Epidemiological studies in humans have shown that perinatal nutrition affects health later in life. We have previously shown that the ratio of n-6 to n-3 polyunsaturated fatty acids (PUFA) in the maternal diet affects serum leptin levels and growth of the suckling pups. The aim of the present study was to investigate the long-term effects of various ratios of the dietary n-6 and n-3 PUFA during the perinatal period on serum leptin, insulin, and triacylglycerol, as well as body growth in the adult offspring. During late gestation and throughout lactation, rats were fed an isocaloric diet containing 7 wt% fat, either as linseed oil (n-3 diet), soybean oil (n-6/n-3 diet), or sunflower oil (n-6 diet). At 3 wk of age, the n-6/n-3 PUFA ratios in the serum phospholipids of the offspring were 2.5, 8.3, and 17.5, respectively. After weaning, all pups were given a standard chow. At the 28th postnatal wk, mean body weight and fasting insulin levels were significantly increased in the rats fed the n-6/n-3 diet perinatally compared with the other groups. The systolic blood pressure and serum triacylglycerol levels were only increased in adult male rats of the same group. These data suggest that the balance between n-6 and n-3 PUFA during perinatal development affects several metabolic parameters in adulthood, especially in the male animals.
- blood pressure
- body weight.
it is now well recognized that impaired fetal and early postnatal growth confers an increased susceptibility for the development of adult chronic disease such as type 2 diabetes and cardiovascular disease (12, 20). This association of early growth and later disease could be a consequence of a genetic predisposition, an effect of environment or a combination of both. Early nutrition is an environmental factor that influences development and can cause adaptive and permanent changes in structure, physiology, and metabolism, that is, programming (19). In this context, nutritional components, such as polyunsaturated fatty acids (PUFA), play an important role in the regulation of many cell functions by influencing membrane fluidity and permeability (13, 23). Furthermore, PUFA and their metabolites have a direct effect on the regulation of gene expression (8). Thus variations in PUFA intake during the perinatal period would condition the development of metabolic and tissue functions, hence causing persistent changes into adulthood.
The ratio of n-6/n-3 PUFA in the milk affects growth, neurodevelopment, and immunoresponsiveness of the offspring (24). This ratio varies between 5:1 and 15:1 in milk of women from industrialized countries and is to a large extent defined by the maternal diet (10). To date, there are limited data on the long-term metabolic effects caused by variations of dietary fatty acids (FA) in early life. In rats, modification of early dietary PUFA intake has been shown to affect lipid metabolism and liver enzyme activity (7). Deficiency in n-3 PUFA during the first 9 wk of life resulted in hypertension in adult rats, regardless of subsequent supply (25). However, whether the ratio of n-6/n-3 PUFA in the maternal diet is of importance for physiological changes in the offspring in adulthood has not been investigated.
Recently, we have shown that the ratio of n-6/n-3 PUFA in the maternal diet affects serum leptin levels and growth of the offspring. Increased maternal intake of n-3 PUFA led to a decreased growth rate, reduced adipose tissue mass, and lower serum leptin levels in the offspring (16). Furthermore, the ratio of n-6/n-3 PUFA rather than the levels of n-6 PUFA in the maternal milk promoted body weight, growth of inguinal white adipose tissue and adipocyte size in the offspring. Modulation of dietary n-6/n-3 PUFA ratio and/or early leptin levels might have long-term effects on later metabolic parameters. This is of particular interest because it has recently been shown that the early leptin surge was important for long-term development (1, 4).
The aim of the present study was to investigate the long-term metabolic effects of various ratios of the dietary n-6 and n-3 PUFA during the perinatal period.
MATERIALS AND METHODS
Pregnant Sprague-Dawley rats (BK Universal, Stockholm, Sweden) were received on day 7 of gestation and kept in our animal facility under constant conditions of humidity (70–80%), temperature (22–25°C), and light (12:12-h light-dark cycle). The rats were housed individually in plastic cages with food and water ad libitum. Ten days before delivery, the rats were randomly assigned to one of three groups (9–10 animals in each group) receiving either the diet containing both n-6 and n-3 PUFA (n-6/n-3 diet), an n-3 PUFA-enriched (n-3 diet), or an n-6 PUFA-enriched (n-6 diet) diet. To obtain a substantial difference in n-6/n-3 FA ratios of the tissue phospholipids (PL) within the short neonatal period in rats, we used dietary intervention with extreme n-6/n-3 ratios in the diet, which at 3 wk of age resulted in the n-6/n-3 FA ratios in the serum PL of offspring at levels with reasonable physiological differences. Litter size was adjusted to 10 pups per litter at birth. Some pups from each litter were studied for the perinatal effects of dietary FA (16), and the rest of the pups were studied for possible long-term effects. Suckling pups (10 animals in each group) randomized from each litter were killed by decapitation at 3 wk of age (mixed gender). Truncal blood was collected, and sera were kept frozen (−20°C) until analyses of FA. At 3 wk of age, male and female pups (8–10 animals in each group) were randomly selected (one from each litter), separated and weaned onto ordinary chow and followed until 28 wk of age. Body weights were recorded regularly. Blood samples were collected from the tip of the tail of the rats at 28 wk of age on two occasions, and sera were kept at −20°C. Blood samples from the first occasion were taken in the morning for analyses of leptin, protein, cholesterol, and triacylglycerol. On the second occasion, blood samples for insulin and glucose determinations were taken in the morning after overnight fasting. The study was approved by the Animal Ethics Committee of Göteborg University.
The dams were fed one of three experimental pellet diets (Morinaga Milk Industry, Tokyo) for the last 10 days of gestation and throughout lactation. The diets differed only by lipid composition: 7 wt% soybean oil (n-6/n-3 diet), sunflower oil (n-6 diet), or linseed oil (n-3 diet) (Table 1). The data on major components, salt, and vitamins have been obtained from the manufacturer. The total metabolizable energy of each of these diets was 13.9 MJ/kg. The ordinary chow (rat and mouse standard diet; B and K Universal, Grimston, Aldbrough, Hull, UK) contained 19% protein, 5% fat, 4% crude fiber, and 5.5% ash. The fat composition of ordinary chow is given in Table 1. The total metabolizable energy for this diet was 14.0 MJ/kg.
When the rats were 28 wk of age, food consumption for each cage was recorded once a day for 8 days (n-6/n-3 group, n = 8, 2 or 3 rats per cage; n-6 group, n = 10, 2 or 3 rats per cage; n-3 group, n = 10, 2 or 3 rats per cage). They were given the same amount of food, and the food intake was measured the following day by subtracting the uneaten food. Food intake was calculated in grams per rat and per day.
Fatty acid analysis.
Total serum lipids were extracted according to Folch et al. (11), fractionated on a single SEP-PAK aminopropyl cartridge (Waters, MA), according to the method described previously (17), and PL fractions were collected. The fractions of PL were transmethylated in methanolic-HCl-3N at 90°C over 4 h. The FA methyl esters were separated by capillary gas-liquid chromatography in a Hewlett-Packard 6890 gas chromatograph equipped with a 30 m × 0.25 mm SP-2380 column; film thickness was 20 μm. Helium at 1.1 ml/min was used as a carrier gas. The injector and detector temperatures were 250°C. The column oven temperature was programmed from 60°C to 230°C at a heating rate of 8°C/min up to 155°C, 1.5°C/min up to 180°C, and thereafter 6°C/min. The separation was recorded with Hewlett Packard GC Chem Station software (HP GC, Wilmington, DE). With 21:1 used as an internal standard, the FA methyl esters were identified by comparison with retention times of pure reference substances (Sigma Aldrich Sweden, Stockholm, Sweden). The inter-assay coefficients of variation were 3.0 and 3.5% for palmitic and linoleic acids (n = 10), respectively.
Plasma glucose was determined by glucose oxidase/PAP assay (PAP refers to peroxidase, aminoantipyrine, and phenol), serum protein by biuret reaction, serum cholesterol by Infinity cholesterol reagent (Sigma Diagnostics, St. Louis, MO), and serum triacylglycerol by glycerol peroxidase/PAP; and all samples were analyzed in duplicate in a biochemical analyzer Cobas MIRA (ABX Diagnostics, Parc Evromedicine, Montpellier, France). Plasma insulin was determined with an enzymatic immunoassay (Rat insulin ELISA kit, Mercodia, Uppsala, Sweden). Leptin concentrations in serum were measured by a rat leptin radioimmunoassay (RIA; Linco Research, St. Charles, MO), and all samples from one experiment were analyzed in duplicate in the same assay.
Systemic arterial pressure and heart rates in conscious rats were measured at 28 wk of age by tail-cuff plethysmography (6). The tails were warmed 15 min before measuring. Six tail-cuff measurements were obtained for each data point. The highest and the lowest readings were discarded, and at least three clear readings were averaged to obtain each data point.
Values are presented as means ± SD. The data were analyzed by one-way ANOVA followed by Fisher's post hoc paired least significant difference test. A value of P < 0.05 was considered statistically significant.
FA composition of the serum PL.
Table 2 shows the FA composition of the serum PL in the rat offspring at 3 wk of age from dams receiving diets with different n-6/n-3 PUFA ratios. The total n-6 to n-3 ratios of serum PL differed between the n-3, n-6/n-3, and n-6 diets being 2.5, 8.3, and 17.5, respectively (Table 2). The FA composition of the serum PL in the offspring of the dams on the n-6/n-3 and on the n-6-diet deviated less from each other compared with the n-3 diet; with the exception of significantly higher levels of α-linolenic 18:3(n-6) and lower levels of docosahexaenoic 22:6(n-3) acids in the n-6-group compared with the other two diets. Consequently, the ratios of 20:4(n-6)/22:6(n-3) in the n-6-group were higher compared with the n-6/n-3 group. The levels of n-6 FA in serum PL in the n-3 group of rat offspring were significantly decreased, with an increase in the levels of myristic 14:0, palmitic 16:0, α-linolenic 18:3(n-3), and eicosapentaenoic 20:5(n-3) acids compared with the other diet groups. In the n-3 group, the proportions of saturated fatty acid and monounsaturated fatty acid were higher, whereas the level of PUFA was lower compared with the other two groups. The unsaturation index (UI) was elevated in the serum PL of the pups from the n-6/n-3 group compared with both the n-6 and n-3 groups.
At 28 wk of age, there were no significant differences in the FA composition of the serum PL between the adult rats originally fed the n-6, n-3, or n-6/n-3 diets during the suckling period (data not shown).
Body weight and food intake.
At 3 wk of age, the mean body weight of the male pups of the dams fed the n-3 diet was significantly lower (P < 0.05) than that of the pups suckling the dams on the n-6/n-3, or the n-6 diets (39.0 ± 3.0 g vs. 51.4 ± 4.5 and 50.6 ± 3.6 g, respectively). In the female 3-wk-old pups, a significant difference (P < 0.05) was found in the mean body weight among all three diet groups 41.1 ± 2.9 vs. 53.7 ± 4.2 and 46.5 ± 3.1, respectively. At 8 wk of age, the male offspring in the n-6 group showed a significantly decreased body weight compared with the n-6/n-3 group (Table 3). The mean body weight of the offspring receiving the n-6 diet or the n-3 diet during the perinatal period did not differ until 24 wk of age when the female offspring from the n-6 group had a lower body weight than those from the n-3 group (P < 0.05).
At 28 wk of age, the average food intake for the male n-6/n-3 rats was significantly higher (31.4 ± 0.5a g/day per rat) compared with the n-3 rats (27.7 ± 2.0b g/day per rat, where values with unlike letters are significantly different) and the n-6 rats (28.0 ± 1.9b g/day per rat) (P < 0.05). In the female offspring a similar significant difference in food intake was observed, being 20.0 ± 0.4a, 19.1 ± 0.3b, and 17.4 ± 0.8c g/day per rat in the n-6/n-3, the n-3, and the n-6 groups, respectively. However, there was no difference in food intake related to 100 g body wt in either diet groups or gender (data not shown).
Circulating protein, triacylglycerol, cholesterol, leptin, glucose, and insulin levels.
No differences were observed in the serum levels of protein, glucose, and leptin between the dietary groups in the adult male and female rats at 28 wk of age (Table 4). The triacylglycerol levels were significantly higher in the n-6/n-3 male rats compared with the n-3 and n-6 groups, while there were no differences in the female rats. The cholesterol levels were similar in the male animals and significantly higher in the n-6 female rats compared with females from the n-6/n-3 and n-3 groups (Table 4). The fasting plasma insulin levels in the n-6/n-3 group of adult rats were significantly higher for both males and females compared with those in the other groups. There was no difference in the insulin levels between the n-3 and the n-6 diet groups (Table 4).
Blood pressure and heart rate.
The systolic blood pressure was significantly increased in the adult male rats, but not in the females, receiving the n-6/n-3 diet during the perinatal period compared with the other groups (Table 4). The blood pressure did not differ between the rats fed the n-6 or the n-3 diets during the perinatal period. The heart rates did not differ between the groups (in females 356 ± 34.6, 357.2 ± 52.6, and 336.9 ± 21.3 beats/min and in males 352.0 ± 36.0, 325.7 ± 32.1, and 346.6 ± 32.1 beats/min in the n-3, n-6/n-3, and n-6 groups, respectively).
Our study shows that the dietary supply of PUFA during the perinatal period in rats had different long-term effects in adulthood, associated with the specific ratio of the n-6/n-3 PUFA in the maternal diet. A higher body weight and fasting insulin levels in the adult rat offspring of both genders were associated with n-6/n-3 FA ratio of 8.3 in serum PL at 3 wk of age. Furthermore, the adult male rats of this diet group also had higher serum triacylglycerol levels and elevated systolic blood pressure compared with the females of the same diet group, compared with the other groups. In addition, we have previously shown that the offspring of the n-6/n-3 group had higher weight at 1 wk of age compared with the other groups (16). This is an interesting observation with respect to the observed difference in the pattern of development of obesity, diabetes, and blood pressure in humans born with high or low birth weight (9, 18).
The composition of nutrients during early life has been shown to have an impact on later growth and development of the offspring (24). The potential role of PUFA has not been fully explored. Variations in the maternal milk PUFA levels lead to rapid adaptations of the FA compositions of tissue PL in the offspring (5, 16). Although at 3 wk of age there were marked differences of serum PL ratio of n-6/n-3 FA in the n-3, n-6/n-3, and n-6 diet groups (2.5, 8.3, and 17.5, respectively), the concentrations were in physiological range, suggesting that the pups were neither n-6 nor n-3 deficient in any group at 3 wk of age (22). In our previous report, we also found differences in serum leptin levels and body and adipose tissue mass in the pups. The n-6/n-3 group exhibited lower mRNA for leptin in adipose tissue and higher serum leptin levels than the n-3 group, whereas the n-6 group were largely similar to the n-6/n-3 group (16). The lower weight in the n-3 group of suckling offspring might be due to reduced milk yield (2) and/or altered milk FA composition. We did not evaluate directly milk yield due to methodological difficulties; however, there were no differences in serum glucose, protein, triacylglycerol, and cholesterol levels in 3-wk-old pups from different dietary groups, indicating indirectly their similar nutritional status (16). Therefore, we assume that the most likely cause of the decreased body weight at 3 wk of age in the pups of the n-3 group is the result of a different n-6/n-3 ratio rather than difference in milk yield. From 8 wk of age the male and female rats from the n-6/n-3 group gained weight at a higher rate compared with the other dietary groups. Furthermore, the n-3 group caught up in body weight with the n-6 group at 8 wk and 12 wk for the females and males, respectively. The differences in weight gain and other parameters observed in the adult rats might be the long-term consequences of the perinatal n-6/n-3 PUFA ratio, but a relationship to the early leptin levels has to be further investigated. There was no difference in leptin levels between the groups at wk 28 in either gender, but we cannot exclude the possibility that the early disturbances in leptin expression and serum levels might have induced changes in leptin receptors resulting in changed regulation. In a recent study by Bouret et al. (4), neonatal deficiency of leptin in ob/ob mice induced reduced expression of leptin receptors in the hypothalamus, which was not reversible if leptin was administered after the neonatal period. Disturbances in the hypothalamic satiety regulation in parallel to those changes reported by Bouret et al. (4) would result in increased food intake, accelerating body growth. Increased body weight could also be due to changes in food efficiency or energy expenditure, which have to be analyzed in further studies.
The increased weight gain was associated with higher fasting insulin levels in adult rats from the n-6/n-3 group compared with the n-3 and n-6 dietary groups both in male and female rats. These higher insulin levels in combination with higher serum triacylglycerol levels indicate alterations in metabolic state of the n-6/n-3 males. In this context, it has to be considered that in humans the males are more prone to develop metabolic syndrome than females (26). The n-6 females, however, showed higher serum cholesterol levels. Hence, the various dietary programs had diverse outcomes in the adult animals differentiated by gender.
Evidence from chronic insulin infusion studies in rats suggests that hyperinsulinemia induces hypertension (15), and a similar association has also been reported in humans (14). Further studies have to be directed to clamp studies to investigate insulin resistance, since that has been shown in rats to precede the development of hypertension (3), but the precise mechanisms are unknown. The absence of hypertension in the female rats from the n-6/n-3 group despite the higher insulin levels may reflect a protective effect of estrogen, which is known to delay the onset of diet-induced hypertension caused by endothelial dysfunction (21).
Thus an increased body weight with enhanced fasting insulin and triacylglycerol levels and higher blood pressure in adult male rats were linked to specific dietary n-6/n-3 PUFA ratio in the perinatal period and possibly to early leptin homeostasis. These data suggest that the adult metabolism can be programmed by the balance of n-6/n-3 PUFA in the perinatal period, possibly by influencing gene expression. The developmental order of events, as well as the mechanisms involved, remains to be defined.
This study was supported by grants from the Swedish Medical Research Council (4995), Göteborg Masonic Order, the Royal Society of Arts and Sciences in Göteborg, Wera Ekström's foundation, and Swedish Nutrition Foundation.
We thank Y. Yamashiro and Morinaga Milk Industry Co. for advice and support in obtaining the diets. The authors also express their appreciation to T. Elfverson and B. Holmberg for excellent technical assistance.
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