LEPTIN, A HORMONE secreted from adipose tissue (1),

Transcription

1 /01/$03.00/0 Vol. 142, No. 1 Endocrinology Printed in U.S.A. Copyright 2001 by The Endocrine Society Transgenic Mice Overexpressing Leptin Accumulate Adipose Mass at an Older, But Not Younger, Age* J. QIU, S. OGUS, R. LU, AND F. F. CHEHAB Department of Laboratory Medicine, University of California, San Francisco, California ABSTRACT Sensitivity to leptin is associated with a normal regulation of the adipose mass, whereas decreased leptin sensitivity results in elevated adipose tissue stores. To address whether the effects of chronic hyperleptinemia are sustained with age, we generated transgenic mice that overexpress leptin under the control of the fat specific ap2 promoter/enhancer. At 6 9 weeks of age, transgenic mice overexpressed 5-fold more human leptin than endogenous mouse levels and had consistently low body weights, with reduced brown and white fat depots characterized by adipocytes either devoid of or containing minute lipid droplets. However, at weeks, despite continuous secretion of human leptin, the transgenic mice showed a rebound effect characterized by an increase in body weight, accumulation of adipose mass, and lipid-filled adipocytes. Thus, this mouse model exhibits a two-stage phenotype, with respect to fat accumulation. In LEPTIN, A HORMONE secreted from adipose tissue (1), is the subject of intense investigations. When injected into leptin-deficient obese (ob/ob) mice, it corrects all their metabolic and physiological defects (2 6). Leptin enters the brain via a saturable transport system (7) and binds to its receptor (8) in the arcuate region of the hypothalamus, where it activates a specific signal transducer and activator of transcription (STAT-3) (9) as part of the Janus kinase signaling pathway. Animal models deficient in the synthesis of a longform of the leptin receptor, such as db/db mice and fa/fa rats (10), have a severe leptin resistance that results in an obese phenotype similar to ob/ob mice. Leptin has been implicated in various biological pathways and plays distinct roles in energy expenditure and fasting (11). Mutations in the human leptin gene are rare; however, a morbidly obese individual who was found to be homozygous for a frameshift mutation in the leptin gene (12) was successfully treated with recombinant human leptin (13). The phenotypes that result from the absence of leptin or its signaling-competent receptor prove that this pathway is not redundant and is therefore critical to the organism. Whereas common obesity in humans does not result from mutations in the leptin gene or its receptor, most obese individuals have considerably elevated circulating leptin levels (termed hyperleptinemia) as a result of increased fat mass (14). It has been hypothesized that such Received July 17, Address all correspondence and requests for reprints to: F. Chehab, Department of Laboratory Medicine, 505 Parnassus Avenue, University ofcalifornia,sanfrancisco,california chehabf@labmed2. ucsf.edu. * This work was funded by NIH Grant HD Supported by a postdoctoral fellowship from NIH Training Grant T32-DK addition, plasma glucose, triglycerides, and cholesterol levels were markedly depressed in young, but not older, transgenic mice. A detrimental consequence of early hyperleptinemia was a failure of the transgenic mice to acclimatize to the cold, as a result of depleted fat stores within their brown adipocytes. Cold exposure was tolerated after a 2-week high-fat diet or at an older age when fat depots had naturally accumulated. Treatment of the older transgenic mice with large doses of leptin stimulated weight loss, demonstrating that the leptin pathway still responds to pharmacological levels of leptin. Overall, these studies show that moderate hyperleptinemia in normal mice results in a sensitivity of the adipose mass to leptin at a younger (but not older) age. The mechanisms that lead to the accumulation of fat at an older age remain largely unknown, and this hyperleptinemic mouse model will allow the uncovering of at least some of these mechanisms. (Endocrinology 142: , 2001) individuals are in a state of leptin resistance, because they fail to respond to their endogenous supraphysiological levels of leptin. Presumably, increasing the brain permeability to leptin in obese subjects would stimulate the leptin pathway via the sympathetic nervous system (15) and result in adipose tissue lipolysis. In this study, we investigated the chronic lifelong effects of hyperleptinemia in the presence of a functional and intact leptin pathway, using a transgenic animal model that expresses, in a tissue-specific fashion, a moderately elevated human leptin level. Materials and Methods Generation of transgenic mice The complementary DNA (cdna) for human leptin spanning the initiation to termination codons was amplified by one round of RT-PCR from human adipose tissue messenger RNA (mrna) with primers containing BamHI sites at their 3 ends. The amplified product was cleaved with BamHI and subcloned into the prokaryotic expression vector pqe30 (QIAGEN, Chatsworth, CA) to yield the plasmid hxp1.2. Human leptin cdna was amplified from hxp1.2 with primers containing SmaI sites, and the amplified product was inserted into the SmaI site of the expression vector pba [obtained from Dr. Richard Weiner, University of California, San Francisco (UCSF)]. The leptin cdna was thus placed downstream of a rabbit -globin intron and upstream of the human GH (hgh) polyadenylation (polya) signal to generate pba-hxp1.2. The -globin intron, leptin cdna, and hgh polya sequences were then recovered as a single SacI fragment and inserted into the SacI site of pbluescript SK. Finally, the construct was recovered from Bluescript as a SpeI fragment and ligated to the SpeI site downstream of the ap2 promoter/enhancer Bluescript-based plasmid obtained from Dr. Bruce Spiegelman (Harvard Medical School, Boston, MA). The integrity of the leptin cdna sequence in the final construct was verified by DNA sequencing. The final construct containing the ap2 promoter/enhancer, rabbit -globin, leptin cdna, and hgh polya signal (Fig. 1A) was recovered as a KpnI/NotI fragment and microinjected into C57BL6J/DBA2J embryos according to standard protocols. The presence of the transgene in founder and F1 mice was detected by SpeI digestion of tail genomic DNA followed by Southern blotting and hybridization with a 32 P-labeled human 348

2 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS 349 FIG. 1. A, Schematic diagram of the transgenic construct spanning the mouse ap2 promoter/enhancer, a rabbit -globin intron (R G), the cdna for human leptin (hlep), and the hgh polya signal. B, Autoradiograph of a genotyping Southern blot from mouse tail genomic DNA cleaved with SpeI and hybridized with a 32 P-labeled human leptin cdna probe. Transgenic mice show a three-banding pattern that arises from the integration of the construct into host genomic DNA. The additional band common to all lanes is attributable to cross-hybridization, most likely from the endogenous mouse leptin gene, and serves as an internal control band. C and D, Expression of transgenic human leptin mrna in adipose tissue (C) and organs (D). The autoradiographs show prominent expression of the 700 nucleotides (nt) transgenic human leptin mrna in WAT and BAT of transgenic (T) but not nontransgenic (N) littermates. Weak-to-moderate expression was detected in the thymus, spleen, and heart. Below each autoradiograph is the ethidium bromide stain of total RNA showing the 28S and 18S ribosomal RNAs (rrna). SubQ, subcutaneous; Kidn., kidney; Stom., stomach. leptin cdna probe. Final posthybridization washes were at 65 C in 0.1 SSC, 0.1% SDS. Animals were housed in colony cages, maintained on 12-h light, 12-h dark cycle and fed a chow (Formulab Diet 5008, Purina-Mills, St. Louis, MO) containing 6.5% fat. When placed on a high-fat diet, mice were fed formula TD (Harlan-Teklad, Madison, WI) containing 42% of calories from fat. Body weights were regularly determined, to the nearest 0.1 g, on a precision balance. Food intake was monitored daily for a 7-day period in males (n 3 for each group) and females (n 4 for each group) from two age groups (6 9 weeks and weeks). All animal procedures were in agreement with institutional guidelines and approved by the UCSF Animal Care Committee. Plasma chemistry, fat pad weights, and histology Ad lib-fed mice were anesthetized with 0.2 ml of 2.5% Avertin and bled from the retroorbital sinus into EDTA-coated tubes. The plasma was immediately separated and frozen at 20 C until use. Glucose, triglycerides, and cholesterol assays were performed on an LX-1 automated clinical chemistry analyzer (Beckman Coulter, Inc., Fullerton, CA) in the UCSF Clinical Laboratories. Human leptin, mouse leptin, and insulin levels were quantitated separately and in duplicate with RIA kits available from Linco Research, Inc. (St. Charles, MO). Because the human and mouse leptin RIAs have less than 0.2% cross-reactivity with each other, we could quantitate murine and human leptin from the plasma of transgenic mice. Fat depots were dissected, and their wet weights were determined on a Metler (Highstown, NJ) analytical balance, to the nearest 0.01 g. For histology, tissues were fixed in 10% phosphate-buffered formalin, then processed for embedding, sectioning, and staining with hematoxylin and eosin, according to standard methods. Northern blot analysis Mouse tissues and organs were harvested immediately after dissection and were either processed for total RNA extraction by the Trizol method (BRL, Life Technologies, Inc., Rockville, MD) or stored at 85 C for later RNA extraction. Thirty micrograms of RNA per lane were fractionated

3 350 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS Endo 2001 Vol. 142 No. 1 FIG. 2. Plasma levels of endogenous mouse and transgenic human leptin in 6- to 9- and 33- to 36-week transgenic (black bars) mice and nontransgenic (white bars) littermates. n 5 for each group except for the human leptin assay in young transgenic males (n 11) and transgenic females (n 14). on denaturing formaldehyde gels according to standard procedures. DNA probes for ap2 and mouse leptin were synthesized by RT-PCR from white adipose tissue (WAT), whereas uncoupling protein 1 (UCP1) cdna was amplified from brown adipose tissue (BAT). The mouse specific PCR primers based on GenBank sequences were: lep-f 5 TAG GGA TGG GTA GAG CCT TTG G 3, lep-r 5 TAA CAC ATC CTC TAC CCT CAG GTG C 3 ap2-f: 5 ATG TGT GAT GCC TTT GTG GGA 3 ; ap2-r; 5 TGC CCT TTC ATA AAC TCT TGT 3 ; UCP1-F; 5 AGT ACC ATT AGG TAT AAA GGT GTC 3 ; UCP1-R 5 AGT GTG GTG CAA AAC CCG GCA ACA 3. PCR products were purified by agarose gel electrophoresis and labeled with - 32 P-deoxycytidine triphosphate, by random priming. Northern blots were hybridized overnight at 65 C and washed extensively at 65 C in 0.1 SSC and 0.1% SDS. Northern blots hybridized with the transgenic leptin probe were boiled for 5 min. in sterile distilled water before rehybridization with the ap2 cdna probe. Cold challenge The core temperature of transgenic mice and their sex- and agedmatched nontransgenic littermates was determined at room temperature, before housing each mouse in an individual cage at 4 C. Hourly rectal temperature recordings were determined with a precision thermometer (YSI, Inc., Yellow Springs, OH) equipped with a 0.2-mm probe (YSI, Inc. model 451). Leptin treatment Human recombinant leptin was prepared as previously described (6) and injected ip daily, at a dose of 30 g/g BW for 13 days, into 33- to 36-week-old nontransgenic and transgenic mice. As a control for the leptin preparation, the same amount of leptin protein was injected into ob/ob mice. Control mice from the three groups were treated with a vehicle saline solution. Body weights were periodically determined, on a precision balance, to the nearest 0.1 g. Statistics Significance levels were based on unpaired Student s t test and derived with the Statistica software package for the Macintosh microcomputer. All data are expressed in means and sem. Results Detection of transgenic founders and F1 progenies Three female founders were initially identified, by Southern blot analysis, to carry the transgene. However, when analyzed with a human leptin RIA, only one founder (F8) expressed human leptin in the plasma at 15 ng/ml. Although all three founders transmitted the transgene to their F1 progenies, only mice that resulted from F8 consistently secreted the transgenic protein in their bloodstream. Litters from the other two founders were repeatedly negative with the human leptin RIA and were thus not further investigated. All the data described in this study represent progenies of the single founder mouse F8. Pups were genotyped, by Southern blot analysis of SpeI digested tail genomic DNA, to yield the

4 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS 351 FIG. 3. Body weights (A D) and food intake (E and F) of transgenic mice and nontransgenic littermates. A and B, Body weight distributions, at 6 weeks of age, of nontransgenic male (n 80) and female (n 83) mice (solid line) and of transgenic male (n 50) and female (n 67) mice (dotted line). C and D, Growth curves of transgenic (F) and nontransgenic (E) males (n 16 in each group) and females (n 13 in each group) from 3 32 weeks of age. E and F, Cumulative food intake during 7 days of young and older nontransgenic (3 males and 3 females; open bars) and transgenic (3 males and 3 females; black bars) mice. P 0.005, at 6 9 weeks, for the two groups of males and females. All graphs on the right and left side of the figure are representative for males and females, respectively. DNA pattern shown in Fig. 1B. Each transgenic mouse was also confirmed, by RIA, for human leptin secretion. Expression of the transgene The tissue specificity of the transgene was determined by analyzing human leptin mrna expression in various tissues of transgenic animals and nontransgenic littermates. Thus, hybridization of Northern blots with a human leptin cdna probe revealed overwhelming expression of a 700-nucleotide mrna species in gonadal, brown, and sc fat depots of only transgenic mice (Fig. 1C). The absence of pericardial and perirenal fat pads in transgenic mice precluded recovery of tissue for RNA extraction and analysis. No expression was detected in other organs except for low levels in heart and even lower expression in spleen and thymus (Fig. 1D). Except for the heart, which contains fat, it is likely that signals from

5 352 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS Endo 2001 Vol. 142 No. 1 FIG. 4. Interscapular brown (left panels) and gonadal WAT (right) of 8(top) and 36 weeks (bottom) N and T mice. Note the absence of BAT and reduced- WAT masses in young, but not older, transgenic mice. the spleen and thymus are attributable to low-grade fat contamination. Nevertheless, predominant expression of the transgene in BAT and WAT demonstrates that secretion of human leptin in the transgenic mice originates from the adipose tissue. Plasma leptin levels To determine whether expression of the transgene resulted in secretion of leptin, we assayed the plasma of 6- to 9-weekold and 33- to 36-week-old transgenic mice and nontransgenic littermates for mouse and human leptin expression. We found that mouse plasma leptin levels were not significantly different from each other in transgenic and nontransgenic males and females at both ages (Fig. 2). Although human leptin was consistently undetectable in the plasma of nontransgenic mice, human leptin levels in transgenic males at 6 9 weeks and weeks of age were ng/ml and ng/ml, respectively, accounting for 4- and 4.9-fold elevation over endogenous levels (Fig. 2, A and B). In transgenic females, human leptin levels were ng/ml at 6 9 weeks of age and increased to ng/ml at weeks (P ), accounting, respectively, for 2.5- and 4.2-fold increases over endogenous levels (Fig. 2, C and D). Body weight and food intake The effect of early chronic hyperleptinemia on body weight could be best assessed at 6 weeks of age, when body weight distributions clearly showed a leftward shift of the bell-shaped curves of the transgenic vs. nontransgenic littermate groups (Fig. 3, A and B). Body weight monitoring until 32 weeks of age (Fig. 3, C and D) revealed that the growth curves of transgenic males and females diverged from their nontransgenic littermates as early as 3 weeks to weeks of age, with P values between groups of the same sex consisting of (Student s t test). Interestingly, the body weights of transgenic males and females, after 18 and 22 weeks of age respectively, did not differ significantly from their nontransgenic littermates. Determination of food consumption over 7 days revealed that, at a young age, both transgenic males and females consumed, respectively, 11% and 15% less food than their nontransgenic littermates (P 0.005), but that at weeks of age, this difference was not significant anymore (Fig. 3, E and F). Adiposity At 2 3 weeks of age, transgenic males and females showed a prominent depression in their interscapular region, which, upon dissection at 6 9 weeks of age, was found to result from a reduction in BAT and an absence of any surrounding WAT (Fig. 4). The interscapular depression was, however, not noticeable at weeks of age, because BAT and WAT had regenerated in both males and females transgenics. Similarly, gonadal WAT was either absent or greatly reduced in mass between nontransgenic and transgenic mice (Fig. 4) at the younger (but not older) age.

6 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS 353 FIG. 5. Fat pads, organs, and carcass weights of nontransgenic (white bars) and transgenic (black bars) mice at 6 9 weeks (n 4 in each group) and weeks of age (n 4 in each group). Left and right side panels represent males and females, respectively. Fat pads weighed are BAT, epididymal WAT (EPI-WAT), uterine WAT (UTER-WAT), mesenteric WAT (MES-WAT), and sc WAT (SUBC-WAT). The effect of hyperleptinemia on the accumulation of adipose mass in young and older transgenic mice was determined by weighing the fat pads in both age groups (Fig. 5). At 6 9 weeks, fat pads (such as pericardial, perirenal, mesenteric, and sc fat depots) were either absent or too minute to be accurately weighed. However, BAT and gonadal WAT, although small in size, could be recovered. We found that BAT was reduced 4.2-fold (P 0.002) and 2.6-fold (P ) in young male and female transgenics, respectively, over their nontransgenic littermates. In addition, gonadal WAT in transgenic mice weighed 13.8-fold (P 0.005) and 8.7-fold (P 0.006) less than nontransgenic male and female littermates, respectively. In contrast, at weeks of age, weights of the fat pads were not significantly different between transgenic and nontransgenic groups. Furthermore, except for the liver, which was 1.4-fold smaller in young transgenic males (P ) and females (P 0.03), other organ weights of transgenic and nontransgenic mice were not significantly different from each other at both ages. On the other hand, differences in the carcass weights between young nontransgenic and transgenic males (P 0.04) and females (P 0.06) accounted for most of the differences in body weights. Adipose tissue histology and differentiation Figure 6 shows the cellular changes that occur in the adipose tissue of transgenic mice as a result of short and long-term hyperleptinemia. At 6 9 weeks of age, the brown adipocytes of transgenic mice were devoid of lipid droplets. However, at 36 weeks of age, the brown adipocytes had regained their normal appearance and stored multilocular lipid droplets. Similarly, but not to the same extent, the white adipocytes of transgenic mice, at 6 9 weeks, either lacked or contained minute lipid droplets, an effect that was also abolished at weeks of age. To determine whether delipidated brown and white adipocytes of young transgenic mice had actually differentiated from preadipocytes, we asked whether expression of ap2 mrna, which is only expressed in mature adipocytes, can be detected in their adipose tissue. Indeed, Northern blot analysis revealed that the fat depots of the transgenics did

7 354 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS Endo 2001 Vol. 142 No. 1 FIG. 6. BAT and WAT staining of young (8 weeks) and older (36 weeks) T mice and N littermate. Note that young transgenic mice show delipidated BAT at 8 weeks, but not 36 weeks, of age and have small white adipocytes either devoid or containing minute lipid droplets (H&E, 400 ) FIG. 7. Expression of ap2 mrna in BAT and WAT of transgenic mice. Northern blots shown in Fig. 1 were stripped by boiling and rehybridized with a mouse ap2-specific cdna probe revealing abundant ap2 mrna expression in all mice. The ethidium bromide staining of the corresponding gels show 28S and 18S ribosomal mrna and demonstrates RNA loading. express ap2 mrna abundantly (Fig. 7), concluding that both BAT and WAT had terminally differentiated, but failed to accumulate, lipid stores. Plasma assays The impact of a hyperleptinemia was also assessed on plasma levels of glucose, triglycerides, cholesterol, and insulin from both age groups. At 6 9 weeks of age, the most dramatic difference between the transgenic and nontransgenic littermate groups was the level of insulin, which was decreased 4-fold (P ) and 3-fold (P 0.001), respectively, in transgenic males and females (Fig. 8, A and B). In addition, triglyceride levels were, respectively, reduced 2- and 3-fold lower in transgenic males

8 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS 355 FIG. 8. Determination of plasma glucose (GLUC), triglycerides (TRIGL), cholesterol (CHOL), and insulin levels in ad lib-fed 6- to 9-week (n 5 in each group) and 33- to 36-week (n 5 in each group) transgenic mice (white bars) and nontransgenic littermates (black bars). (P ) and females (P 0.001) over their nontransgenic littermates. Furthermore, glucose and cholesterol levels were modestly, but significantly, depressed in transgenic mice of both sexes. However, at weeks of age, although insulin levels were still decreased by 2.7-fold in transgenic males (P 0.03), but not transgenic females, all other metabolic measurements were not significantly different between the transgenic and nontransgenic groups (Fig. 8, C and D). Acclimatization of transgenic mice to cold exposure To determine whether the lack of lipids in the BAT of young transgenic mice bears any consequence on cold adaptation, the mice were placed at 4 C for 12 h. In contrast to 6- to 9-week-old nontransgenic littermates that regulated their temperature and survived cold exposure, the temperature of aged matched transgenic males and females dropped, respectively, to C (range, C) and 20.4 C 0.9 C (range, C) within hours of cold exposure (Fig. 9A). However, the timing for this drop was variable and ranged from 6 9 h for transgenic males and 2 11 h for transgenic females. On the other hand, at weeks of age, all transgenic mice invariably tolerated the cold and appropriately regulated their body temperature. The increased adiposity of older transgenic mice (as shown in Figs. 4 and 5) suggested that elevated fat stores might have contributed to their cold tolerance. This is also evidenced by differences in body weight between the cold-sensitive young and cold-tolerant older transgenic mice (Fig. 9A; P for males and P for females). To further explore this possibility, 6- to 9-week-old transgenic mice were first fed a 2-week high-fat diet and then exposed to a 12-h cold challenge. Indeed, all transgenic mice fed the high-fat diet tolerated the cold and did not exhibit significant reductions in core temperature (Fig. 9B). To unravel further mechanisms for the cold susceptibility of the young transgenic mice, we asked whether they failed to induce UCP1 mrna expression upon cold exposure. We thus analyzed UCP1 mrna expression in the brown fat of 6- to 9-week transgenic mice before and during the cold challenge. Figure 9C shows that steady-state levels of UCP1 mrna at room temperature were not markedly different

9 356 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS Endo 2001 Vol. 142 No. 1 FIG. 9. Cold exposure of young and older transgenic mice. A, Rectal temperatures of 6- to 9-week and 33- to 36-week-old nontransgenic (white bars) and transgenic (black bars) mice at room temperature (22 C) and during a 4 Ccold challenge. The temperature range during cold exposure of 6- to 9-week-old mice varied from C in normal mice and C in transgenic mice. Body weights of both groups, at the time of each experiment, are shown alongside of the plots (N and T). B, Rectal temperatures of 6- to 9-week-old nontransgenic (n 5 in each group; white bars) and transgenic (n 5 in each group; black bars) mice, maintained at 25 C or 4 C and fed a diet containing either 6.5% or 42% fat. The bar graphs in (A) and (B) represent the mean SEM of the lowest temperature of transgenic mice and their nontransgenic littermates during the cold challenge. C, Profile of UCP1 mrna expression in nontransgenic (N) and transgenic (T) females before and after exposure at 4 C.

10 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS 357 FIG. 10. Differences in body weight after saline (open circles) or leptin treatment (triangles) of 33- to 36-week-old nontransgenic (A), transgenic (B), and ob/ob (C) mice. Each timepoint represents the mean SEM of three mice. between transgenic and nontransgenic littermates. In addition, up-regulation of UCP1 mrna during the cold challenge was similar in both groups, demonstrating that induction of UCP1 mrna expression was not defective in coldchallenged hyperleptinemic transgenic mice. Sensitivity to exogenous leptin The accumulation of adipose mass in older transgenic mice raises the question of whether they are, at all, sensitive to exogenously administered leptin. To address this question, we treated 34- to 36-week-old nontransgenic and transgenic mice with either leptin or vehicle solution for 13 days (Fig. 10). To ensure the effectiveness of the leptin preparation, ob/ob mice were simultaneously treated with the same leptin preparation as that of the transgenic mice. As expected, the body weights of ob/ob mice decreased significantly after the third day and throughout leptin treatment. Similarly, the body weights of the transgenic mice showed significantly decreased body weights, compared with transgenic mice receiving vehicle solution. However, decreases in body weight between leptin-treated transgenic and nontransgenic mice were not statistically different from each other, demonstrating similar leptin sensitivities in both groups. Discussion Numerous studies have shown that, in rodent animal models, leptin exerts pleiotropic effects on body weight, adiposity, glucose homeostasis, and reproduction (2 6). However, because these studies were limited to acute leptin treatment, they did not address the consequences of a lifelong chronic hyperleptinemic state. In this study, we determined the effects of such a state as induced by tissue-specific expression of a human leptin transgene. Early hyperleptinemia, assessed at 6 9 weeks of age, resulted in a reduction of body weight, fat mass, food intake, low plasma glucose, triglycerides, cholesterol, and insulin levels consistent with exogenous leptin treatment in normal and ob/ob mice (2 6); viral induced hyperleptinemia in normal rats (16); and liver-specific transgenic leptin expression in mice (17). In addition, we showed that the lipid-depleted brown and white adipocytes of young transgenic mice are associated with intolerance to cold. This effect does not result from a deficiency in UCP1 expression, but rather from the lack of -oxidation substrates; because accumulation of lipid stores, after a high-fat diet or with age, reverses their cold susceptibility. Most important, the early effects of hyperleptinemia on body weight and fat mass are not sustained in transgenic mice at weeks of age despite continuous expression of transgenic human leptin. The loss of a leptin effect in older transgenic mice is evidenced by the accumulation of adipose mass and by the appearance of lipid droplets within their brown and white adipocytes. The onset of this effect is likely to be gradual and starts to be manifested around 20 weeks of age, when the P values for the differences in body weight between the transgenic and nontransgenic groups start to increase. Unlike the early onset of severe leptin resistance of db/db mice, the leptin insensitivity of older transgenic mice is mild and does not result in obesity, even at 60 weeks of age

11 358 MICE OVEREXPRESSING LEPTIN ACCUMULATE ADIPOSE MASS Endo 2001 Vol. 142 No. 1 (Chehab et al. unpublished observations) but rather produces substantial adipose mass accumulation. The important question is whether the appearance of fat depots in older transgenic mice results from decreased responsiveness of the leptin pathway or from a different pathway that supersedes the effects of leptin. Although the elucidation of these mechanisms will have to await further investigations, the response of the older transgenic mice to exogenous leptin demonstrates that the leptin pathway is still intact, inferring that large doses of leptin are effective in treating moderate agerelated leptin insensitivity. This hyperleptinemic mouse model offers an opportunity to investigate and unravel molecular pathways that are differentially activated between the biphasic states of leptin sensitivity. Although central mechanisms dictate peripheral effects, the ability of fat to change from lipid-depleted to lipid-accumulating states must reflect molecular and cellular changes within the adipose tissue as well. Thus, the uncovering of differentially expressed genes in the hypothalamus and the adipose tissue will further our understanding of the temporal changes that are associated with an age-related accumulation of fat mass. Acknowledgments We thank Drs. Bruce Spiegelman and Richard Wiener for their gifts of the ap2 promoter/enhancer and the pba cloning vector, respectively. References 1. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM 1994 Positional cloning of the mouse obese gene and its human homologue. Nature 372: Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters D, Boone T, Collins F 1995 Effects of the obese gene product on body weight regulation in ob/ob mice. Science 269: Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, Burley SK, Friedman JM 1995 Weight-reducing effects of the plasma protein encoded by the obese gene. Science 269: Campfield LA, Smith FJ, Guisez Y, Devos R, Burn P 1995 Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 269: Weigle DS, Bukowski TR, Foster DC, Holderman S, Kramer JM, Lasser G, Lofton-Day CE, Prunkard DE, Raymond C, Kuijper JL 1995 Recombinant ob protein reduces feeding and body weight in the ob/ob mouse. J Clin Invest 96: Chehab FF, Lim ME, Lu R 1996 Correction of the sterility defect in homozygous obese female mice by treatment with the human recombinant leptin. Nat Genet 12: Banks WA, Kastin AJ, Huang W, Jaspan JB, Maness LM 1996,Leptin enters the brain by a saturable system independent of insulin. Peptides 17: Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards GJ, Campfield LA, Clark FT, Deeds J, Muir C, Sanker S, Moriarty A, Moore KJ, Smutko JS, Mays GG, Woolf EA, Monroe CA, Tepper RI 1995 Identification and expression cloning of a leptin receptor, OB-R. Cell 83: Vaisse C, Halaas JL, Horvath CM, Darnell Jr JE, Stoffel M, Friedman JM 1996 Leptin activation of Stat3 in the hypothalamus of wild-type and ob/ob mice but not db/db mice. Nat Genet 14: Chua Jr SC, Chung WK, Wu-Peng XS, Zhang Y, Liu SM, Tartaglia L, Leibel RL 1996 Phenotypes of mouse diabetes and rat fatty due to mutations in the OB (leptin) receptor. Science 271: Ahima RS, Kelly J, Elmquist JK, Flier JS 1999 Distinct physiologic and neuronal responses to decreased leptin and mild hyperleptinemia. Endocrinology 140: Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H, Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst JA, Cheetham CH, Earley AR, Barnett AH, Prins JB, O Rahilly S 1997 Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 387: Farooqi IS, Jebb SA, Langmack G, Lawrence E, Cheetham CH, Prentice AM, Hughes IA, McCamish MA, O Rahilly S 1999 Effects of recombinant leptin therapy in a child with congenital leptin deficiency. N Engl J Med 341: Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R, Ranganathan S, Kern PA, Friedman JM 1995 Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med 1: Haynes WG, Morgan DA, Walsh SA, Mark AL, Sivitz WI 1997 Receptormediated regional sympathetic nerve activation by leptin. J Clin Invest 100: Chen G, Koyama K, Yuan X, Lee Y, Zhou YT, O Doherty R, Newgard CB, Unger RH 1996 Disappearance of body fat in normal rats induced by adenovirus-mediated leptin gene therapy. Proc Natl Acad Sci USA 93: Ogawa Y, Masuzaki H, Hosoda K, Aizawa-Abe M, Suga J, Suda M, Ebihara K, Iwai H, Matsuoka N, Satoh N, Odaka H, Kasuga H, Fujisawa Y, Inoue G, Nishimura H, Yoshimasa Y, Nakao K 1999 Increased glucose metabolism and insulin sensitivity in transgenic skinny mice overexpressing leptin. Diabetes 48: