INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

INVOLVEMENT OF LEPTIN in the regulation of physiological energy homeostasis has been ascribed to a negative feedback signal of the adipocyte hormone on the central regulation of food consumption and to enhanced thermogenesis under leptin influence (1, 3, 8). Decrease of food intake and body mass in ob/ob mice after intracerebroventricular injection of leptin led to the assumption of a central leptin-specific satiety signal (1); high local concentration of leptin receptor sites in hypothalamic areas (6) and inhibition of neuropeptide Y synthesis after leptin administration (11) favored this hypothesis. Treatment with leptin decreased the body mass of ob/ob mice more than did limitation of food supply to the amount consumed by leptin-treated animals (5). Interpretation of this phenomenon as a leptin-specific increase in the thermogenic activity of ob/ob mice seemed to be confirmed by enhanced transcription of mRNA encoding uncoupling proteins (14).

In this study, energy expenditure in ob/ob mice was quantified from the nutritional energy metabolized by the animals and their change of body composition (lipid and lean mass) during a short (15 days) as well as during a prolonged treatment period (75 days). Thermoneutral conditions were maintained to exclude effects of ambient temperature on the thermogenic response. Surprisingly, no leptin-specific central satiety signal nor a leptin-specific activation of thermogenesis was necessarily required to explain the effects of the hormone on body mass and food intake at thermoneutral conditions.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and body composition analysis. C57BL/6OlaHsd-Lep ob mice were from Harlan; 14- to 18-wk-old male ob/ob mice weighing 55-61 g were investigated. The mice were single housed and acclimated to thermoneutrality (32 ± 1°C) (12) for 3-5 wk at a 12:12-h light-dark cycle. They were given ground rodent chow (maintenance diet Altromin 1320, Lage, Germany) in feeders for powdered food (Ehret, Emmendingen, Germany). Body weight and food intake of the mice were registered every day. Pair-fed mice were given the amount of food consumed by leptin-treated mice at the same time of the experiment. Body composition was determined by drying to constant weight and extraction of the carcass with a boiling 2:1 mixture of chloroform-methanol; constant weight of the dried lean body mass was attained after seven extractions over a period of 4 days. Analysis of the body composition of 12 untreated ob/ob mice revealed no significant difference in percent water, fat, and lean mass. The amount of body components of ob/ob mice at the beginning of each experiment could thus be calculated from the predetermined percentage and the individual body weight. The change of body composition during the experiment was calculated as the difference of these data and the results of whole body analysis at the end of the experiment.

Production of recombinant leptin. NH₂ terminally (His)₆-tagged mouse leptin starting at Val²² of the proleptin sequence (13) was expressed in Escherichia coli and purified as previously described (10). The protein was >95% pure, and the endotoxin level was <0.3 U/mg of leptin.

Leptin treatment and determination. Leptin was administered by osmotic minipumps (Alzet 1002) from Alza, Palo Alto, CA, in a dose of 280 ng leptin/h (110-122 µg leptin · kg¹ · day¹) for 15 and 75 days, respectively. For subcutaneous implantation of the pumps, mice were anesthetized with 2,2,2-tribromoethanol (300 mg/kg) injected intraperitoneally. Controls and pair-fed mice were implanted with pumps filled with vehicle (phosphate-buffered saline). For treatment with leptin over 75 days, minipumps were changed every 2 wk. The leptin concentration in the plasma was determined by a radioimmunoassay from Linco Research Laboratories, St. Charles, MO. Blood samples were taken from the orbits of mice anesthetized with 2,2,2-tribromoethanol. Formation of antileptin antibodies in the plasma of leptin-treated mice was controlled by an ELISA technique.

Metabolizable food energy. Gross energy of the diet, freeze-dried feces, and urine were determined by means of a calorimeter. The metabolizable energy of the diet, as calculated from the amount of consumed food and fecal and urine losses (4), was 13.31 ± 0.08 kJ/g with ob/ob mice and 13.41 ± 0.10 kJ/g with lean +/+ mice (79.3 and 79.8% of the gross energy of the diet; 7 animals in each group). Caloric values used for the calculation of the energy expenditure based on mobilized endogenous substrate reserves were 39.7 kJ/g for lipid and 17 kJ/g for protein and glycogen, respectively. Relative changes of protein and glycogen in the lean mass of mice were of minor importance for energy expenditure calculation (see table 1) and were not differentiated. A metabolized food energy of 1 kJ/24 h corresponds to a heat production of 11.57 mW (1 J = 1 Ws).

Statistical analysis. The statistical significance of differences between data groups was evaluated by unpaired two-tailed t-tests. Values were considered to be statistically significant when P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The leptin concentration in the plasma of ob/ob mice infused with 280 ng leptin/h (110-122 µg leptin · kg¹ · day¹) for 15 days was measured every third day and averaged a value of 17.1 ± 4.3 ng leptin/ml (n = 9). The established plasma leptin level was supraphysiological, because the leptin concentration measured in the plasma of starved and fed lean mice was 1.6 ± 0.8 ng/ml (n = 11) and 0.9-5.3 ng/ml (n = 9), respectively.

The body mass of ob/ob mice was decreased by ~30% within 15 days of leptin treatment (Fig. 1A). Food consumption of the animals was reduced by >95% within 7 days and remained at this level to the end of the experiment (Fig. 1B). About 75% of the observed body mass reduction was due to loss of fat, with only 7% stemming from a decrease of lean body mass (Table 1). With the metabolized food energy included, total energy expenditure in ob/ob mice treated with leptin for 15 days was calculated as 656.5 kJ/animal, equivalent to a thermogenic power of 506.8 mW. Virtually the same energy (479.1 mW) was expended in ob/ob controls of similar body weight. Total food energy metabolized in leptin-treated ob/ob mice during 15 days was 90.8 kJ/animal, compared with 658.3 kJ/control animal. Thus 86% of the energy expended in mice treated with leptin came from endogenous substrate, mostly tissue triglycerides (83%), whereas energy expenditure in controls was supplied from food intake exclusively (Table 1). Only 6% of the total food energy metabolized by controls within 15 days was stored as endogenous substrate, predominantly as fat. In other words, the nutritional energy metabolized by hyperphagic ob/ob controls minus the energy deposited as body reserves equaled the energy dissipated by leptin-treated ob/ob mice from endogenous sources plus that derived from their reduced food consumption.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Effects of leptin administration and pair-feeding on body weight (A) and intake of metabolizable energy (B) in male ob/ob mice acclimated to 31-33°C. Leptin was infused by means of Alzet 1002 osmotic minipumps for 15 days. Pair-fed animals were given amount of food consumed by leptin-treated mice. Controls and pair-fed mice were implanted with pumps filled with vehicle only. Means ± SE are given from 6 and 9 (control) mice in each group.

Limitation of the food supply in ob/ob mice pair-fed to leptin-infused animals caused a significantly smaller decrease of body weight than did treatment with leptin (Table 1). Loss of water was higher, whereas energy expended at the expense of triglyceride over a period of 15 days was 227.9 kJ/pair-fed mouse vs. 543.4 kJ/leptin-treated animal, and the total energy expenditure in pair-fed ob/ob mice was approximately half of that in leptin-treated animals or ad libitum-fed controls. Quantification of lipid mobilization by treatment with leptin and by starvation in pair-fed ob/ob mice, respectively, indicated that at least 50% of the energy expenditure in leptin-treated ob/ob mice was based on a specific lipolytic effect of the hormone independent of starvation ascribed to a centrally mediated satiation signal of leptin (1).

To test if the restriction of food intake under leptin influence was a cause, or rather a consequence, of the leptin-induced mobilization of fat reserves, ob/ob mice were infused with 280 ng leptin/h for 75 days. The leptin concentration established in the plasma of the mice was controlled on days 4 (12.2 ± 3.9 ng/ml), 55 (14.5 ± 4.3 ng/ml), and 75 (16.5 ± 3.3 ng/ml) of the experiment, respectively. No immune response against the applied murine leptin was detected in the plasma of the mice. After rapid reduction of the body mass during the first 30 days of leptin treatment, the body weight remained constant at ~24 g (Fig. 2A). Food energy metabolized by the animals was reduced by >95% within 6 days of leptin influence. However, it increased significantly from day 20 to day 55 of the experiment and attained a value of ~30 kJ/day per mouse, which remained constant to the end of the experiment (Fig. 2B). This value of metabolized food energy corresponded to ~65% of the nutritional energy consumed by the hyperphagic ob/ob mice (46.5 ± 1.9 kJ/day per animal) before the application of leptin.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effects of leptin administration on body weight (A) and consumption of metabolizable food energy (B) in male ob/ob mice. Leptin was infused for 75 days; Alzet 1002 osmotic minipumps and pumps filled with vehicle (pair-fed mice) were changed every 14 days. Means ± SE are given from 6 animals in each group.

About 90% of the loss of body mass in response to long-term leptin infusion was based on breakdown of fat (Table 2), which represented 55.2 ± 3.3% of the total body mass before administration of leptin, but only 8.9 ± 2.3% after 75 days of hormone treatment. No massive breakdown of muscle protein was indicated by the measured amount of lean mass. Total energy expended in ob/ob mice infused with 280 ng leptin/h for 75 days was 379.5 mW/animal (Table 2). High energy expenditure during the first 15 days (506.8 mW; see Table 1) declined when the body mass of the mice approached a constant low value, i.e., when the fat deposits were depleted and food intake became the basis of energy expenditure. Finally, the expended energy remained constant at an output of 347.1 mW/animal during the last 15 days of long-term leptin infusion, as reflected by the constant body mass and food intake of the mice (Fig. 2 and Table 2). The correlation between food intake and mass of degradable lipid reserves in leptin-treated ob/ob mice favored the view that the restriction of food intake observed under the influence of the hormone was caused by excessive energy supply from tissue triglyceride rather than by a central satiety signal of leptin.

As demonstrated before, body mass reduction occurred significantly more slowly in pair-fed mice than under leptin infusion (Fig. 2). Concomitant with minimized food intake, energy expenditure was mainly based on breakdown of endogenous lipid reserves and attained its lowest level during the first 15 days of the experiment (see Table 1). Body mass reduction ceased, when pair-feeding was continued and the metabolized food energy approached 20 kJ/day (Fig. 2; days 40-45). Energy expenditure attained a value of ~347.1 mW/mouse during the last 15 days of the experiment, when the animals were fed 30 kJ/day in parallel to their leptin-treated companions and expended the same amount of energy (Table 2). Food consumption and energy expenditure of ob/ob mice under continued leptin infusion were similar to data measured in lean (+/+) mice acclimated to thermoneutrality. The amount of metabolizable food energy consumed by +/+ mice was 29.4 ± 1.9 kJ/day (n = 6), and energy expenditure was 340.4 ± 29.7 mW/mouse as calculated from the food intake and change of body composition of the animals.

Results in Fig. 2 show that virtually no decrease of body mass, i.e., of fat reserves, was observed in ob/ob mice offered roughly half the amount of food devoured by ad libitum-fed ob/ob mice (20-30 vs. 40-50 kJ/day). Thus a significant relationship became obvious between food intake and energy expenditure in ob/ob mice in the absence of leptin.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

It is widely accepted that leptin performs a specific centrally mediated satiation signal. Our results do not exclude such a signal; it is, however, not necessarily required to explain the observed effects of the hormone on the energy expenditure in ob/ob mice at thermoneutral conditions. Pair-feeding experiments showed that leptin is responsible for at least 50% of the observed breakdown of tissue triglyceride in ob/ob mice, independent of a possible hypothalamic satiation signal. Excess of nutritional energy set free from endogenous sources in response to treatment with leptin established conditions similar to "overfeeding" in hyperphagic ob/ob controls and may well cause a satiating effect (including inhibition of neuropeptide Y synthesis) similar to that caused by hyperphagia.

Food intake in obese ob/ob mice was extremely reduced under leptin influence, and it was unclear if the animals would die from starvation after continued treatment with the hormone. However, food consumption increased when the lipid reserves in leptin-treated ob/ob mice had been depleted and energy homeostasis became similar to that in lean mice. This indicated that either functionality of a central leptin-specific satiation depends on the amount of triglyceride mobilizable in the animal or that the physiological significance of leptin-induced lipolysis independent of satiety-caused starvation was underestimated.

Energy expenditure has been reported to be similar in adult ob/ob mutants and lean mice when measured at thermoneutral conditions (12). This result was based on short-term monitoring of the oxygen consumption of mice acclimated to 22°C and transferred to thermoneutrality for the measurement. In our experiments, energy expenditure in mice was calculated from data covering experimental periods of 15 or 75 days, respectively, and the animals had been acclimated to thermoneutrality for 3-5 wk before the experiment to reduce effects of lowered ambient temperature on the thermogenic activity. We found the energy expenditure in adult ob/ob mice to be significantly higher than in lean +/+ mice, as evidenced by the different amount of food energy metabolized in the two groups. Untreated ob/ob mice metabolized 41.3 ± 1.8 kJ of food energy per day and animal (Table 1), whereas +/+ mice metabolized 29.4 ± 2.6 kJ (Table 2) per day and mouse. In the case of similar energy expenditure, the difference of ingested food energy, i.e., 11.9 kJ/day, would be stored by ob/ob mice as body reserves; 11.9 kJ corresponds to roughly 0.3 g of fat and a body mass increase of ~0.5-0.6 g, which is far more than the average increase of fat and body mass in ob/ob mice per day (Table 1). We found the energy expenditure in obese ob/ob mice ~30% higher than in lean animals adapted to thermoneutrality. Increased energy expenditure in obese ob/ob mice, which is likely a consequence of the hyperphagia of the mutants (see below), was not able to compensate for the excess of energy intake sufficiently to prevent obesity.

Reduction of body mass in leptin-treated ob/ob mice was faster than in pair-fed ob/ob controls and has been ascribed to increased thermogenesis induced by leptin (8). Again, this was concluded from short-term measurement of oxygen consumption and the age of the animals and temperature conditions were possibly not sufficiently taken into account. Our results confirmed that administration of leptin increased the energy expenditure in ob/ob mice relative to that in their pair-fed counterparts, but leptin-induced energy expenditure never exceeded that measured in hyperphagic ob/ob controls. No leptin-specific thermogenic mechanism seemed to be needed to explain the effect of the hormone at thermoneutral conditions, because both the hyperphagic controls and leptin-treated ob/ob mice reacted with a similarly enhanced energy expenditure, regardless of whether the excess of metabolizable energy came from exogenous or, as under leptin, from endogenous sources. Hyperphagia seemed to cause maximal energy expenditure in ad libitum-fed ob/ob controls; only ~6% of the 40-50 kJ of nutritional energy metabolized per day was stored as fat (Table 1). On the other hand, lipid reserves and body mass of pair-fed ob/ob mice weighing ~40 g remained unchanged when they were offered as little as 20- 30 kJ/day (Fig. 2A), revealing a significant relationship between the energy expenditure in ob/ob mice and the amount of metabolized food. The mechanism of this leptin-independent regulation of the energy dissipation is unknown, but it is obviously responsible for the high energy expenditure in both hyperphagic ad libitum-fed ob/ob controls and in leptin-treated ob/ob mice with nutritional energy mobilized in excess from endogenous sources. No leptin-specific thermogenic mechanism appeared necessary in view of these findings. Increased expression of uncoupling proteins after leptin-administration (14) is likely to indicate increased thermogenic capacity of animals, possibly highly important at low environmental temperatures, but it does not necessarily indicate increased thermogenic activity under thermoneutral conditions.

Perspectives