INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LEPTIN, A 16-KDA PROTEIN secreted by the adipose tissue, has been hypothesized to act as a signal from the periphery to the central nervous system (CNS), indicating the size of energy stores (39). Leptin enters the brain by a saturable transport system (3), where it activates hypothalamic long-form leptin receptors, Ob-Rb. Although five leptin receptor isoforms exist in mice, only Ob-Rb is abundantly expressed in the hypothalamus and appears to possess full signaling capabilities (36). These receptors are responsible for the inhibitory effect of leptin on food intake and other physiological functions including reproduction. Initially it was assumed that leptin would act as a lipostatic signal, inhibiting food intake during periods of positive energy balance and enlargement of body fat mass (39). It has been reported that in young lean rats there is a direct negative correlation between circulating concentrations of leptin and body fat mass when leptin remains within the narrow range normally found in these animals (6). Leptin administration also induces weight loss and a transient inhibition of food intake in lean, wild-type mice (17, 27). Low doses of leptin administered directly into the brain produce effects comparable to those seen with larger doses administered in the periphery, supporting the concept that leptin-induced changes in energy balance are mediated by receptors in the CNS (34). In obese animals (16) and humans (5), circulating concentrations of leptin increase in proportion with body fat mass and the relationship between leptin and adiposity is lost. It is now hypothesized that, rather than increased levels of leptin signaling energy excess, a reduction in circulating concentrations of leptin conserves energy during periods of energy deficit by inhibiting activity of some energy-expensive processes (1).

A majority of human obesity is characterized by high circulating levels of leptin and increased adipose tissue expression of leptin mRNA. These high levels of leptin do not downregulate body fat mass, which has led to the concept of leptin resistance (5), defined as a defect in leptin signaling that allows a dysregulation of energy balance and a failure to decrease body weight and food intake in response to increasing leptin concentrations (7, 9). In addition, the leptin levels in obese humans appear unresponsive to short-term changes in nutritional status, as large changes in body mass (>7%) are required before circulating concentrations of leptin change significantly (19). Several authors have reported that male mice and rats exposed to a high-fat diet also are unresponsive to peripheral injections of leptin (23, 37) and that continued consumption of a high-fat diet eventually results in reduced central leptin sensitivity (24), suggesting that this model accurately reflects the human obese condition. The development of peripheral leptin resistance appears to be rapid, developing within 16 days in 5-wk-old mice on a 45% kcal fat diet (37) and within 5 days in rats fed a 56% kcal fat diet (23).

In a previous experiment, we found that female C57BL/6J mice fed a high-fat (45% kcal fat) diet for 15 wk remained responsive to peripheral administration of leptin both as a constant peripheral infusion and as a single bolus injection (14). The obvious explanation for the conflicting results with high-fat diet-fed mice in our study and those of others (23, 37) was that the gender of the mice influenced sensitivity to leptin, as we have previously reported significant effects of gender on the response to both central and peripheral administration of leptin in mice that overexpress agouti protein (15). The initial objective of this study was to identify factors that contribute to the development of leptin resistance in young mice weaned onto a high-fat diet. Due to the failure to induce leptin resistance in older female mice (14), we included both males and females in this study to determine the importance of gender in the development of leptin resistance in high-fat diet-fed mice. Mice were offered the high-fat diet from 10 days of age, while they were still suckling, so that they were never exposed to a low-fat diet. On the basis of the observation that 5-wk-old mice were leptin resistant after 16 days on a 45% kcal fat diet (37), we anticipated that the high-fat diet-fed mice would be leptin resistant, providing an appropriate model for the study of juvenile, diet-induced obesity. The initial experiments demonstrated that high-fat diet-fed male and female mice were, at least partially, resistant to central injections of leptin but were fully responsive to peripheral infusions of leptin; therefore, additional experiments compared the effects of strain, gender, and housing conditions of the mice on the sensitivity to peripherally infused leptin in 5-wk-old mice weaned onto either low- or high-fat diet.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and diet. Male and female C57BL/6J and NIH Swiss mice were obtained from breeding colonies maintained at the University of Georgia. Mice were housed at 73°F with lights on 12 h/day from 7:00 AM. They had free access to food and water except where specified. Dams and their litters were fed either low-fat diet containing 10% kcal as fat (Diet 12450B; Research Diets) or high-fat diet containing 45% kcal as fat (Diet 12451; Research Diets) starting 10 days postpartum. Pups were weaned at 28 days of age and, at 30 days of age, single-housed mice were housed individually in cages with grid floors to allow for measures of food intake. Group-housed mice were housed three or four mice per cage on bedding. Only two male and two female pups were taken from each litter for incorporation into studies to minimize litter-specific responses. All experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of Georgia and were conducted in conformity with the American Physiological Society's guiding principles for the care and use of animals (2).

Experiment 1: the effects of intracerebroventricular leptin injection on single-housed C57BL/6J mice. This study tested whether young, high-fat diet-fed C57BL/6J mice responded to central leptin injections by reducing food intake or weight gain. Male and female single-housed C57BL/6J mice, aged 34 days, were fitted with bilateral cannulas of the lateral ventricle using the procedure of Guild and Dunn (12). Briefly, a 1-cm saggital incision was made along the skin covering the skull followed by placement of the cannulas at 0.6 mm posterior and ±1.6 mm lateral of the bregma. One week later, mice within each dietary treatment were divided into two groups of 7-9 males or 8-13 females and were injected unilaterally with 5 µg leptin (recombinant murine leptin; R&D Systems) in 1.5 µl of PBS or an equal volume of PBS. This dose was in the middle of the range of doses used by Van Heek et al. (37) to demonstrate development of peripheral but not central leptin resistance in mice fed a 45% kcal fat diet for 56 days. The mice were housed in shoebox cages with an elevated grid floor to allow measurement of food intake. Food was placed on the grid floor, and intake, measured to 0.01 g, was corrected for spillage that collected below the grid. Food intakes and body weights were recorded at 24-h intervals for 3 days before injections. On test days, mice were food deprived from 7:00 AM and were injected at 4:00 PM. Food was returned to the cages 1 h after injection, and food intake was recorded 4, 14, 24, 38, and 48 h postinjection. Body weights were recorded 14 and 38 h postinjection. One week later, the procedure was repeated with treatments switched. At the end of the study, methylene blue dye was injected into the cannulas and placement was verified by visually examining staining of the ventricles.

Experiment 2: the effects of peripheral leptin infusion on single-housed C57BL/6J mice. The results of the previous experiment indicated that the young mice weaned onto high-fat diet were resistant to central injections of leptin. The objective of this experiment was to test whether the high-fat diet-fed mice were also resistant to the effects of peripheral leptin infusion on food intake, body composition, and insulin status. C57BL/6J mice fed low- or high-fat diet were single housed and daily food intakes and body weights were recorded for 3 days. Male and female mice from each dietary treatment were further divided into two weight-matched groups of five or six male mice or seven to nine female mice. At 34 days of age, the mice were fitted with intraperitoneal Alzet miniosmotic pumps (model 1002; Durect). One group was infused with PBS and the other with 10 µg leptin/day for 13 days. Food intakes and body weights were measured daily. On day 10 of infusion, mice were deprived of food from 7:00 AM to 12:00 PM, and small blood samples were collected from the tail for measurement of fasting insulin (Mouse Insulin RIA; Linco Research) and glucose (Accumet glucometer; Boehringer Mannheim). On day 13, mice were decapitated and trunk blood was collected for measurement of serum leptin (Mouse Leptin RIA; Linco Research). Gonads and gonadal, mesenteric, and retroperitoneal fat were weighed and the gut was cleaned. Tissues were returned to the carcass for determination of body composition as described previously (18).

Experiment 3: the effects of peripheral leptin infusion on single-housed NIH Swiss mice. In experiment 2, we found that C57BL/6J mice fed high-fat diet responded to peripheral leptin infusions by reducing body fat content. Therefore, in this experiment we tested whether the retention of leptin responsiveness was strain specific and determined whether NIH Swiss mice became leptin resistant when they were fed a high-fat diet. These mice showed a 96% preference for the high-fat over the low-fat diet when offered a choice between the two diets, and older male NIH Swiss mice significantly increased their body fat mass when offered the high-fat diet used in this study. The experimental design was exactly the same as in experiment 2, except that groups of five male and four or five female single-housed NIH Swiss mice were used in this study and carcass composition was not determined, although fat depot weights were recorded.

Experiment 4: the effects of peripheral leptin infusion on group-housed C57BL/6J mice. In experiments 2 and 3, we found that single-housed mice fed high-fat diet responded to peripheral infusions of leptin, independent of strain. Leptin reduced body fat content without causing any substantial inhibition of food intake, implying that there was a leptin-induced stimulation of thermogenesis. Therefore, we conducted this study with the mice housed in conditions that minimized the need for heat production and tested whether group-housed C57BL/6J mice fed high-fat diet responded to peripheral infusions of leptin. The experimental design was the same as for experiment 2, except that mice were group housed, as described above, so that food intakes could not be recorded. There were eight or nine male mice and seven or eight female mice per treatment group.

Experiment 5: the effects of peripheral leptin infusion on group-housed NIH Swiss mice. Group-housed C57BL/6J mice in experiment 4 responded to leptin by reducing their body fat content; therefore, we tested whether responsiveness was dependent on the strain of the mouse. The experimental design was the same as for experiment 4, except that there were seven to nine male and seven or eight female NIH Swiss mice per treatment group.

Experiment 6: the effects of peripheral leptin injection on single-housed C57BL/6J mice. All of the high-fat diet-fed mice in experiments 2-5 responded to peripheral infusions of leptin by reducing body fat content. Others have reported leptin resistance in high-fat diet-fed mice given daily peripheral injections of leptin (24, 37). Therefore, in this study we tested whether single-housed C57BL/6J mice were resistant to intraperitoneal injections of leptin. Single-housed male and female mice from each dietary treatment were divided into two groups, and daily food intakes and body weights were recorded for 3 days before initiation of injections. Mice were 35 days of age on the first day of injection. On three consecutive test days, each mouse was injected intraperitoneally with either PBS or 30 µg leptin (~1.5 mg/kg). Food intakes and body weights were recorded for 5 days after the first injection. One hour after the last injection, small blood samples were obtained from the tail vein for measurement of serum leptin concentration. The procedure was repeated 1 wk later with treatments switched.

Statistics. Body weight, weight change, and energy intake measures were analyzed by repeated-measure ANOVA with day or time as the repeated measure. Baseline measures of body weight or energy intake were used as covariates in body weight and energy intake analysis, and in experiments where mice served as their own control, mouse was considered a covariate. Organ weights, body composition, and serum measurements were analyzed by ANOVA. Each analysis was initially conducted with data for both males and females (Statistica, StatSoft). The analysis was then repeated independently for each sex. In some instances, the mice were further separated by diet to detect differences between control and leptin-treated mice. Differences between individual groups on a specific day were determined by post hoc Duncan's multiple-range test. Differences were considered significant at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experiment 1: the effects of intracerebroventricular leptin injection on single-housed C57BL/6J mice. This experiment tested the effects of centrally injected leptin on male and female C57BL/6J mice fed low-fat or high-fat diet. Weight gain was influenced by both gender and leptin (gender: P < 0.01; leptin: P < 0.001; leptin × time: P < 0.05; Fig. 1). Leptin had no effect on weight gain of any of the male mice but reduced weight gain 38 h after the injection in female mice fed the low-fat diet, but not those fed high-fat diet (Fig. 1B). The changes in weight gain even of low-fat diet-fed mice were small, but were in the same range (3-5% of body weight) reported by others testing the response of C57BL/6J mice to central injections of leptin (24). Male mice fed the high-fat diet had higher cumulative energy intakes than those fed the low-fat diet at 4, 24, 38, and 48 h postinjection (P < 0.05; Fig. 2A), but leptin did not have any significant effect on intake of male mice in either dietary treatment at any time point. Female mice fed high-fat diet had higher energy intakes than those fed the low-fat diet 24 h after injection (P < 0.01). Leptin reduced energy intake of low-fat diet-fed female mice 38 (P < 0.01) and 48 h (P < 0.04) postinjection but did not have any significant effect on mice fed high-fat diet (Fig. 2B). The reduction in energy intake of low-fat fed mice was small (15.7% for females) but similar to that reported by others (24) for mice receiving lateral ventricle injections of leptin.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Change in body weight of male (A) and female (B) C57BL/6J single-housed mice fed low- or high-fat diet and given a lateral ventricle injection of 5 µg leptin or PBS in experiment 1. Data are means ± SE for groups of 7-9 males or 8-13 female mice. Values for weight change in female mice at 38 h that do not share a common superscript are significantly different at P < 0.05, determined by 2-way ANOVA and post hoc Duncan's multiple-range test.

View larger version (36K): [in this window] [in a new window]   Fig. 2.   Cumulative food intake of male (A) and female (B) C57BL/6J single-housed mice fed low- or high-fat diet and given a lateral ventricle injection of 5 µg leptin or PBS in experiment 1. Data are means + SE for groups of 7-9 males or 8-13 female mice. Values for intake at specific time points that do not share a common superscript are significantly different at P < 0.05, determined by 2-way ANOVA and post hoc Duncan's multiple-range test. The absence of superscripts indicates no difference between groups at that time point.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Weight change of male (A) and female (B) single-housed C57BL/6J mice that received a third ventricle injection of 0.5 µg of leptin or PBS in experiment 1. Data are means ± SE for groups of 5 or 6 mice. Values for weight change of females 48 h after the injection that do not share a common superscript are significantly different at P < 0.05, determined by 2-way ANOVA and post hoc Duncan's multiple-range test.

Experiment 2: the effects of peripheral leptin infusion on single-housed C57BL/6J mice. This experiment tested the effects of peripheral leptin infusion on single-housed male and female C57BL/6J mice. Serum leptin concentrations were higher in female than male mice and increased with leptin infusion (Table 1). Fasting glucose and insulin concentrations on day 10 of infusion were higher in males than females and the high-fat diet further increased fasting insulin in male mice. Leptin reduced fasting glucose levels in female mice fed low-fat diet and fasting insulin levels in male mice fed high-fat diet (Table 1).

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Daily body weights of male (A and B) and female (C and D) single-housed C57BL/6J mice that were infused with 10 µg leptin/day from age 34 days in experiment 2. Data are means ± SE for groups of 5 or 6 male mice or 7-9 female mice. * Significant (P < 0.05) difference between leptin-treated and control mice determined by repeated-measures ANOVA and post hoc t-tests. A and C, low-fat diet-fed mice; B and D, high-fat diet-fed mice.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Daily food intakes of male (A and B) and female (C and D) single-housed C57BL/6J mice that were infused with 10 µg leptin/day from age 34 days in experiment 2. Data are means ± SE for groups of 5 or 6 male mice or 7-9 female mice. * Significant (P < 0.05) difference between leptin-treated and control mice, determined by repeated-measures ANOVA and post hoc t-tests. A and C, low-fat diet-fed mice; B and D, high-fat diet-fed mice.

Experiment 3: the effects of peripheral leptin infusion on single-housed NIH Swiss mice. Results for the leptin-infused single-housed NIH Swiss mice were similar to those found in C57BL/6J mice and are summarized in Table 4. There was no effect of gender on serum leptin concentrations, but leptin infusion significantly increased serum leptin levels in all treatment groups. Fasting glucose concentrations were higher in males than females, and leptin reduced fasting glucose concentrations in female mice fed low-fat diet (P = 0.05). There were no differences in fasting insulin levels between any of the groups.

Experiment 4: the effects of peripheral leptin infusion on group-housed C57BL/6J mice. This experiment tested the effects of peripheral leptin infusion on group-housed male and female mice. Serum leptin concentrations were higher in male than female mice and were approximately doubled with leptin infusion (data not shown). Fasting serum glucose concentrations were higher in male than female mice and neither diet nor leptin influenced fasting glucose or insulin concentrations (data not shown). There was a significant effect of leptin on body weight of all of the mice (gender: NS; diet: NS; leptin: P < 0.0001; day: P < 0.0001; gender × diet: P < 0.05; gender × day: P < 0.0001; leptin × day: P < 0.0001; gender × diet × day: P < 0.005; Fig. 6).

View larger version (35K): [in this window] [in a new window]   Fig. 6.   Daily body weights of male (A and B) and female (C and D) group-housed C57BL/6J mice that were infused with 10 µg leptin/day from age 34 days in experiment 4. Data are means ± SE for groups of 8 or 9 male mice or 7 or 8 female mice. * Significant (P < 0.05) difference between leptin-treated and control mice, determined by repeated-measures ANOVA post hoc t-tests. A and C, low-fat diet-fed mice; B and D, high-fat diet-fed mice.

Experiment 5: the effects of peripheral leptin infusion on group-housed NIH Swiss mice. This experiment was conducted to determine whether failure to consistently induce leptin resistance in group-housed mice fed high-fat diet was due to the strain of the mice. Male and female NIH Swiss mice were group housed and infused with leptin for 13 days and the data are summarized in Table 6. Serum leptin levels were higher in males than females and were increased by leptin infusion. The difference did not reach statistical significance in female mice fed high-fat diet. Serum glucose concentrations were higher in male than female mice and increased in female mice fed a high-fat diet. Leptin reduced fasting glucose concentrations of female mice fed the low-fat diet (P < 0.05). There were no gender, diet, or leptin effects on fasting insulin concentrations.

Experiment 6: the effects of peripheral leptin injection on single-housed C57BL/6J mice. This experiment was conducted to determine if peripheral leptin resistance in high-fat diet-fed mice was dependent on the method of leptin administration. A single intraperitoneal injection of 30 µg leptin produced an ~100-fold increase in circulating leptin concentrations in all mice, measured 1 h after the last injection (data not shown). There were no effects of diet or leptin on weight gain in male mice; however, leptin inhibited weight gain on the second and third day of injection in low-fat diet-fed female mice (gender: NS; diet: NS; leptin: NS; day: P < 0.0001; gender × leptin: P < 0.05; gender × leptin × day: P < 0.05; Fig. 7A). There was no significant effect of diet or leptin on the energy intakes of male mice. In female mice, leptin inhibited intake of low-fat diet-fed females on the days of leptin injection (Fig. 7B)

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Weight change (A) and food intake (B) of single-housed C57BL/6J mice that were injected intraperitoneal with 30 µg of leptin in each of 3 days in experiment 6. Data are means + SE for groups of 5 or 6 mice. * Significant differences (P < 0.05) between control and leptin-treated low-fat diet-fed female mice, determined by repeated-measures ANOVA and post hoc t-test.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The initial objective of these studies was to develop a mouse model of juvenile diet-induced leptin resistance, and we chose to wean mice onto a high-fat diet because it was anticipated that this would accelerate the development of leptin resistance. Other investigators have reported that exposure of 5-wk-old mice to a high-fat diet for 16 days induces peripheral, but not central, leptin resistance (37). Therefore, it was surprising to find that 6- to 7-wk-old mice that had been weaned onto a high-fat diet showed an attenuated response to the central effects of leptin on food intake and weight gain but remained fully responsive to the effects of peripheral infusions of leptin on body weight and body fat mass. These results are consistent with those of Halaas et al. (13), who found that female mice fed a 45% kcal fat diet for 10 wk lost weight in response to daily intraperitoneal injections of leptin, but contrast with those of Van Heek et al. (37), who found that male mice became resistant to the effects of intraperitoneal injections of leptin on food intake and body weight when they were fed a 45% kcal fat diet for only 16 days. In experiment 6, we tested whether the method of peripheral leptin administration influenced our interpretation of whether the mice were leptin resistant. We found that single daily injections of 30 µg leptin, which induced a 100-fold increase in circulating concentrations of leptin, had no significant effect on either food intake or body weight of low-fat or high-fat diet-fed male mice. Thus it is possible that previous studies that found mice resistant to peripheral injections of leptin would find them responsive to peripheral infusions of leptin.

Van Heek et al. (37) reported that high-fat diet-fed C57BL/6 mice became resistant to peripheral injections of leptin faster than high-fat diet-fed AKR mice, implying that background strain influences the development of leptin resistance and that a factor other than dietary fat content, such as body mass or adiposity, is responsible for the leptin resistance (37). In the experiments described here we tested both C57BL/6J and NIH Swiss mice on the high-fat diet and found that both strains of mice were fully responsive to the peripheral infusions of leptin, although the fat pads of some of the high-fat diet-fed mice were double the size of those in low-fat diet-fed mice. The role of body fat mass in determining leptin responsiveness also is challenged by a study that shows that rats become resistant to peripheral leptin within 5 days of being offered a high-fat diet, but that high-fat diet-fed rats responded to leptin within 1 day of being returned to the low-fat diet (23). These changes in sensitivity to peripheral leptin are too rapid to be determined by the size of body fat stores.

Mistry et al. (25) examined the energetic responses to centrally administered leptin in lean and ob/ob pups and found that lateral ventricle injection of leptin did not inhibit food intake until the pups were 28 days old, whereas energy expenditure was stimulated in 17-day-old pups. Therefore, although the centrally mediated stimulation of thermogenesis by leptin develops earlier than centrally mediated inhibition of food intake in mice, this cannot account for the development of central, but not peripheral, leptin resistance in mice in our experiments because they were 5 to 7 wk old at the time that leptin was administered. Older mice fed a high-fat diet for 8 wk become resistant to peripheral injections of leptin but remain responsive to centrally administered leptin. With continued exposure to the high-fat diet, the response to centrally injected leptin is attenuated after 16 wk on the diet (24). In this experiment, we found that consumption of a high-fat diet from weaning attenuated the response to a lateral ventricle and a third ventricle injection of leptin in 6- to 7-wk-old mice. These results indicate that in mice that have never consumed a low-fat diet, sensitivity to centrally administered leptin is compromised more quickly than when older animals are switched from a low-fat (chow) to a high-fat diet.

A previous study in lean C57BL/6J mice reported changes in food intake of neonatal mice as early as 30 min after central injections of leptin (25). In contrast, we did not observe significant changes in food intake or body weight until 38 h after lateral or third ventricle injections. The mice in the previous study were only food deprived for 4 h before the leptin injections, whereas our mice were food deprived for 9 h. The longer period of food deprivation and increased hunger of the mice may have resulted in immediate eating by both leptin-treated and control groups of mice. The observed delay in the response to leptin may also be explained by observations that lateral ventricle injections of leptin do not immediately block feeding but decrease the size of meals and the rate of feeding in rats (21). A study in which male Wistar rats received third ventricle injections of leptin showed that meal size was only reduced with the second meal after injection (8), which may explain why we did not find any differences in food intake 4 h after injection in experiment 1.

Few studies have examined the effects of peripheral leptin infusions in mice fed high-fat diet. One long-term study investigated whether leptin infusions could prevent the development of obesity and diabetes in mice fed a high-fat diet (35). Although subcutaneous infusion of leptin (0.4 mg · kg¹ · day¹, 8-10 µg · mouse¹ · day¹) reduced body weight and food intake during the first 5 wk of infusion, there were no differences in body weights, food intakes, or fat pad weights of leptin-infused and control mice at the end of the 12-wk study. The mice in this study (35) were group housed and the results from experiments described here suggest that the housing conditions and the gender of the mice may influence whether leptin infusions change body fat mass.

The differences between single-housed and group-housed mice may be attributable to differences in thermogenic capacity, because group-housed animals huddle and reduce the requirement for heat production from each animal. The degree of huddling varies according to ambient temperature and gender, with females more likely to huddle than males (4). Because of the reduced requirement for heat production, brown adipose tissue thermogenesis (20, 22) and food intake (28) are inversely related to number of animals per cage. One of the mechanisms by which leptin induces weight loss is by stimulating heat production (25, 27), and leptin has been shown to increase brown adipose tissue mRNA expression of uncoupling proteins (UCP2 and 3) (31). These observations suggest that loss of body fat in single-housed, leptin-treated mice may be, at least partially, dependent on leptin-induced thermogenesis, which is consistent with our observations of decreased efficiency of energy use in experiment 2. In group-housed animals, the need for thermogenesis is reduced and the effect of leptin on body composition is limited. This is consistent with observations by Stehling et al. (32) that the reduced body fat mass of juvenile lean Zucker rats injected subcutaneously with leptin was entirely due to an increase in energy expenditure, but that leptin did not stimulate energy expenditure when the rats were reared in thermoneutral conditions (33). Because group housing in our experiment inhibited leptin activity in high-fat diet-fed, but not low-fat diet-fed, mice, diet composition must directly influence the mechanisms by which leptin induces loss of body fat in mice.

In the experiments described here peripheral infusions of leptin reduced the body fat mass of single-housed, high-fat diet-fed mice that had a reduced sensitivity to central injections of leptin. These results suggest either that the central receptors in high-fat diet-fed, but not low-fat diet-fed, mice need to be chronically stimulated by leptin for there to be an effect on food intake or that the body fat-reducing effects of peripheral leptin are not mediated by the same receptors that are responsible for the central inhibition of food intake by leptin and that some of the response may result from direct action of leptin on peripheral tissues. In vitro (11, 30) and in vivo (10) studies have shown that leptin directly stimulates lipolysis in adipocytes and it has been reported that denervated fat depots are reduced in hyperleptinemic rats (38), implying that the metabolic changes responsible for loss of fat in leptin-treated animals are independent of the activation of central leptin receptors. The mechanisms by which fat is specifically decreased by peripheral infusions of leptin in high-fat diet-fed weanling mice need to be clarified, but our results suggest that leptin may act directly in the periphery and that part of the response is due to an increase in thermogenesis.

As shown in the results, peripheral infusions of leptin produced variable increases in serum leptin concentrations ranging from no change in concentration (low-fat diet-fed group-housed C57BL/6J females) to a fourfold increase (single-housed C57BL/6J females). Although all mice received the same dose of leptin and it was not adjusted for body weight, these differences are unlikely to be due to the size of the animals because the group-housed females were smaller than most of the other animals in this study. In addition, the measured serum leptin concentrations did not necessarily correlate with leptin response because body fat content was significantly lower in low-fat diet-fed, leptin-treated, group-housed female C57BL/6J mice compared with their controls. One possible explanation for this discrepancy is that the RIA kit we used to measure leptin has been reported to measure leptin binding protein in addition to leptin (26); thus no distinction is made between free (bioactive) leptin, bound leptin, and leptin binding protein. The effect of leptin infusions on serum fasting glucose and insulin concentrations also varied between treatment groups, but the reason for the inconsistent changes in this study are unknown and were not investigated.

In summary, the experiments described here show that mice weaned onto a high-fat diet develop an insensitivity toward peripheral and central injections of leptin at 5-7 wk of age; however, these mice respond to peripheral infusions of leptin by specifically reducing body fat mass. The results from the group-housing experiments suggest that leptin partially exerts its effects on body fat in high-fat diet-fed mice through increased thermogenesis and, in situations where thermogenic capacity is reduced, leptin has limited effects on body weight regulation. Because human obesity is associated with "leptin resistance," characterized by maintenance of an enlarged body fat mass in the presence of increased concentrations of endogenous leptin (5), it is important that we find an appropriate animal model to study this condition. Mice fed a high-fat diet have been reported to be resistant to peripheral injections of leptin (24, 37), but the studies described here demonstrate that, in weanling mice fed a high-fat diet, the response to peripherally administered leptin is determined by gender, strain, housing conditions, and method of leptin administration. Therefore, we need to develop a better understanding of the factors that influence the development of leptin resistance before we can evaluate its role in facilitating the development and maintenance of an obese state.