The role of arachidonic acid (AA) on the development of adipose tissue is still controversial since its metabolites, i.e., prostaglandins, can either stimulate or inhibit preadipocyte differentiation in vitro. In the present study, we evaluated the effects of early postnatal supplementation of AA on body weight and adipose tissue development in guinea pigs. Male newborn guinea pigs were fed for 21 days (day 21) with diets (milk and pellet) supplemented (+AA) or not (−AA) with 1.2% (total fatty acids) AA. From day 21 to day 105 both groups were fed a chow diet. The 21-days-old +AA pups showed a twofold higher AA accretion in phospholipids associated with a two- to sixfold increase in several prostaglandins, such as 6-keto PGF1α (the stable hydrolysis product of PGI2), PGF2α, PGE2, and PGD2 in adipose tissue, compared with the −AA group. No difference in fat pad and body weight, aP2, and leptin gene expression in adipose tissue, fasting plasma glucose, free-fatty acids, and triglyceride concentration was observed between groups at day 21 or day 105. These results show that dietary supplementation of AA during the suckling/weaning period increases prostaglandin levels in adipose tissue but does not influence early fat mass development in the guinea pig.
- dietary fat
- polyunsaturated fatty acids
- metabolic programming
environmental factors, including nutrition, during pre- and postnatal life may affect the susceptibility to diseases later in life (17). The growing prevalence of overweight and obesity has raised interest in evaluating the role of early nutrition in the susceptibility to develop an excess of fat mass later in life. The late stage of gestation and the first year of life represent the most important periods for human fat mass development and consequently may be critical windows for nutritional programming of obesity.
Dietary fats during early infancy provide not only energy for growth but also supply essential fatty acids, linoleic acid (LA; C18:2n-6), and linolenic acid (ALA; C18:3n-3) required for a healthy development. LA and ALA serve as the essential precursors for producing their respective long-chain products (LC-PUFAs), arachidonic acid (AA; 20:4n-6), and both eicosapentaenoic acid (EPA; 20:5n-3) and docosahexaenoic acid (DHA; 22:6n-3). AA is the precursor of eicosanoids, including prostaglandins of the n-2 series (PGs), a family of compounds that play an important role in cellular processes (29). In adipose tissue, PGs control adipocyte formation, maturation, and function. PGJ2 derivatives and prostacyclin (PGI2) promote the in vitro differentiation of preadipocyte to adipocyte (8, 22). While several reports indicate that PGF2α inhibits the differentiation of preadipocytes (3, 21, 26, 28), controversial results exist for PGE2 and PGD2 with either neutral (28) or inhibitory (3) effects described. Finally, PGE2 has been shown to inhibit lipolysis in mature adipocyte and then may contribute to the hypertrophic development of adipose tissue (30). Therefore, adipogenesis is subject to both positive and negative regulators derived from AA metabolism and dietary intake of AA, or its precursors may influence fat mass development.
In vitro, the effects of AA supplementation on adipogenesis is still quite controversial since some studies showed promotion of differentiation of preadipocytes into adipocytes (9, 22), while others described inhibition (16, 24, 28). In addition, AA has been shown to inhibit lipogenic gene expression in mature adipocytes, suggesting potential antihypertrophic effects (20). Nevertheless, a high LA/ALA intake during prenatal plus postnatal life was shown to induce obesity in wild-type mouse but not in PGI2-receptor knockout animals, suggesting that PGI2 signaling contributes to adipose tissue development (19). Nevertheless, the impact of AA supplementation during early life on adiposity and body weight has never been investigated.
The use of rodents as relevant models of the human infant for studying the effects of PUFA and LC-PUFA intake early in life on adipose tissue development and obesity risk has some limitations. In rodents, the development of adipose tissue takes place after birth during the suckling period through hyperplasia and hypertrophy of fat cells. Humans diverge from most mammals by depositing significant quantities of body fat in utero and are consequently one of the fattest species on record at birth (14). At birth, fat mass represents 15% of body weight in human infants and 1% or less in rodents (32). Interestingly, the guinea pig (Cavia porcellus) shares with human the deposition of a significant amount of body fat in utero and presents 10–15% fat mass at birth (18, 32). Moreover, like in human, brown adipose tissue rapidly disappears after birth while white adipose tissue continues to expand (10). Therefore, the guinea pig appears an interesting and appropriate animal model for studying the effect of early nutrition on adipose tissue development and the metabolic consequences later in life.
Since PGs of the n-2 series have been shown to modulate adipogenesis in vitro, we hypothesized that AA supplementation will impact adipose tissue homeostasis during its postnatal development with potential long-term consequences. For that purpose, the effects of AA supplementation in the suckling/weaning newborn guinea pigs on body weight and adipose tissue development/metabolism were evaluated until adulthood.
MATERIALS AND METHODS
Animals and diets.
All experimental procedures were reviewed and approved by the ethical committee of the Swiss authorities (authorization no. 1633, “Service vetérinaire du canton de Vaud,” Lausanne, Switzerland). Thirty female guinea pigs (450–500 g, Dunkin-Hartley strain, Charles River, France) were mated with six males. All animals were fed a chow diet (3420.0.25, Provimi Kliba AG, Kaiseraugst, Switzerland). When the fetuses could be palpated, pregnant guinea pigs were housed individually. Offspring were kept with the lactating dam during the first 48 h. The mean of pup birth weight did not differ between groups. At day 2, only male newborns were kept and randomly assigned to each group. From day 2 and day 5, male newborn guinea pigs were in the presence of a nonlactating female guinea pig and were artificially bottle fed with experimental milk supplemented (+AA) or not (−AA) with ARASCO oil (Martek Biosciences, Columbia, MD). The artificial milk composition (Table 1) was designed according to Weaver et al. (31). The average milk intake was 6.6, 9.4, and 11.7 ml at days 2, 3, and 4. No differences between groups were observed. At day 5, pups were removed from the adult guinea pig and placed with other pups from the same group to limit stress. At day 7, animals were housed individually (to record individual food intake) until the end of the experiment (day 105). Between days 5 to 21, pups were fed with the solid experimental −AA or +AA diets only. This solid diet contained the same lipid composition than the experimental milks (Table 1). After day 21, both groups of guinea pigs were reared on a chow diet (Kliba 3420.0.25) until day 105. Pups were weighed every day from birth until day 21 and twice a week until day 105. Intake of solid food was monitored every day from day 8 to day 21 and then every 3 days until day 105. The calorie content of the experimental solid diets and chow diet was 337 kcal/100 g and 258 kcal/100 g, respectively. No statistically significant difference in body weight between the experimental groups and the reference group (pups reared by their lactating mothers on a chow diet) was observed until day 15. From day 15 to day 21, the reference group displayed a 20% higher body weight (P < 0.01) than the animals on the experimental diets.
At day 21 and day 105, guinea pigs were fasted overnight (16 h) and killed for blood and tissue collection. On the day of death, animals were anesthetized with isoflurane, and blood was drawn from the abdominal aorta.
Fatty acid analysis.
Total lipids of diet (25 mg), plasma (400 μl), or adipose tissue (50 mg) were extracted according to Bligh and Dyer (2) in the presence of internal standards: triheptadecanoin (Nu-Chek-Prep), dipentadecanoyl-sn-glycero-3-phosphocholine (Avanti, Switzerland), and heptadecanoin (Nu-Chek-Prep). The extracted lipids were separated by thin-layer chromatography (20 × 20 cm, Silica gel 60 F254; cat. no. 105715; Merck, Switzerland) by sequential elution with hexane-diethylether-acetic acid (70:30:1 ratio, respectively). The fatty acids from PL and TG were transesterified to their methyl esters in a 3% H2SO4 methanolic reagent for 3 h at 80°C, extracted with hexane, dried under N2, and resuspended into hexane. Fatty acid methyl ester separation was performed by automated gas-liquid chromatography (Hewlett-Packard 6890 series) by use of a flame ionization detector (280°C) and a DB-Wax column (30 m × 0.32 mm × 0.25 μm thickness). Authentic standard mixtures (GLC-85 FAMES; C24:0, C23:0, C19:0, C22:4n-6, C22:5n-3 from Nu-Chek-Prep; C20:5n-3 from Sigma, Buchs, Switzerland; C26:0 from Larodan) of fatty acid methyl esters were also injected to identify fatty acid methyl ester peaks. Fatty acid concentrations were determined by comparing gas-liquid chromatography peak area to the internal standards (TG C17:0, PL C15:0, and FFA C17:0).
Frozen adipose tissue was cut in pieces with a scalpel knife, and the required quantity of tissue (50 mg) was added to 3 ml distilled ethanol in siliconized glass tubes. A previously prepared mixture of internal deuterated standards (1 ng of each compound to be analyzed) was added in 10 μl of methanol to each sample. Samples were homogenized in a Polytron tissue homogenizer and then centrifuged at 2,200 rpm for 5 min. The supernatant was transferred to a clean glass tube, the pellets were resuspended in another 3 ml of ethanol, and the homogenization/centrifugation procedure was repeated. The combined supernatants were evaporated to complete dryness with a nitrogen stream, and the residue was dissolved in 50 μl ethanol, and 1 ml double-distilled water was added. The mixture was acidified to pH 3 and extracted with ethyl acetate. The ethyl acetate extract was washed to neutrality with water and taken to dryness. The residue was dissolved in 50 μl of acetonitrile/water (1:1) and stored at −20°C. On the day of analysis, the lower phase was carefully transferred to a siliconized tube and 25 μl (from about 30 μl total) were injected into a liquid chromatography-mass spectrometer (LC-MS/MS). Analysis utilized the selected ion monitoring mode for each analyte investigated based on operating parameters previously set up with authentic standards. Quantitation was carried out by using the Analyst software available with the Sciex 4000 API mass spectrometer and was based on response lines of authentic standards obtained for each batch of samples analyzed (1, 5). All LC-MS/MS analyses were carried out on blinded samples.
Plasma triglycerides and FFA concentrations were analyzed by using kits from Roche Diagnostic (Basel, Switzerland) and Wako Chemicals (Richmond, VA), respectively. Plasma glucose was determined with commercially available kits purchased from Sigma.
Design of aP2 and leptin primers.
The full cDNA sequence of aP2 or leptin gene from different species was aligned to select conserved sequences for both genes. Degenerated primers were designed (aP2: forward: 5′-TGTGYGAYGCHTTTGTRGG-3′, reverse: 5′-CATTCCASCACCA-3′; leptin: forward: 5′-GACACCAAAACCCTCATCAAG-3′, reverse: 5′-ACCTCYGTGGAGTAG-3′) to cover every species sequence. Adipose tissue mRNAs from guinea pig (prepared as described in Real-time PCR) were subjected to RT-PCR using a Taq polymerase (Q-Bio Taq; MP Biomedicals) and the following conditions: initial denaturation at 95°C for 5 min and cycled 40 times (aP2 gene: 94°C for 1-min denaturation, primer annealing at 60°C for 1 min, and extension at 72°C for 1 min; leptin gene: 94°C for 1-min denaturation, primer annealing at 55°C for 1 min followed by 58°C for 1 min, and extension at 72°C for 1 min; for both genes a final extension at 72°C for 5 min was performed). The PCR products were sequenced (aP2: 354 bp; leptin: 346 bp) by a service company (Value Read; MWG-Biotech). New primers were designed from these sequences (aP2: forward: 5′-AAGTGGGAGTGGGCTTTGC-3′, reverse: 5′-TCTTGTTATTGTGGTTGTTTTTCCA-3′). cDNAs from guinea pig adipose tissue were amplified by PCR by using a proofreading enzyme (Isis polymerase; MP Biomedicals) under the following conditions: initial denaturation at 93°C for 5 min and cycled 30 times (aP2 gene: 91°C for 1-min denaturation, primer annealing at 62°C for 1 min, and extension at 72°C for 1 min and 15 s). The PCR products were sequenced again. Finally, primers adapted for real-time PCR amplification (Primer express software; Applied Biosystem) were designed from the newly defined sequence of each gene.
Total RNAs were prepared from 100 to 200 mg adipose tissue using the RNeasy mini kit (Qiagen; Basel; Switzerland) according to the procedure described by the manufacturer and after a delipidation step. Reverse transcription was performed on 0.3 μg total RNA using the first-strand cDNA synthesis kit for real-time PCR (AMV; Roche Biomedical, Basel, Switzerland) with oligo(dT)15 as primer. Real-time RT-PCR analyses were performed in a fluorescent temperature cycler (GeneAmp PCR 5700 Sequence Detection System; Applied Biosystem) as previously described (6). The following sequences were used for guinea pig genes: actin: forward: 5′-GACAGCAGTTGGTTGGAGCAA-3′, reverse: 5′-ATCAAAGTCCTCGGCCACATT-3′; aP2: forward: 5′-GCCAGGAATTCGATGAAGTCA-3′, reverse: 5′- GGGCACCCCCATCTAAGGT-3′; leptin: forward: 5′-ACCACCCTGTCCTCCTCACA-3′, reverse: 5′-TCTGGAGATTTTCCAGGTCGTT-3′. The effects of treatments on gene expression were evaluated by calculating the relative expression level as follows: 2mean Ct genes of interest − mean Ct RPL-p0, using the raw cycle threshold (Ct) values.
All statistics were performed on the GraphPad InStat 3 software. Unpaired Student's t-test was performed with Welch correction if groups have a nonequal SD. Significance was reached when the two-tailed P value is < 0.05.
Dietary intervention and fatty acid compositions.
Male guinea pig pups were fed from 2 to 21 days of age a diet with 1.2% (total fatty acids) (+AA) or without AA supplementation (−AA). The fatty composition of the milk and pellet food, measured by gas chromatography, is presented in Table 2. At day 21, all groups were fed the same chow diet until day 105.
Fatty acid profiles were determined in different lipid fractions of retroperitoneal adipose tissue (r-AT) and plasma. At day 21, the quantity of saturated fatty acids (not shown), linoleic, and α-linolenic (Table 3) in the different lipid fractions did not significantly differ between groups. As expected, an increase of AA concentration was observed in all lipid fractions of the r-AT and plasma of the +AA group (Table 3). The AA concentration in the PL of r-AT was twofold higher (P < 0.01) in the +AA vs. −AA group (Table 3). This difference disappeared after feeding the chow diet. Indeed, the concentration of AA in the PL of r-AT at day 105 was 2.5 ± 0.2 and 1.3 ± 0.5 μg/100 mg AT in the −AA and +AA guinea pigs, respectively.
Eicosanoid profile in adipose tissue.
The impact of AA supplementation on the eicosanoid levels in r-AT was assessed in the 21-days-old pups. As shown in Fig. 1A, the concentration of the prostaglandin 6-keto F1α (6-keto PGF1α), the stable hydrolysis product of PGI2, was increased by twofold (P < 0.01) in the +AA pups, compared with nonsupplemented animals. While PGJ2, the natural ligand of the peroxisome proliferator-activated receptor-γ (8, 22) was barely detectable and was not different between groups, PGE2, PGF2α, and PGD2 were, respectively, approximately threefold (not significant), tenfold (P < 0.05), and twofold (P < 0.05) higher in the adipose tissue of +AA vs. −AA animals. Other products of AA, including lipoxin A4 (LXA4) and leukotriene B4 (LTB4) were also increased by about sixfold (P < 0.05).
At day 105, the concentration of most eicosanoids was markedly decreased in the r-AT of both groups (Fig. 1B), compared with those observed at day 21 (Fig. 1A). Although all of the animals were fed the same chow diet for 84 days, the +AA group still showed higher concentrations in LXA4 (5.7-fold, P < 0.01), LTB4 (8.7-fold, P < 0.01), and 6-keto PGF1α (2-fold, not significant) than the −AA group (Fig. 1B).
Energy intake, body weight, fat pad weight, and gene expression.
Growth curves (Fig. 2A) did not significantly differ between groups, even if body weight tended to diverge around puberty (i.e., 8–10 wk of age in male guinea pigs). Indeed, compared with the −AA group, the +AA young adults tended to present a lower body weight, at the end of the study (793.6 ± 23.7 vs. 854.0 ± 28.3 g), and a reduced body weight gain from day 21 to day 105 (570.1 ± 25.8 vs. 639.2 ± 22.3). During the experimental diet feeding, no difference of daily energy intake was observed between groups (Fig. 2B). The average daily energy intake on the chow diet was slightly but significantly increased between weeks 8 and 11 in the +AA group compared with the −AA group. This difference did not persist afterward (Fig. 2B), and the overall (weeks 4 to 15) energy intake was identical between the groups (data not shown).
As shown in Table 4, the weight of different fat pads and muscles did not differ between groups at day 21. At day 105, a consistent trend toward a decrease in the weight of all fat pads was found in the +AA group. The combined weight of soleus and gastrocnemius tended also to be reduced in the +AA animals but to a lesser extent (−6%) than the adipose depots (−13%).
To get further insight into the effects of AA supplementation on adipose tissue development, markers of adipogenesis, such as aP2 and leptin were analyzed by real-time PCR in the retroperitoneal fat depot. As shown in Fig. 3, A and B, aP2 and leptin mRNA expression significantly increased (P < 0.01) with age but did not differ between groups at both day 21 and day 105. At day 21, leptin mRNA was poorly expressed (relative to actin, the housekeeping gene), and a high variability of expression was observed in the −AA group, rendering difficult the interpretation of the results (Fig. 3C).
The plasma levels of free fatty acids, triglycerides, and glucose were not different between groups, at day 21 or day 105 (data not shown).
To our knowledge this is the first study demonstrating that dietary AA supplementation leads to an increase of eicosanoids in adipose tissue. Previous studies showed that intake of AA- or LA-rich diets increases eicosanoid production in different tissues including liver (23), uterus (4) in animals, and plasma in human (13).
Adipose tissue has been shown to secrete different eicosanoids, including 6-keto PGF1α, PGE2 (7), PGF2α (25), and Thromboxane B2 (12). To our knowledge, the capacity of adipose tissue to produce LXA4 and LTB4, respectively, involved in anti- and proinflammatory responses, has not been previously described. Among the PG family, 6 keto-PGF1α, PGD2, and PGE2 were the three major forms detected in the guinea pig adipose tissue. PGE2 and the stable hydrolysis product of PGI2 have been previously shown to be the two major PGs produced from adipocytes (11, 27).
The eicosanoids are formed by enzymatic oxygenation of free AA after its release from the cell membrane by phospholipase A2. Therefore, the increase of eicosanoid levels in the r-AT of the 21-days-old supplemented guinea pigs is certainly due to the twofold-higher AA accretion in the PLs of this fat depot compared with the nonsupplemented animals. The concentration of eicosanoids in the adipose depot were markedly decreased in all animals at day 105 compared with day 21, probably due to the lower accretion of AA in PLs observed after the chow diet feeding. Surprisingly, despite similar AA levels in the PL of r-AT in both groups of 105-days-old guinea pigs, the increased concentrations of 6-keto PGF1α, LXA4, and LTB4 observed at day 21 (2-, 6.5- and 5.5-fold, respectively; +AA vs. −AA) in the supplemented group were maintained at day 105 (2-, 5.7-, 8.7-fold, respectively: +AA vs. −AA). This phenomenon could be due to a programming effect of AA on the activity of enzymes, such as PGI2 synthase and leukotriene A4-hydrolase, involved in the synthesis of these eicosanoids, but this remains to be investigated. Nevertheless, this result demonstrates that AA supplementation during early life has a long-term effect on eicosanoid homeostasis, which may have physiological consequences.
While dietary supplementation of AA during the 19 first days of postnatal life markedly modifies PG profile in guinea pig adipose tissue, we did not observe any impact on fat mass development. Weight of four different fat pads and gene expression of markers of adipogenesis did not differ between the supplemented and nonsupplemented animals. Lien et al. (15) have previously found that accumulation of AA in different tissues of suckling rat pups was not associated with changes in body weight gain. Unfortunately, the weight of fat pads was not assessed in this study. Nevertheless, it has been recently proposed that AA promotes adipose tissue development and obesity through PGI2 signaling, in mice (19). Our results are not in favor of this hypothesis, since a twofold increase in 6 keto-PGF1α, the surrogate marker of PGI2, did not induce fat mass development. Nevertheless, the difference of animal models, dietary interventions (high LA/ALA ratio vs. AA supplementation) and, more importantly, of the time window of dietary interventions (mating-pregnancy-lactation for the mothers and postweaning for the offspring vs. 19 first days of postnatal life) may explain this discrepancy. It would be interesting to assess the impact of AA supplementation during the gestation period in the development of adipose tissue in the offspring guinea pig, knowing that as in humans, a large amount of fat is deposited during its fetal life.
In conclusion, we demonstrated that dietary supplementation of physiological levels of AA during the suckling/weaning period increases eicosanoids concentrations in adipose tissue but does not influence the early development of fat mass in the guinea pig.
Perspective and Significance
To date, the role of dietary fat and fat types as early determinants of childhood obesity has been poorly investigated, but studies in mice have highlighted potential detrimental effects of n-6 PUFA intake (19). By using an animal model with an adipose tissue development and a PUFA metabolism close to the human term infant we could not demonstrate detrimental effects of AA supplementation during early postnatal life on adipogenesis and obesity. Because of the increase in dietary intake of n-6 PUFAs over the last decades, it appears important to further investigate the consequences of early life exposure (through placenta and breast milk) to these dietary fats, for future recommendations to the pregnant and lactating mother.
The present work has been carried out in the frame of the European Community project entitled “Early Nutrition Programming-Long Term Follow Up of Efficacy and Safety Trials and Integrated Epidemiological, Genetic, Animal, Consumer and Economic Research”, EARNEST (Food-CT-2005–007036 FP 6 Priority 184.108.40.206 Food Quality and Safety).
The authors thank José Sanchez-Garcia, Ribeiro Manuel, and Massimo Marchesini for their assistance with the animal study and Drs. Clara Garcia-Rodenas and Marco Turini for helpful discussions.
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 © 2007 the American Physiological Society