Small size at birth has been associated with an increased risk of central obesity and reduced lean body mass in adult life. This study investigated the time of onset of prenatally induced obesity, which occurs after maternal feed restriction, in the guinea pig, a species that, like the human, develops substantial adipose tissue stores before birth. We examined the effect of maternal feed restriction [70% ad libitum intake from 4 wk before to midpregnancy, then 90% until day 60 gestation (term ∼69 days)] on fetal growth and body composition in the guinea pig. Maternal feed restriction reduced fetal (−39%) and placental (−30%) weight at 60 days gestation and reduced liver, biceps muscle, spleen, and thymus weights, relative to fetal weight, while relative weights of brain, lungs, and interscapular and retroperitoneal fat pads were increased. In the interscapular depot, maternal feed restriction decreased the volume density of multilocular fat and increased that of unilocular fat, resulting in an increased relative weight of interscapular unilocular fat. Maternal feed restriction did not alter the relative weight of perirenal fat or the volume density of adipocyte populations within the depot but increased unilocular lipid locule size. Maternal feed restriction in the guinea pig is associated with decreased weight of major organs, including liver and skeletal muscle, but increased adiposity of the fetus, with relative sparing of unilocular adipose tissue. If this early-onset obesity persists, it may contribute to the metabolic and cardiovascular dysfunction that these offspring of feed-restricted mothers develop as adults.
- adipose tissue
epidemiological studies, in various communities around the world, have shown that small size at birth is associated with an increased incidence of a number of adult-onset diseases, including cardiovascular disease, hypertension, type 2 diabetes, and the insulin resistance syndrome (5). It has been hypothesized that perturbation of prenatal development can permanently alter the structure, function, and metabolism of major tissues and organs and program an increased susceptibility to disease in adult life (5). Obesity is a major risk factor for adult cardiovascular and metabolic disease (12, 34), and studies suggest that an adverse prenatal environment can have long-term influences on adipose tissue development (37, 38, 40, 45, 48).
In adults and children of several populations, low weight at birth has been associated with an increased tendency to a central or truncal fat distribution (6, 31, 40, 47). Deficits in muscle mass may also accompany increases in relative adiposity in people who were small at birth, with a number of studies reporting a direct correlation between proportional lean body mass and size at birth (18, 32, 40, 42). These changes in body composition can be detected at an early age, with a persistent deficit in muscularity detected in small-for-gestational age children at 2–47 mo of age, accompanied by an increase in fatness, as indicated by arm skinfold measures, up to 1 yr of age (19). Furthermore, studies of babies born in India indicate that increases in adiposity, and deficits in skeletal muscle and abdominal viscera, are present at birth in this population (52–55) and may contribute to the adult phenotype of increased body fat, but reduced muscle mass, observed in adult Indians (4, 52, 53). In addition, these prenatally induced changes in fatness and lean body mass may contribute to the development of various related adult-onset disorders, especially if body composition is altered early in life.
Growth before birth is dependent on an adequate supply of oxygen and nutrients crossing the placenta from the mother (36). Factors that perturb fetal substrate supply and are known to be responsible for much fetal growth restriction, such as placental insufficiency or poor maternal nutrition, are implicated in the long-term programming of adult disease, including obesity (5). Direct support for a role for maternal nutrition in programming increased adiposity is provided by the observation that exposure to maternal famine during early gestation is associated with higher rates of obesity in 19-yr-old men and 50-yr-old women, although no differences were detected in men at 50 yr of age (37, 38). Experimental studies in other species also indicate that perturbation of maternal nutrition can increase postnatal adiposity (1, 22, 23, 25, 48). In addition, maternal undernutrition influences fetal adipose tissue development (2, 7, 8, 45, 51). In sheep, both increased and decreased fetal adiposity is observed in late gestation, dependent on the timing of nutrient restriction (7, 8, 45). In guinea pigs, restriction of maternal nutrient intake by 50–66% during the second half of pregnancy is associated with relative increases in retroperitoneal and interscapular fat weight in fetuses and neonates (2, 51). The guinea pig, like the human fetus, lays down both brown and white adipose tissue in utero and has a fat content of ∼10% at birth compared with 2% in the fetal rat or sheep (13). Thus experimental animal studies, including those in the guinea pig, a species that resembles humans in the extent of fat deposition before birth, support the suggestion that adipose tissue development and longer-term outcomes for adiposity can be influenced by perturbation of maternal nutrition.
However, these experimental studies, in the guinea pig (2, 51) and other species (1, 8, 22, 23, 48), have largely focused on the effects of maternal undernutrition during particular periods of pregnancy (1, 2, 8, 22, 23, 51), and often of a severe degree (1, 2, 22, 23, 51), thus resembling the human famine studies. In contrast, in populations in developing countries, chronic reduction of energy intake commencing before and continuing throughout gestation, and related indicators, such as low prepregnancy body mass index and low gestational weight gain, account for a significant proportion of intrauterine growth restriction (29, 30). These maternal nutrient deficits may contribute to the altered birth phenotype, and subsequent alterations in adult body composition, that occurs in the Indian population (52–55). In the guinea pig, we have shown that chronic maternal feed restriction from before and throughout pregnancy retards fetal and placental growth (39, 43) and alters adult glucose homeostasis, blood pressure, and cholesterol metabolism in young adult male offspring (25–27). Adult male offspring of mothers placed on a moderately restricted diet also have altered body composition, with an increase in the relative weight of the retroperitoneal fat depot and reduced relative weight of biceps muscle (25). The aim of the current study was to determine whether changes in body composition induced by moderate maternal feed restriction, including increased adiposity, are evident before birth in the guinea pig, and thus could contribute to the development of related disorders throughout postnatal life. We therefore examined the effect of chronic moderate maternal feed restriction in the guinea pig on fetal growth, body composition, including the weight of major adipose tissue depots, and metabolic markers of nutritional state, in late gestation. In addition, adipose tissue development was further analyzed by measuring volume densities of unilocular and multilocular adipocytes in specific fat depots.
Nulliparous 3- to 4-mo-old female guinea pigs (IMVS colored, Gillies Plains Animal Resource Centre, Gillies Plains, SA, Australia) were housed in individual wire-bottomed cages under 12:12-h light-dark conditions and constant temperature. The animals were fed a guinea pig/rabbit ration (Milling Industries Stockfeeds, Murray Bridge, South Australia) with an increased content of vitamin E (165 mg/kg) and had free access to tap water supplemented with vitamin C (400 mg/l). Control animals (n = 20) were fed ad libitum. Feed-restricted animals (n = 20) were given 70% of the average daily food intake of the ad libitum-fed animals, calculated per kilogram body weight, as previously described (25, 43). Body weights were measured three times per week in both groups, and food intake was monitored three times per week in the ad libitum-fed group for calculation of the rate of food intake. Feed-restricted animals were fed between 0800 and 0930 each morning. After 4 wk of controlled feeding, guinea pigs were mated. Females in estrus were placed with a male overnight, and pregnancy was detected by the presence of a vaginal copulatory plug and a failure to return to estrus in the subsequent cycle.
Feed-restricted animals continued to receive 70% of the ad libitum food intake per day per kilogram body weight until day 34 of pregnancy (term ∼69 days). From day 35 to day 60 of pregnancy, the food ration was increased to 90% of the average ad libitum intake per day per kilogram body weight (25, 43). Ad libitum-fed and feed-restricted dams were killed on day 30 (n = 10 in each group) and day 60 (n = 10 in each group) of pregnancy by intraperitoneal overdose of pentobarbital sodium. Blood (20 ml) was collected from the pregnant dams by cardiac puncture and centrifuged at 2,500 rpm for 10 min at 4°C, and plasma was recovered and stored at −20°C. Feed-restricted animals were killed 3–4 h after they received their daily ration, while the ad libitum-fed animals were killed at 0900, after having ad libitum access to food all night.
The entire uterine contents were removed and weighed, and then individual fetuses and placentas were removed and weighed at day 30 or day 60 of gestation. In day 60 fetuses, fetal blood (1–2 ml) was collected by cardiac puncture. Fetal crown-rump length, abdominal circumference, head width, and head length were measured. Major fetal organs and tissues, including individual adipose tissue sites (retroperitoneal, perirenal, interscapular), were then dissected out on ice and weighed. Fat depots were immediately immersion fixed in 4% paraformaldehyde-0.2% glutaraldehyde-2.5% polyvinylpyrolidone in 70 mM phosphate buffer (pH 7.4). All studies were approved by the University of Adelaide Animal Ethics Committee.
Fat cell size and volume densities of unilocular and multilocular fat.
After a minimum of 3 days fixation at 4°C, perirenal and interscapular adipose tissues were washed in four changes of phosphate-buffered saline (pH 7.4) over 2 days, processed, and embedded in paraffin wax. Blocks were cut into 7-μm sections with random orientation and stained with hematoxylin and eosin (HE) or light green. Sections were examined with a 20× objective lens, on an Olympus BH2 microscope equipped with a Video Image Analysis system using Video Pro software (Leading Edge, Adelaide, Australia). Sections of fetal adipose tissue stained with HE were morphometrically analyzed for the volume densities of unilocular and multilocular adipocytes and blood vessels using point counting with an L-36 Merz grid (50). Ten fields (360 points), 1 mm apart, were counted in a section from each fetus, with the first field location chosen at random. The number of points necessary to achieve an SE of 10% was calculated from a preliminary study using a nomogram relating test point number and volume density (50). The weight of multilocular or unilocular fat for each depot was calculated by multiplying the fat depot weight by the appropriate volume density. Morphometric analyses in fat tissues were performed for 21 fetuses from seven different litters from ad libitum-fed mothers and 15 fetuses from nine different litters from feed-restricted mothers.
In sections for which light green was used to stain the cell membranes, the mean area, maxima, and minima of unilocular adipocytes in each of 10 fields per slide were measured with Video Pro software.
Albumin, cholesterol, urea, triglycerides, and protein were measured in plasma from individual fetuses by enzymatic analysis, using a Cobas Mira automated centrifugal analyzer (Roche Diagnostic Systems) and commercial kits. Plasma levels of metabolites in the mothers have been reported (43).
Data were analyzed using BMDP Statistical Software (Los Angeles, CA). The effect of maternal undernutrition on fetal size, body composition, and metabolism was analyzed by repeated-measures ANOVA, using a mixed model (BMDP 5V). Litter size was included as a covariate. Ratios of placental to fetal weight were log transformed before analysis. The effect of fetal gender on various parameters was analyzed by one-way ANOVA with litter size as a covariate. Correlation analysis was used to examine relationships between variables. Differences between groups were considered significant if P < 0.05. All data are expressed as means ± SE.
The effect of moderate maternal feed restriction on maternal feed intake, weight gain, and body composition has been described previously (25, 39, 43). Briefly, maternal feed restriction reduced weight at mating [ad libitum: 603 ± 11 g (n = 20); feed restricted: 532 ± 12 g (n = 20), P < 0.001] and reduced weight gain during pregnancy (day 30 group: pregnancy × feed restriction, P < 0.003; day 60 group: pregnancy × feed restriction, P < 0.001). Maternal body weights were lower in feed-restricted animals killed at day 30 [ad libitum: 722 ± 20 g (n = 10); feed restricted: 568 ± 17 g (n = 10), P < 0.001] and day 60 of pregnancy [ad libitum: 962 ± 20 g (n = 10); feed restricted: 666 ± 14 g (n = 10), P < 0.001]. Feed intake (in g/day, without correction for body weight) in the feed-restricted mothers averaged 63% of the ad libitum intake (g/day) from before and to day 35 of gestation, and 68% ad libitum intake (g/day) from day 35 to day 60 of gestation.
Fetal size and placental weight.
Fetal and placental weights were reduced in the feed-restricted group by 17% and 10% at day 30 of gestation, respectively, and by 39% and 30% at day 60 of gestation, respectively (Ref. 43; Table 1). Maternal feed restriction increased the placental-to-fetal weight ratio at day 60 of gestation (Ref. 43; Table 1). Fetal weight and placental weight were positively associated at day 30 (r = 0.50, P < 0.0001, n = 54) and day 60 (r = 0.83, P < 0.0001, n = 51) of gestation.
Head width, crown rump length, abdominal circumference, body weight/crown rump length, and ponderal index (body weight/crown rump length3) were reduced in fetuses of feed-restricted mothers at day 60 of gestation (Table 1).
At day 60 of gestation, fetal weight (P < 0.04), placental weight (P < 0.0001), and abdominal circumference (P < 0.001) were greater in males than females when both maternal feeding groups were combined. The effect of sex on fetal and placental weights and birth measurements was then examined in fetuses of ad libitum-fed mothers separately (16 male, 13 female, gender not recorded for 1 fetus) because the fetal gender distribution in the feed-restricted group (5 male, 16 female) may have influenced the comparison of gender effects when both groups were combined. In offspring of ad libitum-fed mothers, fetal weight (male: 71.9 ± 1.7 g, female: 65.3 ± 1.6 g, P < 0.03), placental weight (male: 4.8 ± 0.2 g, female: 4.0 ± 0.2 g, P < 0.01), and abdominal circumference (male: 9.0 ± 0.2 cm, female: 8.3 ± 0.1 cm, P < 0.002) were greater in male than female fetuses. No differences were detected in fetal size measures between male (n = 5) and female (n = 16) fetuses of feed-restricted mothers.
Maternal feed restriction reduced litter size (Table 1; Ref. 43). In fetuses from ad libitum-fed mothers, where litter size varied from 2 to 4 pups per litter, fetal (P < 0.01) and placental weight (P < 0.01) decreased as litter size increased. In the feed-restricted group, litter sizes of 1 to 3 were observed. The trend for decreasing fetal and placental weight with increasing litter size was not significant in the feed-restricted group.
Fetal body composition.
The weights of fetal brain, liver, biceps brachii, lungs, heart, kidneys, adrenals, thymus, and spleen were reduced in fetuses of feed-restricted mothers at day 60 of pregnancy (Table 2). Maternal feed restriction reduced the weights of liver, biceps, thymus, and spleen, when expressed relative to fetal body weight, while the weights of brain and lungs increased relative to fetal body weight (Table 2). The relative weights of fetal heart, kidneys, and adrenal were not altered by maternal feed restriction (Table 2). Fetal liver weight at day 60 of gestation was positively associated with fetal abdominal circumference (r = 0.89, P < 0.0001, n = 42).
The effect of sex on fetal tissue weights was examined in the fetuses of ad libitum-fed mothers (16 male, 13 female). Fetal brain and heart were heavier in male than female fetuses (brain: male 2.45 ± 0.03 g, female 2.34 ± 0.04 g, P < 0.04; heart: male 0.44 ± 0.02 g, female 0.37 ± 0.02 g, P < 0.03). The relative weights of the fetal adrenals were lower in male than female fetuses (male 0.031 ± 0.001 g%, female 0.036 ± 0.001 g%, P < 0.02). No differences in fetal tissue weights were detected between male (n = 5) and female (n = 16) fetuses of feed-restricted mothers.
Adipose tissue depot weights.
The absolute weights of the interscapular (−27%), perirenal (−40%), and retroperitoneal (−28%) fat depots were reduced in the fetuses of feed-restricted mothers (Table 3). When expressed relative to fetal body weight, the weights of the interscapular and retroperitoneal fat pads were increased by 19% and 15%, respectively, by maternal undernutrition (Table 3). Maternal feed restriction did not alter the relative weight of perirenal fat in the fetus.
In fetuses of ad libitum-fed mothers, retroperitoneal fat (male 0.67 ± 0.03 g, n = 13; female 0.54 ± 0.02 g, n = 13, P < 0.001) and retroperitoneal fat as a percentage of body weight (male 0.92 ± 0.03 g%, female 0.83 ± 0.04 g%, P < 0.02) were higher in male than female fetuses, while perirenal and interscapular fat weights were not different. No differences in fetal fat weights were detected between male (n = 5) and female (n = 16) fetuses of feed-restricted mothers.
Adipose tissue morphometry.
Maternal feed restriction decreased the volume density of multilocular fat in the fetal interscapular fat depot (−15%) (Table 4). Although the volume density of multilocular fat in perirenal fat tended to be lower, this was not significant (P = 0.06, Table 4). Weight of multilocular fat in each depot was calculated by multiplying tissue weight by volume density of the fat. The weight of multilocular fat in the perirenal (−52%) and interscapular depots (−37%) was reduced in fetuses of feed-restricted mothers in absolute terms (Table 4) but was not altered in either depot when expressed relative to fetal weight (Table 4).
The volume density of unilocular adipocytes in the fetal interscapular fat depot was increased by maternal feed restriction (+31%, Table 4). The weight of unilocular fat in the interscapular depot was not altered by maternal feed restriction (Table 4). However, interscapular unilocular fat weight, expressed as a percentage of fetal weight, was increased in the feed-restricted fetuses (+54%, Table 4). Maternal feed restriction did not alter the volume density of unilocular fat in the fetal perirenal fat depot (Table 4). The weight of unilocular fat in the fetal perirenal depot decreased in the feed-restricted group by 36% (P < 0.01); however, perirenal unilocular fat weight, as a percentage of fetal body weight, was not altered by maternal feed restriction (Table 4).
Maternal feed restriction increased the mean unilocular lipid locule size in the fetal perirenal fat pad (+21%, Table 4). Mean unilocular lipid locule size in the interscapular fat pad also tended to be increased by maternal feed restriction, but this was not significant (P = 0.06, Table 4).
Maternal feed restriction reduced the concentration of albumin (−10%) and triglycerides (−58%) in fetal plasma (Table 5). Plasma urea (+55%) and cholesterol (+21%) concentrations increased in fetuses of feed-restricted mothers (Table 5). Maternal feed restriction did not alter fetal plasma protein concentrations (Table 5). Fetal gender did not influence fetal metabolite concentrations.
Restricting maternal nutrition from before and throughout pregnancy in the guinea pig substantially alters fetal and placental growth, fetal metabolic state, and body composition. Late-gestation fetuses of feed-restricted mothers were lighter, shorter, and thinner than offspring of ad libitum-fed mothers. Development of major organs, including skeletal muscle and liver, was adversely affected in prenatally growth-restricted guinea pigs. Maternal feed restriction increased the relative adiposity of the fetus and altered the relative proportions of unilocular and multilocular adipocytes in interscapular fat. The proportion of unilocular adipose tissue suggests that white or storage adipose tissue may be relatively preserved, at the expense of multilocular, or potentially thermogenic, brown adipose tissue in the interscapular depot. Thus the growth and development of major tissues responsible for thermogenesis at birth and for the maintenance of glucose and cholesterol homeostasis and mediating insulin action in postnatal life are perturbed before birth in the growth-restricted guinea pig. We have previously reported that maternal feed restriction in the guinea pig increases the relative weight of retroperitoneal fat and reduces the relative weight of biceps muscle, as well as inducing hyperinsulinemia, glucose intolerance (25), increased systolic blood pressure (27), and perturbed cholesterol metabolism (26) in young adult male offspring. Importantly, increased adiposity together with reduced lean body mass and altered hepatic development are present from before birth with moderate maternal feed restriction and have the opportunity to contribute to the development of these other disorders throughout postnatal life.
Moderate maternal feed restriction in the guinea pig from before mating results in a fetal phenotype that resembles the “thin-fat” neonatal phenotype of newborn babies in India (15, 52–55). Compared with babies born in England, Indian babies were light and thin, with reduced ponderal index, but had a relative sparing of body fat, as indicated by relative maintenance of subscapular skin fold thickness (15, 54, 55). Thinness in these babies is associated with a relative deficit in muscle mass and abdominal viscera, which include the liver (54, 55). Relative sparing of head circumference, in the smallest Indian babies, suggests that brain growth has been spared (54), as occurs in many experimental models of intrauterine growth restriction (36). This phenotype is maintained throughout life, with reduced lean muscle mass and a higher percentage of body fat, particularly visceral fat, for a given body mass index, in adult Indians (4, 52, 53). Indian men and women who were short and relatively fat at birth have increased adult rates of type 2 diabetes (14), while insulin resistance can be detected in 8-yr-old children that were small at birth, particularly when combined with rapid postnatal growth (53). We have demonstrated that adult offspring of moderately feed-restricted guinea pigs are at increased risk of altered glucose and lipid metabolism in adult life and that the perturbation of adiposity and lean body mass persists into adult life (25–27). Chronic deficiency of macro- and micronutrients, beginning from early in life in Indian women, is likely to contribute to the altered birth phenotype observed in Indian populations (15, 16, 52, 53). Caloric intake in Indian mothers during pregnancy was ∼75% that of British mothers, while protein intake was 50% lower (15, 52). In the current study, fetal albumin and triglycerides levels were reduced by maternal feed restriction, consistent with chronic nutrient deficiency. In contrast to experimental studies that restrict nutrition from the time of conception, or for particular periods of pregnancy, our study may more closely reflect the conditions experienced by women exposed to chronic undernutrition.
Other studies support a role for maternal undernutrition in enhancing adiposity in the fetal guinea pig (2, 51). Limiting maternal food intake by 50% during the second half of pregnancy increases relative interscapular and retroperitoneal fat mass in fetal and neonatal guinea pigs but does not alter the relative weight of two subcutaneous fat depots (2). Similarly, relative interscapular fat weight was increased in fetal guinea pigs whose mothers were fed 33% ad libitum intake for the second half of pregnancy (51). In the current study, increased relative weight of the interscapular and retroperitoneal fat pads was observed when maternal nutrition was moderately restricted from before and throughout pregnancy. In contrast, perirenal fat weight was not altered by chronic maternal nutrient restriction. Thus, as observed in other studies (2), the effect of maternal nutrient restriction on adipose tissue development in the fetal guinea pig varies between depots.
The interscapular fat pad contained ∼50% multilocular tissue, while 28% of the perirenal fat depot was multilocular in the guinea pig fetus. This is consistent with interscapular fat having a significant brown adipose tissue (BAT) content in the fetal or neonatal guinea pig (2, 21). Ashwell et al. (2) proposed that BAT development in the guinea pig fetus was preferentially spared when maternal nutrient intake was restricted in the second half of pregnancy, because this perturbation increased the relative weight of interscapular and retroperitoneal fat, depots that contain brown fat. However, they also reported an increased proportion of unilocular tissue in the interscapular fat pad after maternal undernutrition in the second half of pregnancy (2), suggesting that white rather than brown adipose tissue may be relatively spared. Similarly, in the current study, restricted nutrition from before and throughout pregnancy increased the proportion of unilocular adipose tissue, while decreasing multilocular fat, in the interscapular depot. Further analysis, in addition to these morphological analyses, is required to more completely characterize brown and white adipocyte populations in the current study (9, 20). Nevertheless, the current and previous studies (2, 51) indicate that maternal feed restriction in the guinea pig results in a relative sparing of the growth of the interscapular and retroperitoneal fat depots in the fetus, and this sparing occurs in the unilocular adipocyte population.
In sheep, fat accounts for ∼2% of body weight at term (13, 45), and perirenal fat is the major adipose tissue site (8, 33, 45). Maternal nutrient restriction commencing 8 days after conception, and continuing throughout gestation, reduces fetal weight and increases relative perirenal fat weight in the fetal sheep (8). In contrast, undernutrition from before and throughout pregnancy, similar to the period of nutrient restriction in the current study, does not alter relative perirenal fat weight, while late-gestation feed restriction reduces fetal adiposity (8). Therefore, in sheep, nutrient restriction commencing shortly after, but not before, conception is associated with an increased adiposity of the fetus (8). In guinea pigs, both maternal undernutrition from before and throughout pregnancy and during only the second half of pregnancy increase relative fetal adiposity (2, 51). Studies to determine the effect of nutrient restriction during early gestation on adipose tissue development are required to further investigate the importance of timing of nutrient restriction in the guinea pig.
The mechanism through which the relative weight of major fat depots is increased in fetuses of feed-restricted mothers is unclear. Endocrine factors known to influence fetal tissue development (17, 36), including that of adipose tissue (7, 45) and skeletal muscle, are influenced by maternal nutrition. Acute maternal starvation reduces fetal plasma concentrations of insulin, IGF-I, and thyroid hormones in the guinea pig (24). Chronic maternal feed restriction in the guinea pig reduces fetal plasma levels of IGF-I, but not thyroid hormones, however (11). Leptin is produced by fetal adipose tissue (56) and correlates with fetal weight (28). Infusion of leptin into fetal sheep increases multilocular and decreases unilocular adipocyte proportions in perirenal fat (58). Maternal undernutrition in late gestation does not alter leptin mRNA in perirenal fat, or circulating levels, in fetal sheep (57). However, Indian babies, despite being significantly lighter, have leptin concentrations similar at birth to those of English babies, which may reflect their increased adiposity (55). In addition, when adjusted for birth weight, both insulin and leptin are higher at birth in Indian babies (55). The effect of chronic maternal undernutrition on plasma levels and/or tissue production of fetal hormones and growth factors, in the current study, remains to be determined.
Altered adipose tissue development in the fetus may have immediate and long-term consequences. Increased relative size of adipose tissues in the fetus may be associated with increased adiposity in later life. Adult male guinea pig offspring of mothers feed-restricted according to the same regimen as this study have increased relative retroperitoneal fat weight, although interscapular fat weight is not altered (25). In Indian populations, where babies preserve body fat before birth, increased relative fatness is observed in adult life (52, 53). Higher rates of obesity are present in 19-yr-old men and 50-yr-old women exposed to maternal famine during early pregnancy (37, 38). Exposure to undernutrition during early pregnancy has been suggested to program susceptibility to obesity by perturbing development of the hypothalamic centers regulating food intake (37) or through alterations in endocrine mechanisms that regulate body fat accumulation postnatally (38). Adult offspring of rats feed-restricted throughout pregnancy are hyperphagic, hyperinsulinemic, and hyperleptinemic and have diminished locomotor activity, suggesting that alterations in neurendocrine appetite regulation and altered behavior patterns may contribute to their increased obesity (48, 49). However, the results of the current and other studies (2, 7, 8, 45, 51) indicate that perturbation of maternal nutrition can alter adipose tissue development before birth. The relative role of prenatal effects on adipose tissue development compared with that of prenatally programmed perturbations in postnatal neuroendocrine and endocrine regulators of adiposity remains to be determined.
Reduced abundance of BAT in undernourished fetal guinea pigs may decrease the capacity for neonatal thermogenesis. Lambs delivered by caesarean section to low-body-weight ewes have less BAT, impaired capacity for thermogenesis, and reduced survival (44). As in humans, adipose depots containing BAT in the guinea pig are transformed after the neonatal period into tissues with the characteristics of white adipose tissue (9, 21). Cold acclimation can reactivate BAT in adult guinea pigs (3, 21). Uncoupling protein-1 mRNA is detected in adipose tissue in adult humans, suggesting that small numbers of brown adipocytes are present (9, 10, 20, 35). However, whether BAT can contribute to energy expenditure and thermogenesis in adult humans, or can be pharmacologically induced to do so, is debated (9, 20, 46). Recruitment of BAT, and the associated increase in capacity for thermogenesis, is a mechanism through which rats adapt to increased energy, high-fat, or high-carbohydrate diets (9, 41), particularly when the increase in energy occurs at the expense of nutrients such as protein (9). The influence of maternal feed restriction on BAT development may differ in altricial species, such as the rat. Whether prenatal perturbation of adipose tissue development could limit the postnatal capacity for activation of BAT and contribute to increased obesity, when prenatally growth-restricted rats are exposed to hypercaloric diets (48), is not known.
In summary, late-gestation fetal guinea pigs exposed to chronic moderate maternal feed restriction are lighter and thinner, and despite decreased relative weight of major tissues such as skeletal muscle and liver, their relative weight of major adipose tissue sites is increased. Moderate maternal feed restriction from before and throughout pregnancy in the guinea pig results in a fetal phenotype that resembles that observed in thin babies, that are born relatively fat, in India (52, 53). Increases in relative adiposity, and defects in lean body mass, together with altered development of the major tissues responsible for glucose and lipid metabolism, are present at birth in these offspring, may continue into adult life, and have the capacity to contribute to the development of subsequent disease from early in postnatal life.
We thank L. Mundy, Dr. J.-J. Erwich, K. Lucero, J. Lang, and K. Irvine for assistance with this work.
Present address for K. Kind: Research Centre for Reproductive Health, Department of Obstetrics and Gynaecology, University of Adelaide, Adelaide, South Australia 5005, Australia.
Present address for A. Sohlstrom: The Swedish National Food Administration, P.O. Box 622, S-751 26 Uppsala, Sweden.
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