SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Maternal obesity due to long-term high-fat diet (HFD) consumption leads to faster growth in offspring during suckling, and increased adiposity at 20 days of age. Decreased expression of the orexigenic neuropeptide Y (NPY) and increased anorexigenic proopiomelanocortin (POMC) mRNA expression were observed in the fed state. However, hunger is the major drive to eat and hypothalamic appetite regulators change in response to meals. Therefore, it is important to compare both satiated and fasting states. Female Sprague–Dawley rats (8 weeks old) were fed a cafeteria-style HFD (15.33 kJ/g) or chow for 5 weeks before mating, with the same diet continuing throughout gestation and lactation. At postnatal day 20, male pups were killed either after overnight fasting or in the fed state. Pups from obese dams were hyperphagic during both pre- and postweaning periods. Pups from obese dams had higher hypothalamic mRNA expression of POMC and NPY Y1 receptor, but lower hypothalamic melanocortin-4 receptor (MC4R) and its downstream target single-minded gene 1 (Sim1), in the fed state. Overnight fasting reduced circulating glucose, insulin, and leptin and increased hypothalamic NPY Y1 receptor mRNA in pups from both lean and obese dams. Hypothalamic NPY and agouti-related protein (AgRP) were only increased by fasting in pups from obese dams; reductions in MC4R and Sim1 were only seen in pups from lean dams. At weaning, the suppressed orexigenic signals in offspring from obese dams were normalized after overnight fasting, although anorexigenic signaling appeared impaired in these animals. This may contribute to their hyperphagia and faster growth.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Appetite is regulated by a complex homeostatic network comprising central and peripheral components to maintain energy balance (1,2). The hypothalamus is considered the main integrator and processor of peripheral metabolic information. The flow of signals regulating metabolism within the brain relies on neuronal interactions that affect the production or release of neurotransmitters (3). The hypothalamic arcuate nucleus (Arc) contains neurons expressing orexigenic peptides neuropeptide Y (NPY) and agouti-related protein (AgRP), and those expressing anorexigenic peptide proopiomelanocortin (POMC) and cocaine- and amphetamine-regulated transcript, that interact with each other (1). Input from hormones reflecting adiposity and energy balance (adipose-derived leptin and pancreatic insulin) is integrated within the hypothalamus to regulate feeding.

Studies on mutant mice indicate that overexpression of orexigenic neuropeptides contributes to obesity or a hyperphagic phenotype. Hypothalamic expression of the powerful feeding stimulator NPY is increased several fold in leptin deficient ob/ob and leptin receptor–deficient db/db mice, and there is also marked overexpression of AgRP in both ob/ob and db/db mice (4,5). However, in adult-onset dietary obese rodent models, particularly those employing palatable diets, there is an apparent discrepancy between a hyperphagic phenotype and levels of both orexigenic and anorexigenic appetite regulators. In both rats and mice, long-term high-fat diet (HFD) feeding-induced marked increases in plasma leptin levels, with reduced concentrations of hypothalamic NPY and increased peptide cleavage product of POMC, α-melanocyte-stimulating hormone (α-MSH) (6,7,8). It seems that hyperphagia in these animals was not regulated by these essential central regulators.

Hunger invokes a complex hormonal and neural response. NPY mRNA expression is increased in response to fasting or chronic food restriction, and decreased within 6–24 h of ad libitum refeeding (9,10,11,12,13). AgRP expression is also further increased in obese ob/ob and db/db mice during fasting (4,5,12). POMC mRNA levels are reduced markedly in fasted animals and restored by refeeding, or increased by exogenous administration of leptin (12,14).

The obesity epidemic is affecting the next generation (15). Maternal obesity has been shown to predispose individuals to overeating, increased adiposity, and glucose intolerance (16,17). However, the underlying neural mechanisms are unclear. Previously, we found reduced hypothalamic NPY and increased POMC mRNA levels in offspring from obese dams at postnatal day 20 in the free-feeding state (18), which is similar to the changes in adult-onset dietary-obesity (6,7,8). Hunger is the major drive for feeding behavior. Therefore, it is important to understand how maternal obesity affects the response of these appetite regulators and hormones to fasting. We hypothesized that the increase in hypothalamic orexigenic neuropeptides in response to overnight fasting would be greater in offspring of obese dams than lean dams, leading to hyperphagia.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Maternal obesity

Outbred female Sprague–Dawley rats (8 weeks, Animal Resources Centre, Canning Vale, Australia) were housed at 20 ± 2 °C, and maintained on a 12:12 h light/dark cycle (lights on at 0600 hours). Rats were assigned to two groups of equal body weight. The control group was fed laboratory chow (Gordon's Specialty Stockfeeds, NSW, Australia; 11 kJ/g energy, 14% fat, 21% protein, 65% carbohydrate), and the second group palatable-HFD (15.33 kJ/g, average energy 34% fat (saturated fat 17%), 18% protein, 50% carbohydrate). The HFD consisted of high-fat modified chow (laboratory chow, milk powder, sweetened condensed milk, and saturated animal fat) and highly palatable cafeteria-style food such as meat pies, cakes, and biscuits (4–5 different types/day) of known caloric content (6,7,18). Female rats were fed either chow or HFD for 5 weeks before mating with adult males obtained from the same source. Dams continued on the same diet during lactation. The study was approved by the Animal Ethics Committee of the University of New South Wales.

Fasting and sample collection

Dams littered normally. At postnatal day 19, half of the male pups from each litter were separated from their mothers at 1700 hours and housed with their littermates next to their initial home cage. They were allowed free access to water without food. The other rats remained with the dams. The following morning, pups fasted overnight were harvested between 0800–0900 hours, and those who were not fasted between 0930 and 1030 hours. Rats were deeply anesthetized (ketamine/xylazine 180/32 mg/kg, intraperitoneal). After measurement of naso-anal (N-A) length, blood was removed by cardiac puncture and blood glucose measured (Accu-Chek meter; Roche Diagnostics, Nutley, NJ); plasma was stored at −20 °C. Pups were killed by decapitation and the whole hypothalamus removed for mRNA measurement following coronal sections at the optic chiasm and the caudal hypothalamic sulcus. The hypothalamic area was dissected by cutting at the hypothalamic sulci, then just above the third ventricle. Various adipose tissue beds were dissected and weighed, as well as organs (heart, liver, kidney) and skeletal muscles (extensor digitorum longus, soleus, and tibialis). Tibia length was measured as a marker of growth.

Milk intake and postweaning chow intake measurement

Milk intake was measured in a separate cohort, because separation of pups from their mothers may affect both dams and pups. At postnatal day 10, body weights of pups were recorded after 2-h separation from their mothers and they were kept with their littermates at 31 °C. Weight increase recorded 2 h after return to the mothers served as an indirect measure of milk ingestion, represented as milliliter drunk. Postweaning chow intake was measured in another cohort of 22-day-old animals (2 days postweaning) by carefully weighing all food remaining after a 24-h period.

Plasma hormone measurements

Plasma leptin and insulin concentrations were measured using commercially available radioimmunoassay kits (Linco, St Charles, MO). The homeostasis model assessment (HOMA) parameter was calculated: HOMA = fasting plasma insulin (ng/ml) × fasting plasma glucose (mmol/l)/(22.5 × 0.0417); the greater the HOMA value, the greater the level of insulin resistance (19).

Quantitative real-time PCR

Total hypothalamic mRNA was isolated using TRIzol reagent (Invitrogen Australia, Melbourne, Australia) according to the manufacturer's instructions. The purified total RNA was used as a template to generate first-strand complementary DNA synthesis using M-MLV Reverse Transcriptase, RNase H Minus, Point Mutant Kit (Promega, Madison, WI). Preoptimized Applied Biosystem probe/primers (Foster City, CA) were used for quantitative real-time PCR (Eppendorf Realplex 2; Eppendorf AG, Hamburg, Germany). The target mRNA was labeled with FAM and 18S rRNA was labeled with VIC. Gene expression was quantified in a single multiplexing reaction, where gene of interest was standardized to 18S rRNA. An individual sample from the control group was then arbitrarily assigned as a calibrator against which all other samples are expressed as fold difference.

Statistical methods

Results are expressed as mean ± s.e.m. Body weight of dams and HOMA of pups were analyzed by Student's unpaired t-test. Other data were analyzed using two-way ANOVA followed by post hoc Fisher's Least Significance Difference tests.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Maternal effect

At day 1 of life, pups born from obese dams were 10% heavier than those from lean dams (Table 1, P < 0.05). At postnatal day 10, the 2-h milk intake was more than doubled in pups from obese dams (1.05 ± 0.10 vs. 0.40 ± 0.08 ml in offspring from obese and lean dams, respectively, P < 0.05 n = 22 in each group). After weaning, pups from obese dams (35.75 ± 0.55 kJ/rat/24 h) also consumed more food than those from lean dams (32.31 ± 0.14 kJ/rat/24 h, P < 0.05). In the free-feeding state, maternal obesity resulted in nearly doubled body weight in offspring at 20 days compared with those from lean dams (Table 1). Pups from obese dams were significantly larger as shown by 14% increase in body length and 8% increase in tibia length (Table 1). All the major organs (liver, kidney, pancreas, and heart) and fat pads (BAT, retroperitoneal, epididymal, and mesenteric fat) sampled were significantly heavier in pups from obese dams, as well as the weights of skeletal muscles. Remarkably, retroperitoneal and epididymal fat masses were >9 and 5 times greater in pups from obese dams. When standardized by body weight, all these differences remained significant. Maternal obesity had little impact on blood glucose and plasma insulin levels in offspring (Figure 1a,b), whereas the plasma leptin levels were more than tripled in pups from obese dams (Figure 1c).

Table 1.  Body weight and parameters in pups from lean and obese dams at 20 days
inline image
image

Figure 1. Blood glucose and plasma insulin and leptin in pups from lean and obese dams at baseline (open bars, n = 8) and after overnight fasting (closed bars, n = 8). Results are expressed as mean ± s.e.m. Data were analyzed by two-way ANOVA, followed by a post hoc LSD test. *P < 0.05, maternal effect; #P < 0.05, fasting effect. LSD, Least Significance Difference.

Download figure to PowerPoint

When animals were in the fed state, hypothalamic NPY and AgRP mRNA expression in pups from obese dams were slightly lower than that in pups from lean dams, without reaching statistical significance (NPY P = 0.07, Figure 2a,b), while POMC mRNA expression was significantly increased in pups from obese dams relative to lean dams (Figure 2c). NPY Y1 receptor expression (Figure 3a) was increased in pups from obese dams, with no significant changes in mRNA expression of the receptor of α-MSH, melanocortin-4 receptor (MC4R) (P = 0.07, Figure 3b). However, mRNA expression of the downstream target of MC4R, single-minded gene 1 (Sim1), was significantly reduced by 45% in pups from obese dams (P < 0.05, Figure 3c). mRNA expression of Ob-Rb and STAT3 was also significantly reduced by 40% in pups from obese dams (P < 0.05, Figure 4).

image

Figure 2. mRNA expression of hypothalamic (a) NPY, (b) AgRP, and (c) POMC in pups from lean and obese dams at baseline (open bars, n = 8) and after overnight fasting (closed bars, n = 8). Results are expressed as mean ± s.e.m. Data were analyzed by two-way ANOVA, followed by a post hoc LSD test. *P < 0.05, maternal effect; #P < 0.05, fasting effect. AgRP, agouti-related protein; LSD, Least Significance Difference; NPY, neuropeptide Y; POMC, proopiomelanocortin.

Download figure to PowerPoint

image

Figure 3. mRNA expression of hypothalamic NPY Y1 receptor, MC4R, and Sim1 in pups from lean and obese dams at baseline (open bars, n = 8) and after overnight fasting (closed bars, n = 8). Results are expressed as mean ± s.e.m. Data were analyzed by two-way ANOVA, followed by a post hoc LSD test. *P < 0.05, maternal effect, #P < 0.05, fasting effect. LSD, Least Significance Difference; MC4R, melanocortin-4 receptor; NPY, neuropeptide Y; Sim1, single-minded gene 1.

Download figure to PowerPoint

image

Figure 4. mRNA expression of hypothalamic Ob-Rb and STAT3 in pups from lean and obese dams at baseline (open bars, n = 8) and after overnight fasting (closed bars, n = 8). Results are expressed as mean ± s.e.m. Data were analyzed by two-way ANOVA, followed by a post hoc LSD test. *P < 0.05, maternal effect; #P < 0.05, fasting effect. LSD, Least Significance Difference.

Download figure to PowerPoint

Fasting effect

Overnight fasting led to 18 and 12% reductions in body weight in pups from lean and obese dams, respectively (Table 1). Of the organs sampled, only liver was affected by fasting, with reductions of 35 and 30% in pups from lean and obese dams, respectively (P < 0.05 fasting effect, Table 1). Overnight fasting also caused some reduction in fat and muscle mass; however, significance was only observed in mesenteric fat and extensor digitorum longus (Table 1). Blood glucose and insulin concentrations were significantly reduced to similar levels in pups from lean and obese dams after overnight fasting (Figure 1a,b). HOMA was not different between pups from lean (1.34 ± 0.26) and obese dams (1.59 ± 0.25). Plasma leptin was reduced by almost 80% in pups from both lean and obese dams, remaining three times higher in the latter (Figure 1c).

After overnight fasting, hypothalamic NPY mRNA expression was not significantly altered in pups from lean dams, but was markedly increased above baseline in pups from obese dams, to a level similar to that of pups from lean dams (Figure 2a). Although hypothalamic AgRP mRNA expression was increased by 20% after fasting in pups from lean dams, this did not reach statistical significance, while expression nearly doubled in pups from obese dams (P < 0.05, fasting effect). POMC mRNA expression was not altered by fasting in either group. Hypothalamic Y1 receptor mRNA was increased in response to fasting by 77 and 65% in pups from lean and obese dams, respectively; however, the expression in fasting pups from obese dams was 30% higher than that in pups from lean dams (Figure 3a). MC4R was significantly reduced in pups from lean, but not obese dams (Figure 3b). Sim1 mRNA expression was also reduced by nearly 40% after overnight fasting in pups from lean dams, with no change in pups from obese dams (Figure 3c). Little change in hypothalamic Ob-Rb mRNA expression was seen on fasting (Figure 4a). STAT3 mRNA expression was significantly reduced by fasting only in pups from lean dams (P < 0.05).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Maternal obesity is a significant risk factor for childhood obesity. Human studies suggest that intrauterine factors may be more important than genetic factors in changing gene expression in response to HFD, which may have a longstanding influence postnatally. Hunger is the major drive for eating behavior in both humans and animals. This study investigated the impact of pre-existing maternal obesity on offspring central appetite regulators, and the response to overnight fasting.

Long-term maternal obesity led to greater body weight, body size, and more severe adiposity in offspring at weaning, which is consistent with our earlier observations (18). In this study, the higher milk intake after 2-h separation is likely due to increased hunger rather than to any stress related effect, as at postnatal day 8, maternal separation caused significant changes in corticosterone after 8, but not 4 h (20). Thus the faster weight gain in pups from obese offspring could be partly due to hyperphagia. As a result, plasma leptin was significantly increased. NPY, AgRP, and POMC are heavily expressed in the Arc of the hypothalamus. As an adaptation to the increased adiposity and plasma leptin concentration, when animals were in a satiated state, only the anorexigen POMC was significantly upregulated in offspring from obese dams, which is similar to the changes observed in adult-onset dietary obese animals (21). However, Y1 receptor mRNA was also upregulated in the nonfasted state. As the NPY Y1 receptor not only mediates the feeding effects of NPY, but also regulates energy expenditure (22), this may lead to reduced energy expenditure in the satiated state.

One of the novel observations in this study is that hypothalamic Sim1 expression was significantly lower in pups from obese dams, and MC4R showed a similar pattern, in the face of increased POMC expression. MC4R and Sim1 are coexpressed in neurons in the hypothalamic paraventricular nucleus (PVN), a region rich in terminals releasing NPY, AgRP, and α-MSH and thus critical for appetite regulation (23,24). Absence of Sim1 is associated with hyperphagic obesity in both humans and rodents (25) and overexpression of Sim1 can normalize food intake in agouti yellow obese mice and diet-induced obesity (26). Heterozygous Sim1 knockout mice have been reported to be resistant to melanocortin signaling, and displayed normal energy expenditure and elevated hypothalamic POMC expression, yet are hyperphagic (27). The appetite suppressive effect of melanocortin agonist melanotan-2 was also blunted (27). Therefore, Sim1 is suggested to be the essential downstream target for MC4R signaling in feeding regulation (26). In this study, appetite regulation in offspring from obese dams showed similar phenotype as Sim1 knockout mice. Thus, we suggest that maternal “junk eating” inhibits melanocortin signaling via both MC4R and its downstream target Sim1. Increased NPY Y1 receptor together with low Sim1 in the satiated state may also drive feeding behavior in offspring from obese dams even if they are not hungry.

In offspring from obese dams, changes in appetite regulators in the satiated state are not sufficient to explain the increased milk consumption during the suckling period and energy intake soon after weaning. In adult rats, NPY concentrations are elevated before a meal and decreased significantly during the course of eating, while α-MSH counteracts NPY to inhibit feeding (28). Thus it is necessary to compare changes in appetite regulators in the hungry state between pups from lean and obese dams.

Rats are nocturnal animals, consuming the majority of their daily intake during the night. After overnight fasting, body weight, liver, and muscle masses, together with blood glucose and plasma leptin and insulin levels were significantly reduced in all pups. The effect of fasting on body weight is related to the growth rate at weaning age, and interruption of nocturnal feeding. Because excess glucose is stored as glycogen in the liver and muscle and used as fuel during times of energy deficiency, the reduced liver and muscle mass could reflect mobilization of stored glycogen to maintain minimum glucose levels.

The major finding of this study is the differential changes in appetite regulators induced by overnight fasting in pups from lean and obese dams. An 80% reduction in plasma leptin levels after fasting in pups from both lean and obese dams and downregulated STAT3 in those from lean dams after fasting suggests reduced actions of leptin in this situation. Although it is possible that some of the effects we observed relate to a stress response induced by overnight maternal deprivation rather than fasting, we feel this is unlikely. First, the animals were studied very close to weaning age. Second, maternal separation for 24 h was only effective when applied between postnatal day 1 and 12, but not on postnatal day 13 (29). Furthermore, in a study by Kowalski and colleagues, maternal deprivation at postnatal day 15 caused increased in NPY mRNA expression in the Arc, which was reversed by continuous gastric infusion of milk (30), suggesting altered NPY was more related to nutritional deficiency, rather than the stress of separation. In this study, NPY and AgRP were not changed after overnight fasting in offspring from chow-fed dams, neither was POMC mRNA, suggesting 16 h may be not long enough to induce significant changes in their production, in line with previous work (31,32). However, NPY and AgRP mRNA were almost double baseline after 16-h fasting in pups from obese dams, suggesting that they are more sensitive to food withdrawal than those from lean dams; further work would be required to confirm this. One may argue that the increase in NPY and AgRP mRNA expression in offspring from obese dams would not be sufficient to cause hyperphagia because the levels were similar to those in pups from lean dams. However, the concomitant upregulated Y1 receptor mRNA expression was greater than that in pups from lean dams, which may allow stronger orexigenic effects of NPY in the face of similar levels of fasting NPY mRNA. This agrees with our previous finding of hyper-responsiveness to exogenous NPY in dietary obese rats although their hypothalamic NPY peptide was low (6).

The Y1 receptor is expressed in Arc, PVN, lateral hypothalamic nucleus, dorsomedial, and ventromedial hypothalamic nucleus (33). Few studies have measured Y1 receptor expression, with one report of reduced Y1 receptor number, distribution and mRNA expression in the Arc during fasting (34). This may reflect regional differences. In this study, we measured total hypothalamic Y1 receptor mRNA expression, including the PVN where NPY exerts potent effects on feeding.

In pups from lean dams, although POMC was not altered by fasting, both MC4R and Sim1 were downregulated, suggesting that the anorexigenic pathway was dampened. In regional knockout experiments, PVN MC4R was shown to be involved in feeding regulation, whereas MC4R in other hypothalamic areas affects energy expenditure (23). Sim1 is expressed exclusively in the PVN (23). Although a more detailed subregional analysis is needed, the reduction in Sim1 together with MC4R in pups from lean dams reflects a reasonable physiological response to fasting. These changes were absent in pups from obese dams, suggesting food deprivation for 16 h is not strong enough to activate anorexigenic signaling in them. In other words, after refeeding, the anorexigenic cascade may be less likely to terminate eating. Moreover, AgRP is a potent selective antagonist of MC4R, inhibiting binding of α-MSH (5,35), while NPY can inhibit the melanocortin system via Arc Y1 receptors (21,36,37). In offspring from obese dams, the fasting induced upregulation of hypothalamic NPY and AgRP mRNA expression, together with Y1 receptor, can further weaken the already blunted anorexigenic effects of the melanocortin pathway. This could directly contribute to their hyperphagia. One of the limitations of fresh dissection was that whole hypothalamus was used, which means that the mRNA changes represent an overall response. Further work would be necessary to explore subregional changes.

Reduced baseline Ob-Rb and downstream STAT3 may be an initial step toward leptin resistance commonly observed in obese animals and humans. Indeed it has been shown that maternal obesity led to leptin resistance in offspring even before they were exposed to HFD (38). This explains why in the satiated state, the high leptin levels in offspring from obese dams did not significantly inhibit both NPY and AgRP. Although Ob-Rb and STAT3 were not significantly altered after fasting, there was even more marked reduction in plasma leptin levels during fasting. As stated above, this may contribute to the increased NPY and AgRP mRNA expression.

At weaning, pre-existing maternal obesity led to greater increases in hypothalamic orexigenic regulators, with little response of anorexigenic regulators to overnight fasting, compared with the offspring of lean dams. These changes could directly contribute to the hyperphagia and faster growth of offspring from obese dams.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

This work received project grant funding of the National Health and Medical Research Council of Australia to Margaret Morris.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES