A peptide leptin antagonist reduces food intake in rodents - Nature

International Journal of Obesity (1999) 23, 463±469 ß 1999 Stockton Press All rights reserved 0307±0565/99 $12.00 http://www.stockton-press.co.uk/ijo

A peptide leptin antagonist reduces food intake in rodents L Brunner1, S Whitebread1, I Leconte1, A Stricker-Krongrad1, Frederic Cumin1*, M Chiesi1 and N Levens1 1

Metabolic and Cardiovascular Diseases, Novartis Pharma AG, Basel, CH 4002, Switzerland

OBJECTIVE: The purpose of the present study was to investigate the continuing validity of the hypothesis that leptin is a physiologically important regulator of food intake, using the human leptin mutant R128Q leptin. DESIGN: In a cellular proliferation assay, based on BAF-3 cells transfected with the murine ObRb receptor, R128Q leptin was shown to be devoid of agonistic activity and to competitively inhibit the proliferative effects of leptin. To determine whether R128Q leptin was also an antagonist of leptin in vivo, the leptin mutant was injected intracerebroventricularly (i.c.v.) into rats in the absence and presence of leptin. R128Q was also injected intraperitoneally (i.p.) into ob==ob and into db==db mice expressing, respectively, either normal or defective ObRb receptors. RESULTS: R128Q was shown to be a competitive antagonist of leptin induced cellular proliferation in vitro. Surprisingly, in vivo R128Q leptin produced a strong dose-dependent decrease in food intake, and was only slightly less potent than leptin itself. In fasted rats, the inhibitory effects of leptin and R128Q leptin (i.c.v.) on post-fast refeeding were additive. Finally, R128Q leptin produced the same inhibition of food intake as leptin when injected i.p. in ob==ob mice and, like leptin, was inactive after i.p. injection to db==db mice. CONCLUSION: R128Q leptin is a leptin agonist in vivo, but behaves as an antagonist against leptin induced proliferation in vitro. The data demonstrate that the human leptin mutant R128Q leptin is not a suitable tool for investigating the physiological actions of leptin. Keywords: leptin physiology; obesity; feeding behaviour; agonist; BAF-3 cells

Introduction Leptin, the cytokine product of the ob gene, is released into the circulation mainly by adipocytes, but is also produced by other tissues, such as the placenta and stomach.1 ± 7 Injection of leptin systemically or directly into the ventricular system of the brain reduces food intake and body weight.8 ± 15 The hypothalamus appears to be the major target tissue for the hormone, since functional leptin receptors are present within the appetite control centers of this region.13,16 ± 19 Although leptin can induce a decrease in food intake and body weight when injected either centrally or peripherally, the role of the endogenous peptide in the control of food intake and body weight remains unclear. In the only study to have directly addressed this question,22 polyclonal antibodies directed against leptin were injected intracerebroventricularly (i.c.v.) into free-feeding rats, at the beginning of the light phase, when these animals normally do not eat. In these studies, rats treated with the anti-leptin antibodies increased their food intake for more than 24 h. In contrast, obese Zucker fa=fa rats, expressing a defective ObRb receptor,20,21 were unresponsive to *Correspondence: Frederic Cumin, Novartis Pharma AG, K125.10.57, CH-4002 Basel, Switzerland. E-mail: [email protected] Received 30 July 1998; revised 9 November 1998; accepted 2 December 1998

central antibody administration. These observations suggest, that in fed rats, endogenous leptin plays a physiologically important role to suppress food intake.22 However, additional studies using other means of blocking the actions of leptin are needed before this conclusion can de®nitely be accepted. Recently, a human leptin mutant has been reported with an arginine to glutamine substitution at position 128 of the molecule.23 This leptin mutant binds to the functional ObRb receptor without intrinsic activity and is believed to be an antagonist of leptin induced proliferation in cultured BAF-3 cells. With these properties, R128Q leptin would be a useful tool for blocking the actions of endogenous leptin. The purpose of the present study was to investigate the continuing validity of the hypothesis that leptin is a physiologically important regulator of food intake, using the human leptin mutant R128Q leptin.

Methods Animals

Male Sprague Dawley rats, weighing 200 ± 260 g, were obtained from Novartis AG (Tif:RA25; Novartis AG, Sisseln, Switzerland). Female ob=ob mice, weighing 55 ± 70 g, were purchased from Jackson laboratories (C57BL=6J; Bar Harbor, ME). Female db=db mice, weighing 30 ± 40 g, were bred in our

A peptide leptin antagonist reduces food intake L Brunner et al

464

animal facilities (C57BL=6J, Novartis). All animals were individually housed in macrolon cages under temperature (22 2 C) and light-controlled conditions (12 h:12 h light=dark cycle with lights on at 06.00 h). The animals were given tap water to drink and had access to a standard rat diet ad libitum (Nafag AG, Gossau, Switzerland). Binding of leptin and R128Q leptin to the ObRb receptor

Preparation of membranes. COS-7 cells transiently expressing the human ObRb receptor were washed in ice-cold phosphate buffered saline (PBS) and harvested using a plastic scraper. The cells were centrifuged at 200 g for 10 min and the pellet resuspended in ice-cold Tris ± HCl (50 mM), EDTA (0.5 mM), pH 7.4. The suspension was sonicated for 25 s and centrifuged at 200 g for 10 min at 4 C. The supernatant was then centrifuged at 55 000 g for 18 min at 4 C and the resulting pellet resuspended in the same buffer. Protein concentration was measured by the method of Bradford, with bovine serum albumin (BSA) as standard.24 The crude membrane preparation was aliquoted, ¯ash-frozen in liquid nitrogen and stored at 7 80 C. Shortly before use, 1% (w=v) BSA and 0.2% Tween 20 was added. Assay conditions.125I-human leptin (Anawa, Wangen, Switzerland, spec. act. approx. 1800Ci=mmol), unlabelled human leptin and R128Q leptin were dissolved in distilled water and further diluted in binding buffer: Tris ± HCl (50 mM), pH 7.4, containing NaCl (100 mM), EDTA (1 mM), BSA (1%) and Tween 20 (0.2%). Incubations were performed in Multiscreen FB ®lter plates (Millipore, Bedford, MA). The ®lters in each well were rinsed once with 300 ml PBS before use. The following were pipetted into each well: 60 ml binding buffer, 20 ml 125I-human leptin (750 pM), 20 ml test compound (or binding buffer for the controls), 100 ml crude membrane suspension (approx. 10 mg protein). Incubations were performed at room temperature for 3 h. Non-speci®c binding was de®ned as the binding remaining in the presence of 1 mM human leptin. The incubations were terminated by rapid ®ltration and washing four times with 200 ml PBS containing 0.2% Tween 20. The ®lters were placed into plastic tubes and counted in a g-counter. Tritiated thymidine uptake assay in BAF-3 cells

Cell culture. Interleukin-3 (IL-3) -dependent murine lymphocyte BAF-3 cells (kindly provided by R. Skoda25) stably expressing the mouse ObRb receptor, were cultured in RPMI1640 medium (GibcoBRL, Paisley, UK) supplemented with 1% L-glutamine (GibcoBRL), 1% penicillin-streptomycin (GibcoBRL,

Paisley, UK), 10% fetal bovine serum (GibcoBRL, Paisley, UK), 2.1 ml=500 ml b-mercaptoethanol (BioRad, Glattbrugg, Switzerland), 0.6 mg=ml geneticin sulfate (GibcoBRL, Paisley, UK) and 10% WEHI cell conditioned medium (a natural source of IL-3). The medium was changed every 2 ± 3 d. Dose response curves to leptin and R128Q leptin on proliferation. BAF-3 cells were washed twice with culture medium (without WEHI conditioned medium) and seeded at a density of 20 000 cells=well in 96-well plates (Nunc, Paisley, UK) with various concentrations of recombinant mouse leptin or R128Q leptin (10 pM ± 1 mM). The total incubation volume was 200 ml. After 24 h of incubation, 1 mCi (methyl-3H) thymidine (Amersham, Zurich, Switzerland) was added for a further 16 h. The reaction was stopped by ®ltration through a ®lter mat A using a cell harvester (Tomtec Harvester 96, Wallac, Huenenberg, Switzerland). The ®lter mats were dried, heat sealed with MeltiLex and counted in a b-counter (MicroBeta, Wallac, Huenenberg, Switzerland). Dose response curves to leptin in the presence of different concentrations of R128Q leptin. Full dose response curves to mouse leptin (1 pM ± 1 mM) were performed in the presence of 0 nM, 30 nM, 100 nM or 300 nM R128Q human leptin, in order to test for competitive antagonism of the mutant leptin. Effects of leptin and R128Q leptin on food intake in rodents

Implantation of i.c.v. cannulas. Under pentothal anesthesia (50 mg=kg, i.p.) guide cannulas, aimed at the right lateral ventricle, were stereotaxically implanted and correct placement was tested as previously described.22 Injections were made in conscious free-moving rats. Comparison of the effect of an i.c.v. injection of leptin or R128Q leptin in free-feeding rats. Either mouse leptin (0.5 mg, 5.0 mg or 20.0 mg; n 5 animals per group), R128Q human leptin (0.5 mg, 5.0 mg, 20.0 mg or 50.0 mg; n 6 ± 9 animals per group) or an equal volume of vehicle (10 ml, PBS, pH 7.4; n 28 animals) were injected into the right lateral ventricle. The drugs were administered in a randomized block design. This experimental paradigm ensured that food intake in each animal was assessed in response to the three different treatments: vehicle, leptin and R128Q leptin. Injections were performed between 09.40 ± 10.00 h. Immediately after the i.c.v. injection, the rats were returned to their cages and weighed food was given. Food intake was measured 1 h, 2 h, 3 h and 4 h after the injection and then every second hour for the following 20 h period.


Effect of prior treatment with R128Q leptin on the response to exogenous leptin in fasted rats. The rats used in this study were deprived of food for 24 h prior to study. Thereafter, 0.5 mg R128Q leptin or an equal volume of vehicle (5 ml, 3.3 mM Tris ± HCl, pH 7.4) were injected i.c.v. at 09.30 h. Then 30 min later, either 0.5 mg leptin or an equal volume of vehicle (5 ml, PBS) were injected i.c.v. so that there were four groups: controls, a group treated with R128Q leptin, a group treated with the leptin and a group treated with R128Q leptin leptin (n 7 ± 8 animals). Two hours after the second injection, weighed food was given back and the quantity consumed measured at 1 h, 2 h, 3 h and 4 h, and then every second hour for the following 20 h after refeeding.

Effect of mouse leptin and R128Q human leptin on food intake in free-fed ob=ob and db=db mice. The mice used in this study were fed ad libitum prior to the study. Injections were performed from 16.30 ± 17.00 h, one hour before the lights went off. Either vehicle (Tris HCl, 5 ml=kg; n 16 animals), R128Q leptin (1 mg=5 ml=kg; n 5 animals) or leptin (1 mg=5 ml=kg; n 16 animals) were injected intraperitoneally (i.p.). Immediately after the injection, weighed food was given.

Materials

Recombinant mouse leptin, human leptin and human R128Q leptin were cloned, and the proteins expressed in Escherichia Coli by Novartis AG (Basel, Switzerland) as previously described.26 Both of the proteins used here were from a single batch. The mutation was introduced by using the Quick Change Site Directed Mutagenesis Kit as described by the manufacturer (Stratagene, La Jolla, CA). Sequencing of R128Q leptin cDNA showed the correct mutation. Lyophilized mouse leptin was either dissolved in PBS or, for the i.p. studies, in 3.3 mM Tris ± HCl. Lyophilized R128Q leptin was dissolved in 10 mmol Tris ± HCl, and then diluted 1:2 with PBS to a ®nal concentration of 3.3 mmol Tris ± HCl, pH 7.4. Rats receiving 3.3 mmol Tris ± HCl, pH 7.4 showed no difference in food intake to rats receiving PBS (data not shown).

Results Binding of leptin and R128Q leptin to the ObRb receptor

Binding curves of 125I-human leptin to the hObRb receptor in the presence of increasing concentrations of either human leptin or R128Q human leptin are shown in Figure 1. The results show that both leptin and R128Q leptin displayed similar af®nities for the ObRb receptor. Stimulation of cell proliferation by leptin and R128Q leptin

Cellular proliferation studies in BAF-3 cells, expressing the ObRb receptor, in response to leptin and R128Q leptin are shown in Figure 2A. In contrast to leptin, which showed a potent response to stimulate cellular proliferation, R128Q leptin was without intrinsic activity. This notable difference occurred despite both compounds having essentially equivalent af®nities to the ObRb receptor (Figure 1). When incubated with increasing concentrations of R128Q leptin, the dose response curves to leptinstimulated cellular proliferation was shifted progressively to the right (Figure 2B). The rightward shift in the dose-response curve occurred without suppression of the maximum response to leptin. Comparison of the effect of an i.c.v. injection of leptin and R128Q leptin in free-feeding rats

The above in vitro studies clearly showed R128Q leptin to competitively antagonize the effect of leptin without intrinsic activity in BAF-3 cells. The experiments in Figure 3 illustrate the comparative effect of leptin and R128Q leptin, when given into the lateral ventricle on 24 h food intake in normal rats. The results show that leptin produced a dose-related inhibition of food intake. Surprisingly, R128Q leptin

Statistical analysis

Data are expressed as mean s.e.m. The data from the food intake studies were analyzed by one way analysis of variance. Post hoc comparisons were made using Fisher's protected least signi®cant difference test and Dunnett's test (two-tailed). Probability values P < 0.05 were considered to be statistically signi®cant.

Figure 1 Binding of 125I-human leptin to COS-7-hObRb membranes in the presence of human leptin and R128Q human leptin. Calculated IC50s for human leptin are 0.3 nM and for R128Q human leptin 1.13 nM.

465


466

also dose-dependently decreased food intake, although with an apparently slightly lower potency than leptin. Effect of prior treatment with R128Q leptin on the response to exogenous leptin in fasted rats

To test additivity of the two compounds, a low dose of leptin and R128Q leptin were injected into the lateral ventricle of fasted rats. Fasted rats were used in these experiments to reduce endogenous leptin to low concentrations. Low endogenous leptin concentrations minimize the possibility that changes in food intake in response to R128Q are due to interactions with endogenous leptin. The results in Figure 4 represent cumulative food intake during the 24 h period after refeeding in animals injected i.c.v. with either leptin (0.5 mg) or R128Q leptin (0.5 mg). When compared to controls, both leptin and R128Q leptin injected alone or in combination signi®cantly reduced food intake. The decrease in food intake seen in the rats treated with both drugs was, if anything, stronger than for the animals that received one drug alone. Effect of leptin and R128Q leptin on food intake in freefeeding ob=ob and db=db mice

Figure 2 A. Effect of mouse leptin and R128Q human leptin on proliferation in BAF-3 cells. The calculated EC50 for mouse leptin is 0.15 nM. B. Effect of mouse leptin in the presence of different concentrations of R128Q human leptin on proliferation in BAF-3 cells. Calculated IC50s for leptin in the presence of increasing concentrations of R128Q human leptin are: 0 nM R128Q leptin: IC50 0.17 nM; 30 nM R128Q leptin: IC50 3.0 nM; 100 nM R128Q leptin: IC50 16.4 nM; 300 nM R128Q leptin: IC50 36.8 nM.

Figure 3 Comparison of the effect of an intracerebroventricular (i.c.v.) injection of mouse leptin or R128Q human leptin in freefeeding rats. Satiated rats were injected either with vehicle, mouse leptin or R128Q human leptin at the beginning of the light phase. Cumulative food intake was then measured during the 24 h period after injection of either vehicle, leptin or R128Q leptin. Results are expressed as mean s.e.m.; *P < 0.05, ** P < 0.01 vs control.

ob=ob mice. After treatment with leptin (1 mg=kg, i.p.) or R128Q leptin (1 mg=kg, i.p.), food intake in leptin treated animals, as well as in R128Q leptin treated mice, declined signi®cantly and to an equivalent level. (Figure 5A) db=db mice. In contrast to the situation in ob=ob mice, neither leptin nor R128Q leptin affected food intake in db=db mice when given i.p. (Figure 5B)

Figure 4 Effect of prior treatment with R128Q human leptin on the response to exogenous mouse leptin. Fasted rats were ®rst either injected intracerebroventricularly (i.c.v.) with vehicle or R128Q (0.5 mg), and 30 min later with either vehicle or leptin (0.5 mg). Food was given back to the animals 2 h after the second injection. Data show cumulative food intake during the 24 h period after refeeding. Results are expressed as mean s.e.m.; oP < 0.05, xP < 0.01, *P < 0.005 vs control.


Figure 5 Effect of mouse leptin and R128Q human leptin on food intake in free-feeding ob=ob and db=db mice. Free-fed ob=ob (A) or db=db (B) mice were injected intraperitoneally (i.p.) with either vehicle, leptin (1 mg=kg) or R128Q (1 mg=kg). Food intake over the 8 h period after injection is shown. Results are expressed as mean s.e.m.; *P < 0.05, **P < 0.01 vs control.

Discussion The initial purpose of these studies was to investigate the continuing validity of the hypothesis that leptin is a physiologically important regulator of food intake, using R128Q leptin, a putative leptin receptor antagonist.23 The ®rst series of experiments sought to con®rm the antagonistic activity of R128Q leptin, observed previously in in vitro experiments. Leptin bound with high af®nity to the recombinant ObRb receptor expressed in BAF-3 cells, and induced a marked proliferative response. However, despite displaying a similarly high af®nity for the ObRb receptor, R128Q leptin was unable to induce a proliferative response over a similar dose-range in this cell line. This would indicate that R128Q leptin is devoid of signal transducing activity. These ®ndings are very similar to those described in the ®rst report of this mutant.23 Extending these initial ®ndings, the present studies also demonstrate that in the presence of increasing concentrations of R128Q leptin, the dose-response curve to leptin induced proliferation was shifted progressively to the right without affecting the maximum response to the protein. Thus, R128Q leptin appears to be a potent and competitive inhibitor of the effects of leptin in this cellular system. Previous studies in free-feeding lean rats have shown that the injection of rabbit anti-mouse leptin antibodies into the brain markedly increases food intake during the light phase.22 No increase in food intake was observed in obese Zucker fa=fa rats expressing a defective functional leptin receptor. These observations imply that the presence of leptin in the brain contributes tonically to the inhibition of food intake observed at this time. Based upon these previous observations, initial experiments were designed to determine whether a similar increase in food intake could be observed with the putative leptin

receptor antagonist R128Q leptin. Surprisingly, given the lack of intrinsic activity and the clear antagonistic nature of the compound in vitro, injection of R128Q leptin directly into the ventricular system of the brain actually decreased food intake in free-feeding lean rats. The potency of the leptin mutant to affect food intake was only slightly less than that induced by leptin, suggesting that in contrast to its effects in vitro R128Q leptin retains strong agonistic properties in vivo. To investigate more clearly the agonist activity of R128Q leptin in comparison to leptin, rats fasted for 24 h were used. Fasting has been shown to decrease circulating leptin levels27,28 and therefore should minimise any interaction of R128Q leptin with the endogenous protein. In fasted animals, both leptin and R128Q leptin produced a similar decrease in food intake after refeeding. Since, in this model, the combined administration of R128Q leptin and leptin produced a greater fall in food intake than either compound alone, antagonistic properties are an unlikely expression of the mutant in vivo. When given daily at a dose of 15 mg=mouse i.p. R128Q leptin has been shown not to affect body weight in ob=ob mice.23 This observation has been taken as evidence that the mutant is devoid of agonist activity. However, the results of the present studies demonstrate that when administered i.p. to ob=ob mice, at a dose of 1 mg=kg, R128Q leptin produced a decrease in food intake which was of similar magnitude to that produced by the same dose of leptin. These observations con®rm R128Q leptin as a potent leptin agonist on food intake in vivo. The ®nding that R128Q leptin and leptin similarly reduce food intake in an animal model lacking endogenous leptin suggests that the mutant is able to exert its full effects on feeding in the absence of leptin. The inability of R128Q leptin to affect food intake in db=db mice lacking a functional leptin receptor29 further con®rms the speci®c agonistic activity of this protein through the ObRb receptor. The activity of the mutant in ob=ob mice also rules out the possibility that the observed agonistic effect of R128Q leptin is due to inhibition of the binding of endogenous leptin to a soluble plasma binding protein which would result in an increase in free circulating leptin. From previous studies, it was shown that daily i.p. injection of 200 mg=mouse R128Q leptin produced a marked increase in body weight in wild type mice over a period of 15 d.23 The increase in body weight in these studies was associated with an increase in white fat deposits. Leptin has been shown to decrease food intake and increase energy expenditure. Thus the increase in body weight produced by R128Q leptin is thought to be the result of a blockade of the effects of endogenous leptin by R128Q leptin, which results in a net increase in white adipose tissue mass and hence body weight. Since our results clearly show R128Q leptin to be an agonist in vivo, how can these prior results with the mutant be explained?

467


468

Differences in the 3-D structure of the expressed proteins could be invoked to explain the differences between the present and previous studies. However, we believe that the most likely explanation centers on the fact that in previous studies R128Q leptin was coadministered with a large dose of monoclonal antibodies directed against human leptin. Unfortunately in this study, no data on the cross-reactivity of the antibody to human and mouse leptin or on the effect of the antibody alone were reported. One potential explanation may be that the monoclonal antibody, injected in a 10-fold molar excess compared to the leptin mutant, bound endogenous mouse leptin and thereby created a state of leptin depletion, which led to the observed increase in food intake. This conclusion is in accord with previous ®ndings, showing that polyclonal antibodies directed against leptin increase food intake in free-feeding lean rats.22 The results of the present studies clearly show R128Q leptin to be an antagonist of leptin in vitro, but to behave as an agonist in vivo. The reason for this paradox is unclear at the present time. Differences in the conformation of the expressed protein in the COS7 cells and in the whole animals could be evoked to explain the observed results. However, a further speculative explanation could involve the relative ability of R128Q leptin to activate different intracellular pathways through the ObRb receptor in the hypothalamus and in BAF-3 cells. In this regard, it has previously been demonstrated that the leptin receptor (ObRb) can activate different intracellular pathways within the same cell line. For example, the ObRb receptor is linked to both the Jak=Stat and the MAPK pathways in transfected COS cells.30 Therefore, leptin acting on the same receptor, in different cell types can trigger distinct intracellular pathways. It is plausible that the mutant could bind to the ObRb receptor in BAF-3 cells, but be unable to induce a proliferative response, whereas it could still activate the pathway leading to food intake inhibition in the hypothalamus, most probably via the Stat3 signalling pathway.31 Clearly, further experiments are needed to test this hypothesis.

Conclusion In summary, we have shown that the leptin mutant R128Q leptin behaves as a competitive antagonist of leptin induced proliferation in BAF-3 cells. In contrast, the leptin mutant is a full agonist at stimulating food intake. The data demonstrate that the human leptin mutant R128Q leptin is not suitable for investigating the physiological actions of leptin in vivo and strongly suggest that care should be taken when extrapolating the action of leptin mimics from the in vitro to the in vivo situation.

Acknowledgements

We thank K. Wyss, S. di Bello and O. Peter for help with the animal studies and A. Crossthwaite and A. Schmid for their help with the cell cultures. References

1 Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 1994; 372: 425 ± 432. 2 Masuzaki H, Ogawa Y, Sagawa N, Hosoda K, Matsumoto T, Mise H, Nishimura H, Yoshimasa Y, Tanaka I, Mori T, Nakao K. Nonadipose tissue production of leptin: leptin as a novel placenta-derived hormone in humans. Nat Med 1997; 3: 1029 ± 1033. 3 Senaris R, Garcia Caballero T, Casabiell X, Gallego R, Castro R, Considine RV, Dieguez C, Casanueva FF. Synthesis of leptin in human placenta. Endocrinology 1997; 138: 4501 ± 4504. 4 Kawai M, Yamaguchi M, Murakami T, Shima K, Murata Y, Kishi K. The placenta is not the main source of leptin production in pregnant rat: gestational pro®le of leptin in plasma and adipose tissues. Biochem Biophys Res Commun 1997; 240: 798 ± 802. 5 Lepercq J, Cauzac M, Lahlou N, Timsit J, Girard J, Auwerx J, Hauguel de Mouzon S. Overexpression of placental leptin in diabetic pregnancy: a critical role for insulin. Diabetes 1998; 47: 847 ± 850. 6 Smith Kirwin SM, O'Connor DM, De Johnston J, Lancey ED, Hassink SG, Funanage VL. Leptin expression in human mammary epithelial cells and breast milk. J Clin Endocrinol Metab 1998; 83: 1810 ± 1813. 7 Bado A, Levasseur S, Attoub S, Kermorgant S, Laigneau JP, Bortoluzzi MN, Moizo L, Lehy T, Guerre Millo M, Lemarchandbrustel Y, Lewin MJM. The stomach is a source of leptin. Nature 1998; 394: 790 ± 793. 8 Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters D, Boone T, Collins F. Effects of the obese gene product on body weight regulation in ob=ob mice. Science 1995; 269: 540 ± 543. 9 Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, Burley SK, Friedman JM. Weightreducing effects of the plasma protein encoded by the obese gene. Science 1995; 269: 543 ± 546. 10 Weigle DS, Bukowski TR, Foster DC, Holderman S, Kramer JM, Lasser G, Lofton Day CE, Prunkard DE, Raymond C, Kuijper JL. Recombinant ob protein reduces feeding and body weight in the ob=ob mouse. J Clin Invest 1995; 96: 2065 ± 2070. 11 Erickson JC, Clegg KE, Palmiter RD. Sensitivity to leptin and susceptibility to seizures of mice lacking neuropeptide Y. Nature 1996; 381: 415 ± 421. 12 Camp®eld LA, Smith FJ, Guisez Y, Devos R, Burn P. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 1995; 269: 546 ± 549. 13 Schwartz MW, Seeley RJ, Camp®eld LA, Burn P, Baskin DG. Identi®cation of targets of leptin action in rat hypothalamus. J Clin Invest 1996; 98: 1101 ± 1106. 14 Seeley RJ, van Dijk G, Camp®eld LA, Smith FJ, Burn P, Nelligan JA, Bell SM, Baskin DG, Woods SC, Schwartz MW. Intraventricular leptin reduces food intake and body weight of lean rats but not obese Zucker rats. Horm Metab Res 1996; 28: 664 ± 668. 15 Halaas JL, Boozer C, Blair West J, Fidahusein N, Denton DA, Friedman JM. Physiological response to long-term peripheral and central leptin infusion in lean and obese mice. Proc Natl Acad Sci USA 1997; 94: 8878 ± 8883. 16 Hakansson ML, Brown H, Ghilardi N, Skoda RC, Meister B. Leptin receptor immunoreactivity in chemically de®ned target neurons of the hypothalamus. J Neurosci 1998; 18: 559 ± 572.


17 Huang XF, Koutcherov I, Lin S, Wang HQ, Storlien L. Localization of leptin receptor mRNA expression in mouse brain. Neuroreport 1996; 7: 2635 ± 2638. 18 Hakansson ML, Hulting AL, Meister B. Expression of leptin receptor mRNA in the hypothalamic arcuate nucleus ± relationship with NPY neurones. Neuroreport 1996; 7: 3087 ± 3092. 19 Baskin DG, Seeley RJ, Kuijper JL, Lok S, Weigle DS, Erickson JC, Palmiter RD, Schwartz MW. Increased expression of mRNA for the long form of the leptin receptor in the hypothalamus is associated with leptin hypersensitivity and fasting. Diabetes 1998; 47: 538 ± 543. 20 Iida M, Murakami T, Ishida K, Mizuno A, Kuwajima M, Shima K. Substitution at codon 269 (glutamine ? proline) of the leptin receptor (OB-R) cDNA is the only mutation found in the Zucker fatty (fa=fa) rat. Biochem Biophys Res Commun 1996; 224: 597 ± 604. 21 Chua SC Jr, White DW, Wu Peng XS, Liu SM, Okada N, Kershaw EE, Chung WK, Power Kehoe L, Chua M, Tartaglia LA, Leibel RL. Phenotype of fatty due to Gln269Pro mutation in the leptin receptor (Lepr). Diabetes 1996; 45: 1141 ± 1143. 22 Brunner L, Nick HP, Cumin F, Chiesi M, Baum HP, Whitebread S, Stricker Krongrad A, Levens N. Leptin is a physiologically important regulator of food intake. Int J Obes 1997; 21: 1152 ± 1160. 23 Verploegen SA, Plaetinck G, Devos R, Van der Heyden J, Guisez Y. A human leptin mutant induces weight gain in normal mice. FEBS Lett 1997; 405:

24 Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: 248 ± 254. 25 Ghilardi N, Skoda RC. The leptin receptor activates janus kinase 2 and signals for proliferation in a factor-dependent cell line. Mol Endocrinol 1997; 11: 393 ± 399. 26 Rentsch J, Levens N, Chiesi M. Recombinant ob-gene product reduces food intake in fasted mice. Biochem Biophys Res Commun 1995; 214: 131 ± 136. 27 Boden G, Chen X, Mozzoli M, Ryan I. Effect of fasting on serum leptin in normal human subjects. J Clin Endocrinol Metab 1996; 81: 3419 ± 3423. 28 Trayhurn P, Thomas ME, Duncan JS, Rayner DV. Effects of fasting and refeeding on ob gene expression in white adipose tissue of lean and obese (ob=ob) mice. FEBS Lett 1995; 368: 488 ± 490. 29 Chen H, Charlat O, Tartaglia LA, Woolf EA, Weng X, Ellis SJ, Lakey ND, Culpepper J, Moore KJ, Breitbart RE, Duyk GM, Tepper RI, Morgenstern JP. Evidence that the diabetes gene encodes the leptin receptor: identi®cation of a mutation in the leptin receptor gene in db=db mice. Cell 1996; 84: 491 ± 495. 30 Bjorbaeck C, Uotani S, Dasilva B, Flier JS. Divergent signaling capacities of the long and short isoforms of the leptin receptor. J Biol Chem 1997; 272: 32686 ± 32695. 31 Vaisse C, Halaas JL, Horvath CM, Darnell JE Jr, Stoffel M, Friedman JM. Leptin activation of Stat3 in the hypothalamus of wild-type and ob=ob mice but not db=db mice. Nat Genet 1996; 14: 95 ± 97.

469