sign in sign up
<<
Home , Leptin

International Journal of (1998) 22, 1159±1163 ß 1998 Stockton Press All rights reserved 0307±0565/98 $12.00 http://www.stockton-press.co.uk/ijo Commentary Leptin resistance in obese humans: does it exist and what does it mean?

JRS Arch1, MJ Stock2 and P Trayhurn3

1SmithKline Beecham Pharmaceuticals, New Frontiers Science Park (North), Harlow, Essex, CM19 5AD; 2Department of , St Georges Hospital Medical School, Tooting, London, SW17 0RE; and 3Division of Biomedical Science, Rowett Research Institute, Bucksburn, Aberdeen, AB21 9SB, Scotland, UK

Keywords: leptin; leptin resistance; obese humans; diabetes

Introduction Raised leptin levels in obesity

It is more than two years since it was ®rst reported1 Clearly the concentration of any molecule that is that most obese people have elevated serum leptin involved in the regulation of body adiposity will concentrations and numerous subsequent reports have alter in a direction opposing the deposition of excess con®rmed these original ®ndings, sometimes describ- , unless it is part of a defective regulatory pathway ing concentrations far higher in the obese than the that is responsible for the excessive fat deposition in lean. Given the obvious failure of these high concen- the ®rst place. We suggest that leptin concentrations trations of leptin to produce changes in energy intake are not raised in the majority of obese individuals or expenditure that restore fat mass to normal, it was because of leptin resistance, but because leptin is reasonable to propose that `leptin resistance' is a opposing other forces that promote obesity. primary cause of obesity, analogous to resis- It has been argued that obese subjects who have high tance in . Alternatively, leptin resis- leptin concentrations must be leptin resistant or they tance might, like , be a secondary would not be obese. However, all that the high leptin consequence of obesity. There is, however, scant concentrations may show us is what we already know: direct evidence for leptin resistance in humans, that body fat content is not tightly regulated in most whereas there is rather better evidence that inadequate people, mechanisms that prevent being leptin secretion is a factor pre-disposing some indivi- weak. In this sense, all humans may be resistant to duals to obesity. Aside from a few rare examples, leptin: obese subjects are not necessarily any more where massive obesity is associated with a single resistant to the than lean subjects. If we could mutation in the OB gene2 or the OB gene,3 measure the concentration of other messenger mole- there are at least four linkage studies which suggest cules involved in weight regulation as easily as we can that the OB or an adjacent gene might be involved in measure leptin, we might ®nd that they also change in human obesity.4,5 There are also studies which link obesity in a direction that opposes further weight gain. lower leptin concentrations than would be expected So obese subjects might equally be described as from the body mass index (BMI) to a pre-disposition `resistant' to these other molecules. By way of a to weight gain.6±8 Evidence for leptin resistance, on speci®c example, it is interesting that in four studies the other hand, has been based solely on the ®nding of Caucasian subjects sympathetic activity was corre- that obese humans generally have elevated serum or lated with body fat content.9 Perhaps, therefore, many plasma leptin concentrations compared to lean sub- obese individuals are also `resistant' to the regulation jects. We suggest that this is not evidence for leptin of obesity by the sympathetic in the resistance, unless one extends the meaning of the term same sense that they are `resistant' to leptin. `Resis- such that it becomes of little value. tance' in this context does not, however, mean that the leptin or noradrenaline is having no effect, or even less effect than in lean individuals. One only has to con- sider the gross obesity of individuals who lack func- tional leptin,2 to appreciate that leptin must play some role in resisting weight gain in most obese subjects. Correspondence: JRS Arch, SmithKline Beecham If leptin resistance is a cause rather than a conse- Pharmaceuticals, New Frontiers Science Park (North), Coldharbour Road, Harlow, Essex, CM19 5AD, UK. quence of obesity, leptin concentrations would be Received 12 June 1998; accepted 1 September 1998 expected to be raised in subjects who are prone to Leptin resistance JRS Arch et al 1160 obesity but have slimmed. No studies have been part from an inappropriate analogy with diabetes: described on subjects who have slimmed to a BMI Type 1 diabetes is due to a failure of insulin secretion, in the normal range, but what evidence there is does whereas Type 2 diabetes results from insulin resis- not support this proposition. In one study a mean tance coupled with inadequate insulin secretion to weight loss of 21% achieved over 74 ± 110 d was cope with the insulin resistance. In the early stages associated with a fall in leptin concentrations of of Type 2 diabetes, elevated blood glucose is 76%.10 In another study over 10 months in 52 obese associated with elevated insulin concentrations and women, a mean fall in BMI of 2.7 kg=m2 was asso- this has been shown to be a consequence of insulin ciated with a reduction in plasma leptin from 30 to resistance. Therefore, the argument goes, elevated 22 ng=ml serum, much as would be predicted. More- body fat associated with elevated leptin concentra- over there was no evidence that subjects with the tions must indicate leptin resistance. highest baseline leptin concentrations ± those that Why, however, should one restrict the analogy might be called most `leptin resistant' ± had the great- between diabetes and obesity to the fasting state? est dif®culty in losing weight.11 After a , both blood glucose and insulin are If the association of elevated leptin concentrations raised, not because of insulin resistance, but because with obesity is an argument for leptin resistance, what insulin is responding to an in¯ux of glucose. Simi- are we to infer from the low leptin concentrations larly, leptin may be raised in obesity, not because of found in nervosa? The authors of one leptin resistance, but because fat stores are high. It is report12 do not exclude the possibility that patients the stimulus to hormone secretion that is high. Of with have an increased sensitivity to course fat stores in humans build up over a period of leptin. It seems more likely to us, however, that leptin months or years, whereas blood glucose concentra- concentrations fall in a largely unsuccessful attempt to tions rise rapidly after a single meal. Similarly, insulin prevent a disease whose origins are primarily cultural acts rapidly to control blood glucose concentrations, and cognitive. Indeed, one can actually argue that whereas leptin does not have a signi®cant impact on anorexic patients have an underlying problem of body fat content within a day. The different timescales leptin resistance. Thus when weight gain was of the insulin=glucose and leptin=fat systems make it achieved by behavioural therapy, their plasma leptin dif®cult to draw simple analogies. concentrations increased to levels well in excess of Timescale is not the only difference between how those of controls matched for BMI.13 A more reason- blood glucose and body fat are regulated. Insulin is able explanation of this ®nding is, however, that these overwhelmingly the dominant regulator of blood glu- patients had adapted to a low body fat content, which cose concentration. There is no `Type 3' diabetes with they defended when in positive energy balance with the absence of involvement of either defective insulin hypersecretion of leptin. Conversely, obese subjects secretion or action. Moreover, insulin acts directly on may adapt to their high body fat content and respond the major glucose utilising tissues and other factors to fasting with a rapid fall in leptin secretion.14 that effect insulin action ultimately modulate the The rapid decreases in plasma leptin following action of insulin on these same tissues. In contrast, short-term fasting or food restriction can be seen energy balance may well be regulated by molecules when there is little or no detectable change in body that are not involved in the leptin pathway (gut fat, indicating that leptin secretion is not only deter- , for example), or (like peripheral noradrena- mined by the size of the fat reserves.10,14 This has been line) are suf®ciently removed from the site of leptin clearly demonstrated in animal studies in which a action that to describe a defect in their production or range of factors which acutely affect leptin production action as leptin resistance is to divert attention from are recognised.15 Such regulatory in¯uences could also the real problem. More important, however, than be responsible for the disproportionate response of leptin or any other physiological mechanism that leptin to increases in fat stores.16 Raised leptin con- might predispose humans to obesity is the in¯uence centrations can also be observed in normal lean sub- of culture or environment. Obesity resembles anorexia jects. A recent study17 found that subjects habitually nervosa in being primarily due to cultural factors, with consuming a high-fat have raised leptin concen- only about 25 ± 40% of the variance in BMI being trations and a higher (BMR) than determined by our genes.4 To explain the rapid subjects with the same BMI and adiposity habitually increase in the incidence of obesity over the last 40 consuming a low-fat diet, suggesting that overnutri- years as an increase in leptin resistance would be tion, like undernutrition, can in¯uence leptin secretion merely to argue that leptin is relatively impotent in independently of its effects on adiposity. controlling our liking for palatable, high fat, readily available food or, bearing in mind that we are actually less than we used to,18 leptin does not increase Analogy with diabetes? our desire to . If we consider how insulin resistance is determined, we can envisage equivalent ways by which leptin The presumption that obesity must be a consequence resistance might be demonstrated. By analogy with of a de®cit in leptin secretion or leptin action stems in the hyperinsulinaemic euglycaemic clamp, leptin Leptin resistance JRS Arch et al 1161 could be infused to the same level in obese and lean Central administration of leptin is more effective subjects and , food intake or metabolic rate than peripheral administration in diet-induced obese measured. Alternatively, tissues from obese indivi- mice.28,29 A possible explanation for this is that duals might be shown to be less responsive to leptin centrally administered leptin produces a suf®ciently than those from lean individuals. Recent papers strong stimulus to make an impact upon diet-induced describe lipolytic responses to leptin in rodent adipose hyperphagia. However, this ®nding has raised the tissue and adipocytes19,20 and so it is possible that possibility that there is a defect in the transport of lipolytic responses can be compared in leptin into the . A similar proposal has been put from obese and lean humans. Should a difference be forward to explain `leptin resistance' in humans, on found, it will be important to compare adipocytes the basis that the relationship between plasma and from post-obese and lean humans to determine CSF leptin concentrations is not linear.30,31 This whether leptin resistance is a primary cause of obesity probably re¯ects the fact that the transport of leptin or a secondary consequence. into the brain is by a saturable mechanism.32 It is possible that in obesity the transporter is close to saturation with leptin and so the plasma leptin con- Insights from rodents centration is near the top of the concentration- response curve. Further administration of peripheral leptin would have little effect. However, since leptin Leptin resistance clearly does occur in some labora- would be exerting a near-maximal effect, it would tory rodents. In fa=fa rats and db=db mice, it is the seem inappropriate to describe this situation as one of primary cause of obesity, since these mutants have leptin resistance. Of course, it is possible that the defective leptin receptors. Other animal models of transport system is down-regulated in obesity, but this obesity may, however, offer greater insights into why has not been demonstrated. In fact Considine and obesity in man tends to be associated with raised Caro33 have recently commented that saturation of plasma leptin concentrations. the transport of leptin into the CSF is probably not the Lethal yellow (Ay=a) mice, in which the Agouti initiating cause of obesity in humans, since saturation is ectopically expressed, are resistant to both only occurs at blood levels that are associated with central and peripheral administration of leptin. Never- extreme obesity. theless, the leptin signal appears to be sensed in these The brown fat de®cient (UCP-DTA) mouse, in mice, since their hypothalami contain low concentra- which expression of diphtheria toxin is driven by the tions of the mRNA for Agouti-related peptide, a gene -1 promoter, provides a particularly that is overexpressed in ob=ob mice.21 Lethal yellow intriguing example of leptin resistance in the rodent. It obese mice also express normal levels of is not surprising that leptin fails to stimulate thermo- Y (NPY) mRNA in the , whereas genesis in these mice since they lack brown fat, but ob=ob and db=db mice contain high concentrations.22 why does leptin also fail to reduce food intake? It seems likely that the concentration of leptin is so Moreover, resistance is present even before hyperlep- high in lethal yellow obese mice that exogenous leptin tinaemia develops. Therefore resistance is not second- can have no further effect. Interestingly, if these mice ary to hyperleptinaemia or simply due to leptin levels are also made leptin de®cient by introducing the ob being beyond the concentration-response range.34 gene, they become even more obese, but are then However, as in the lethal yellow mouse and the sensitive to leptin.23 This is easily explained as the diet-induced obese rat, hypothalamic NPY mRNA leptin system is now operating from a zero baseline. concentrations are reduced after hyperleptinaemia Diet-induced obesity using high-fat or `cafeteria' develops, suggesting that the leptin signal is diets is perhaps the best available model for most sensed.35 A clue to the leptin resistance of the human obesity, particularly as it demonstrates that the brown fat de®cient mouse lies in the link between hedonic qualities of the diet are quite capable of its hyperphagia and defective brown fat thermogen- overriding the effects of any physiological satiety esis. When housed at an environmental temperature signal, and result in hyperphagia in rats and humans below thermoneutrality (29=33C for rats=mice), ani- alike.24 Leptin is relatively ineffective in reducing mals increase their food intake to help maintain body food intake and body weight in rodents fed on a high temperature via non-shivering (including diet- fat diet,25,26,27 but when rats that have been fed on induced) . Thus, with a reduced capa- such a diet are returned to standard laboratory diet, city for brown fat thermogenesis, it is not surprising their food intake immediately falls below that of rats that the UCP-DTA mouse is hyperphagic, and its fed on laboratory diet throughout.26 This suggests that apparent leptin resistance is simply due to the fact sensitivity to leptin was unaltered on the high-fat diet, that maintenance of homeothermy is more important but leptin failed to repress or restrain intake due to its for survival than resisting obesity. inability to compete with the highly-palatable, energy- It has been proposed that leptin resistance is dense diet. As soon as the animals were restored to the induced by overactivity of the hypothalamic-pitui- monotonous chow diet, intake responded immediately tary-adrenal axis and raised concentra- to the elevated leptin levels. tions.36 Thus adrenalectomy enhances the effects of Leptin resistance JRS Arch et al 1162 leptin on food intake and body weight in rats, and energy intake and increase the amount of exercise that restores the sensitivity to normal.37 they take. In these subjects, there is a reduction in Although this is an appealing theory, it must be leptin levels10,11,41,42 which will oppose their efforts, remembered that adrenalectomy also ameliorates or but they will also become more sensitive to exogenous arrests obesity in db=db mice38 and fa=fa rats,39 both leptin, and indeed to other anti-obesity agents that act of which are resistant to leptin due to mutations in the via physiological mechanisms. Drugs which are most gene. The impressive effects of adre- effective in those who help themselves would nalectomy and their reversal by glucocorticoid repla- obviously be more acceptable ethically and less cement, together with studies manipulating central liable to abuse than those which might endorse a corticotropin (CRH) levels,40 strongly implicate life of gluttony and sloth. CRH as a mediator of leptin's effects that acts down- stream from the leptin receptor. Finally, it should be noted that animals that lack References leptin (for example, ob=ob mice) or have obvious 1 Considine RV, Sinha MK, Heiman ML, Kriauciunas A, leptin resistance (for example, db=db mice, fa=fa Stephens TW, Nyce MR, Ohannesian JP, Marco CC, McKee rats) fail to achieve and are infertile. How- LJ, Bauer TL, Caro JF. Serum immunoreactive-leptin concen- ever, there is no evidence for sterility or failure to trations in normal-weight and obese humans. N Engl J Med 1996; 334: 292 ± 295. reach puberty in any of the so-called `leptin resistant' 2 Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H, diet obese rodent models, and neither is there in the Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst majority of obese humans. JA, Cheetham CH, Earley AR, Barnett AH, Prins JB, O'Rahilly S. Congenital leptin de®ciency is associated with severe early-onset obesity in humans. Nature 1997; 387: 903 ± 908. Conclusion 3 Clement K, Vaisse C, Lahlou N, Cabrol S, Pelloux V, Cassuto D, Gourmelen M, Dina C, Chambaz J, Lacorte J, Basdevant A, Bougneres P, Lebouc Y, Froguel P, Guy-Grand B. A mutation Does leptin resistance occur in man? Of course, to in the human leptin receptor gene causes obesity and pituitary some degree it must. As in other biological systems dysfunction. Nature 1998; 392: 398. 4 Bray J, Bouchard C. Genetics of human obesity: research there is likely to be a genetically-based variability directions. FASEB J 1997; 11: 937 ± 945. between individuals in their sensitivity to leptin. 5 Francke S, Clement K, Dina C, Inoue H, Behn P, Vatin Moreover, as the leptin concentration rises within an V, Basdevant A, Guy-Grand B, Permutt MA, Froguel P, individual, there is likely to be some down-regulation Hager J. Genetic studies of the leptin receptor gene in of leptin receptors or those mechanisms that couple morbidly obese French Caucasian families. Hum Genet 1997; 100: 491 ± 496. the receptor to their response. More simplistically, as 6 Scholz GH, Englaro P, Thiele I, Scholz M, Klusmann T, the leptin concentration rises it will reach a ¯atter part Kellner K, Rascher W, Blum WF. Dissociation of serum leptin of the concentration-response curve. What this article concentration and body fat content during long term dietary challenges is the facile description of human obesity intervention in obese individuals. Horm Metab Res 1996; 28: as being due to `leptin resistance,' simply because it is 718 ± 723. 7 Ravussin E, Pratley RE, Maffei M, Wang H, Friedman JM, associated with raised plasma leptin concentrations. Bennett PH, Bogardus C. Relatively low plasma leptin con- Obesity has a number of causes, cultural in¯uences centrations precede weight gain in Pima Indians. Nat Med being of particular importance, and to combine them 1997; 3: 238 ± 240. all into the term `leptin resistance' is misleading. Is 8 Clement K, Lahlou N, Ruiz J, Hager J, Bougneres P, Basde- `leptin resistance' a primary cause of obesity, a vant A, Guy-Grand B, Froguel P. Association of poorly controlled diabetes with low serum leptin in morbid obesity. secondary phenomenon that exacerbates the main Int J Obes 1997; 21: 556 ± 561. cause, or in practice a term that indicates that body 9 Macdonald IA. Advances in our understanding of the role of weight is not well-regulated by leptin or anything the sympathetic nervous system in obesity. Int J Obes 1995; 19 else? (Suppl 7): S2 ± S7. Despite our concerns with the widespread and 10 Weigle DS, Duell PB, Connor WE, Steiner RA, Soules MR, Kuijper JL. Effect of fasting, refeeding, and dietary fat indiscriminate use of the term `leptin resistance', it restriction on plasma leptin levels. J Clin Endocrinol Metab does raise one important issue: is exogenous leptin 1997; 82: 561 ± 565. likely to be very effective as an anti-obesity agent in 11 Wing RR, Sinha MK, Considine RV, Lang W, Caro JF. those subjects in whom blood levels are already Relationship between weight loss maintenance and changes raised? If leptin receptors are already strongly stimu- in serum leptin levels. Horm Metab Res 1996; 28: 698 ± 703. 12 Grinspoon SK, Askari H, Landt ML, Nathan DM, Schoenfeld lated, giving more may be of little bene®t. The same DA, Hayden DL, Laposata M, Hubbard J, Klibanski A. Effects question can also be asked of targets downstream of of fasting and glucose infusion on basal and overnight leptin leptin. For example, if NPY levels are already low, concentrations in normal-weight women. Am J Clin Nutr how effective will NPY receptor antagonists be? Of 1997; 66: 1352 ± 1356. course, there may still be ample potential for such 13 Hebebrand J, Blum WF, Barth N, Coners H, Englaro P, Juul A, Ziegler A, Warnke A, Rascher W, Remschmidt H. Leptin agents to achieve a therapeutic bene®t, but perhaps levels in patients with anorexia nervosa are reduced in the leptin or NPY receptor antagonists will be of most acute stage and elevated upon short-term weight. Mol Psychiat bene®t in those who are attempting to restrain their 1997; 2: 330 ± 334. Leptin resistance JRS Arch et al 1163 14 Boden G, Chen X, Mozzoli M, Ryan I. Effect of fasting on 29 Van Heek M, Compton DS, France CF, Tadesco RP, Fawzi serum leptin in normal human subjects. J Clin Endocrinol AB, Graziano MP, Sybertz EJ, Strader CD, Davis HRJ. Metab 1996; 81: 3419 ± 3423. Diet-induced obese mice develop peripheral, but not 15 Trayhurn P, Duncan JS, Hoggard N, Rayner DV. Regulation central, resistance to leptin. J Clin Invest 1997; 99: of leptin production: A dominant role for the sympathetic 385 ± 390. nervous system? Proc Nutr Soc 1998; 57: 413 ± 419. 30 Schwartz MW, Peskind E, Raskind M, Boyko EJ, Porte D Jr. 16 Luke AH, Rotimi CN, Cooper RS, Long AE, Forrester TE, Cerebrospinal ¯uid leptin levels: relationship to plasma levels Wilks R, Bennett FI, Ogunbiyi O, Compton JA, Bowsher RR. and to adiposity in humans. Nat Med 1996; 2: 589 ± 593. Leptin and body composition of Nigerians, Jamaicans and US 31 Caro JF, Kolaczynski JW, Nyce MR, Ohannesian JP, Open- blacks. Am J Clin Nutr 1998; 67: 391 ± 396. tanova I, Goldman WH, Lynn RB, Zhang PL, Sinha MK, 17 Cooling J, Barth J, Blundell J. Functional phenotypes: Considine RV. Decreased cerebrospinal-¯uid=serum leptin Increased leptin and basal metabolic rate (BMR) as adaptive ratio in obesity: a possible mechanism for leptin resistance. responses to a habitual high fat (high energy) diet. Proc Nutr Lancet 1996; 348: 159 ± 161. Soc 1998; 57: 127A 32 Banks WA, Kastin AJ, Huang W, Jaspan JB, Maness LM. 18 Prentice AM, Jebb SA. Obesity in Britain: gluttony or sloth? Leptin enters the brain by a saturable system independent of BMJ 1995; 311: 437 ± 439. insulin. Peptides 1996; 17: 305 ± 311. 19 Siegrist-Kaiser CA, Pauli V, Juge-Aubry CE, Boss O, Pernin 33 Considine RV, Caro JF. Leptin and the regulation of body A, Chin WW, Cusin I, Rohner-Jeanrenaud F, Burger AG, Zapf weight. Int J Biochem Biol 1997; 29: 1255 ± 1272. J, Meier CA. Direct effects of leptin on brown and white 34 Hamann A, Busing B, Kausch C, Ertl J, Preibisch G, Greten H, . J Clin Invest 1997; 100: 2858 ± 2864. Matthael S. Chronic leptin treatment does not prevent the 20 FruÈhbeck G, Aguado M, Martinez JA. In vitro lipolytic effect development of obesity in transgenic mice with brown fat of leptin on mouse adipocytes: evidence for a possible auto- de®ciency. Diabetologia 1997; 40: 810 ± 815. crine=paracrine role of leptin. Biochem Biophys Res Commun 35 Mantzoros CS, Frederich RC, Qu DQ, Lowell BB, Maratos- 1997; 240: 590 ± 594. ¯ier E, Flier JS. Severe leptin resistance in brown fat-de®cient 21 Ollmann MM, Wilson BD, Yang YK, Kerns JA, Chen Y, uncoupling protein promotor driven diphtheria-toxin-A mice Gantz I, Barsh GS. Antagonism of central despite suppression of hypothalamic and receptors in vitro and in vivo by agouti-related protein. Science circulating corticosterone concentrations. Diabetes 1998; 47: 1997; 278: 135 ± 138. 230 ± 238. 22 Kesterson RA, Huszar D, Lynch CA, Simerly RB, Cone RD. 36 Ur E, Grossman A, DespreÂs JP. Obesity results as a conse- Induction of neuropeptide Y in the dorsal quence of glucocorticoid induced leptin resistance. Horm medial hypothalamic nucleus in two models of the Agouti Metab Res 1996; 28: 744 ± 747. obesity syndrome. Mol Endocrinol 1997; 11: 630 ± 637. 37 Zakrzewska KE, Cusin I, Sainsbury A, Rohner-Jeanrenaud F, 23 Boston BA, Blaydon KM, Varnerin J, Cone RD. Independent Jeanrenaud B. as counterregulatory and additive effects of central POMC and leptin pathways on of leptin: toward an understanding of leptin resistance. Dia- murine obesity. Science 1997; 278: 1641 ± 1644. betes 1997; 46: 717 ± 719. 24 Stock MJ. Models of nutrient partitioning. In: Bouchard C, 38 Shimomura Y, Bray GA, Lee M. Adrenalectomy and steroid Bray GA (eds). Regulation of Body Weight. Wiley: Chichester, treatment in obese (ob=ob) and diabetic (db=db) mice. 1996, 33 ± 43. Hormone Metab Res 1987; 19: 295 ± 299. 25 Van Heek M, Compton DS, France CF, Tedesco RP, Fawzi 39 Marchington D, Rothwell NJ, Stock MJ, York DA. Energy AB, Graziano MP, Sybertz EJ, Strader CD, Davis HR Jr. Diet- balance, diet-induced thermogenesis and induced obese mice develop peripheral, but not central, in lean and obese (fa=fa) Zucker rats after adrenalectomy. resistance to leptin. J Clin Invest 1997; 99: 385 ± 390. J Nutr 1983; 113: 1395 ± 1402. 26 Widdowson PS, Upton R, Buckingham R, Arch J, Williams G. 40 York DA. Lessons from animal models of obesity. Obesity Inhibition of food response to intracerebroventricular injection 1996; 25: 781 ± 800. of leptin is attenuated in rats with diet-induced obesity. 41 Hickey MS, Houmard JA, Considine RV, Tyndall GL, Mid- Diabetes 1997; 46: 1782 ± 1785. gette JB, Gavigan KE, Weidner ML, McCammon MR, Israel 27 Rayner DV, Walbank H, Trayhurn P. Leptin insensitivity in RG, Caro JF. Gender-dependent effects of exercise training on C57B1=6J mice fed high fat diets of differing composition. serum leptin levels in humans. Am J Physiol 1997; 272: Proc Nutr Soc 1998; 57: 126A. E562 ± E566. 28 Halaas JL, Boozer C, Blair-West J, Fidahusein N, Denton DA, 42 Kohrl WM, Landt M, Birge SJ Jr. Serum leptin levels are Friedman JM. Physiological response to long-term peripheral reduced in response to exercise training, but not hormone and central leptin infusion in lean and obese mice. Proc Natl replacement therapy, in older women. J Clin Endocrinol Acad Sci USA 1997; 94: 8878 ± 8883. Metab 1996; 81: 3980 ± 3985.