¹Max-Planck-Institut für physiologische und klinische Forschung, W.G. Kerckhoff-Institut, Bad Nauheim, Germany

²Institut für Experimentelle Endokrinologie, Medizinische Fakultät (Charité) der Humboldt-Universität zu Berlin, Berlin, Germany

Correspondence to: I Schmidt, Max-Planck-Institut für physiologische und klinische Forschung, W.G. Kerckhoff-Institut, Parkstrasse 1, D-61231 Bad Nauheim, Germany. E-mail: I.Schmidt@kerckhoff.mpg.de

OBJECTIVE: To identify the role of hyperleptinaemia in mediating the effects of early postnatal overfeeding in a rat strain known to be prone to manipulations of the early environment which result in predispositions for obesity and associated metabolic and cardiovascular disturbance in later life.

DESIGN: Wistar rats were reared in normal litters (NL, 10-12 pups) or small litters (SL, four pups) from postnatal day 3 and killed for determination of body composition and plasma leptin and insulin concentrations on day 7 or day 21 after having been treated with recombinant leptin (2´50 (pmol/g)/day) or saline from day 1.

RESULTS: Rearing in SL doubled the body fat content and plasma leptin levels in comparison to NL pups by 21 days of age. Under leptin-treatment throughout suckling age, NL pups remained leptin responsive, ie the difference in body fat content was progressively reduced relative to the controls. Until 7 days of age, despite the body fat content of untreated SL pups being 2-fold higher and their plasma leptin level 7-fold higher than that of NL pups, leptin treatment caused the same percentage decreases in body fat in SL than in NL pups. But in contrast to NL pups, the SL pups became leptin resistant thereafter. Plasma insulin levels in 7-day-old leptin-treated SL pups were 3-fold higher than in untreated littermates and 5-fold higher than in the NL groups.

CONCLUSION: Prophylactic leptin treatment does not prevent hyperinsulinaemia and excessive fat deposition in SL pups. On the other hand, selective hyperleptinaemia during suckling age does not trigger leptin resistance and obesity in NL pups. Rather than hyperleptinaemia per se, other factors associated with early postnatal overnutrition, for example, the concurrent hyperinsulinaemia, seem to play a pivotal role for the development of leptin-resistance and life-long obesity risk in SL rats.

International Journal of Obesity (2001) 25, 1168-1174

insulin; early overnutritition; leptin resistance

Introduction

Experimental and clinical studies have shown that overweight at birth and during early postnatal life, as induced, for example, by early overnutrition, may represent a risk factor for later obesity and associated metabolic and cardiovascular disturbances.^1,2,3 Rats raised in small litters (SL) have been used for many years as experimental models to study the immediate and long-term consequences of overnutrition during the early postnatal period,^4,5,6 but the factors responsible for the development of abnormalities in rats reared in SL are still unclear.

Fiorotto et al showed that Sprague-Dawley rats reared in small litters (four pups) from within 24 h after birth developed a 20% higher body fat content than rats reared in normal litters (10 pups) already by 5 days of age.⁷ By 9 days of age, this difference had increased to 60%, and pups had also developed a greater lean body mass. Acceleration of growth and fat deposition were associated with an increase in the daily volume of milk ingested, and in the amount of fat and protein consumed¾with the increase in the fat uptake exceeding the increase in protein uptake.⁷ Detailed studies on Wistar pups reared in small litters demonstrated the development of hyperleptinaemia and hyperinsulinaemia in addition to pronounced overweight during the suckling period^3,8 and, thus, suggested a similarly high susceptibility of this strain to early overfeeding. Changes of the hypothalamic NPY content in response to undernutrition but not to overnutrition reported for 21-day-old Wistar pups point to the development of resistance to leptin and/or insulin of this brain-intrinsic peptidergic system,⁸ which is importantly involved in the control of food intake and assumed to be already functional at this age.^9,10 Possible aberrations of body composition under the condition of overnutrition have not yet been analyzed in the Wistar strain, but they may differ from what is known for wild-type (+/+) Zucker rats because the 30-40% difference in body fat content found at an age of 21 days between pups of this strain raised in small and normal litters was associated with a degree of hyperleptinemia only half of that in overfed Wistar rats at the same age.^8,11

If Zucker rats, heterozygous for the leptin receptor defect 'fa' were, however, reared in small litters, they had developed by 21 days of age a 60% higher fat content and correspondingly higher leptin levels than pups of the same genotype reared in normal litters.¹¹ Moreover, female+/+Zucker rat pups reared in small litters could be maintained at a normal body fat content when being treated prophylactically with 100 (pmol/g)/day leptin, starting 2 days before litter size was reduced. Equally treated female+/fa pups, on the other hand, became leptin resistant, developing a similarly high body fat content as their saline-treated+/fa littermates.¹¹ Taken together, these findings suggest a strong interaction of early postnatal overfeeding caused by rearing in small litters with a defined genetic lesion (in the leptin receptor) as well as with undefined strain-related differences in the genetic background.

Here we tried to use these strain differences to more closely define the conditions causing the development of leptin resistance in juvenile animals by investigating the effect of 'prophylactic' leptin treatment from day 1 to 21 on Wistar rats reared in small litters. For this purpose, we treated Wistar rat pups identically to the Zucker rat pups in a previous study¹¹ until day 21 or until day 7, and determined the effects of rearing in small and normal litters with or without leptin treatment on body composition and plasma concentrations of leptin and insulin.

Methods

Animals

We used male offspring from Wistar rats originally purchased from Charles River (Sulzfeld, Germany). Apart from the brief treatment and weighing procedures the pups were left undisturbed until the end of the experiments at 21 days of age with their mothers in colonies maintained at 22°C on a 12:12 h light:dark cycle. A second group of animals was used for determination of body composition, plasma leptin and plasma insulin concentrations at 7 days of age. On the day of birth (day 0) each pup was marked with subcutaneous injections of India ink. At the beginning of the light phase, when the litters (10-14 pups at birth) were 3 days old, pups were assigned either to litters containing 10-12 pups (normal litters, NL) or to litters containing only four pups (small litters, SL) with starting weights of pups as closely as possible matched for pups reared in NL and SL. Only male pups were used in this study, but females were left with their mothers to obtain the desired number of pups per litter.

Leptin treatment

In some of the litters, half of the pups were treated with recombinant his-tagged murine leptin (17 600 Da) produced as described previously.¹² Starting when the animals were 1 day old (that is, on the second postnatal day), half of the littermates were subcutaneously injected twice daily at light transitions with a dose of 50 pmol/g until the end of the experiment; the other half received control injections (phosphate-buffered saline). Within each litter the pups were randomly assigned to the treatment and control groups in such a way that the average body masses of the leptin-treated animals before the first injection matched those of their control littermates.

Sample preparation and body composition analysis

At the end of the light phase when pups were 7 or 21 days old, litters were removed from their mothers, pups anesthetized for 30 s by CO₂ and decapitated. Blood was collected on ice in tubes containing heparin as anticoagulant. Dilution was determined by weighing, and concentrations appropriately corrected. Plasma was collected after centrifugation and stored at -80°C for RIA measurements. Carcass mass was determined after stomach and intestine had been removed, and body composition¾fat content and fat-free dry mass (FFDM)¾was evaluated by drying to constant weight, followed by whole-body chloroform extraction in a Soxhlet apparatus.¹³

Plasma determinations

Leptin concentration were determined with a commercial RIA-kit for murine leptin (Linco, St Charles, MO). From repeated measurements against rat leptin as standard we determined conversion factors for the estimation of rat leptin concentration. Inter- and intraassay coefficients of variance were 8% and 6%, respectively. To decrease variability due to interassay-variability, we measured pooled samples in each assay and used the deviation from the long-term mean value of the pool for appropriate correction. Plasma insulin concentrations were determined by a commercial RIA-kit (Serono Diagnostics, Freiburg, Germany) with rat insulin (Novo Nordisk, Denmark) as the standard. The intra-assay variability was <10%, samples of all pups were run in the same assay.

Statistical evaluation

Regression analysis was used to analyze the effect of litter size on body fat content independent from interlitter variations in growth.¹⁴ Differences in the mean values of final body mass, body composition data and plasma leptin concentrations of untreated pups reared in small or normal litters were evaluated by t-testing, respectively U-testing if data were not normally distributed. As in previous studies,¹⁴ differences in body composition of leptin-treated and control pups were determined by two-way-ANOVA with litter and treatment as the factors. In this way we were able to clearly separate the leptin treatment effects relevant for this study from the significant (P<0.05) interlitter differences in body composition. Mean values were consequently presented as least-square means (±s.e.). We also determined the changes in fat mass after leptin treatment by calculating for each pup the percentage difference from the mean value of the control littermates, because this procedure normalizes interlitter differences and permits comparison of the leptin effects across ages.^14,15,16,17

Results

Effect of litter size on body fat content and growth of lean body mass

Carcass mass as well as fat mass of 21-day-old Wistar rats reared in SL are much higher than that of pups reared in NL (Figure 1), and the correlation between the two variables is significantly steeper in the SL pups (P<0.01). The mean values of body mass, body fat and FFDM of 21-day-old Wistar rats reared in NL and SL are summarized in Figure 2. Rearing in SL increases both FFDM and fat mass, but the effect on fat deposition is nearly three-times as large as that on growth of FFDM. At the time of weaning, SL pups, thus, have a nearly 3-fold greater total fat mass and twice the percentage body fat than NL pups.

Effect of litter size on plasma leptin concentration

Figure 3 shows the effects of the rearing conditions on plasma leptin concentration and its relation to body fat. Plasma leptin concentration is nearly three times as large in SL pups, ie plasma leptin concentration increases proportional to the increased total fat mass of the carcasses so that leptin per gram body fat is identical for NL and SL pups. The higher total fat mass is, however, partially a consequence of a larger body size and, thus, not an unequivocal indicator of adiposity. Therefore, it should be noted that leptin levels increase more than proportional to the increase in percentage body fat caused by rearing in SL (right part of Figure 3).

Effect of litter size on the responsiveness to leptin treatment

In Figure 4(A) growth of control and leptin treated pups from normal and SL are compared. Body mass of NL pups increased much slower than that of SL pups, and the growth curve obtained for the leptin-treated NL pups lags slightly behind that of the non-treated controls (P=0.06 for body mass difference at day 21), while this effect is not observed for the leptin-treated SL pups. The corresponding changes in percentage body fat were analyzed at 21 days of age and in additional experimental groups killed at 7 days, this being the earliest age at which leptin treatment has been demonstrated to have a significant effect on the body fat content of Zucker rat pups from normal litters.¹⁴ Figure 4(B) shows that in the NL animals percentage body fat in the leptin-treated animals was already significantly lower at day seven and the difference to the control pups increased markedly until day 21. For the SL group body fat content of the leptin-treated animals was also significantly lower than that of their control littermates on day seven, however, this difference became smaller in the further course of treatment and was no longer significant at day 21. To visualize the age dependence of leptin effects more clearly, responsivity to leptin is indicated in Figure 4(C) by the percentage change of fat mass relative to the respective control pups. From day seven to 21, the percentage decrease in fat mass had become more pronounced in NL pups, whereas it had disappeared in SI rats.

Early abnormalities in plasma leptin and insulin concentrations of rats from small litters

By 7 days of age, the 2-fold higher body fat content of SL control pups (see Figure 3(B)) was accompanied by a more than 7-fold higher plasma leptin concentration in comparison to NL pups (22.0±1.9 vs 3.0±0.5 ng/ml). Because early hyperinsulinaemia has been implicated in the development of abnormalities caused by rearing rats in SL, plasma insulin levels were also determined in the 7-day-old leptin-treated and control pups from NL and SL. Figure 5 shows that they indeed developed differently. In the NL animals, the insulin levels were in the low normal range and not influenced by the leptin treatment. In the SL animals, plasma insulin was 5-fold higher and although leptin treatment reduced this level slightly (P=0.11), it remained 3-fold higher than in the NL animals (P<0.001).

Discussion

Hyperleptinaemia per se does not cause leptin resistance

The salient point of this study is that Wistar pups reared in small litters develop a pronounced leptin resistance between 7 and 21 days of age, whereas pups reared in litters of normal size remain leptin sensitive when their temporal mean of plasma leptin concentration is experimentally increased to levels similar to or higher than that of pups from untreated small litters. In weanlings of this frequently used strain not affected by a defined genetic lesion, rearing in SL thus resulted¾even when combined with 'prophylactic' leptin-treatment¾in a similar body fat content and a similar degree of hyperleptinaemia and leptin inresponsiveness as observed in weanlings homozygous for the leptin receptor defect 'fa'.^15,16

Previous studies have shown that the leptin doses used in this study result in plasma leptin levels of about 30-40 ng/ml if constantly applied via miniosmotic pumps to 2 to 3-week-old rat pups,¹⁷ and, although measurements in 7-day-old pups showed that plasma leptin concentrations of suckling-age pups show marked oscillations when doses straddeling this range were delivered by two daily injections rather than continuously, plasma leptin levels in leptin-treated NL pups remained for most of the day in or above the range occurring spontaneously in SL pups.¹⁴ In this context, it is especially noteworthy that the body fat content of 7-day-old SL Wistar pups is already twice that of NL pups, and this early increase in body fat content is associated with several-fold increases of plasma leptin and insulin concentrations. Since the NL pups exposed to experimentally induced high plasma leptin levels remained responsive to the hormone, hyperleptinaemia is, by itself, not sufficient to induce leptin resistance, though with the reservation that oscillating hyperleptinaemia caused by the bolus injections might fail to mimic a constantly elevated leptin due to sustained high endogenous hormone production. This theoretical possibility could be eliminated only by studying the effects of continuous leptin infusions, which can, however, presently not be performed for methodological reasons in pups younger than 15 days.

Divergent physiological abnormalities caused by the leptin receptor defect 'fa' and early postnatal overfeeding

Control of milk intake in suckling rats differs considerably from that of independent ingestion,¹⁸ and during the first two postnatal weeks, milk intake of sucklings seems not to be influenced by leptin treatment or impairment of the leptin receptor.^19,20 Thus, the increased supply of milk energy caused by rearing in SL⁷ probably results in an increase in body fat content at an age at which energy intake is not yet under the control of leptin as it is in mature animals. On the other hand, leptin is already controlling the size of body fat stores during the first postnatal week. This statement is derived from the observation that 7-day-old+/fa and fa/fa Zucker rats, in which the leptin signalling is more or less impaired by the receptor defect, show gene-dose-dependent differences in body fat^21,22 and, moreover, during the first postnatal week, wild-type Zucker rats in which the leptin signal is enhanced by treatment with recombinant leptin, dose-dependently deposit less body fat in comparison to control littermates.¹⁴ In addition, impairment of the leptin receptor results in a gene-dose dependent increase of fat-mass-corrected plasma leptin concentrations and leptin expression in WAT by 10 days of age, suggesting the involvement of the leptin receptor in the control of circulating leptin levels at this early age.^23,24 In contrast to the findings in our 7-day-old SL Wistar pups, the abnormally increased leptin levels in 10-day-old+/fa and fa/fa pups were not associated with significantly increased insulin levels.²³ A further important difference between these two experimental models of early onset excessive fat deposition seems to be leptin receptor binding in the hypothalamus being clearly reduced at 21 days of age in normally reared animals carrying the fa-defect but not in wild-type animals reared in SL.^11,25 The physiological changes associated with the onset of excess fat deposition in SL pups, thus, differ decisively from those in fa/fa Zucker rats reared in NL, even though the most obvious resultant variables, body fat content and leptin responsiveness at weaning, are very similar.

Selective hyperinsulinaemia but not hyperleptinaemia during suckling age increases the risk for obesity-associated abnormalities

Leptin-treatment was not able to prevent a several-fold increase in plasma insulin concentration in 7-day-old SL pups. This might either indicate the beginning development of leptin resistance or reflect a generally lacking effect of leptin on insulin secretion at this early age. The lacking effect of leptin treatment also on plasma leptin concentrations of NL pups suggests the latter, but this not completely conclusive because the plasma concentrations of insulin in the NL pups are so low at this age that it is not clear whether a further reduction might be conceivable. In any case, rats from SL litters, whether leptin treated or not, were thus exposed to several-fold increases in insulinaemia at least from the age of 7 days.

Experimental evidence has been provided suggesting a pivotal role of hyperinsulinaemia during the second postnatal week in provoking lifelong alterations predisposing to obesity and syndrome X-like metabolic aberrations in adult animals observed after early postnatal overnutrition.^3,26,27 Since, however, early postnatal overnutrition is not only associated with hyperinsulinaemia but also with hyperleptinaemia, the question remained, whether the abnormal leptin levels per se might also be critically involved in triggering leptin resistance and programming syndrom X-like abnormalities in later life. However, as demonstrated in this study, functional leptin resistance in overfed pups emerges only after the first postnatal week, and only in animals simultaneously developing hyperinsulinaemia and not in the normoinsulinaemic NL pups made hyperleptinaemic by treatment with exogenous leptin.

This developmental pattern seems particularly interesting because the time around postnatal days 10/11 has been demonstrated to be of crucial importance for the differentiation, functional maturation and 'programming' of neuroendocrine systems.^28,29,30,31 Subcutaneous or intrahypothalamic insulin treatment during this particularly critical period of early development resulted in a markedly increased body mass,^26,27 suggesting increased body fat content at weaning. This was followed by persisting overweight and syndrome X-like abnormalities throughout life. Subcutaneous treatment with high doses of leptin throughout suckling-age, on the other hand, decreased body fat content at weaning by more than 30% (this study), and when wild-type Zucker rats were leptin-treated with an even higher dose (600 (pmol/g)/day) from day 7 to 16, they still did not show abnormalities in body fat content at 3 months of age (Ingrid Schmidt, unpublished results). On the other hand, triggering endogenous hyperinsulinaemia in combination with hyperleptinaemia by rearing in SL was observed to predispose for lasting overweight, hyperinsulinaemia and syndrome X-like changes,³ accompanied by hypothalamic alterations, suggesting some kind of central resistance to leptin and/or insulin.^3,8 The immediate and lasting changes observed in perinatally overnourished animals seem, thus, not due to the concomitant change in circulating leptin levels by itself but rather to associated changes causing alterations in neural circuits, with the consequences of increases in peripheral and central insulin levels being presently the experimentally best documented example for such malprogramming processes.^3,32,33,34

Irrespective of the question which nutritional and/or metabolic changes fuel the increased fat deposition of SL pups during suckling age^7,11 or thereafter,^5,6 we have, thus, to consider the question which factors might be able to perpetuate ('program') an obesity disposition in vigorously growing animals which are increasing their body mass and fat mass several-fold until adulthood, while being offered from weaning the same food as NL pups. Peripheral factors, like changes in adipose tissue cellularity and metabolism,⁴ might contribute to the permanent abnormalities seen in SL rats, but assuming additional central nervous changes seems inevitable to explain the concerted metabolic, hormonal and behavioral differences observed in adulthood.³³ While central changes caused by hyperinsulinaemia are presently the best documented changes that might cause a malprogramming by early overnutrition, other hormonal or metabolic factors associated with early-onset obesity, particularly alterations that might, like hyperinsulinaemia, result in permanent changes in the central sympathetic system, must receive further experimental consideration.^32,33,34,35

Acknowledgements

We are indebted for help with the experiments to many people, particularly to Diana Fuchs (RIA measurements) and Knut Schwarzer (statistics). Aventis Pharma, Frankfurt, is gratefully acknowledged for the generous supply of leptin. This work was supported by DFG (PL 241/1-3).

1 Kramer MS, Barr RG, Leduc DG, Boisjoly C, Pless IB. Infant determinants of childhood weight and adiposity. J Pediatr 1985; 107: 104-107, MEDLINE

2 Dietz WH. Critical periods in childhood for the development of obesity. Am J Clin Nutr 1994; 59: 955-959, MEDLINE

3 Plagemann A, Harder T, Rake A, Voits M, Fink H, Rohde W, Dörner G. Perinatal elevation of hypothalamic insulin, acquired malformation of hypothalamic galaninergic neurons, and syndrome X-like alterations in adulthood of neonatally overfed rats. Brain Res 1999; 836: 146-155, MEDLINE

4 Knittle J, Hirsch J. Effect of early nutrition on the development of rat epididymal fat pads: cellularity and metabolism. J Clin Invest 1968; 47: 2091, MEDLINE

5 Miller DS, Parsonage SR. The effect of litter size on subsequent energy utilization. Proc Nutr Soc 1972; 31: 30A, MEDLINE

6 Oscai LB, McGarr JA. Evidence that the amount of food consumed in early life fixes appetite in the rat. Am J Physiol 1978; 235: R141-R144, MEDLINE

7 Fiorotto ML, Burrin DG, Perez M, Reeds PJ. Intake and use of milk nutrients by rat pups suckled in small, medium, or large litters. Am J Physiol 1991; 260: R1104-R113, MEDLINE

8 Plagemann, A, Harder T, Rake A, Waas T, Melchior K, Ziska T, Rohde W, Dörner G. Observations on the orexigenic hypothalamic neuropeptide Y-system in neonatally overfed weanling rats. J Neuroendocrinol 1999; 11: 541-546, Article MEDLINE

9 Capuano CA, Leibowitz SF, Barr, GA. Effect of paraventricular injection of neuropeptide Y on milk and water intake of preweanling rats. Neuropeptides 1993; 24: 177-182, MEDLINE

10 Kowalski TJ, Houpt TA, Jahng JW, Okada N, Chua SC, Smith GP. Ontogeny of neuropeptide Y expression in response to deprivation in lean Zucker rat pups. Am J Physiol 1998; 44: R466-R470,

11 Schmidt I, Schoelch C, Ziska T, Schneider D, Simon E, Plagemann A. Interaction of genetic and environmental programming of the leptin system and of obesity disposition. Physiol Genomics 2000; 3: 113-120, MEDLINE

12 Müller G, Ertl J, Gerb M, Preibisch G. Leptin impairs metabolic actions of insulin in isolated rat adipocytes. J Biol Chem 1997; 272: 10585-10593, MEDLINE

13 Markewicz B, Kuhmichel G, Schmidt I. Onset of excess fat deposition in Zucker rats with and without decreased thermogenesis. Am J Physiol 1993; 265: E478-E486, MEDLINE

14 Kraeft S, Schwarzer K, Eiden S, Nuesslein-Hildesheim B, Preibisch G, Schmidt I. Leptin responsiveness and gene dosage for leptin receptor mutation (fa) in newborn rats. Am J Physiol 1999; 276: E836-E842, MEDLINE

15 Olbort M. Auswirkungen von einer und zwei Kopien des Leptinrezeptordefektes fa (fatty) auf die Körperzusammensetzung und den Plasma-Leptinspiegel von Ratten im Säuglingsal-ter. Thesis, Justus-Liebig-Universität Giessen. 1998;

16 Stehling O, Döring H, Nuesslein-Hildesheim B, Olbort M, Schmidt I. Leptin-doses halving the body fat of lean pups are not effective in genetically obese (fa/fa) rat pups>. In: Blum WF, Kiess W, Rascher W (eds). Leptin¾the voice of the adipose tissue. Barth: Heidelberg, 1997, 140-147.

17 Eiden S, Preibisch G, Schmidt I. Leptin responsiveness of juvenile rats: proof of leptin function within the physiological range. J Physiol 2000; 530: 131-139,

18 Hall WG. What we know and don't know about the development of independent ingestion in rats. Appetite 1985; 6: 333-356, MEDLINE

19 Stehling O, Döring H, Nuesslein-Hildesheim B, Olbort M, Schmidt I. Leptin does not reduce body fat content but augments cold defense abilities in thermoneutrally reared rat pups. Pflügers Arch 1997; 434: 694-697, MEDLINE

20 Buchberger P, Schmidt I. Is the onset of obesity in suckling fa/fa rats linked to a potentially larger milk intake? Am J Physiol 1996; 271: R472-R476, MEDLINE

21 Schwarzer K, Döring H, Schmidt I. Different physiological traits underlying increased body fat of fatty (fa/fa) and heterozygous (+/fa) rats. Am J Physiol 1997; 272: E100-E106, MEDLINE

22 Meierfrankenfeld B, Abelenda M, Jauker H, Klingenspor M, Kershaw EE, Chua SC, Leibel RL Jr, Schmidt I. Perinatal energy stores and excessive fat deposition in genetically obese (fa/fa) rats. Am J Physiol 1996; 270: E700-E708, MEDLINE

23 Zhang Y, Olbort M, Schwarzer K, Nuesslein-Hildesheim B, Nicolson M, Murphy E, Kowalski TJ, Schmidt I, Leibel R. The leptin receptor mediates apparent autocrine regulation of leptin gene expression. Biochem Biophys Res Commun 1997; 240: 492-495, Article MEDLINE

24 Zhang Y, Hufnagel C, Eiden S, Guo K, Diaz P, Leibel R, Schmidt I. Mechanisms for LEPR-mediated regulation of leptin expression in brown and white adipocytes in rat pups. Physiol Genomics 2001; 4: 189-199, MEDLINE

25 Ziska T, Waas T, Harder T, Rohde W, Plagemann A. Hypothalamic leptin binding in neonatally over- or underfed weanling rats. Exp Clin Endocrinol Diabetes 1999; 107: (Suppl 1) 75, MEDLINE

26 Harder T, Plagemann A, Rohde W, Dörner G. Syndrome X-like alterations in adult female rats due to neonatal insulin treatment. Metabolism 1998; 47: 855-862, MEDLINE

27 Plagemann A, Heidrich I, Götz F, Rohde W, Dörner G. Lifelong enhanced diabetes susceptibility and obesity after temporary intrahypothalamic hyperinsulinism during brain organization. Exp Clin Endocrinol 1992; 99: 91-95, MEDLINE

28 Hill DJ, Strutt B, Arany E, Zaina S, Coukell S, Graham CF. Increased and persistent circulating insulin-like growth factor II in neonatal transgenic mice suppresses developmental apoptosis in the pancreatic islets. Endocrinology 2000; 141: 1151-1157, MEDLINE

29 Eriksson P, Ahlbom J, Fredriksson A. Exposure to DDT during a defined period in neonatal life induces permanent changes in brain muscarinic receptors and behaviour in adult mice. Brain Res 1992; 582: 277-281, MEDLINE

30 Ahima RS, Prabakaran D, Flier JS. Postnatal leptin surge and regulation of circadian rhythm of leptin by feeding. Implications for energy homeostasis and neuroendocrine function. J Clin Invest 1998; 101: 1020-1027, MEDLINE

31 Plagemann A, Harder T, Rake A, Janert U, Melchior K, Rohde W, Dörner G. Morphological alterations of hypothalamic nuclei due to intrahypothalamic hyperinsulinism in newborn rats. Int J Dev Neurosci 1999; 17: 37-44, MEDLINE

32 Dörner G, Plagemann A. Perinatal hyperinsulinism as possible predisposing factor for diabetes mellitus, obesity and enhanced cardiovascular risk in later life. Horm Metab Res 1994; 26: 213-221, MEDLINE

33 Levin B. The obesity epidemic: metabolic imprinting on genetically suspectible neural circuits. Obes Res 2000; 8: 342-347, MEDLINE

34 Schmidt I. The role of juvenile thermoregulatory thermogenesis in the development of normal energy balance or obesity. In: Kosaka M, Sugahara T, Schmidt KL, Simon E (eds). Thermotherapy for neoplasia, inflammation, and pain. Springer: Tokyo, 2000, 215-225.

35 Young JB, Morrison SF. Effects of fetal neonatal environment on sympathetic nervous system development. Diabetes Care 1998; 21: (Suppl 2) B156-B160, MEDLINE

Figure 1 Total body fat mass as a function of carcass mass in 21-day-old male Wistar rats reared either in litters containing 10-12 pups (normal) or four pups (small litters). The correlation coefficients are 0.58 and 0.73, respectively.

Figure 2 Mean values (±s.e.) of body mass, fat-free dry-mass (FFDM), body fat mass and percentage body fat in 21-day-old male Wistar rats reared in normal (n=29) or small (n=14) litters. White bars, normal litters; black bars, small litters. ***P<0.001 (U-test).

Figure 3 Mean values (±s.e.) of plasma leptin concentration (absolute and per gram or percentage body fat) in 21-day-old male Wistar rats for which body composition data are shown in Figure 2. White bars, normal litters; black bars, small litters. **P<0.01; ***P<0.001 (U-test).

Figure 4 Part A shows growth (least square means±s.e.) of leptin-treated (thin lines) and control (thick lines) Wistar rats reared in normal (NL) and small (SL) litters from day 3 to 21. Animals were treated with leptin starting on the second postnatal day and sacrificed for body composition analysis at 21 days of age (n=5-6 per group). Part B shows the body fat content (least square means±s.e.) of 21-day-old leptin-treated (white) and control (black) animals and of a second group of animals, treated identically, but sacrificed already at 7 days of age (n=8-14 per group). To facilitate comparison across ages, the mean (±s.e.) of changes in body fat content of 7 and 21-day-old pups are shown in part (C); these values are expressed as the percentages by which body fat mass of each leptin-treated pup differs from the mean values of its respective control littermates and were tested for significance from zero by t-testing. *P<0.05; **P<0.01; ***P<0.001.

Figure 5 Plasma insulin concentrations (±s.e.) of 7-day-old leptin-treated (white) and control (black) pups from normal (NL) and small (SL) litters; n=8-14 per group. ***P<0.001 (U-test) for differences between pups from normal litters and the respective small litter groups.