Whether a high-unsaturated-fat, high-protein (HFP), and low-carbohydrate (CHO) diet during gestation has long-lasting beneficial effects on lipid metabolism in the offspring was investigated using a mouse model. Female mice were fed either a standard (CHO rich) chow diet or a CHO HFP diet, before and during gestation and lactation. All offspring were weaned onto the same chow until adulthood. Although liver cholesterol concentration and fasting plasma triglyceride (TG), cholesterol, and free fatty acid concentrations were not affected in either male or female HFP offspring, hepatic TG concentration was reduced by ∼51% (P < 0.05) in the female adult offspring from dams on the HFP diet, compared with females from dams on the chow diet (a trend toward reduced TG concentration was also observed in the male). Furthermore, hepatic protein levels for CD36, carnitine palmitoyltransferase-1 (CPT-1), and peroxisomal proliferator activated receptor-α (PPAR-α) were increased by ∼46% (P < 0.001), ∼52% (P < 0.001), and ∼14% (P = 0.035), respectively, in the female HFP offspring. Liver TG levels were negatively correlated with protein levels of CD 36 (r = −0.69, P = 0.007), CPT-1 (r = −0.55, P = 0.033), and PPAR-α (r = −0.57, P = 0.025) in these offspring. In conclusion, a maternal HFP diet during gestation and lactation reduces hepatic TG concentration in female offspring, which is linked with increased protein levels in fatty acid oxidation.
epidemiological and animal studies have supported the “fetal programming” hypothesis linking adult diseases, such as the metabolic syndrome, Type 2 diabetes, and cardiovascular disease with events occurring during early development (8, 17, 18, 25). The fetal programming hypothesis is supported by data from animal models that have manipulated maternal diet (23, 26, 35). Whereas these studies provide data showing a detrimental effect on the offspring (i.e., linking increased risk of development of the metabolic syndrome and cardiovascular disease to early development), emerging evidence also suggest a long-lasting beneficial effect on the offspring may occur if a “good” manipulation or stimulus is applied to the mother during gestation and lactation (we suggest the term of “beneficial fetal programming”). For example, the addition of vitamin E or cholestyramine to the maternal diet during gestation markedly reduces the progression of atherogenic lesions in the aorta of offspring in rabbits (28).
There is increasing interest in low-carbohydrate (CHO) diets among obese individuals who are trying to lose weight (6). The Atkins diet is a low-CHO, high-lipid, and high-protein diet (6). In randomized controlled studies, this diet generated comparable or slightly better results, in terms of weight loss and improvement in lipid profile, than a conventional low-fat, high-CHO, low-calorie diet after 3 and 6 mo of consumption (13, 32). Because more people are interested in losing weight with the use of these low-CHO diets, their use will increase, and this may result in their consumption during pregnancy. Therefore, there is a need to determine whether there are long-term benefits to the offspring of maternal consumption of these diets during pregnancy and lactation.
We have used a murine model to assess whether a low-CHO, high-unsaturated-fat diet (as unsaturated fat produces favorable effects on lipid metabolism) during pregnancy has any long-term beneficial effect on the offspring with a focus on the metabolic parameters and hepatic lipid metabolism. We have examined whether the low-CHO, high-unsaturated fat, and protein (HFP) diet affected body weights of the dams, and those of her offspring. We have also measured hepatic fat concentrations and protein levels of key molecules involved in hepatic fat metabolism, as well as fasting plasma lipid [triglyceride (TG), cholesterol, and nonesterified fatty acids] concentrations in the adult offspring.
MATERIALS AND METHODS
Animals and experimental protocols.
All animal procedures carried out in this study were in accordance with the British Home Office Animals (Scientific Procedures) Act, 1986. This study was approved by the University ethics committee. Virgin female BALB/c mice (3 wk old) were randomly assigned to two dietary groups. They were fed ad libitum with a low-CHO, high-fat, and high-protein (HFP; n = 9) diet (Tables 1 and 2) containing 16.5% CHO, 26.9% protein, and 52.6% lipid in energy, respectively, or a standard laboratory chow diet (n = 6) for 6 wk before conception (Fig. 1). The ingredients for both diets were purchased from commercial sources and prepared in-house. At 9 wk old, all animals were time-mated, and pregnancy was determined by the presence of vaginal plug (defined as day 0). There were four or six pregnancies from mice on the chow or HFP diets, respectively. The diet assigned to an animal before conception was also given throughout the gestation and lactation period. Food intake and body weights of the dams were measured every 3 days until their offspring had been weaned. All offspring were weaned at 3 wk of age onto the standard chow diet and maintained on this diet ad libitum until they reached adulthood (Fig. 1). All mice had free access to water throughout the study. We refer to the adult offspring born to dams fed the HFP diet during gestation and lactation as “HFP offspring,” and to the adult offspring born to dams fed the chow diet as “control offspring.” Offspring were examined at 8 wk old, as the onset of puberty of BALB/c mice start at ∼4 wk of age (7) and sexual maturity at ∼5 wk of age (4). The offspring (5 males and 6 females from the dams on the chow diet and 6 males and females randomly selected from each litter born to dams on the HFP diet) were fasted for ∼12 h and euthanized the next day by CO2 inhalation and cervical dislocation. Blood was collected by cardiac puncture, and liver tissue was dissected, snap-frozen in liquid nitrogen, and stored at −80°C for later analysis. Plasma glucose, nonesterified fatty acid (NEFA), total cholesterol, and TG concentrations were determined with the use of an autoanalyzer (Konelab 20, Thermo Electron).
Liver lipid extraction and analysis.
Livers (∼60 mg) were homogenized in phosphate-buffered saline and protein concentration determined (24). Homogenate (300 μl) was extracted with 5 ml of chloroform/methanol (2:1) and 0.5 ml of 0.1% sulfuric acid (12). The organic phase was dried under nitrogen and resuspended in ethanol. Hepatic TG and cholesterol content were determined using commercially available kits (Randox Laboratories). Data were normalized for differences in protein concentration.
Western blot analysis.
Liver tissues (∼100 mg) were added to a homogenizing buffer containing 50 mM Tris·HCl (pH7.6), 0.25% Triton X-100, 0.15 M NaCl, 10 mM CaCl2, 0.1 mM PMSF, 10 μM leupeptin, 10 μM pepstatin A, 0.1 mM iodoacetamide, 25 μg/ml aprotinin, and 0.1 mM PMSF. The cell homogenates were spun at 15,000 g for 10 min at 4°C. The supernatants were transferred to Eppendorf tubes and stored at −70°C. Protein concentrations were measured (24). Equal amounts (60 μg) proteins for each sample were mixed with ×5 sample buffer (5 ml of sample buffer contains 2.5 ml of glycerol, 1.25 ml of 2-mercaptoethanol, 0.5 g of SDS, 1.04 ml of 1.5 M Tris, pH 6.8, and 1.25 mg of bromphenol blue), boiled for 4 min, and loaded on 7% SDS-polyacrylamide minigels. The gels were run at 200 V for 1 h, and transferred on an Immobilon-P transfer membrane (Millipore, Bedford, MA) in 25 mM Tris, 150 mM glycine, and 20% methanol. After being transferred, the membranes were blocked in 10 ml of 20 mM Tris, 0.15 M NaCl, and 0.1% Tween (TBST) and 5% dry milk for 1 h and then incubated for 16–18 h at 4°C in 10 ml of TBST and 5% dry milk containing primary antibody (see results). At the end of the incubation period, the membranes were washed in TBST for 3 × 10 min at room temperature and then incubated for 1 h in TBST and 5% dry milk containing secondary antibodies conjugated to horseradish peroxidase (Santa Cruz, 1:4,000). The membranes were washed in TBST for 3 × 10 min at room temperature and processed with the ECL (Perbio Science, United Kingdom) detection system. Specific protein bands were then detected by exposing the membrane on a Kodak BioMax Light film (Sigma). To quantify the protein bands, the image of the membrane on the film was analyzed with the use of imaging software (1 D advanced version 4.01; Phoretix, Newcastle upon Tyne, United Kingdom). Values are presented as the relative intensity of the protein bands (calculated as volume).
All statistical calculations were performed with the use of SPSS (version 10.1) software. Differences in mean values were examined with unpaired Student’s t-test. Results were presented as means ± SD. Non-normally distributed data was normalized with transformation, and parametric statistical analyses undertaken (t-tests and Pearson univariate regression).
Female mice were randomly assigned to be fed either a chow (n = 6) or HFP diet (n = 9) and their body weights measured every 3 days. There was no difference in body weights between female mice on the HFP diet and those on the chow diet on all measurements before mating. Interestingly, mice on the HFP diet ate on average ∼1.58 g of diet/day (1.58 ± 0.25), which was ∼21% less by weight than those on the chow diet (2.00 ± 0.23 g/day, P = 0.002). However, the average daily energy intake was similar for mice on the two types of diets (chow vs. HFP: 8.74 ± 0.95 vs. 8.65 ± 1.36 kcal/day, P = 0.85), as the energy density of the HFP diet was ∼25% greater than that of the chow diet (Table 2). Thus mice on the HFP diet ate on average 57.5% less CHO, 23% more protein, and 153% more lipid (by calorie) without increasing their total calorie intake, compared with mice on the chow diet (Table 2).
Of the total pups born, 6 males and 17 females were born to 4 dams on the chow diet (average litter size was 5.75), and 13 males and 20 females were born to 6 dams on the HFP diet (average litter size was 5.5; Table 3). The mean birth weight of the HFP offspring was ∼10.8% greater than that of the control offspring (P < 0.01; Table 3), although the litter size was similar between the two groups (Table 3). However, the mean body weight of HFP offspring (nursed by dams on the HFP diet) and that of the control offspring (nursed by dams on the chow diet) was not significantly different at the time of weaning for both male and female offspring (Table 3). From postweaning, all offspring were fed the same chow diet for 5 wk (i.e., to 8 wk old). There was no difference in body weights of adult offspring between experimental groups (Table 3).
Plasma lipid concentrations.
All offspring were euthanized 55–56 days after birth and plasma samples were analyzed. Fasting plasma TG, total cholesterol, and NEFA concentrations were similar between the HFP and control groups for both male and female offspring (Table 4).
Liver lipid analysis.
Analysis of liver lipids and protein levels were undertaken using the same piece of liver tissue divided in half from adult offspring. There was a trend toward reduced liver TG concentration in male HFP (n = 6) compared with male control offspring (n = 5, P = 0.22; Fig. 2A). Interestingly, liver TG concentrations in the female HFP offspring (n = 6) was markedly lower compared with female control offspring (n = 6, P < 0.05; Fig. 2B). Liver cholesterol concentrations were similar between the HFP and control groups for both male and female offspring (Fig. 2, C and D).
Level of protein expression key to lipid metabolism.
We then determined whether protein levels of key genes in hepatic lipid metabolism were altered in the female offspring (as no significant change in hepatic TG concentration was observed in the male offspring), in association with altered hepatic TG concentration.
Protein levels of CD36, a key long-chain fatty acid transporter (1, 19) were markedly increased in the HFP offspring (n = 6) compared with the controls (n = 6, P < 0.001; Fig. 3A). Protein levels of CPT-1, the rate-limiting enzyme in fatty acid β-oxidation (10), were increased in the female HFP offspring (P < 0.001, n = 6 for both groups; Fig. 3B). Protein levels of PPAR-α, a transcription factor regulating genes in hepatic fatty acid oxidation (5, 20), were also increased in the HFP offspring (P < 0.05, n = 6 for both groups; Fig. 3C).
Correlation between liver TG concentrations and protein levels of CD36, CPT-1, PPAR-α.
We then examined whether there was any association between observed changes in hepatic TG concentrations and levels of key metabolic proteins. Data of protein levels from female offspring born to dams on both the chow and HFP diets (n = 12) were pooled and analyzed against hepatic lipid concentrations obtained from these mice. Liver TG concentrations were negatively correlated with protein levels of CD36 (r = −0.69, P = 0.007; Fig 4A), CPT-1 (r = −0.547, P = 0.033; Fig. 4B) and PPAR-α (r = −0.574, P = 0.025; Fig 4C).
We have presented novel data showing that feeding female mice a CHO high-protein, high-fat diet for 6 wk before conception and during gestation and lactation leads to a marked reduction in hepatic TG concentration in the female offspring. In parallel to reduced hepatic TG concentration, hepatic protein levels of CD36, CPT-1, and PPAR-α were increased in these offspring. Furthermore, hepatic TG concentrations were negatively correlated with hepatic protein levels of CD36, CPT-1, and PPAR-α in these offspring. Because reduced hepatic TG concentration is linked with increased hepatic insulin sensitivity (36), our data favor a beneficial long-lasting effect on the female offspring born to dams fed the HFP diet during gestation and lactation.
The most striking effect in giving the HFP diet to the dam during pregnancy and lactation is the long-term reduction in hepatic TG concentration in the adult female offspring, the magnitude of this reduction is over twofold. Importantly, this reduction in liver TG concentration occurs despite the HFP offspring being fed the standard laboratory chow diet from weaning until adulthood.
Hepatic TG concentration is affected by de novo lipogenesis and fatty acid oxidation in the liver (15). Because only small quantities of hepatic TG are derived from de novo lipogenesis (2, 14), our study focused on molecules relevant to fatty acid oxidation. PPAR-α is a master transcription factor regulating a number of proteins involved in fatty acid oxidation (3, 16, 27, 30, 34, 38). CPT-1 is the rate-limiting enzyme in fatty acid β-oxidation (10) and CD36 is an 88-kDa membrane glycoprotein that belongs to the class B scavenger receptor family (21). CD36 binds with many ligands including native and oxidized lipoproteins (21) and long-chain fatty acids, transports long-chain fatty acids across the cell membrane (1, 19). Both CD36 and CPT-1 are responsive to PPAR-α activation because the promoter of CD36 contains a PPAR response element that binds to PPAR-α (33), and PPAR-α activation or overexpression stimulates CPT-1 mRNA in human hepatocytes (22). Previous data support a link between increased PPAR-α expression and reduced hepatic lipid content. For example, PPAR-α agonists prevented fatty liver in ethanol-fed mice by stimulating fatty acid β-oxidation via upregulating mRNA levels of PPAR-α target genes (11). In contrast, in mice with fatty liver dystrophy, hepatic fatty acid oxidation is markedly reduced with altered expression of several peroxisome proliferator regulated proteins (29). Furthermore, in PPAR-α-deficient mice, etomoxir (a CPT-1 inhibitor) induced a much greater increase in hepatic lipid concentration compared with the effect of etomoxir in the wild-type mice (9). These data are consistent with our data showing a reduced hepatic TG concentration with increased hepatic protein levels of PPAR-α, CD36, and CTP-1 in the HFP offspring, suggesting that increased protein levels of PPAR-α may be relevant to reduced hepatic TG concentration in the HFP offspring. Indeed, our data show that hepatic TG concentrations were negatively correlated with hepatic protein levels of CD36, CPT-1, and PPAR-α in female offspring. Taken together, these data suggest that feeding the HFP diet during gestation and lactation programs reduced hepatic TG concentration, which is associated with increased hepatic protein levels of PPAR-α and its response genes, including CD36 and CPT-1.
The Atkins diet is gaining popularity in reducing body weight among people with obesity or diabetes (6). Interestingly, our study has shown that feeding female mice with a HFP diet 6 wk before conception did not affect their body weights, despite dietary fat intake being ∼53% of the total energy intake (∼2-fold more than the amount of lipid consumed by the control mice). The actual calorie intake in mice on the HPF diet was similar to that on the chow diet, although the energy density of the HFP diet was markedly greater. Our data also show that the plasma lipid profile in the male and female adults are similar between those born to dams on the HFP diet and the dams on the chow diet. Thus these data suggest that the HFP diet during gestation and lactation does not have any adverse effect in terms of body weight and plasma lipid profiles in the adult offspring. Further investigation is required to assess what happens to the plasma lipid profile and hepatic lipid metabolism in these mice with the effects of aging.
It is worth noting that the HFP diet we used was enriched in mono- and polyunsaturated fats. The HFP diet contained a much higher proportion of unsaturated fatty acids compared with the chow diet. This makes our HFP diet distinct from most other high-fat diets used in previous animal studies that contained predominantly saturated fatty acids. It is well established that treatment with unsaturated fatty acids produces a favorable effect on TG metabolism in adult human studies. It is worth noting that the HFP diet contains ∼77.5% less sucrose than the chow diet. A sucrose-rich diet increases hepatic TG deposition (37) and plasma TG concentration in rats (31). Thus a low sucrose in our HFP diet may also contribute to reduced hepatic TG levels in the HFP offspring. Although we have not tested the effects of unsaturated fatty acids and low sucrose per se, our data suggests that consumption of a diet that is enriched with unsaturated fat and is low in sucrose during pregnancy and weaning, induces a lasting reduction in liver TG in the offspring. Furthermore, the changes we observed were purely due to dietary modification in the mother, as we have used a genetically homogeneous strain to eliminate any potential confounding effect due to genetic background.
Our data showing that reduced hepatic TG concentrations occurred in the female HFP offspring (although there was a trend toward reduced TG concentrations in the male HFP offspring) suggest that there maybe a gender-specific effect in the reduction of hepatic TG in female offspring, because current evidence suggests that sex hormone affects lipid metabolism in skeletal muscle and the liver. For example, 17β-estradiol increases maximum activity of CPT-1 (1) and upregulates the expression of PPAR-α and CPT-1 genes in skeletal muscle (2), but suppresses expression of CPT-1 in the liver (4). Hepatic expression of PPAR-α is regulated in a gender-specific manner in the liver (3, 6, 10). The mechanism underlying the interaction between gender and fetal programming requires further investigation. However, these data suggest that liver TG in the female HFP offspring may be affected by sex hormone and not entirely due to maternal dietary modification.
In conclusion, we have presented novel data showing that a maternal diet enriched with unsaturated fat and low in sucrose during gestation and lactation reduces hepatic TG concentration in adult female offspring. Reduced hepatic TG content is associated with upregulation of levels of key proteins regulating fat oxidation.
J. Zhang and P. L. Terroni are supported by the School of Medicine and the Developmental Origins of Health and Disease Centre, University of Southampton, respectively. This research is supported by the Welcome Trust and British Heart Foundation.
We thank C. J. Gelauf for measuring plasma glucose, NEFA, triglyceride, and cholesterol concentrations.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society