Objective: The suckling period is one potentially “critical” period during which nutritional intake may permanently “program” metabolism to promote increased adult body weight and insulin resistance in later life. This study determined whether fructose introduced during the suckling period altered body weight and induced changes in fatty acid transport leading to insulin resistance in adulthood in rats.
Methods and Procedures: Pups were randomly assigned to one of four diets: suckle controls (SCs), rat milk substitute formula (Rat Milk Substitute), fructose-containing formula (Fructose), or galactose-containing formula (Galactose). Starting at weaning, all pups received the same diet; at 8 weeks of age, half of the SC rats began ingesting a diet containing 65% kcal fructose (SC-Fructose). This continued until animals were 12 weeks old and the study ended.
Results: At weeks 8, 10, and 11, the Fructose group weighed more than SC and SC-Fructose groups (P < 0.05). At weeks 8 and 10 of age, the Fructose group had significantly higher insulin concentrations vs. rats in the SC-Fructose group. 3H-Palmitate transport into vesicles from hind limb skeletal muscle was higher in Fructose vs. SC rats (P < 0.05). CD36 expression was increased in the sarcolemma but not in whole tissue homogenates from skeletal muscle from Fructose rats (P < 0.05) suggesting a redistribution of this protein associated with fatty acid uptake across the plasma membrane. This change in subcellular localization of CD36 is associated with insulin resistance in muscle.
Discussion: Consuming fructose during suckling may result in lifelong changes in body weight, insulin secretion, and fatty acid transport involving CD36 in muscle and ultimately promote insulin resistance.
Obesity continues to be a major health problem worldwide. At an individual level, total caloric intake, together with decreased physical activity, is clearly among the key factors that contribute to the increased prevalence of obesity (1,2). The alarming increases in body weights have been thought to be due, at least in part, to high consumption of dietary fat and accompanying increased energy intake. However, in the United States, and likely similarly in Canada, the absolute and relative intake of fat is declining (3). On the other hand, the amount of fructose in the diet has increased by fourfold to fivefold over the past few decades (4,5).
Prior to the introduction of high fructose corn syrup in the mid1970s, the main sources of fructose were fruits, fruit juices, honey, and sucrose. Today, the major sources of fructose are soft drinks, sweetened beverages, and foods such as baked goods, cereals, and prepared desserts (5) in which high fructose corn syrup is used as the sweetener. In 1978, average fructose intake in children (>1 year of age), adolescents, and adults in the United States was 36–40 g/day, of which ∼15 g/day was from naturally occurring sources (fruits, vegetables), and the remaining was from high fructose corn syrup and sucrose (5). Today, fructose intake from fruits and vegetables remains at ∼15 g/day (5), but total intake is estimated at 97 g/day; the per capita use of high fructose corn syrup has increased 100-fold from 0.23 kg in 1970 to 28.4 kg in 1997.
The specific effects of fructose intake on weight gain, particularly during childhood, have not been established. Children's consumption of sugar-sweetened beverages increased by 150% from 1965 to 1996 (6) and there are now several studies reporting that such drinks contribute a higher proportion of total energy intake in overweight vs. normal-weight children and adolescents (3,7,8,9). Whether fructose specifically is causal in the path to overweight or obesity in these children is not known (10).
Animal models could help to shed light on the question of whether dietary fructose introduced early in life leads to excessive weight gain. In a series of studies using an artificial rearing of rat pup model, Patel et al. have shown that introduction of a high-carbohydrate diet during the suckling period contributes to the development of obesity, insulin resistance, reduced insulin secretion, and increased risk of type 2 diabetes (11,12,13,14). These changes are an example of “metabolic programming.” In this context, programming is defined as lasting metabolic effects of a nutritional change initiated during a sensitive period of the lifespan. Whether fructose intake during this early life period could have programming effects that are comparable to those observed with alteration in the macronutrient composition of the diet is not known.
Fructose feeding in weanling and young adult rats and other rodents (15,16,17) is a well-established model of the metabolic syndrome, with its hallmark sequelae of insulin resistance (18,19,20,21). Generally, insulin resistance is highly associated with elevated intramyocellular triglyceride content (22,23,24) and in obese Zucker rats (25), high fat-fed rats (26) and obese, insulin-resistant humans (27), the increased lipid deposition in muscle is secondary to increased fatty acid transport. In healthy muscle, long-chain fatty acid transport involves the combined action of at least two structurally unrelated proteins, i.e., CD36 (also known as fatty acid translocase; 88 kd) and FABPpm (plasma membrane fatty acid-binding protein; 43 kd) (28). The molecular mechanism by which CD36 and FABPpm facilitate fatty acid transport is still not known, though based on their membrane topology in muscle, these proteins will not act as classical transporters. Indeed, the area of fatty acid uptake continues to be the focus of intense study and considerable debate (29,30). It has been proposed that proteins including CD36 and FABPpm could be involved in trapping of fatty acids (31), after which a rapid flip-flop rate drives transmembrane passage (32). Several extensive reviews of the mechanisms of fatty acid uptake have been published recently (32,33,34,35,36).
Interestingly, in healthy muscle, fatty acid transport is induced by insulin. This involves the translocation of CD36, but not FABPpm, from intracellular membrane compartments to the sarcolemma (37). Interestingly, in heart and skeletal muscle of prediabetic rodents and humans, there is a permanent redistribution of CD36 to the sarcolemma in skeletal muscle (33,38). Using succinimidyl-esters of long-chain fatty acids, which specifically block CD36, we were able to demonstrate that the resulting relocation of CD36 to the plasma membrane results in increased fatty acid uptake, and this is causal to intramuscular lipid accumulation and the onset of insulin resistance (13).
The objective of this study was to determine the effects of fructose introduced during the suckling period on body weight and tissue specific fatty acid uptake in adult rats. We used an artificial rearing model to introduce fructose to rat pups during the suckling period. We also compared the effects of introducing fructose early in life vs. in the adult period. We hypothesized that exposure to a diet containing fructose during the suckling period would program metabolism in such a way to promote increased body weight in adulthood. Because high-fructose diets fed to rodents are known to cause insulin resistance in skeletal muscle (39), adipose tissue (40), and liver (41), we examined fatty acid transport in these three tissues as possible target tissues of metabolic programming effects.
Methods and Procedures
Animals and diets
Pregnant Sprague-Dawley rat dams (Charles River Laboratories, Wilmington, MA) were housed individually in shoebox cages with free access to water and rat chow (5001 Rodent diet; Canadian Lab Diets, Leduc, Alberta, Canada) under standard conditions (12 h light/dark cycle 22 °C). Two days after birth, litters were culled to 12 pups/l. At 12 days of age, pups from all litters were combined and then randomly assigned to one of the following preweaning diets: dam-reared, suckle controls (SCs), rat milk substitute formula (Rat Milk Substitute), fructose-containing formula (Fructose), or galactose-containing formula (Galactose). The Rat Milk Substitute has been used by others (42,43,44) and is similar in macro and micronutrients to rat breast milk (45). The Fructose and Galactose formulae were rat milk substitute-based diets in which 50% of the lactose was substituted with fructose or galactose, respectively. Rat Milk Substitute, Fructose, and Galactose formulae were fed using the artificial rearing technique described below. All diets contained the same amount of total carbohydrate, the identical amounts and sources of proteins and lipids, and were isocaloric. Animals remained on their assigned preweaning diet until 19 days of age, at which time they were weaned to a purified control diet (AIN 93; Dyets, Bethlehem, PA). In order to compare the effects of feeding fructose in the suckling vs. adult period, one half of the SC group was given a high fructose (65% kcal) purified diets (Dyets, Bethlehem, PA) from 8 to 11 weeks of age, (SC-Fructose). The study was ended when all rats were between 11 and 12 weeks of age. All experimental procedures were approved by the Faculty of Agriculture, Forestry and Home Economics Animal Policy, and Welfare Committee at the University of Alberta, (Edmonton, Alberta, Canada) in accordance with the guidelines of the Canadian Council on Animal Care (Ottawa, Ontario, Canada).
Artificial rearing of rat pups
Rat pups were artificially reared from postnatal days 12 to 19, based on the procedures outlined by Ward et al. (42,46) with minor modifications. In brief, at 12 days of age, pups were lightly anesthetized (isofluorane; Abbott Laboratories, Saint-Laurent, Quebec, Canada) and fitted with a cheek tube made from polyethylene tubing (PE10Clay Adams, Parsippany, NJ). Cheek tubes were connected to syringes containing one of the three preweaning diets on automated syringe pumps (Harvard Apparatus, South Natick, MA) set to deliver diet for 12 min each hour. The volume of diet given to pups fed by artificial rearing was calculated to try to match the growth rates of the SC group. During this time, each rat pup was housed individually in a plastic cup that floated freely in a temperature controlled water bath (38–40 °C). Body weight was measured daily, and the health of each pup was checked frequently throughout this period of the study.
Regularly monitored variables
From weaning (21 days) onward, body weight and food intake were measured weekly and biweekly, respectively. Fed-state blood samples were drawn biweekly from the tail vein, and plasma was separated and stored at −20 °C until analyzed. Glucose (glucose oxidase, Point Scientific, Lincoln Park, MI) and triglyceride (Diagnostic Chemicals, Charlottetown, Prince Edward Island, Canada) concentrations were determined using spectrophotometric methods (SpectraMax 190, Sunnydale, CA) while insulin and leptin concentrations were determined using radioimmunoassay (Linco Research, St. Charles, MO). At the end of the study, when rats were between 11 and 12 weeks of age, rats were killed by an overdose of sodium pentobarbital (MTC Pharmaceuticals, Cambridge, Ontario, Canada) and the pancreas, liver, and retroperitoneal and epididymal fat pads were quickly removed and weighed. Muscle from the hind limb was excised and used, along with liver and retroperitoneal fat, for assessing palmitate uptake into giant membrane vesicles (described below).
Determination of fatty acid transport and membrane-associated proteins that facilitate fatty acid uptake
Giant membrane vesicles were prepared from skeletal muscle, liver, and adipose tissue using an isolation procedure used previously (47,48). These vesicles are solely derived from plasma membrane, and are virtually uncontaminated with intracellular membranes. Previous studies have demonstrated that fatty acids are taken up across the vesicle membrane, with small cytoplasmic fatty acid-binding proteins likely acting as a luminal sink (49). Fatty acid transport into membrane vesicles was carried out as previously described (39,40,49).
CD36 and FABPpm protein were determined in whole tissue homogenates and in aliquots of the vesicles using western blotting as described previously (33,37,47). Data from the western blots were quantified relative to the amount of protein loaded in each lane. The vesicle content of transporters was interpreted as their localization in the sarcolemma. The antibody MO25 to CD36 was kindly provided by Dr Tandon, Otsuka Maryland Medicinal Laboratories Rockville, MD. FABPpm antiserum was kindly provided by Dr Calles-Escandon, Wake Forest University School of Medicine and Baptist Medical Center, Winston-Salem, NC.
All results are presented as means ± s.e.m. For the variables measured regularly throughout the study, a two-way ANOVA, with time as the repeated factor, was used to test differences among dietary treatments. For variables measured only at the end of the study, a one-way ANOVA was used to assess differences among diet groups. Statistical significance was accepted at P < 0.05. Treatment comparisons were carried out by least squares, using the PDIFF option in SAS v.8 (SAS Institutes, Cary, NC).
Regularly monitored variables: body weight
On the first day of artificial rearing, the body weights of all pups were similar (Figure 1). All artificially reared groups (Fructose, Rat Milk Substitute, and Galactose) grew similarly from days 12 to 19 while pups in the SC group were heavier than those fed the Fructose or Galactose formula on days 14, 15, 17, and 18 (P < 0.05). On day 19, pups in the SC group were heavier compared to all artificial rearing groups (P ≤ 0.001). The differences in body weight between the SC and artificial rearing groups were due to technical difficulties related to pump operation and subsequent delivery of the diet.
There were no significant differences in body weight among any of the groups from 3 to 7 weeks of age (Figure 2). Between weeks 8 and 11, the Fructose group weighed more than SC and SC-Fructose groups and the difference reached statistical significance at weeks 8, 10, and 11 (P < 0.05). Throughout the study, the Rat Milk Substitute and Galactose groups had similar body weights and they were not significantly different from any of the other groups.
Food intake was not significantly different among diet groups at any point during the postweaning period (data not shown). Between weeks 4 and 8 of age, food intake increased to ∼30 g/day and remained there for the duration of the study. Food intake as a function of body weight (feed efficiency) did not differ significantly between groups at any time during the study (data not shown).
Fed-state plasma glucose, insulin, triglyceride, and leptin concentrations
Plasma glucose concentrations were similar among the diet groups throughout the study except at week 11 when glucose was higher in the SC-Fructose group vs. all other groups (Table 1; P < 0.05). At weeks 8 and 10 of age, rats in the Fructose group had significantly higher insulin concentrations vs. rats in the SC-Fructose group (Table 1; P < 0.05 at both time points). Although the rats from the Fructose group generally had higher plasma insulin concentrations compared to all other groups, the differences did not reach statistical significance. There were no significant differences in plasma triglyceride concentrations among the different diet groups at any time during the study. Plasma leptin concentrations were higher in the Fructose group compared with all other groups at weeks 8 and 10 (P < 0.05).
Table 1. Fed-state plasma glucose, insulin, triglyceride, and leptin concentrations of rats from one of five diet groups (suckle controls (SCs), Rat Milk Substitute, Fructose formula, Galactose formula, SC-Fructose diet from 8 to 12 weeks) at different times throughout the postweaning period
The average pancreas weight of rats from the Fructose group was significantly less than the Rat Milk Substitute and SC-Fructose groups (1.4 ± 0.13 g vs. 1.8 ± 0.12 g vs. 1.8 ± 0.13 g, respectively; P < 0.05). Epidydymal fat pads from rats in the Fructose group were heavier than those from rats in the SC-Fructose group (12.44 ± 1.72 g vs. 7.46 ± 0.71 g; P < 0.01); these differences were not statistically significant when corrected for body weight. Liver weights did not differ significantly among any of the groups.
Fatty acid transport and membrane-associated proteins that facilitate fatty acid uptake
3H-Palmitate transport into giant vesicles prepared from skeletal muscle of rats in the Fructose group was markedly higher (∼1.7-fold) than from rats in the SC group (P < 0.05). Differences in transport were not statistically different between rats in the Fructose vs. other diet groups (Figure 3), and were similar between the Rat Milk Substitute, SC, Galactose, and SC-Fructose groups. There were no differences in 3H-palmitate transport into membrane vesicles derived from liver or adipose tissue of any of the diet groups
CD36 expression was also higher (i.e., 1.8-fold) in muscle giant vesicles from the Fructose group relative to the SC group. In the other diet groups (Rat Milk Substitute, Galactose, and SC-Fructose), there was no difference with the SC group in CD36 expression (P < 0.05; Figure 4). In whole muscle tissue homogenates, there was no difference in CD36 expression among different diet groups, suggesting that there was proportionately more CD36 residing in the sacrolemma than the intracellular compartment. CD36 expression was not different in giant vesicles or tissue homogenates from liver or adipose tissue of any of the diet groups (data not shown).
FABPpm expression was unaltered by diet treatment in giant vesicles from skeletal muscle (expression relative to SC = 1.0: Rat Milk Substitute = 1.01, Fructose = 1.14, Galactose = 0.94, SC-Fructose = 1.34) and in the whole muscle tissue homogenates (expression relative to SC = 1.0: Rat Milk Substitute = 0.90, Fructose = 1.14, Galactose = 0.94. SC-Fructose = 1.34). FABPpm in homogenates and giant vesicles from liver and adipose tissue did not differ among diet treatment groups (data not shown).
Controversy to the role of dietary fructose in promoting excessive body weight gain has come to the forefront, with particular interest in whether fructose intake in children may be particularly problematic (5,8,9). Results from this study suggest that in an animal model using artificial rearing to introduce fructose to rat pups during the suckling period, body weight was increased in adulthood as compared with rats suckled by their mothers. Introduction of dietary fructose starting in adulthood did not promote excessive weight gain, suggesting that the timing of the fructose ingestion affects body weight gain differently.
Rats that were artificially reared but given the Rat Milk Substitute or Galactose diet during suckling had body weights that were not statistically different from rats in the SC and the Fructose groups throughout the study. The method of feeding could also contribute to this increase. Because body weights of animals that were artificially reared were similar at weaning, and food intake throughout the postweaning period was similar in all groups, the ingestion of fructose during suckling may have promoted energy storage, rather than promoting increased energy intake. It is not possible from the data collected to know whether the increased energy storage is due to changes in basal metabolism or reflects a reduction in expenditure due to reduced activity by these rats. Vickers et al. have noted that maternal undernutrition during gestation can lead to offspring with reduced locomotor activity in later life (50). The hypothesis that ingestion of fructose in the suckling period reduces activity levels has important implications for humans and should be studied in more detail.
Other studies in which a diet high in fructose is fed or a fructose solution was given in the drinking water to rodents have reported increased weight gain (18,51,52,53,54,55), no change in body weight (17,56,57,58) or reduced weight gain (59) relative to those fed a control diet. Results from the current study suggest that fructose introduced at different ages may have different effects on weight gain, because higher final body weights were observed in those fed fructose early in life rather than as adults. Variation in weight gain with fructose feeding reported in the literature could reflect different ages for introducing a high-fructose diet. The fact that in this study younger animals gained more weight than older animals when given fructose could have importance in the childhood obesity area, as has been suggested by others (6,7,8,9).
The body composition of rats given fructose in the suckling diet is not known from this study. The absolute weight of epididymal fat pads was significantly higher in rats fed fructose in the suckling period vs. those fed fructose in adulthood (SC-Fructose group), although these differences disappeared when fat pad weight was corrected for body weight. The higher fat mass was associated with increased leptin concentrations at 8 and 10 weeks in the Fructose group, confirming that total body fat mass was partly responsible for the increased body weight. Only fat accumulated into “typical” fat depots was measured, but additional lipid may have also been deposited into areas outside of such depots. Detailed body composition analysis was not part of this study because of the tissue requirements for other measures. Future studies that measure the location of lipid deposition, for example using nuclear magnetic resonance spectroscopy, would be helpful in assessing this possibility.
An increased plasma insulin concentration in the Fructose vs. the SC-Fructose group was observed after 8 weeks of age. The origin of the increased insulin concentrations is not clear, because it could reflect changed insulin secretion or reduced insulin sensitivity. Although the insulin response to acute fructose ingestion is small, we and others (60,61) have reported that chronic feeding of a high-fructose diet to young rats leads to increased insulin concentrations in the absence of any differences in body weight (51). This suggests that dietary fructose may lead to β-cell dysfunction and hypersecretion of insulin independently of increases in insulin resistance. Such changes would set the stage for permanent CD36 relocation in skeletal muscle, which could then promote intracellular lipid accumulation and contribute to the insulin resistance that results from chronic fructose feeding (15,17,62,63,64,65). The earliest blood samples taken in this study were at 8 weeks of age, but samples collected from a limited number of animals in the 3–6 week age range (n = 3–4/time point) suggest that hyperinsulinemia is present earlier and persists after cessation of dietary fructose intake.
Fatty acid transport into skeletal muscle, as measured with giant membrane vesicles, was elevated in Fructose vs. SC rats, and the sarcolemmal content of CD36 in skeletal muscle from Fructose rats was increased. Other studies have shown this combination of effects to be associated with insulin resistance in muscle (33,37). Coort et al. (25) proposed a hypothetical model for the development of insulin resistance in cardiomyocytes, whereby permanent relocation of CD36 to the plasma membrane is likely to facilitate fatty acid uptake and lead to intracellular accumulation of fatty acid metabolites, such as diacylglycerols and ceramides. In turn, these metabolites would activate ser/thr kinases, such as protein kinase C isoforms, that are known to inhibit tyrosine phosphorylation of the insulin receptor substrate and insulin-stimulated glucose uptake (66). Using this model as a basis, one could envision a sequence of events whereby fructose feeding during the suckling period leads to chronically increased insulin concentrations and promotes the permanent redistribution of CD36 to the sarcolemma in skeletal muscle followed by insulin resistance. The tissue-specificity of our results (i.e., increased fatty acid uptake associated with permanent redistribution of CD36 in skeletal muscle, but not in liver or adipose tissue) suggest that different mechanisms could be acting in different tissues, even with the same stimulus (29,30).
It is possible that in the early fructose-fed rodent model, the redistribution of CD36 in skeletal muscle is accompanied by changes to additional factors that increase or reduce the movement of fatty acids across the plasma membrane. For example, the composition of lipid rafts in muscle could be altered resulting in further augmentation of fatty acid uptake. Meshulam et al. recently conducted a series of in vitro experiments using an adipocyte cell line and observed that altering the cholesterol and caveolin-1 content of the caveolae, a subtype of lipid raft, modulated fatty acid movement across the membrane (67). Chronic high fructose feeding in rodents is known to induce changes in lipid availability to cells by increasing triglyceride and fatty acid synthesis; both hepatic and intestinal lipoproteins are over-secreted with this type of diet (68). Several groups have demonstrated that the n −6:n −3 ratio of the plasma membrane of cardiomyocytes and skeletal muscle is altered under these dietary conditions (69,70,71). This suggests that current dietary carbohydrate intake may affect the lipid profiles of tissues. Such changes could result in increased fatty acid uptake through any of the potential mechanisms described in the literature (15,32,34,36,72,73,74). It remains to be determined whether such changes to the membrane composition persist beyond the time of acute ingestion of the experimental diet. Determination of the lipid profile of the lipid microdomains such as the rafts or caveolae from different tissues was beyond the scope of this study but could be an important end point in future investigations. As outlined by Pilch et al. (75), such studies could shed light on important interactions between glucose and lipid uptake in a variety of cell types.
There were no differences in FABPpm expression among diet groups either in whole tissue homogenates or in membrane vesicles. Other studies have noted that FABPpm expression does not change in the presence of insulin resistance (27,33,38,76). The development of insulin resistance in skeletal muscle as a result of increased fatty acid transport does not involve changes in FABPpm.
In summary, introduction of dietary fructose during the suckling period resulted in increased circulating insulin and leptin concentrations in adulthood along with increased body weight by ∼12 weeks of age. Increased fatty acid transport into giant vesicles from skeletal muscle was also observed at 12 weeks of age, which is consistent with the presence of insulin resistance. Since the largest difference in body weights and hormonal concentrations were between animals in the Fructose group (that were artificially reared) and SC animals (suckled by their mothers), it is likely that subtle effects of the mode of feeding during the suckling period may contribute to the effects observed in adulthood. The period(s) in the lifespan that result in metabolic programming following nutritional modification, leading to increased adult body weight and/or adiposity are not well defined. The suckling period is one that is malleable to changes in the source of carbohydrate. Data from this study suggest that physiological changes incurred by consuming fructose during the suckling-weaning period may result in lifelong changes in cellular function through a manner that involves changes to the distribution of CD36, promotes increased body weight, and could add to risk for conditions such as metabolic syndrome and type 2 diabetes.
We extend our sincere gratitude to Abha Hodel and Joan Turchinsky for their expert technical assistance during this project. These experiments were supported by grants from the Canadian Institutes of Health Research (Obesity Strategy Pilot Projects) and from the Natural Science and Engineering Research Council, Canada. M.H. was the recipient of a student award from Merck-Frosst, Canada. J.J.J.P.L. is the recipient of a VIDI-innovational Research grant from the Netherlands Organization for Scientific Research (ZonMw grant no. 016.036.305).