Clinical implications of leptin and its potential humoral regulators in long-term low-calorie diet therapy for obese humans

ORIGINAL COMMUNICATION (2002) 56, 593 600 ß 2002 Nature Publishing Group All rights reserved 0954 3007/02 $25.00 www.nature.com/ejcn and its potential humoral regulators in long-term low-calorie diet therapy for obese humans T Miyawaki 1, H Masuzaki 1 *, Y Ogawa 1, K Hosoda 1, H Nishimura 1, N Azuma 1, A Sugawara 1, I Masuda 1, M Murata 2, T Matsuo 2, T Hayashi 1, G Inoue 1, Y Yoshimasa 1 and K Nakao 1 1 Department of Medicine and Clinical Science, Kyoto University Graduate School of Medicine, Kyoto, Japan; and 2 Department of Internal Medicine, Osaka Saiseikai Nakatsu Hospital, Osaka, Japan Objective: To address the clinical implications of leptin and to re-examine the relationship between leptin and its potential humoral regulators such as insulin, nonesterified fatty acids (NEFA) and triiodothyronine (T3) in low-calorie diet (LCD) for obese humans. Design: Longitudinal study. Setting: University and foundation hospitals. Subjects: Ten obese men and 10 premenopausal obese women. Interventions: Five men and five women took 800 kcal=day LCD and another five men and five women took 1400 kcal=day balanced deficit diet (BDD) during 4 weeks. Results: Plasma leptin levels in the LCD group decreased more markedly (46.2 14.6 to 13.2 3.6 ng=ml) than that expected for the decrement in percentage fat (39.0 1.7 to 35.9 1.7%) and body mass index (BMI; 35.4 2.4 to 33.1 2.2 kg=m 2 ), while that in the BDD group did not decrease significantly (14.9 3.5 to 13.4 2.8 ng=ml). The ratio of the decrease in leptin levels to that of BMI during the first week was significantly greater than that during the following 3 weeks (39.5 2.7 vs 29.3 2.1%, P ¼ 0.017). The plasma insulin and T3 levels also fell substantially in the first week and continued to decrease during the entire course. Plasma leptin levels measured weekly in each subject were correlated well with insulin (r ¼ 0.586, P ¼ 0.0003) and T3 (r ¼ 0.785, P ¼ 0.0004). Multiple regression analyses after adjustment for the time course and BMI revealed that serum levels of T3 were independently correlated with plasma leptin levels (r ¼ 0.928, P < 0.0001). The plasma NEFA level was markedly elevated during the first 2 weeks and decreased thereafter. Conclusions: A rapid fall in leptin during the first week of LCD, coordinated by insulin, T3 and NEFA, should be beneficial for responding to decreased energy intake. Inversely, in view of the powerful effect of leptin on energy dissipation, the present findings suggest the potential usefulness of leptin in combination with caloric restriction for the treatment of obesity. Sponsorship: The Ministry of Education, Culture, Sports, Science and Technology of Japan and the Ministry of Health, Labour and Welfare of Japan. (2002) 56, 593 600. doi:10.1038=sj.ejcn.1601363 Keywords: leptin; insulin; triiodothyronine (T3); nonesterified fatty acids (NEFA); diet therapy *Correspondence: H Masuzaki, Division of Endocrinology and Metabolism, Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, 99 Brookline Avenue, Research north 345, Boston, MA, 02215, USA. E-mail: hmasuzak@caregroup.harvard.edu Guarantor: H Masuzaki. Received 6 June 2001; revised 4 October 2001; accepted 17 October 2001 Introduction Obesity is an increasingly prevalent, costly and important health problem throughout the world (Anderson & Kannel, 1992). Although novel therapeutic modalities against obesity have been expected, the low calorie diet therapy (LCD) is still one of the most effective and generally employed therapeutic regimens for weight reduction, which provides 12 20 kcal=kg of ideal body weight per day (Atkinson, 1989). Resting energy demands range from 1100 to 1500 kcal for

594 females and from 1100 to 2000 kcal for males, therefore an LCD providing 800 to 1100 kcal=day is a kind of semistarvation therapy for weight reduction (Bray, 1988; Flier & Foster, 1998; Henry & Gumbiner, 1991). Restricting intake of food leads to several nutritional changes. Serum levels of humoral factors such as insulin (Maxwell et al, 1994), nonesterified fatty acids (NEFA; Bryson et al, 1996), and triiodothyronine (T3; Gelfand & Hendler, 1989) are altered acutely and substantially by caloric restriction. Plasma norepinephrine, a sensitive marker for sympathetic nerve activity in vivo, is also elevated during the early phase of severe caloric restriction (Maxwell et al, 1994), all of which may contribute to the metabolic response to reduced energy intake. However, an adipocyte-derived hormone=cytokine, leptin, plays a crucial role in the regulation of food intake and energy expenditure (Spiegelman & Flier, 1996, Masuzaki et al, 1995; Ogawa et al, 1995, 1999; Satoh et al, 1998). Synthesis and secretion of leptin is increased in most obese humans, and the plasma leptin level is correlated tightly with the percentage of body fat (Considine et al, 1996; Maffei et al, 1995; Masuzaki et al, 1997). These observations strongly indicate the presence of leptin resistance in human obesity (Caro et al, 1996; Heek et al, 1997; Arch et al, 1998; Beaufrere & Morio, 2000). In addition to the regulation associated with adipose tissue mass, leptin production is known to be regulated by such factors as peripheral nutritional status (Kolaczynski et al, 1996a,b; Boden et al, 1996; Reseland et al, 2001), sympathetic nerve activity (Auwerx & Staels, 1998; Slieker et al, 1996; Mantzoros et al, 1996), and a series of humoral factors (Auwerx & Staels, 1998). Nevertheless, little is known about the clinical implications of leptin in terms of caloric restriction therapy in obese humans, and its relevance to other humoral factors related to long-term caloric restriction in human obesity (Dubuc et al, 1998). In this context, the present study was designed to reexamine the precise change in plasma leptin levels in relation to its potential humoral regulators such as insulin, NEFA, and T3 during 4 weeks of LCD for obese humans using multiple regression analysis for the treatment of obesity and reconsidered the potential usefulness of leptin administration during LCD. Subjects and methods Subjects and protocol of dietary management Twenty obese subjects were hospitalized to participate in the present study. Ten obese subjects (five men and five premenopausal women; age 33 13 y old) underwent 800 kcal=day LCD for the same period. Table 1 shows the profiles of subjects participating in LCD. Figure 1 demonstrates the protocol of LCD. After consuming a standard diet of 2000 kcal=day (75 g protein, 55 g fat and 320 g carbohydrate) for 5 days during an observation period, subjects undertook a 2 day 1200 kcal fixed-formula prepared food diet (63 g protein, 38 g fat and 153 g carbohydrate) for adaptation, and subsequently received 800 kcal=day LCD (47 g protein, 18 g fat and 114 g carbohydrate) for 26 days. During LCD, they were recommended to drink 1.0 1.5 l of water per day and Table 1 Profiles of patients before LCD Case Age (y) Gender BW (kg) BMI (kg=m 2 ) %Fat Leptin (ng=ml) 1 51 M 84.6 30.0 37.4 29.6 2 21 M 154.2 48.7 40.0 72.6 3 30 M 69.6 33.6 39.6 15.3 4 50 M 79.0 28.0 26.7 8.3 5 21 M 130.8 47.4 41.0 44.6 6 18 F 107.6 42.0 48.0 166.0 7 41 F 80.8 31.1 37.0 26.5 8 48 F 71.3 28.7 42.9 54.1 9 24 F 77.7 31.1 37.4 26.1 10 28 F 87.2 33.6 39.7 19.7 BW, body weight; BMI, body mass index; %fat, percentage of body fat. Figure 1 The design of LCD in the present study. From admission, the subjects consumed a normal diet of 2000 kcal=day for 5 days during an observation period, and subsequently took a 1200 kcal=day fixed-formula prepared food diet for 2 days and then received LCD (800 kcal=day) for 26 days. After LCD, calorie intake was increased to 1200 kcal=day. Blood samples were obtained to measure plasma leptin and metabolic parameters every week from day 0 to day 28.

encouraged to maintain their usual level of physical activity. After LCD for 26 days, caloric intake was increased to 1200 kcal=day. In addition, 10 obese subjects (five men and five premenopausal women; age 31 11 y old; body mass index (BMI; 32.3 2.1 kg=m 2, percentage fat (%fat), 35.1 1.8%; plasma leptin level, 14.9 3.5 ng=ml) consumed a 1400 kcal=day balanced deficit diet (BDD; 66 g protein, 38 g fat and 197 g carbohydrate) for 4 weeks. No subjects had either diabetes mellitus or any metabolic complications, and took no medications to affect adipocyte metabolism. Informed consent was obtained from each subject. This study was approved by the ethical committee on human research of Kyoto University Graduate School of Medicine. Blood and urine examination To measure plasma leptin levels, blood samples were obtained at 9:00 am after an overnight fast on days 0 (day 0isdefined as the first day of the 1200 kcal diet in LCD), 7, 14, 21 and 28. The blood samples were immediately transferred to chilled siliconized glass tubes containing Na 2 EDTA (1 mg=ml) and centrifuged at 4 C, and plasma was immediately frozen and stored at 7 20 C until assay. Plasma leptin levels were measured by the radioimmunoassay for human leptin as described previously (Hosoda et al, 1996). The sensitivity and inter- and intra-assay coefficients of variations (CV) of plasma leptin levels were 0.2 ng=ml, 5.3 and 5.9%, respectively. Serum glucose levels were measured using the hexokinase method (International Reagent, Kobe, Japan). Serum NEFA and triglyceride were measured by standard enzymatic methods (NEFA-HRII, Wako Pure Chemical Industries, Osaka, Japan, and N-assay TG-L, Nittobo, Tokyo, Japan, respectively). Insulin and T3 were measured using a radioimmunoassay kit (Dainabot, Tokyo, Japan). The sensitivity and inter- and intra-assay CV of these biochemical factors were as follows: glucose, 1 mg=dl, 0.6%, 1.2%; NEFA, 0.01 meq=l, 0.2%, 0.9%; triglyceride, 1 mg=dl, 1.6%, 0.6%; insulin, 1.00 mu=ml, 4.3%, 2.5%; T3, 10 ng=dl, 9.2%, 8.1%, respectively. Urinary ketone body was determined every day in a qualitative assay by the nitroprusside method (Multistix, Bayer-Sankyo, Tokyo, Japan). pretreatment plasma leptin levels per BMI in each group accounted for a significant component of leptin changes during the treatment. To analyze leptin levels and other parameters during LCD, one-way repeated measures ANOVA or Wilcoxon signed-ranks test were used, where applicable. To determine the independent effects of parameters for predicting the plasma leptin levels, a multiple regression model was employed. All analyses were calculated using a statistical software package (StatView 4.5 and Super ANOVA; Abacus Concepts Inc., Berkeley, CA, USA). Results Plasma leptin levels before diet therapy were correlated with BMI, both LCD (male Y ¼ 2.304X 7 52.406 (r ¼ 0.88, P ¼ 0.048), female Y ¼ 12.495X 7 365.363 (r ¼ 0.94, P ¼ 0.017)) and BDD (male Y ¼ 2.012X 7 56.885 (r ¼ 0.88 P ¼ 0.049), female Y ¼ 0.71X 7 7.491 (r ¼ 0.93 P ¼ 0.023)). In both LCD and BDD groups, differences of pretreatment plasma leptin levels were not significant after adjustment for BMI (ANCOVA, F ¼ 0.808, P ¼ 0.382). During 28 days of LCD, the mean body weight of all subjects decreased substantially by 6.9 0.9 kg (P ¼ 0.005), which corresponded to 7.1 0.5% of the initial value. Accordingly, BMI and %fat decreased markedly from 35.4 2.4 to 33.1 2.2 kg=m 2 (P ¼ 0.005), and from 39.0 1.7 to 35.9 1.7% (P ¼ 0.005), respectively. As shown in Table 2, plasma leptin levels during LCD decreased substantially from 46.2 14.6 to 13.2 3.6 ng=ml (P ¼ 0.005). Serum levels of glucose, insulin, triglyceride and T3 were also reduced significantly (P ¼ 0.008, P ¼ 0.008, P ¼ 0.011, P ¼ 0.011, respectively). Serum NEFA levels were increased but not significantly (P ¼ 0.41) at the end point. The urinary ketone body in all subjects was continuously positive during LCD. As shown in Figure 2, in both male and female subjects, the percentage decrease in plasma leptin levels during LCD was much greater than predicted for the linear regression between %fat and plasma leptin levels (Masuzaki et al, 1997), therefore, the ratio of D leptin to D %fat during the entire course was much larger than that in the BDD group (11.0 2.6 vs 0.8 0.4 (ng=ml=%), P ¼ 0.005). 595 Measurement of %fat Percentage of body fat (%fat) in all subjects was measured by the dual energy X-ray absorptiometry (DXA) method using the Hologic QDR-2000 (Hologic Inc., USA; Tataranni & Ravussin, 1995) on days 0 and 28. Statistical analysis All values were expressed as mean s.e., except age (mean s.d.). The relationship between plasma leptin levels and BMI was evaluated by Pearson s correlation test. Analysis of covariance (ANCOVA) was used to determine whether Table 2 Clinical and metabolic characteristics on LCD day 0 and day 28 Day 0 Day 28 Body weight (kg) 94.3 8.9 87.4 8.1** Body mass index (kg=m 2 ) 35.4 2.4 33.1 2.2** Percentage of body fat 39.0 1.7 35.9 1.7** Leptin (ng=ml) 46.2 14.6 13.2 3.6** Glucose (mg=dl) 98.6 6.8 78.7 3.5** Insulin (mu=ml) 14.3 2.7 10.2 2.0** Nonesterified fatty acids (meq=l) 545 71 611 55 Triglyceride (mg=dl) 147 35 92 14* Triiodothyronine (T3; ng=dl) 119 8 83 4* Values are expressed as mean s.e.; *P < 0.05; **P < 0.01.

596 Figure 2 Plasma leptin levels and %fat in males (a) and females (b) before and after LCD (closed circle) and BDD (open circle). Solid lines with arrowheads represent plasma leptin levels and %fat on day 0 and day 28 in LCD and BDD. Broken lines represent the linear regression between plasma leptin levels and %fat in 57 males (a) and 84 premenopausal females (b) with neither endocrine nor metabolic diseases (Masuzaki et al, 1997). The linear regression in males is Y ¼ 0.534 X 7 6.364 (r ¼ 0.68, P < 0.05) and in females is Y ¼ 1.505X 7 27.252 (r ¼ 0.75, P < 0.05). The inset in (b) shows the findings in case 6 in Table 1. In contrast, comparing day 0 with day 28 for BDD, mean body weight, BMI, and %fat in all subjects showed slight decreases from 88.0 9.7 to 84.9 8.9 kg (P ¼ 0.005), from 32.3 2.1 to 31.0 2.5 kg=m 2 (P ¼ 0.011), from 35.1 1.8 to 33.1 1.6% (P ¼ 0.005), respectively. Plasma leptin levels did not change significantly (14.9 3.5 to 13.4 2.8 ng=ml, P ¼ 0.083; open circle in Figure 3 (a)), nor did the relative change in the leptin=bmi ratio during BDD (4.6 4.8%, P ¼ 0.27; (open circle in Figure 3(b)). There were also no significant differences in serum levels of insulin (10.2 2.1 to 8.0 1.5 mu=ml, P ¼ 0.07), T3 (100 1 to 98 1ng=dl, P ¼ 0.52), NEFA (464 44 to 390 49 meq=l, P ¼ 0.36; open circles in Figure 3 (d) (f)), glucose (89.3 4.4 to 85.3 2.6 mg=dl, P ¼ 0.07), and triglyceride (136.5 14.0 to 120.0 8.3 mg=dl, P ¼ 0.12) between days 0 and 28. The urinary ketone body was continuously negative in all subjects during the time course during BDD. To closely analyze the exaggerated decrement in plasma leptin levels during LCD, we examined the values of plasma leptin levels and BMI measured every week in each subject. Plasma leptin levels were significantly decreased at all time points (P ¼ 0.005) compared with the initial value, furthermore, they dropped markedly during the first week (41.0 2.5% of the initial value) and then decreased mildly over the following 3 weeks (closed circle in Figure 3(a)). The leptin=bmi ratio also dropped markedly during the first week (closed circle in Figure 3(b)). The ratio of the decrease in plasma leptin levels to that of BMI (D leptin to D BMI) during the first week was significantly greater than those in the following 3 weeks (39.5 2.7 vs 29.3 2.1%, P ¼ 0.017). These findings clearly demonstrate that plasma leptin levels fall substantially during the early phase of LCD, compared with the decrease in BMI. Furthermore, as shown in Figure 3(c), D leptin to D BMI during the first week of LCD was markedly greater than those predicted from the linear regression between plasma leptin levels and BMI in BDD (male Y ¼ 2.012X 7 56.885 (r ¼ 0.88 P ¼ 0.049), female Y ¼ 0.71 X 7 7.491 (r ¼ 0.93, P ¼ 0.023; 7 16.2 3.9 vs 7 1.4 0.2 (ng=ml=kg=m 2 ), P ¼ 0.0002). This result indicates that obese subjects showing greater drop in circulating leptin could attain much lower reduction in body weight. To further elucidate the regulatory mechanism of plasma leptin levels during LCD, the relationship between leptin and its potential regulators such as insulin, T3 and NEFA was analyzed. Values measured weekly in each subject are shown in Figure 3 (d) (f). The change in insulin and T3 levels paralleled those in leptin (closed circles in Figure 3(d) and (e)). Plasma leptin levels measured weekly during LCD in each subject correlated well with insulin (r ¼ 0.586, P ¼ 0.0003; Figure 4 (a)) and T3 (r ¼ 0.785, P ¼ 0.0004; Figure 4 (b)), respectively. Multiple regression analyses revealed that serum levels of T3, after adjustment of within-subjects variation (time course) and BMI, correlated with plasma leptin levels (r ¼ 0.928, P < 0.0001). Serum NEFA levels were elevated substantially during the first 2 weeks and decreased thereafter (closed circle in Figure 3(f)). Serum levels of glucose and triglyceride in all subjects were not correlated with plasma leptin levels (r ¼ 0.101, P ¼ 0.51 and r ¼ 7 0.054, P ¼ 0.73, respectively).

597 Figure 3 Time course in plasma leptin levels (a), relative change in the leptin=bmi ratio (b), insulin (d), T3 (e) and FFA (f) during diet therapy (open circle BDD; closed circle LCD; *P < 0.05, **P < 0.01, vs the initial value). The left bar in (c) shows the decrease of leptin to BMI during the first week of LCD ((leptin at day 7) 7 (leptin before LCD)=(BMI before LCD) 7 (BMI at day 7)). The right bar in (c) represents the predicted ratio of leptin to BMI in BDD group estimated by the linear regression between BMI and plasma leptin levels before treatment of BDD (male Y ¼ 2.012X 7 56.885 (r ¼ 0.88 P ¼ 0.049), female Y ¼ 0.71X 7 7.491 (r ¼ 0.93 P ¼ 0.023)), because the data for plasma leptin and BMI at day 7 was not available. Discussion The findings of the present study demonstrated that, during an LCD, plasma leptin levels in obese humans decreased more than expected for the decline in %fat and BMI, while those in BDD were not reduced significantly. This finding suggests that, in addition to adiposity itself, factors related to energy homeostasis such as sympathetic nerve activity and humoral factors (Atkinson, 1989; Henry & Gumbiner, 1991; National Institutes of Health, 1993) are involved in the substantial fall in plasma leptin levels. Falling plasma leptin was shown to be a critical signal that initiates the neuroendocrine response to dietary energy reduction (Ahima et al, 1997). For example, an inverse relationship between plasma leptin levels and the activity of the hypothalamic pituitary adrenal axis was reported (Licinio et al, 1997; Spinedi & Gaillard, 1998), suggesting that leptin is not only an adipostatic hormone but also a stress-related factor potentially crucial for survival (Bornstein, 1997; Yanovski, 1997). Taken together, it is probable that a substantial fall in plasma leptin levels during an LCD serves as an adaptation to reduced calorie intake. The findings of the present study demonstrated that, during an LCD, plasma leptin levels were substantially decreased, especially during the first week (41.0 2.5% of the initial value in all obese subjects) and diminished gradually over the following 3 weeks. In view of the robust effect of leptin on energy dissipation (Auwerx & Staels, 1998), a rapid fall in plasma leptin levels during the early stages of an LCD should be beneficial for the adaptation to reduced energy intake. It is interesting to note that, during the early stage of an LCD, plasma leptin levels required for the maintenance of the same body weight were lowered considerably compared with those before an LCD, suggesting that leptin resistance in obese humans is ameliorated during the early stages of an LCD. A long-term caloric restriction therapy is employed as a standard therapeutic strategy for obesity, but sometimes in vain (National Institutes of Health, 1993; Wadden et al, 1989; Doucet & Tremblay, 1997). A therapeutic failure is due, at least in part, to loss of motivation to sustain behaviors required for continuing weight reduction (Wadden et al,

598 Figure 4 Relationship between plasma leptin levels and insulin (a) or T3 (b) at days 0, 7, 14, 21 and 28. Each point indicates the mean value s.e. measured weekly. 1989). Considering the potency of leptin for energy expenditure, a substantial decline in plasma leptin levels during caloric restriction may also be involved in such a failure. A recent study of long-term leptin administration to mice suggested that leptin acts to blunt the decrease in energy expenditure which typically follows reduced calorie intake (Halaas et al, 1997). Taken together, it is suggested that continuous leptin treatment during an LCD is more effective than caloric restriction alone for the treatment of obesity. A recent leptin trial with reduced-energy diet clearly demonstrated a dose response relationship between body fat loss and subcutaneous leptin injections in obese humans, suggesting that exogenous leptin administration with energy restriction is effective even in obese subjects with elevated endogenous plasma leptin levels or leptin resistance (Heymsfield et al, 1999). However, safety and efficacy of long-term leptin administration in combination with caloric restriction in humans awaits further investigations. The present study demonstrated that the change in plasma leptin levels paralleled those in insulin during an LCD. Several studies have demonstrated that the leptin gene expression is positively regulated by insulin both in vivo (Saladin et al, 1995; Kolaczynski et al, 1996c) and in vitro (Kolaczynski et al, 1996c; Leroy et al, 1996). Plasma leptin levels decrease after fasting and increase by refeeding in humans (Kolaczynski et al, 1996a,b; Boden et al, 1996; Dubuc et al, 1998; Geldszus et al, 1996; Wadden et al, 1998) and rodents (Frederich et al, 1995), which parallels the change in plasma insulin levels (Kolaczynski et al, 1996b). Accordingly, it is possible that acutely decreased insulin levels during the early phase of an LCD may contribute to a rapid fall in plasma leptin levels. The present study demonstrated that serum T3 also changed in parallel with the plasma leptin levels during an LCD. With normal thyroid function, serum T3 levels in humans are reported to fall when calorie intake decreases or energy expenditure increases (Spaulding, 1976; Chomard et al, 1988). In view of a potent effect of T3 on energy expenditure and a notion that leptin potentially stimulates hypothalamic pituitary thyroid axis (Ahima et al, 1997), a concomitant decrease in serum T3 with leptin may serve as a response to negative energy balance. To determine whether decreased T3 levels during LCD is only a reflection of low calorie intake or a kind of positive regulation for plasma leptin levels (Shintani et al, 1999) must await further investigation. Noteworthy in the present study is the finding that only serum T3 levels were independently correlated with plasma leptin levels (r ¼ 0.928, P < 0.0001) in multiple regression analyses after adjustment of the time course and BMI, raising the possibility that T3 levels during LCD may play a crucial role in regulating plasma leptin levels. However, this does not exclude the possibility that insulin and NEFA may affect plasma leptin levels (Kolaczynski et al, 1996c; Rentsch & Chiesi, 1996). The present study revealed that, during an LCD, plasma NEFA levels increased substantially during the first 2 weeks and decreased thereafter, while levels of insulin and T3 continued to decrease during the entire course in obese humans. In rodents, it has been reported that sympathetic activation, camp and b-receptor agonist down-regulate leptin gene expression and lower the plasma leptin level accompanied by an increase in lipolysis (Auwerx & Staels, 1998; Slieker et al, 1996; Mantzoros et al, 1996). In vitro studies demonstrated that NEFA inhibits leptin gene transcription in 3T3-L1 adipocytes (Rentsch & Chiesi, 1996), and that elevated intracellular NEFA induced by acyl-coa synthase inhibitor down-regulates leptin synthesis and secretion in primary cultured rat adipocytes (Shintani et al, 2000). Taken together, these findings raise the possibility that acutely elevated NEFA during the early phase of LCD contributes to the rapid fall in plasma leptin levels. In summary, the present study is the first to investigate minutely the relationship between leptin and its potential humoral regulators including insulin, T3 and NEFA. A rapid fall in plasma leptin especially during the first week of LCD, coordinated by insulin, T3 and NEFA, should be beneficial for responding to decreased energy intake. The present study, using multiple regression analyses after adjustment of the time course and BMI, suggests that T3 levels may play a crucial role in regulating plasma leptin levels. Inversely, in view of the powerful effect of leptin on energy dissipation, the present findings suggest the potential usefulness of

leptin in combination with energy restriction for the treatment of obesity. Acknowledgements We thank the clinical staff of Kyoto University Hospital and Osaka Saiseikai Nakatsu Hospital for their assistance. References Ahima RS, Prabakaran D, Mantzoros C, Qu D, Lowell B, Maratos-Flier E & Flier JS (1997): Role of leptin in the neuroendocrine response to fasting. Nature 382, 250 252. Anderson KM & Kannel WB (1992): Obesity and disease. In: Obesity, ed. Björntorp P & Brodoff BN, pp 465 473. Philadelphia, PA: JB Lippincott. Arch JR, Stock MJ & Trayhurn P (1998): Leptin resistance in obese humans: does it exist and what does it mean? Int. J. Obes. Relat. Metab. Disord. 12, 1159 1163. Atkinson, RL (1989): Low and very low calorie diets. Med. Clin. N. Am. 73, 203 215. Auwerx J & Staels B (1998): Leptin. Lancet 351, 737 741. Beaufrere B & Morio B (2000): Fat and protein redistribution with aging: metabolic considerations. Eur. J. Clin. Nutr. 54(Suppl 3), S48 S53. Boden G, Chen X, Mozzoli M & Ryan I (1996): Effect of fasting on serum leptin in normal human subjects. J. Clin. Endocrinol. Metab. 81, 3419 3423. Bornstein SR (1997): Is leptin a stress related peptide? Nature Med. 3, 937. Bray GA (1988): Treatment of obesity: an overview. Diabetes Metab. Rev. 4, 653 679. Bryson JM, King SE, Burns CM, Baur LA, Swajaj S & Caterson ID (1996): Changes in glucose and lipid metabolism following weight loss produced by a very low calorie diet in obese subjects. Int. J. Obes. Relat. Metab. Disord. 20, 338 345. Caro JF, Kolaczynski JW, Nyce MR, Ohannesian JP, Opentanova I, Goldman WH, Lynn RB, Zhang PL, Sinha MK & Considine RV (1996): Decreased cerebrospinalfluid=serum leptin ratio in obesity: a possible mechanism for leptin resistance. Lancet 348, 159 161. Chomard P, Vernhes G, Autissier N & Debry G (1988): Serum concentrations of total and free thyroid hormones in moderately obese women during a six-week slimming cure. Eur. J. Clin. Nutr. 42, 285 293 Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephens TW, Nyce MR, Ohannesian JP, Marco CC, McKee LJ, Bauer TL & Caro JF (1996): Serum immunoreactive-leptin concentrations in normalweight and obese humans. New Engl. J. Med. 334, 292 295. Doucet E & Tremblay A (1997): Food intake, energy balance and body weight control. Eur. J. Clin. Nutr. 51, 846 855 Dubuc GR, Phinney SD, Stern JS & Havel PJ (1998): Changes of serum leptin and endocrine and metabolic parameters after 7 days of energy restriction in men and women. Metabolism 47, 429 434. Flier JS & Foster DW (1998): Eating disorders: obesity, anorexia nervosa, and bulimia nervosa. In Williams Textbook of Endocrinology, 9th edn, eds, JD Wilson, DW Foster, HM Kronenberg & PR Larsen, pp 1061 1083. Philadelphia, PA: JB Lippincott. Frederich RC, Lollmann B, Hamann A, Napolitano-Rosen A, Kahn BB, Lowell BB & Flier JS (1995): Expression of ob mrna and its encoded protein in rodents. Impact of nutrition and obesity. J. Clin. Invest. 96, 1658 1663. Geldszus R, Mayr B, Horn R, Geisthovel F & Muhlen A (1996): Serum leptin and weight reduction in female obesity. Eur. J. Endocrinol. 135, 659 662. Gelfand RA & Hendler R (1989): Effect of nutrient composition on the metabolic response to very low calorie diets: learning more and more about less and less. Diabetes Metab. Rev. 5, 17 30 Halaas JL, Boozer C, Blair-West J, Fidahusein N, Denton DA & Friedman JM (1997): Physiological response to long-term peripheral and central leptin infusion in lean and obese mice. Proc. Natl Acad. Sci. USA 94, 8878 8883. Heek MV, Compton DS, France CF, Tedesco RP, Fawzi AB, Graziano MP, Sybertz EJ, Strader CD & Dabis HR Jr (1997): Diet-induced obese mice develop peripheral, but not central, resistance to leptin. J. Clin. Invest. 99, 385 390. Henry RR & Gumbiner B (1991): Benefits and limitations of very-lowcalorie diet therapy in obese NIDDM. Diabetes Care 14, 802 823. Heymsfield SB, Greenberg AS, Fujioka K, Dixon RM, Kushner R, Hunt T, Lubina JA, Patane J, Self B, Hunt P & McCamish M (1999): Recombinant leptin for weight loss in obese and lean adults. A randomized, controlled, dose-escalation trial. J.A.M.A. 282, 1568 1575. Hosoda K, Masuzaki H, Ogawa H, Miyawaki T, Hiraoka J, Hanaoka I, Yasuno A, Nomura T, Fujisawa Y, Yoshimasa Y, Nishi S, Yamori Y & Nakao K (1996): Development of radioimmunoassay for human leptin. Biochem. Biophys. Res. Commun. 221, 234 239. Kolaczynski JW, Considine V, Ohannesian J, Marco C, Opentanova I, Nyce MR, Myint M & Caro JF (1996a): Responses of leptin to shortterm fasting and refeeding in humans. Diabetes 45, 1511 1515. Kolaczynski JW, Ohannesian JP, Considine RV, Marco CC & Caro JF (1996b): Response of leptin to short-term and prolonged overfeeding in humans. J. Clin. Endocrinol. Metab. 81, 4162 4165. Kolaczynski JW, Nyce MR, Considine RV, Boden G, Nolan JJ, Henry R, Mudaliar SR, Olefsky J & Caro JF (1996c): Acute and chronic effects of insulin on leptin production in humans: studies in vivo and in vitro. Diabetes 45, 699 701. Leroy P, Dessolin S, Villageois P, Moon BC, Friedman JM, Ailhaud G & Dani C (1996): Expression of ob gene in adipose cells. Regulation by insulin. J. Biol. Chem. 271, 2365 2368. Licinio J, Mantoros C, Negrao AB, Cizza G, Wong ML, Bongiorno PB, Chrousos GP, Karp B, Allen C, Flier JS & Gold PW (1997): Human leptin levels are pulsatile and inversely related to pituitary-adrenal function. Nature Med. 3, 575 579. 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. Nature Med. 1, 1155 1161. Mantzoros CS, Frederich RC, Susulic VS, Lowell BB, Maratos-Flier E & Flier JS (1996): Activation of b-3 adrenergic receptors suppresses leptin expression and mediates a leptin-independent inhibition of food intake in mice. Diabetes 45, 909 914. Masuzaki H, Ogawa Y, Isse N, Satoh N, Okazaki T, Shigemoto M, Mori K, Tamura N, Hosoda K, Yoshimasa Y, Jingami H, Kawada T & Nakao K (1995): Human obese gene expression: adipose-specific expression and regional differences in the adipose tissue. Diabetes 44, 855 858. Masuzaki H, Ogawa Y, Hosoda K, Miyawaki T, Hanaoka I, Hiraoka J, Yasuno A, Nishimura H, Yoshimasa Y, Nishi S & Nakao K (1997): Glucocorticoid regulation of leptin synthesis and secretion in humans: elevated plasma leptin levels in Cushing s Syndrome. J. Clin. Endocrinol. Metab. 82, 2542 2547. Maxwell MH, Heber D, Waks AU & Tuck ML (1994): Role of insulin and norepinephrine in the hypertension of obesity. Am. J. Hypertens. 7, 402 408. National Institutes of Health (1993): Very low-calorie diets. J.A.M.A. 270, 967 974. Ogawa Y, Masuzaki H & Nakao K (1999): Clinical implication of leptin, the obese gene product. Intern. Med. 38, 210 212. Ogawa Y, Masuzaki H, Isse N, Okazaki T, Mori K, Shigemoto M, Satoh N, Tamura N, Hosoda K, Yoshimasa Y, Jingami H, Kawada T & Nakao K (1995): Molecular cloning of rat obese cdna and augmented gene expression in genetically obese Zucker fatty (fa=fa) rats. J. Clin. Invest. 96, 1647 1652. Rentsch J & Chiesi M (1996): Regulation of ob gene mrna levels in cultured adipocytes. F.E.B.S. Lett. 379, 55 59. 599

600 Reseland JE, Anderssen SA, Solvoll K, Hjermann I, Urdal P, Holme I & Drevon CA (2001): Effect of long-term changes in diet and exercise on plasma leptin concentrations. Am. J. Clin. Nutr. 73, 240 245 Saladin R, De Vos P, Guerre-Millo M, Leturque A, Girard J, Staels B & Auwerx J (1995): Transient increase in obese gene expression after food intake or insulin administration. Nature 377, 527 529. Satoh N, Ogawa Y, Katsuura G, Numata Y, Masuzaki H, Yoshimasa Y & Nakao K (1998): Satiety effect and sympathetic activation of leptin are mediated by hypothalamic melanocortin system. Neurosci. Lett. 249, 107 110. Shintani M, Nishimura H, Akamizu T, Yonemitsu S, Masuzaki H, Ogawa Y, Hosoda K, Inoue G, Yoshimasa Y & Nakao K (1999): Thyrotropin decreases leptin production in rat adipocytes. Metabolism 48, 1570 1574. Shintani M, Nishimura H, Yonemitsu S, Masuzaki H, Ogawa Y, Hosoda K, Inoue G, Yoshimasa Y & Nakao K (2000): Downregulation of leptin by free fatty acids in rat adipocytes: effects of triacsin C, palmitate and 2-bromopalmitate. Metabolism 49, 326 330. Slieker LJ, Sloop KW, Surface PL, Kriauciunas A, LaQuier F, Manetta J, Bue-Valleskey J & Stephens TW (1996): Regulation of expression of ob mrna and protein by glucocorticoids and camp. J. Biol. Chem. 271, 5301 5304. Spaulding SW (1976): Effect of caloric restriction and dietary composition of serum T3 and reverse T3 in man. J. Clin. Endocrinol. Metab. 42, 197 200. Spiegelman BM & Flier JS (1996): Adipogenesis and obesity: rounding out the big picture. Cell 87, 377 389. Spinedi E & Gaillard RC (1998): A regulatory loop between the hypothalamo-pituitary adrenal (HPA) axis and circulating leptin: a physiological role of ACTH. Endocrinology 139, 4016 4020. Tataranni PA & Ravussin E (1995): Use of dual-energy X-ray absorptiometry. Am. J. Clin. Nutr. 61, 274 278. Wadden TA, Sternberg JA, Letizia KA, Stunkard AJ & Foster GD (1989): Treatment of obesity by very low calorie diet, behavior therapy, and their combination: a five-year perspective. Int. J. Obes. 13, 39 46. Wadden TA, Considine RV, Foster GD, Anderson DA, Sarwer DB & Caro JS (1998): Short- and long-term changes in serum leptin dieting obese women: effects of caloric restriction and weight loss. J. Clin. Endocrinol. Metab. 83, 214 218. Yanovski JA, Yanovski SZ, Gold PW & Chrousos GP (1997): Differences in corticotropin-releasing hormone-stimulated adrenocorticotropin and cortisol before and after weight loss. J. Clin. Endocrinol. Metab. 82, 1874 1878.