Mannheim, Germany) fitted with individual filter sets for each channel. Image production was then performed with Adobe Photoshop (Adobe System, San Jose, CA, USA).
The quantification of pHH3-positive (+) cells in the dentate gyrus was performed as previously reported (23)
. Serial sections were cut (40-µm coronal sections) through the entire dentate gyrus in its rostrocaudal extension. Every sixth section (240 µm apart) was processed for immunohistochemistry (n=6/group/age). All pHH3+ cells located in the granule cell and subgranular layers were counted in each section. Eight sections per animal were examined, and the total number of pHH3+ cells per dentate gyrus was obtained by multiplying the value by 6. To count DCX+ cells, sections at 440-µm intervals (every 12th section) containing dorsal hippocampus were selected from each animal (n=6/group), and stained with anti-DCX antibody as reported previously (44
, 45)
. All DCX+ cells located in the granule cell and subgranular layers were counted in each section. Four sections per animal were examined, and the total number of DCX+ cells per dentate gyrus was obtained by multiplying the value by 12. All analyses were done in a masked fashion regarding the experimental groups. Cells were counted at x630 (x63 objective with a x10 ocular) or x1000 (x100 objective with an x10 ocular) on a Leica confocal microscope (Leica).
For the detection of neutral lipids in the liver, Oil Red O staining was performed. Mice were perfused with 4% PFA as described above. The 20-µm frozen sections of liver were postfixed in 4% PFA for 10 min. The sections were dipped in 60% isopropanol and then stained in Oil Red O working solution (60% 3 mg/ml Oil Red O in isopropanol) for 30 min. After rinsing in 60% isopropanol and H2O, the sections were counterstained with Mayer hematoxylin for 1 min and mounted. Lipid droplets were visualized by Olympus BX51 light microscope (Olympus).
BrdU administration and quantification of BrdU-labeled cells
BrdU administrations were performed according to the previously described protocol (23
, 43)
. The DNA base analog BrdU (Sigma) was dissolved in PBS with 0.007 N NaOH. The injection of BrdU was performed between 10:00 and 12:00 AM. To label proliferating cells in the hippocampus of the embryo, pregnant dams (n=3/group) carrying E18 fetuses were given an injection of BrdU (100 mg/kg; i.p.), and the fetuses were removed 2 h later (see
Fig. 6A
). The fetal brains were removed, and then were fixed in 4% PFA. Six embryos from 3 dams (2 embryos/dam) were processed for immunohistochemistry as described above. To label postnatal hippocampal progenitor cells, male offspring (n=6/group) were given an injection of BrdU (100 mg/kg, i.p.) at P21 and P70. For analysis of progenitor proliferations, mice were sacrificed at 2 h after the BrdU injection at P21 and P70 (see
Fig. 6A
). For analysis of neuronal differentiation, mice were sacrificed at 28 d after the BrdU injection at P21 and P70 (see
Fig. 7A
). Six age-matched male offspring from 4 dams (1-2 offspring/dam) were used in each group.
|
|
|
|
The quantification of BrdU-labeled cells in the dentate gyrus was performed as previously reported (23)
. Serial sections were cut (40-µm coronal sections) through the entire dentate gyrus in its rostrocaudal extension. Every sixth section (240 µm apart) was processed for immunohistochemistry. At E18, all BrdU-labeled cells in the dentate gyrus were counted in each section. Five sections per animal were examined, and the total number of BrdU-labeled cells per dentate gyrus was obtained by multiplying the value by 6. At P21, P49, P70, and P98, all BrdU-labeled cells located in the granule cell and subgranular layers were counted in each section. All analyses were done in a masked fashion regarding the experimental groups. Eight sections per animal were examined, and the total number of BrdU-labeled cells per dentate gyrus was obtained by multiplying the value by 6. BrdU-labeled cells were counted at x630 (x63 objective with an x10 ocular) or x1000 (x100 objective with an x10 ocular) on a Leica confocal microscope.
Western blot analysis
Western blot analysis was carried out as described previously (46)
. The tissues from hippocampus, cerebral cortex, and cerebellum were homogenized in ice-cold RIPA buffer (Tris-HCl, 50 mM; NaCl, 150 mM; EDTA, 5 mM; and Triton X-100, 1%; pH 7.5) containing proteinase inhibitors (Complete, EDTA-free; Roche Applied Science, Indianapolis, IN, USA). The homogenates were subjected to SDS-PAGE, transferred to a nitrocellulose membrane, blocked with 3% BSA in PBS, and reacted with anti-MDA (NOF Corporation), 4-hydroxy-2-hexenal (4-HHE) (NOF Corporation), and anti-β-actin (Sigma) antibodies. Immunoreactive signals were visualized with SuperSignal West Dura Extended Duration Substrate (Pierce Biotechnology, Rockford, IL, USA) and detected with a chemiluminescence imaging system (FluorChem; Alpha Innotech, San Leandro, CA, USA).
Cell culture and proliferation assay
Neural progenitor cells were isolated from C57BL/6J mice at P1. The hippocampus were dissected free of meninges and enzymatically digested with 0.1% trypsin-EDTA for 5 min at 37°C, followed by a wash with 0.014% w/v trypsin inhibitor. Single-cell suspension was plated into a 12-well culture plate with untreated surfaces (CellSeed, Tokyo, Japan). The culture medium was composed of Dulbecco’s modified Eagle medium/Ham’s F12 (1:1; Sigma), B27 supplement (Invitrogen, Carlsbad, CA, USA), 0.5 mM L-glutamine (Invitrogen), 100 µg/ml penicillin (Invitrogen), 100 µg/ml streptomycin (Invitrogen), 10 ng/ml basic fibroblast growth factor (bFGF) (Pepro Tech, Rocky Hill, NJ, USA), and 20 ng/ml epidermal growth factor (Pepro Tech). Neural progenitor cells were grown as free-floating neurospheres for 7 d. Neurospheres were collected and then digested with trypsin. Digested cells were plated onto poly-L-ornithine/fibronectin-coated 6-well cell-culture dishes in serum-free neurobasal (NB) medium (Invitrogen) containing B27 supplement, 0.5 mM L-glutamine, 100 µg/ml penicillin, 100 µg/ml streptomycin, and 10 ng/ml bFGF as reported previously (47)
. For the experiments investigating the peroxidized lipid modification of neural progenitor cells, 3-d-old adherent cultured cells were washed with PBS and incubated at 37°C for 120 min with MDA (Sigma) in PBS containing 5 mM glucose, 0.3 mM CaCl2, and 0.62 mM MgCl2. For the BrdU incorporation assay, medium was changed for NB medium containing 20 µM BrdU, and the cells were cultured for an additional 12 h. For immunocytochemistry, cultured cells were fixed with 4% PFA and then immunostained with anti-nestin (1:200; mouse-IgG; Chemicon) and anti-BrdU (1:200; rat-IgG; Chemicon) antibodies. To analyze the cell proliferations, at least 100 nestin-positive (+) cells were observed in each condition. Data were obtained from three independent experiments for each condition. For Western blot analysis, after 12-h incubations with NB medium, cells were homogenized with ice-cold RIPA buffer, and then the homogenates were subjected to SDS-PAGE as described above. The primary antibody was proliferating cell nuclear antigen (PCNA) (1:2000; mouse-IgG; Chemicon).
Statistical analysis
Data were expressed as means ± SE. In each experiment, a direct comparison between the scores for a pair of groups was made using an unpaired Student’s t test. With multiple comparisons, data were analyzed by one-way analysis of variance (ANOVA), with a Dunnett’s post hoc test. P < 0.05 was considered statistically significant. N.S. indicates no significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 9)
. Age-matched, adult female mice (5 wk old) were fed either an ND or an HFD for 6 wk. One week after HFD feeding, HFD-fed female mice showed increases in body weight that surpassed those seen in ND-fed female mice (Fig. 1A
). Six weeks after HFD feeding, white adipose tissue from HFD-fed female mice showed increased adipocyte size relative to ND-fed female mice (Fig. 1B
). We next measured the serum lipid concentrations, glucose tolerance, and food intake of ND- and HFD-fed female mice at 11 wk of age. We observed significant increases in serum cholesterol and FFA concentrations in HFD-fed female mice compared with ND-fed female mice at 11 wk of age (Fig. 1C, D
). However, there were no significant differences in serum triglyceride between the two groups (Fig. 1E
). To address the response to glucose administration, we performed an i.p. glucose-tolerance test. We observed no differences in fasting blood glucose levels between HFD- and ND-fed female mice. However, HFD-fed female mice had higher levels of blood glucose than ND-fed female mice 15 and 30 min after i.p. glucose injection (Fig. 1F
). We then measured food intake in HFD- and ND-fed female mice. No significant difference was observed in daily food intake between ND- and HFD-fed female mice at 11 wk of age (Fig. 1G
). There were also no differences in the maternal daily food intake during lactations between two groups of dams (Fig. 1G
). These results demonstrated that HFD-fed female mice developed obesity, hyperlipidemia, and abnormal glucose tolerance, which are characteristic of metabolic syndrome.
Effect of maternal HFD on fetal and postnatal offspring development
We next investigated the effect of a maternal HFD on fetal and postnatal offspring development. On E18, fetal body weights were not significantly different between ND and HFD offspring (Fig. 2A
). Serum cholesterol (Fig. 2C
), FFAs (Fig. 2D
), triglyceride (Fig. 2E
), and glucose levels (Fig. 2G
) were also not different between the two groups.
To analyze postnatal development, pregnant ND-fed and HFD-fed females were allowed to deliver their offspring. The average fetal litter size was not significantly different between the two groups (ND, 7.46±0.28, n=35 dams; HFD, 6.85±0.31, n=35 dams). However, there was a significant reduction in the ratio of males to females in HFD group (ND, 1.10±0.07, n=191 offspring; HFD, 0.85±0.07, n=167 offspring, P < 0.05). Offspring were reared by their natural mothers (continuing on their assigned diets) after adjustment of litter size to 6–8 pups/mother. The mean body weights of HFD offspring were similar to those of ND offspring up to P2. Subsequently, HFD offspring showed significant increases in body weight and adipocyte size during suckling (Fig. 2A, B
). The biochemical characteristics of the HFD offspring during suckling (at P10 and P21) indicated significant increases in serum cholesterol (Fig. 2C
), FFAs (Fig. 2D
), triglyceride (Fig. 2E
), and blood glucose levels (Fig. 2G
) compared with the value of age-matched ND offspring. The histological analyses also revealed major hepatic abnormalities in nursed HFD offspring (at P10 and P21) that corresponded to excessive lipid accumulations, as assessed by the lipid-specific Oil Red O staining of liver sections (Fig. 2F
). These results indicate that HFD offspring developed obesity and hyperlipidemia during postnatal development.
Although increases in body weight were observed even at P70 (Fig. 2A
), serum cholesterol (Fig. 2C
), FFAs (Fig. 2D
), and the glucose-tolerance test (Fig. 2H
) were not different in HFD offspring compared with ND offspring. However, HFD offspring showed higher levels of serum triglyceride (Fig. 2E
) and slightly more Oil Red O-positive lipid drops in the liver (Fig. 2F
) compared with ND offspring. No differences were observed in daily food intake between ND offspring and HFD offspring during postweaning development (Fig. 2G
). These results suggest that the abnormal blood glucose and lipid metabolism in the HFD offspring are transient changes, and some metabolic parameters may not be sustained in the adulthood under the condition of normal diet feeding after weaning.
Peroxidized lipid accumulation in sucking pups from HFD-fed dams
A previous report demonstrated that the increased level of plasma concentrations of FFAs caused oxidative stress (48)
. Oxidative stress markers, such as lipid peroxidation, were in fact increased in the plasma of an animal model of metabolic syndrome (49)
. We next examined whether peroxidants are increased in HFD offspring during development by using two oxidative-stress markers, TBARS, an index of lipid peroxidation, and protein carbonyl contents, a biomarker of oxidized proteins.
At E18, serum TBARS were not different between ND and HFD offspring, but serum TBARS were significantly increased in HFD offspring as compared with ND offspring during suckling (at P10 and P21) (Fig. 3B
). There were no significant differences in serum TBARS levels between ND and HFD offspring at P70 (Fig. 3B
). We next examined whether the elevation of TBARS in offspring during suckling is dependent on whether mothers are fed HFD until delivery, only during lactation, or both; thus we prepared an additional two groups. One group was as follows: Mothers were fed an ND until delivery and then fed an HFD from lactational days 0 to 16 [HFD (lactation)]. In the other group, dams were fed an HFD until delivery and then fed an ND during lactation [HFD (gestation)] (see Fig. 3A
). Offspring from these two groups showed a slight elevation of serum TBARS levels compared with ND offspring during suckling (at P10 and P21), but the elevation was not significant (Fig. 3B
). Serum protein carbonyl contents were not different between each group at different developmental stages (Fig. 3C
). From these results, we concluded that long-term (gestation to lactation) HFD feeding in females leads to the accumulation of peroxidized lipids in offspring during postnatal development.
To examine the cellular accumulation of peroxidized lipids in the brain, we next performed immunoblotting analysis with MDA antibody in the cerebral cortex, the hippocampus, and the cerebellum. MDA is a major compound generated by lipid peroxidation. Immunoblotting analysis showed that MDA accumulations in the hippocampus were higher in HFD offspring than ND offspring at P21 (Fig. 4A
). Proteins ranging from 75 to 150 kDa were especially modified with MDA in HFD offspring. Also, in the cerebral cortex and the cerebellum, there were somewhat higher MDA accumulations in HFD offspring compared with ND offspring (Fig. 4A
). However, at P70, there were no significant MDA accumulations in the hippocampus of HFD offspring (Fig. 4B
). Furthermore, we performed immunoblotting using another lipid peroxidation marker, 4-HHE, and obtained similar results showing more oxidized damage in the hippocampus of HFD offspring (data not shown). To clarify the distribution of MDA accumulations, we performed immunohistochemistry. Immunostaining showed that higher MDA accumulations were observed in the hippocampal dentate gyrus of HFD offspring at P21 (Fig. 4C
). The strong immunoreactivity for MDA was detected from the subgranular zone toward the molecular layer of HFD offspring at P21 (Fig. 4C
). In the subgranular zone, neural stem/progenitor cells exist and produce new neurons throughout postnatal life. Therefore, we then examined whether peroxidized lipids accumulated in these progenitor cells in this area. Pax6 is a transcription factor and a marker for neural progenitor cells in postnatal hippocampal dentate gyrus (50)
. By double staining for Pax6 and MDA, we examined the peroxidized lipids accumulated in hippocampal progenitor cells. Whereas we could not detect the MDA signals in Pax6+ progenitor cells in ND offspring (Fig. 4E
), we could confirm significant MDA signals in those cells in HFD offspring at P21 (Fig. 4F, G
). At P70, immunohistochemical and Western blot analysis demonstrated no significant differences in peroxidized lipid accumulations in hippocampus between ND and HFD offspring (Fig. 4B, D
). These results indicated that maternal HFD-induced obesity leads to lipid-mediated oxidative stress in their offspring during early development. Our findings also raise the possibility that the peroxidized lipid accumulations may affect progenitor cell mobility, such as cell proliferations and differentiation during postnatal development.
Impaired hippocampal neurogenesis in HFD offspring
We next examined the toxic effects of MDA on the proliferation of hippocampal progenitor cells in vitro. Neural progenitor cells were isolated from hippocampus on P1. Neurosphere-forming cells were collected and then expanded to the adherent monolayer cultures. These progenitor cells were treated with 0, 50, 100, 500, or 1000 µM MDA for 2 h. To measure the proliferation rate, we next added BrdU into the growth medium and then incubated for 12 h (Fig. 5A
). As shown in Fig. 5B, C
, the rate of BrdU-labeled nestin+ cells was decreased in the MDA-treated cells in a dose-dependent manner. Furthermore, we confirmed a decrease in the expression of PCNA, a marker for cell proliferation, after MDA treatments (Fig. 5D
). These data demonstrated that MDA treatment reduced the proliferation of the hippocampal progenitor cells in vitro.
To address the question of whether postnatal neurogenesis is impaired in the dentate gyrus of HFD offspring in vivo, we next determined the total number of BrdU-labeled cells in the dentate gyrus in ND offspring and HFD offspring at E18, P21, and P70. To determine the number of proliferating progenitor cells, measurements were carried out 2 h after a single injection (i.p.) of 100 mg/kg BrdU (34
, 37)
(Fig. 6A
). We found no difference in the number of BrdU+ cells between ND offspring and HFD offspring at E18 (Fig. 6D
). However, we found a significant reduction in the number of BrdU+ cells in HFD offspring at P21 and P70 as compared with ND offspring (Fig. 6B, C, D
). We further examined whether progenitor cell proliferations are reduced in HFD offspring at P49 and P98 (Fig. 6A
) by immunostaining with pHH3, a marker for mitotic cells. We found that the significant reduction in the number of pHH3+ cells in dentate gyrus of HFD offspring at P49 compared with age-matched ND offspring (Fig. 6E, H
), but not at P98 (Fig. 6H
). From these data and BrdU-pulse labeling study, we conclude that the maternal obesity could reduce the hippocampal cell proliferations in offspring at postnatal young ages; however, they did not affect cell proliferations in offspring at older ages. Furthermore, at different developmental stages, there were no changes in the basal levels of serum corticosterone and cortisol (Fig. 6I, J
), indicating that decreased hippocampal neurogenesis was not because of stress hormones.
Finally, we examined the neuronal differentiation of newborn cells in the dentate gyrus. Mice were injected with BrdU at P21 or P70 and then killed 28 d after the BrdU injection (at P49 or P98) (23
, 35)
(Fig. 7A
). The total number of newborn neurons in each group was counted by double immunostaining with BrdU and calbindin, a marker for mature granule cells (51)
(Fig. 7B
). We found a significant decrease in the number of BrdU+/calbindin+ cells (new granule cells) in HFD offspring compared with ND offspring at P49 (Fig. 7C
). At this point, the total number of surviving BrdU+ cells was also reduced in the HFD offspring (Fig. 7D
). There were no significant differences in the number of new neurons (Fig. 7C
) and the surviving BrdU+ cells (Fig. 7D
) between ND and HFD offspring at P98. At each developmental stage, there were no differences in the rate of neuronal differentiation (% of calbindin+ cells within BrdU+ cells) in both groups, indicating that maternal HFD did not affect the cell-fate determination of newborn cells in the dentate gyrus (Fig. 7E
). At this older age, we further found no differences in the number of doublecortin-positive newborn immature neurons between two groups (Fig. 7F-H
), suggesting that the maternal obesity could not affect hippocampal neurogenesis in offspring at older ages. These results demonstrated that maternal HFD-induced obesity reduced hippocampal neurogenesis during the early life of offspring.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, only lactation (15)
, or both periods (5
6
7
8
9)
on progeny. In the present study, we focused on the consequences of long-term HFD feeding in adult female mice because it better represents the present dietary habits of human society. Many women are already overweight or obese by the time they reach childbearing age. Therefore, the HFD-induced obese mouse model has physiological relevance to human situations, and it is a useful model to investigate the consequences in offspring of increased body weight prior to and during gestation/lactation because of chronic consumption of an HFD.
Nutritional status during early life is an important determinant of proper brain development (53
, 54)
. The placenta plays a crucial role in the transport of bioactive molecules from mother to fetus (54)
. As for maternal lipids, cholesterols and FFAs can be transferred across the placenta and are accumulated in the brain and other organs during fetal development (55
, 56)
. However, our findings demonstrated that fetal serum FFAs, triglyceride, and cholesterol levels were not significantly different between ND and HFD groups during the embryonic development. In accordance with our findings, another group has also reported that a maternal HFD did not change fetal blood lipid concentrations during gestational periods (5)
. However, in contrast, it has also reported that maternal HFD could elevate the fetal blood triglyceride and nonesterified fatty acids but not cholesterol levels (16)
. These differences might be due to the dietary fat compositions, maternal HFD-feeding periods, and species.
However, we found that HFD offspring have higher levels of serum lipid concentrations compared with ND offspring during suckling. Milk lipids provide not only calories, but also some bioactive molecules, such as essential fatty acids and derivatives, that support the development of the newborn pups (55
, 56)
. These unique demands make the lactating mammary gland one of the most active lipid synthesis and transport organs in the body. The fatty acids of milk triacylglycerols are derived from de novo synthesis within the mammary gland from lipids of dietary origin or lipids mobilized from adipose tissue (57)
. More interestingly, previous studies have demonstrated that changes in dietary fat content act to modify milk lipid content and daily milk lipid production in lactating animals (13
14
15)
. A high-fat diet fed throughout lactation was found to increase milk lipid levels and metabolizable energy content compared with those in animals fed normal laboratory chow (13
, 14)
. These changes in milk were associated with faster growth of pups during suckling (14)
. Our results showed that there was no difference in the maternal daily food intake during lactations between ND and HFD groups. On the basis of the diet compositions of CE-2 and HFD-32, we could easily predict that the daily fat intake in HFD-fed dams was 
8-fold higher than those in ND-fed dams during these periods. From the previous and current studies, we can postulate that the HFD offspring would intake breast milk with higher levels of lipids throughout lactation in the present study too, and one plausible hypothesis might consider a higher-energy and higher-fat in the milk would lead to obesity and hyperlipidemia in HFD offspring during suckling.
Furthermore, we focused on the elevated serum FFAs levels in HFD offspring. It has been reported that plasma concentrations of FFAs are elevated in metabolic syndrome (59)
, and the increased FFAs result in the induction of oxidative stress (48
, 60)
. In the present study, the serum FFAs were also elevated, and lipid peroxidants were accumulated in HFD offspring. The interesting result was that these peroxidized lipid (TBARS) accumulations resulted from the long-term (throughout pregnancy and the lactation) maternal HFD feeding. Maternal HFD feeding during only gestation or only lactation also led to somewhat (but not significant) higher serum TBARS levels in offspring. On the basis of these findings, the intake of breast milk containing excessive lipids from obese dams would be one possible mechanism underlying obesity and accelerated lipid peroxidation in HFD offspring during their development. After weaning, body weights in HFD offspring were still somewhat higher at older ages; however, most of the metabolic parameters and TBARS levels in HFD offspring were not different from those in ND offspring. Abnormal lipid metabolism and the peroxidized lipid accumulations in young HFD offspring seem to be transient changes that result from the maternal HFD-induced obesity. The normal chow after weaning would improve the diverging body weight gains and some metabolic abnormalities in offspring born from obese dams. From previous research and our data, we can postulate that the effect of breast milk may be important and influential for the development of offspring from HFD-fed obese dams. Taken together, these effects would be because of the long-term (during gestation and lactation) consumption of an HFD in dams.
Maternal obesity impairs postnatal hippocampal neurogenesis in offspring
Interestingly, the process of neuronal production in offspring is also affected by maternal conditions such as nutrients, stress, experience, and physical activity (26
, 28
, 32
33
34
35)
. Here, we demonstrated that maternal obesity impaired neurogenesis during postnatal development of the offspring. In this condition, oxidative stress is promoted in the dentate gyrus of HFD offspring.
It has been reported that ionizing irradiation results in significant alterations in hippocampal neurogenesis that are associated with cognitive impairment (25
, 61
, 62)
. In these conditions, one important factor that may negatively affect hippocampal neurogenesis is oxidative stress-mediated cellular damage (63
, 64)
. Chronic alcoholism also results in the reduction of hippocampal neurogenesis (65)
. The pharmacological mechanisms that mediate ethanol effects may result in the production of oxidative stress, and it has been suggested that ethanol may cause tissue damage through lipid peroxidation (66)
. Furthermore, Sava et al. (67)
have reported that the proliferation of adult hippocampal neural progenitor cells is inhibited by the mycotoxin Ochratoxin A, which causes oxidative stress in in vitro experiments. From these findings, oxidative stress, including excess lipid peroxidation, could negatively affect the viability of hippocampal neural progenitor cells, resulting in the reduction of neurogenesis. Also, in our study, neural progenitor cells in HFD offspring were more immunoreactive for the peroxidized lipids (MDA) compared with those of ND offspring during postnatal development. MDA is a naturally occurring product of lipid peroxidation and is a highly reactive three-carbon dialdehyde produced as a byproduct of polyunsaturated fatty acids (PUFAs) peroxidation and arachidonic acid metabolism. PUFAs have often been implicated as substrates in peroxidative damage because of the abundance of multiple double bonds. These double bonds constitute ideal sites for the attack of free radicals. Furthermore, very interestingly, it has been suggested that adult hippocampal progenitor cells contain more unsaturated fatty acid compared with other neural cell types (neurons, astrocytes, and oligodendrocytes) (68)
. In this study, we found that MDA was accumulated in hippocampal neural progenitor cells of HFD offspring, and MDA reduced the proliferation of progenitor cells. From these findings, we postulated that hippocampal progenitor cells may be vulnerable to peroxidative damage.
It has been reported that corticosteroids strongly inhibit hippocampal progenitor cell proliferation (69)
and are also transferred to the pups from the lactating mother (70)
. Furthermore, it has also been shown that HFD intake leads to the reduction of hippocampal neurogenesis, largely because of the elevation of corticosterone (36
, 37)
. However, interestingly, our data showed no difference in the basal levels of serum corticosterone and cortisol between ND and HFD offspring. These findings suggest that the reduction of neurogenesis in HFD offspring is not dependent on stress hormone-mediated cascades. This is not surprising, given that long-term feeding of an HFD increased serum corticosterone levels in males, but not females (36)
. In females, HFD feeding during pregnancy and lactational periods may not have an effect on the developing offspring via an adrenal hormone-dependent mechanism. This result suggests a differential mechanism for hippocampal neurogenesis impairment induced by an HFD between adults and offspring. Furthermore, it has been reported that diabetic rodent models also show lower rates of hippocampal neurogenesis (37
, 38)
. In diabetic animals, the basal hypothalamo-pituitary-adrenocortical function is up-regulated, resulting in elevated circulating glucocorticoids (71)
. Stranahan et al. (37)
have clearly demonstrated that diabetes impaired hippocampal neurogenesis and functions, and corticosterone contributed to these adverse effects by using the type 1 and type 2 diabetic rodents. In our study, we observed the somewhat higher blood glucose levels in HFD offspring during the early developments; however, there were no increases in the glucocorticoids levels. From these findings, the mechanism of reduced hippocampal neurogenesis in HFD offspring may be different from those in the diabetic models.
Because of their unique properties and higher plasticity, the functional roles of new hippocampal neurons have been highlighted. New neurons have been shown to exhibit robust long-term potentiation, unlike the preexisting mature granule neurons (72
, 73)
. These properties may enable new neurons to affect hippocampal network plasticity. Very interestingly, several papers have recently demonstrated that the ablation and modification of adult-born neurons could contribute to the hippocampus-dependent cognitive functions, such as spatial learning memory (27
28
29
30
31)
, contextual fear memory (25)
, and anxiety-related behaviors (74)
. On the basis of these reports, our data suggest that the decreases in the number of new neurons in HFD offspring may show some cognitive deficits during postnatal development.
From current and previous findings in animal models, maternal obesity can affect not only the metabolic functions, but also hippocampal neurogenesis of offspring. In contrast, a previous study showed that the physical activity of pregnant or lactating mice led to an increase in neurogenesis during early postnatal development (35)
. In humans, the increase in the prevalence of obesity has been attributed to increased dietary intake of high-fat foods and reduced physical activity (2
, 75)
. The prevalence of obesity among children is also increasing (75)
. Moreover, it has been reported that oxidative stress is increased in obese children (76)
, and childhood obesity involves cognitive deficits (77)
. Therefore, intake of the proper quality and quantity of food and moderate exercise in adult woman of childbearing age is beneficial to the brain development of their children.
In the present study, we focused on maternal obesity, which is a public health concern, and we demonstrated that maternal metabolic conditions can impair the postnatal hippocampal neurogenesis of offspring. In addition to a genetic analysis, a comprehensive analysis of the brain regarding external signals, such as environmental factors, lifestyle, metabolic functions, and the relationship between parents and offspring, is necessary. A clarification of these organic relationships would help us better understand the precise mechanisms of human brain function and neural development.
|
ACKNOWLEDGMENTS |
|---|
Received for publication October 30, 2008. Accepted for publication January 8, 2009.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. D. Bilbo and V. Tsang Enduring consequences of maternal obesity for brain inflammation and behavior of offspring FASEB J, June 1, 2010; 24(6): 2104 - 2115. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
; n=30) and HFD (
; n=30). Data are from 4 studies, with females maintained on diets for 6 wk, starting at 5 wk old. B) Representative photomicrographs showing retroperitoneal adipose tissues of ND- or HFD-fed dams at 11 wk old. Scale bar = 50 µm. C–E) Serum lipid concentration of 11-wk-old dams on ND (n=9) and HFD (n=9). Total cholesterol (C), serum-free fatty acids (FFAs; D), and triglyceride (E) are shown. F) Intraperitoneal glucose-tolerance test in 11-wk-old dams on ND (
; n=6) and HFD (•; n=6). Changes in blood glucose levels after glucose administration are shown. G) Daily food intake of 11-wk-old dams on ND (n=8) and HFD (n=8). Daily food intake of dams during lactations between two groups (n=8/group) are also presented. *P < 0.05, **P < 0.01, ***P < 0.001.
