- Email: [email protected]
Circulating leptin did not associate with the development of the hyperglycemia accompanied by insulin insensitivity in spontaneous noninsulin dependent diabetes mellitus model Otsuka–Long–Evans–Tokushima–Fatty rats
Regulatory Peptides 77 (1998) 141–146
Circulating leptin did not associate with the development of the hyperglycemia accompanied by insulin insensitivity in spontaneous noninsulin dependent diabetes mellitus model Otsuka–Long–Evans– Tokushima–Fatty rats Mitsuru Iida a ,c , Takashi Murakami a , Masako Sei a , Masamichi Kuwajima a , Masayo Yamada b , b a, Toshihiro Aono , Kenji Shima * a
b
Department of Laboratory Medicine, School of Medicine, The University of Tokushima, Kuramoto-cho 3 chome, Tokushima 770 -8503, Japan Department of Obstetrics and Gynecology, School of Medicine, The University of Tokushima, Kuramoto-cho 3 chome, Tokushima 770 -8503, Japan c Otsuka Pharmaceutical Co. Ltd., Kawauchi-cho, Tokushima 771 -0195, Japan Received 20 January 1998; received in revised form 10 June 1998; accepted 30 June 1998
Abstract Leptin, the product of the ob gene, has been reported to regulate feeding behavior and energy metabolism. Plasma leptin concentration was strongly correlated with body fat content in humans. It is well known that increased body fat content is accompanied by insulin insensitivity. In order to study the relationship between serum leptin level and metabolic variables, we performed caloric restriction on Otsuka–Long–Evans–Tokushima–Fatty (OLETF) rats, a model of noninsulin dependent diabetes mellitus. The male OLETF rats were allocated at random to three groups: 100% group, and 85% and 70% groups (which consumed 85% and 70% of the amount of food consumed by the 100% group, respectively). A significant correlation between serum leptin level and the body fat content, body weight, triglyceride, and fasting plasma glucose was observed. Using a partial correlation analysis to control for body fat content, however, the correlation between serum leptin and these variables disappeared. No significant changes in serum leptin levels were observed before and after a 1 h hyperinsulinemic euglycemic clamp test. In conclusion, serum leptin was significantly correlated with body fat content rather than fasting plasma glucose, serum insulin and insulin sensitivity. This suggests that circulating leptin per se may not result in hyperinsulinemia and insulin insensitivity in the OLETF rat. 1998 Elsevier Science B.V. All rights reserved. Keywords: Leptin; RIA; OLETF rat; Body fat content; Hyperglycemia; Insulin sensitivity
1. Introduction Obesity is causally associated with the development of noninsulin dependent diabetes mellitus (NIDDM). Recently, several genes responsible for obesity, ob [1,2], db [3,4], fat [5] and fa [6], have been identified in obese rodent models. The ob gene encodes leptin, a circulating protein, which is produced in adipose tissues [1,2,7], causes a reduction in food intake and regulates energy expenditure *Corresponding author. Tel.: 1 81 886 337184; fax: 1 81 886 339245; e-mail: [email protected]
[8], and the expression of the ob mRNA is related to body fat content [9,10]. Increasing body fat content is known to be accompanied by reduced insulin sensitivity [11,12] which leads to increase in insulin secretion and circulating leptin levels [7,13,14]. Acute [15] and chronic [16] administration of insulin has been reported to upregulate ob mRNA in rodents, while in humans, the short term administration of insulin failed to affect either the ob mRNA level or circulating leptin level [17]. Chronic leptin administration reduces circulating insulin in ob /ob mice, but is ineffective in normal mice [18]. Thus, the reciprocal regulation between insulin and leptin remains controver-
0167-0115 / 98 / $ – see front matter 1998 Elsevier Science B.V. All rights reserved. PII: S0167-0115( 98 )00108-6
142
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146
sial. In addition, the regulation of the circulating level and the physiological role of leptin is largely unknown in rodents. When body weight is lost in obese subjects, serum insulin and glucose levels are decreased as a result of amelioration of insulin resistance in parallel with circulating leptin levels. Therefore, it may be postulated that leptin might be causally associated with the development of insulin resistance. In a recent report, leptin attenuates some insulin-induced signals in hepatic cell lines [19], suggesting the possible role of leptin in obesity-associated insulin resistance. In the Otsuka–Long–Evans– Tokushima–Fatty (OLETF) rat, a model of NIDDM, obesity is usually associated with insulin resistance and hyperinsulinemia and the prevalence of diabetes mellitus increases with body weight gain and fat deposition [20,21]. Therefore, this animal model represents a way to study the relationship between circulating leptin and metabolic variables, which seems to be closely related to the development of obesity and diabetes, without the complicating factors of differences in genetic background and environmental conditions which affect similar studies in humans. In order to clarify the possibility that circulating leptin is involved in the development of diabetes, we examined the relationships between circulating leptin level and metabolic variables in OLETF rats whose diets were restricted to various extents.
2. Materials and methods
2.1. Animals and experimental design A spontaneously diabetic strain of male OLETF [20,21] rats, 5 weeks of age, were obtained from the Tokushima Research Institute (Otsuka Pharmaceutical, Tokushima, Japan) and were maintained until 22 weeks of age in our animal facilities under specific pathogen-free conditions at controlled temperature (21628C), humidity (5565%) and lighting (7:00–19:00) with air conditioning and supplied with standard rat chow with a caloric composition of 27% protein, 59% carbohydrate and 14% fat (Oriental Yeast, Tokyo, Japan) and tap water ad libitum until the end of the experiments. All animals were randomly assigned to three groups: 100% group (n 5 12) and 85% (n 5 6) and 70% (n 5 6) groups (which consumed 85% and 70% of the amount of food consumed by the 100% group, respectively). The dietary intervention continued for 18 weeks from 5 to 22 weeks of age.
2.2. Measurement of in vivo glucose disposal by euglycemic clamp test We measured insulin-mediated whole body glucose uptake in rats at 22 weeks of age by a euglycemic clamp technique as described previously [21]. In brief, after overnight fast, rats were infused insulin at a rate of 70
pmol / kg / min for 60 min, while maintaining a plasma glucose level at approximately 6.1 mM by variable infusion of a 100 g / l glucose solution. Insulin sensitivity was represented as a glucose infusion rate (GIR) during the last 20 min (mmol glucose / kg / min). Before the euglycemic clamp test, blood (1.0 ml) was withdrawn from the jugular vein for determination of serum insulin, glucose, total cholesterol and triglyceride concentrations. For determination of serum leptin concentration the blood (1.0 ml) was collected after the euglycemic clamp test, to avoid the effect of excessive blood loss on the clamp test.
2.3. Acute effect of insulin on serum leptin concentration To verify the effect of short term hyperinsulinemia on serum leptin concentration, we measured the serum leptin concentration before and after the euglycemic clamp test in the randomly chosen six male OLETF rats of the 100% fed group at 22 weeks of age. (The data obtained from this group were not used for the correlation study.)
2.4. Radioimmunoassay ( RIA) for rat leptin The recombinant rat leptin was produced in Escherichia coli using QIA expressionist TM (Qiagen, Hilden, Germany) in forms of NH 2 -terminal fusion to the His-Tag sequence, and purified and refolded from inclusion bodies according to the manufacturer’s protocols. Specific antisera to rat leptin were obtained from rabbits after eight biweekly injections of 100 mg recombinant rat leptin (the first immunization with Freund’s complete adjutant followed by Freund’s incomplete adjuvant) into the femoral hypodermis. The leptin was radiolabeled with iodine 125 ( 125 I) using Iodo-Gen (Pierce, IL, USA) as the iodination reagent. The labeled leptin was purified by gel filtration chromatography on a 1.0 3 10.0 cm column of Sephadex G-25. The specific activity of this 125 I labeled rat leptin was about 20 mCi / mg. One hundred ml of standard or test serum were incubated with 200 ml of phosphate buffered saline (pH 7.2) containing, per liter, 10 g bovine serum albumin, 1 g of Tween 20, 5 mM of EDTA-2Na, 100 000 U of aprotinin and 0.5 g of sodium azaide, and with 50 ml of antisera (at a dilution of 1:1000) for 24 h at 48C. 125 I labeled rat leptin (about 15 000 cpm in 50 ml) was then added and the incubation continued for an additional 24 h. The antibody–antigen complex was separated by the double antibody method reported previously [22]. In the absence of unlabeled leptin the antisera in this assay precipitated about 30% of the added labeled leptin. Recombinant human leptin was prepared from E. coli with human leptin cDNA using a procedure similar to that described above [22]. Recombinant mouse leptin was obtained from Alpha Diagnostic (TX, USA).
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146
2.4.1. Assay precision The within and between-assay coefficient of variation by measurements of three different rat serum samples were 5.0–7.4% and 5.6–10.9%, respectively. 2.4.2. Analytical recovery and dilution We prepared two sets of serum samples with and without the addition of various amounts of recombinant rat leptin. The recovery against expected leptin concentration ranged from 97.1% to 101.4% over a range of 5.98 to 40.4 ng / ml. Three different rat sera (initial leptin concentrations 5.3, 16.3, 53.4 ng / ml) were serially diluted two- and fourfold (a typical dilution pattern is shown in Fig. 1, n). Recovery against calculated leptin concentrations ranged from 101.3 to 114.2%. A typical standard curve of RIA for rat leptin is shown in Fig. 1 (d). Under these conditions, the lowest detectable concentration of rat leptin was 0.5 ng / ml, as estimated from the concentration that produced a response of 2S.D. from the zero dose response. The antiserum used in this RIA system showed low crossreactivity with recombinant mouse leptin, which was only 1% that of rat leptin when evaluated at the 50% B /Bo point (Fig. 1, s). It scarcely crossreacts with recombinant human leptin (Fig. 1, h). The serial dilution curve of rat serum was parallel to the standard curve in the RIA (Fig. 1, n). 2.5. Other analytical methods Plasma glucose values were determined by the glucose oxidase method (Fuji Dry Chem 2000; Fuji Medical System, Tokyo, Japan). Insulin was measured with a
Fig. 1. Crossreactivity of rat leptin RIA. (d), Recombinant rat leptin (typical standard curve); (s), recombinant mouse leptin; (h), recombinant human leptin; (n), a typical dilution pattern of rat serum, serially dilution two-, four-, eightfold (indicated horizontal upper axis) with phosphate buffered saline described in Section 2.4.2. B: count in bound form in the presence of cold leptin. Bo: count in bound form in the absence of cold leptin.
143
commercial kit (Eiken Chemical, Tokyo, Japan) using rat insulin (Novo Nordisk, Denmark) as a standard. Total cholesterol and triglyceride were measured using autoanalyzer type-736 (Hitachi, Tokyo, Japan). Body fat content was measured with EM-SCAN model SA-2 (EMScan, IL, USA) which is a noninvasive analysis device using electromagnetic scanning technology for measurement of the body composition of small animals, and the values are presented as the percentages of body weight.
2.6. Statistics Log-transformation of serum leptin and triglyceride concentration were performed to normalize the distribution for subsequent analysis. Quantitative variables were reported as mean6S.D. The statistical significance of difference was evaluated using the paired Student’s t-test and Scheffe’s method for multiple comparison, if the three groups were evaluated by analysis of variance. All tests were two-tailed and the level of significance was set at P , 0.05. Pearson’s correlation and partial correlation analysis were performed using the software package SPSS 6.1J for Macintosh.
3. Results The serum leptin levels in the 100%, 85% and 70% groups were 30.262.7 ng / ml, 28.067.5 ng / ml and 17.065.2 ng / ml, respectively (Table 1). The value in the 70% group was significantly lower than those in the others. Similar trends were observed in cases of body weight and body fat content among the three groups (Table 1, Fig. 2a,b,c). Insulin-stimulated glucose disposal in vivo, presented as GIR, improved with decrease in body weight and body fat content. The GIR in the 70% group was twice that in the 100%, and the GIR in the 85% group was intermediate between those in the 70% and 100% groups (Fig. 2d). However, serum insulin levels and fasting plasma glucose (FPG) were not significantly different among the three groups. Table 2 also shows correlations between leptin and several anthropometric and metabolic variables. There were statistically significant correlations between leptin level and body weight (r 5 0.626, P 5 0.005), body fat content (r 5 0.728, P 5 0.001), serum triglyceride (r 5 0.614, P 5 0.007) and FPG (r 5 0.515, P 5 0.029). The highest correlation was observed between serum leptin and body fat content. Cholesterol and insulin levels correlated positively and GIR negatively with the serum leptin level, but the correlation coefficients did not reach a statistically significant level. The significance of the correlation to leptin was abolished when the correlations were adjusted for body fat content as a confounding factor by partial correlation analysis. Thus, these significant correlations to leptin were due to dependence on differences in the amounts of body fat content. An acute insulin administra-
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146
144 Table 1 Characteristics of three groups of OLETF rat
BW (g) Body fat (%) GIRc FPG (mmol / l) Insulin (pmol / l) TG (mmol / l) T-cho (mmol / l) Leptin (ng / ml)
100% (n 5 6)
85% (n 5 6)
70% (n 5 6)
598623.3 15.462.12 36.1611.1 6.4961.97 4136135 1.1560.61 2.3160.31 30.262.7
515618.1 a 15.061.54 63.2614.3 5.6860.87 3906107 0.6260.29 2.1860.15 28.067.5
431613.4 a,b 10.763.62 a,b 76.4629.9 a 4.8061.22 3956147 0.2960.15 a 2.1960.13 17.065.2 a,b
Statistical significance was evaluated by Scheffe’s method. a P , 0.05 compared to 100% group, b P , 0.05 compared to 85% group. c Values are presented as mmol glucose / kg / min. BW, body weight at 22 weeks of age; FPG, fasting plasma glucose; TG, triglyceride; T-cho, total cholesterol.
Fig. 2. Effects of caloric restriction on body weight (a), body fat content (b), serum leptin concentration (c) and GIR (d). Statistical significance was evaluated by Scheffe’s method for multiple comparison and data are presented as mean6S.D. ** P , 0.01; NS, not significant.
Table 2 Correlation coefficients and partial correlation coefficients between variables and serum leptin concentration Variables
Correlation coefficient (n 5 18)
Partial correlation coefficient controlled for body fat content (n 5 18)
BW (g) Body fat (%) GIRa FPG (mmol / l) Insulin (pmol / l) TG (mmol / l) T-cho (mmol / l)
0.626 0.728 2 0.378 0.515 0.367 0.614 0.366
0.356 – 2 0.225 0.458 0.037 0.474 0.356
(0.005) (0.001) (0.149) (0.029) (0.148) (0.007) (0.135)
(0.161) (0.419) (0.065) (0.892) (0.055) (0.161)
Correlation and partial correlation coefficient were calculated using the values together with three groups (n 5 18) against the log-transformed leptin concentration; data are presented as r (P-value). a Values are presented as mmol glucose / kg / min. BW, body weight at 22 weeks of age; FPG, fasting plasma glucose; TG, triglyceride; T-cho, total cholesterol.
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146
Fig. 3. Serum leptin concentration before and after a hyperinsulinemic euglycemic clamp test. Statistical significance was evaluated by paired Student’s t-test using log-transformed values and data are presented as mean6S.D. NS, not significant.
tion had no effect on serum leptin levels as shown in Fig. 3 (28.064.8 ng / ml before vs. 27.364.2 ng / ml after a 1 h hyperinsulinemic clamp).
4. Discussion In this study, we developed a RIA for determining circulating rat leptin concentrations, which performed adequately in evaluating circulating leptin levels from the standpoint of both accuracy and sensitivity. The relative crossreactivities of the antiserum used in this RIA system to mouse and human leptin were very low compared with that to rat leptin. Judging from amino acid homology between the leptin from these species, it is conceivable that the epitopes of this antiserum are located in the peptide including amino acid at either the 25th, the 54th, the 99th and the 123rd position (or combination thereof) in the leptin molecule, since these amino acids are identical between mouse and human leptin but different from those of rat leptin [2]. OLETF rats are reported to manifest hyperinsulinemia with insulin resistance in accordance with an increase in body fat deposition, and these metabolic derangements ameliorate in parallel with weight loss by either caloric restriction or exercise [21,23]. The caloric restriction study makes it possible to investigate the animals in different physical and metabolic states, and also unnecessary to use a large number of animals to get satisfactory correlation results. Indeed, we observed that five of the six rats in the free eating group (100% group), one of the six in the 85% group (15% food restricted) and none of the six in the 70% group (30% food restricted) were diabetic, as judged from the results of an oral glucose tolerance test performed at 20 weeks of age (data not shown). Thus, this model animal is suitable for investigating the relationship between circulat-
145
ing leptin and metabolic derangements related to the development of diabetes. It is well known that obesity causes hyperleptinemia [7,13,14] and is accompanied by insulin insensitivity [11,12]. The latter is a stimulus for insulin secretion, resulting in hyperinsulinemia in cases where the pancreatic b-cells are capable of responding to this stimulus. Furthermore, leptin may cause insulin resistance through modulating insulin signal transduction, such as the attenuation of tyrosine phosphorylation of the insulin receptor substrate-1 [19]. Leptin receptors are also expressed in pancreatic b-cells and leptin modulates the secretion of insulin from the isolated islet [24,25]. However, the fact that one modulates the secretion of the other remains controversial. Our study showed that circulating leptin levels closely mirror body fat content in OLETF rats and that serum leptin levels were not altered before and after hyperinsulinemic euglycemic clamp, as described in humans [13]. Furthermore, the partial correlation analysis revealed that there were no significant correlations between circulating leptin levels and various metabolic variables, including insulin and insulin sensitivity, after controlling for body fat content, suggesting that leptin per se did not associate with the development of the hyperglycemia with insulin insensitivity in the OLETF rats. In our study, serum leptin levels were reduced in the 70% food consumption group whose body fat content was also reduced, but the 85% chronic food consumption group did not result in a significant reduction of body fat content or plasma leptin level compared to the free eating group. These results suggest that chronic food restriction per se has no direct effect on plasma leptin levels but reduces them indirectly through its effect on decrease in body fat content, although acute restriction of food intake has been reported to reduce plasma leptin levels [13,14,26]. In any case, as reported by Collins and Surwit [7], there may not be an obligate relationship between plasma leptin levels and food intake.
Acknowledgements We thank Ms. Noriko Okauchi for her excellent work in maintaining the OLETF rats and for valuable advice during this study.
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;372:425–32. [2] Murakami T, Shima K. Cloning of rat obese cDNA and its expression in obese rats. Biochem Biophy Res Commun 1995;209:944–52. [3] Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards JG, Campfield LA, Clark FT, Deeds J, Muir C, Sanker S,
146
[4]
[5]
[6]
[7] [8]
[9]
[10]
[11]
[12]
[13]
[14]
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146 Moriarty A, Moore KJ, Smutko JS, Mays GG, WoolF EA, Monroe CA, Tepper RI. Identification and expression cloning of a leptin receptor (OB-R). Cell 1995;83:1236–71. Chen H, Charlat O, Tartaglia LA, Woolf EA, Weng X, Ellis SJ, Lakey ND, Culpepper J, Moore KJ, Breitbart RF, Duyk GM, Tepper RI, Morgenstern JP. Evidence that the diabetes gene encodes the leptin receptor: Identification of a mutation in the leptin receptor gene in db /db mice. Cell 1996;84:491–5. Naggert J, Fricker L, Varlamov D, Nishina P, Roaille Y, Steiner D, Carrol R, Paigen B, Leiter E. Hyperproinsulinemia in obese fat /fat mice associated with carboxypeptidase E mutation which reduce enzyme activity. Nat Genet 1995;10:134–42. Iida M, Murakami T, Ishida K, Mizuno A, Kuwajima M, Shima K. Phenotype-linked amino acid alteration in leptin receptor cDNA from Zucker fatty ( fa /fa) rat. Biochem Biophy Res Commun 1996;222:19–26. Collins S, Surwit RS. Pharmacological manipulation of ob expression in a dietary model of obesity. J Biol Chem 1996;271:9437–40. Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Robinowitz D, Lallone RL, Burley SK, Friedman JM. Weight-reducing effect of the plasma protein encoded by the obese gene. Science 1995;269:543– 6. Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R, Ranganathan S, Kern PA, Friedman JM. Leptin levels in human and rodent: measurement of plasma leptin and ob mRNA in obese and weight-reduced subjects. Nat Med 1995;1:1156–61. ¨ Frederich RC, Hamman A, Andersson S, Lollmann B, Lowell BB, Flier JS. Leptin levels reflect body lipid content in mice: evidence for diet-induced resistance to leptin action. Nat Med 1995;1:1311–4. Bergman RN, Phillips LS, Cobelli C. Physiologic evaluation of factors controlling glucose tolerance in man: measurement of insulin sensitivity and b-cell glucose sensitivity from the response to intravenous glucose. J Clin Invest 1981;68:1456–67. Zunigo-Guajardo S, Jimenez J, Angel A, Zinman B. Effect of massive obesity on insulin sensitivity and insulin clearance and metabolic response to insulin as assessed by the euglycemic clamp technique. Metabolism 1986;35:278–82. ¨ Frederich RC, Lollmann B, Hamann A, Napolitano-Rosen A, Kahn BB, Lowell BB, Flier JS. Expression of ob mRNA and its encoded protein in rodents: impact of nutrition and obesity. J Clin Invest 1995;96:1658–63. Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephans TW, Nyce MR, Ohannesian JP, Marco CC, Mckee LJ, Bauer TL, Caro JF. Serum immunoreactive-leptin concentrations in normalweight and obese humans. New Engl J Med 1996;334:292–5.
[15] Cusin I, Sainsbury A, Doyle P, Rohner-Jeanrenaud F, Jeanrenand B. The ob gene and insulin: a relationship leading to clues to the understanding of obesity. Diabetes 1995;44:1467–70. [16] Saladin R, De Vos P, Guerre-Millo M, Leturque A, Girard J, Stales B, Auwerx J. Transient increase in obese gene expression after food intake or insulin administration. Nature 1995;377:527–9. [17] Kolaczynski JW, Nyce MR, Considine RV, Boden G, Nolan JJ, Henry R, Mudaliar SR, Olefsky J, Caro JF. Acute and chronic effect of insulin on leptin production in humans: studies in vivo and vitro. Diabetes 1996;45:699–701. [18] 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;269:540–3. [19] Choen B, Novick D, Rubinstein M. Modulation of insulin activities by leptin. Science 1996;274:1185–8. [20] Kawano K, Hirashima T, Mori S, Saitho Y, Kurosumi M, Natori T. Spontaneous long-term hyperglycemic rat with diabetic complications: Otsuka Long–Evans Tokushima Fatty (OLETF) strain. Diabetes 1992;41:1422–8. [21] Ishida K, Mizuno A, Zhu M, Sano T, Shima K. Which is the primary etiologic event in Otsuka Long–Evans Tokushima Fatty rats, a model of spontaneous, non-insulin-dependent diabetes mellitus, insulin resistance or impaired insulin secretion? Metabolism 1995;44:940–5. [22] Iida M, Murakami T, Yamada M, Sei M, Kuwajima M, Mizuno A, Noma Y, Aono T, Shima K. Hyperleptinemia in chronic renal failure. Horm Metab Res 1996;28:724–7. [23] Okauchi N, Mizuno A, Zhu M, Ishida K, Sano T, Noma Y, Shima K. Effect of obesity and inheritance on the development of non-insulindependent diabetes mellitus in Otsuka Long–Evans Tokushima Fatty rats. Diabetes Res Clin Pract 1995;29:1–10. [24] Emilsson V, Lin YL, Cawthorne MA, Morton NM, Davenport M. Expression of the functional leptin receptor mRNA in pancreatic islets and direct inhibitory action of leptin on insulin secretion. Diabetes 1997;46:313–6. [25] Ishida K, Murakami T, Mizuno A, Iida M, Kuwajima M, Shima K. Leptin suppresses insulin secretion from rat pancreatic islet. Regul Pept 1997;70:179–82. [26] Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R, Ranganathan S, Kern PA, Friedman JM. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med 1995;1:1155–61.
Circulating leptin did not associate with the development of the hyperglycemia accompanied by insulin insensitivity in spontaneous noninsulin dependent diabetes mellitus model Otsuka–Long–Evans– Tokushima–Fatty rats Mitsuru Iida a ,c , Takashi Murakami a , Masako Sei a , Masamichi Kuwajima a , Masayo Yamada b , b a, Toshihiro Aono , Kenji Shima * a
b
Department of Laboratory Medicine, School of Medicine, The University of Tokushima, Kuramoto-cho 3 chome, Tokushima 770 -8503, Japan Department of Obstetrics and Gynecology, School of Medicine, The University of Tokushima, Kuramoto-cho 3 chome, Tokushima 770 -8503, Japan c Otsuka Pharmaceutical Co. Ltd., Kawauchi-cho, Tokushima 771 -0195, Japan Received 20 January 1998; received in revised form 10 June 1998; accepted 30 June 1998
Abstract Leptin, the product of the ob gene, has been reported to regulate feeding behavior and energy metabolism. Plasma leptin concentration was strongly correlated with body fat content in humans. It is well known that increased body fat content is accompanied by insulin insensitivity. In order to study the relationship between serum leptin level and metabolic variables, we performed caloric restriction on Otsuka–Long–Evans–Tokushima–Fatty (OLETF) rats, a model of noninsulin dependent diabetes mellitus. The male OLETF rats were allocated at random to three groups: 100% group, and 85% and 70% groups (which consumed 85% and 70% of the amount of food consumed by the 100% group, respectively). A significant correlation between serum leptin level and the body fat content, body weight, triglyceride, and fasting plasma glucose was observed. Using a partial correlation analysis to control for body fat content, however, the correlation between serum leptin and these variables disappeared. No significant changes in serum leptin levels were observed before and after a 1 h hyperinsulinemic euglycemic clamp test. In conclusion, serum leptin was significantly correlated with body fat content rather than fasting plasma glucose, serum insulin and insulin sensitivity. This suggests that circulating leptin per se may not result in hyperinsulinemia and insulin insensitivity in the OLETF rat. 1998 Elsevier Science B.V. All rights reserved. Keywords: Leptin; RIA; OLETF rat; Body fat content; Hyperglycemia; Insulin sensitivity
1. Introduction Obesity is causally associated with the development of noninsulin dependent diabetes mellitus (NIDDM). Recently, several genes responsible for obesity, ob [1,2], db [3,4], fat [5] and fa [6], have been identified in obese rodent models. The ob gene encodes leptin, a circulating protein, which is produced in adipose tissues [1,2,7], causes a reduction in food intake and regulates energy expenditure *Corresponding author. Tel.: 1 81 886 337184; fax: 1 81 886 339245; e-mail: [email protected]
[8], and the expression of the ob mRNA is related to body fat content [9,10]. Increasing body fat content is known to be accompanied by reduced insulin sensitivity [11,12] which leads to increase in insulin secretion and circulating leptin levels [7,13,14]. Acute [15] and chronic [16] administration of insulin has been reported to upregulate ob mRNA in rodents, while in humans, the short term administration of insulin failed to affect either the ob mRNA level or circulating leptin level [17]. Chronic leptin administration reduces circulating insulin in ob /ob mice, but is ineffective in normal mice [18]. Thus, the reciprocal regulation between insulin and leptin remains controver-
0167-0115 / 98 / $ – see front matter 1998 Elsevier Science B.V. All rights reserved. PII: S0167-0115( 98 )00108-6
142
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146
sial. In addition, the regulation of the circulating level and the physiological role of leptin is largely unknown in rodents. When body weight is lost in obese subjects, serum insulin and glucose levels are decreased as a result of amelioration of insulin resistance in parallel with circulating leptin levels. Therefore, it may be postulated that leptin might be causally associated with the development of insulin resistance. In a recent report, leptin attenuates some insulin-induced signals in hepatic cell lines [19], suggesting the possible role of leptin in obesity-associated insulin resistance. In the Otsuka–Long–Evans– Tokushima–Fatty (OLETF) rat, a model of NIDDM, obesity is usually associated with insulin resistance and hyperinsulinemia and the prevalence of diabetes mellitus increases with body weight gain and fat deposition [20,21]. Therefore, this animal model represents a way to study the relationship between circulating leptin and metabolic variables, which seems to be closely related to the development of obesity and diabetes, without the complicating factors of differences in genetic background and environmental conditions which affect similar studies in humans. In order to clarify the possibility that circulating leptin is involved in the development of diabetes, we examined the relationships between circulating leptin level and metabolic variables in OLETF rats whose diets were restricted to various extents.
2. Materials and methods
2.1. Animals and experimental design A spontaneously diabetic strain of male OLETF [20,21] rats, 5 weeks of age, were obtained from the Tokushima Research Institute (Otsuka Pharmaceutical, Tokushima, Japan) and were maintained until 22 weeks of age in our animal facilities under specific pathogen-free conditions at controlled temperature (21628C), humidity (5565%) and lighting (7:00–19:00) with air conditioning and supplied with standard rat chow with a caloric composition of 27% protein, 59% carbohydrate and 14% fat (Oriental Yeast, Tokyo, Japan) and tap water ad libitum until the end of the experiments. All animals were randomly assigned to three groups: 100% group (n 5 12) and 85% (n 5 6) and 70% (n 5 6) groups (which consumed 85% and 70% of the amount of food consumed by the 100% group, respectively). The dietary intervention continued for 18 weeks from 5 to 22 weeks of age.
2.2. Measurement of in vivo glucose disposal by euglycemic clamp test We measured insulin-mediated whole body glucose uptake in rats at 22 weeks of age by a euglycemic clamp technique as described previously [21]. In brief, after overnight fast, rats were infused insulin at a rate of 70
pmol / kg / min for 60 min, while maintaining a plasma glucose level at approximately 6.1 mM by variable infusion of a 100 g / l glucose solution. Insulin sensitivity was represented as a glucose infusion rate (GIR) during the last 20 min (mmol glucose / kg / min). Before the euglycemic clamp test, blood (1.0 ml) was withdrawn from the jugular vein for determination of serum insulin, glucose, total cholesterol and triglyceride concentrations. For determination of serum leptin concentration the blood (1.0 ml) was collected after the euglycemic clamp test, to avoid the effect of excessive blood loss on the clamp test.
2.3. Acute effect of insulin on serum leptin concentration To verify the effect of short term hyperinsulinemia on serum leptin concentration, we measured the serum leptin concentration before and after the euglycemic clamp test in the randomly chosen six male OLETF rats of the 100% fed group at 22 weeks of age. (The data obtained from this group were not used for the correlation study.)
2.4. Radioimmunoassay ( RIA) for rat leptin The recombinant rat leptin was produced in Escherichia coli using QIA expressionist TM (Qiagen, Hilden, Germany) in forms of NH 2 -terminal fusion to the His-Tag sequence, and purified and refolded from inclusion bodies according to the manufacturer’s protocols. Specific antisera to rat leptin were obtained from rabbits after eight biweekly injections of 100 mg recombinant rat leptin (the first immunization with Freund’s complete adjutant followed by Freund’s incomplete adjuvant) into the femoral hypodermis. The leptin was radiolabeled with iodine 125 ( 125 I) using Iodo-Gen (Pierce, IL, USA) as the iodination reagent. The labeled leptin was purified by gel filtration chromatography on a 1.0 3 10.0 cm column of Sephadex G-25. The specific activity of this 125 I labeled rat leptin was about 20 mCi / mg. One hundred ml of standard or test serum were incubated with 200 ml of phosphate buffered saline (pH 7.2) containing, per liter, 10 g bovine serum albumin, 1 g of Tween 20, 5 mM of EDTA-2Na, 100 000 U of aprotinin and 0.5 g of sodium azaide, and with 50 ml of antisera (at a dilution of 1:1000) for 24 h at 48C. 125 I labeled rat leptin (about 15 000 cpm in 50 ml) was then added and the incubation continued for an additional 24 h. The antibody–antigen complex was separated by the double antibody method reported previously [22]. In the absence of unlabeled leptin the antisera in this assay precipitated about 30% of the added labeled leptin. Recombinant human leptin was prepared from E. coli with human leptin cDNA using a procedure similar to that described above [22]. Recombinant mouse leptin was obtained from Alpha Diagnostic (TX, USA).
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146
2.4.1. Assay precision The within and between-assay coefficient of variation by measurements of three different rat serum samples were 5.0–7.4% and 5.6–10.9%, respectively. 2.4.2. Analytical recovery and dilution We prepared two sets of serum samples with and without the addition of various amounts of recombinant rat leptin. The recovery against expected leptin concentration ranged from 97.1% to 101.4% over a range of 5.98 to 40.4 ng / ml. Three different rat sera (initial leptin concentrations 5.3, 16.3, 53.4 ng / ml) were serially diluted two- and fourfold (a typical dilution pattern is shown in Fig. 1, n). Recovery against calculated leptin concentrations ranged from 101.3 to 114.2%. A typical standard curve of RIA for rat leptin is shown in Fig. 1 (d). Under these conditions, the lowest detectable concentration of rat leptin was 0.5 ng / ml, as estimated from the concentration that produced a response of 2S.D. from the zero dose response. The antiserum used in this RIA system showed low crossreactivity with recombinant mouse leptin, which was only 1% that of rat leptin when evaluated at the 50% B /Bo point (Fig. 1, s). It scarcely crossreacts with recombinant human leptin (Fig. 1, h). The serial dilution curve of rat serum was parallel to the standard curve in the RIA (Fig. 1, n). 2.5. Other analytical methods Plasma glucose values were determined by the glucose oxidase method (Fuji Dry Chem 2000; Fuji Medical System, Tokyo, Japan). Insulin was measured with a
Fig. 1. Crossreactivity of rat leptin RIA. (d), Recombinant rat leptin (typical standard curve); (s), recombinant mouse leptin; (h), recombinant human leptin; (n), a typical dilution pattern of rat serum, serially dilution two-, four-, eightfold (indicated horizontal upper axis) with phosphate buffered saline described in Section 2.4.2. B: count in bound form in the presence of cold leptin. Bo: count in bound form in the absence of cold leptin.
143
commercial kit (Eiken Chemical, Tokyo, Japan) using rat insulin (Novo Nordisk, Denmark) as a standard. Total cholesterol and triglyceride were measured using autoanalyzer type-736 (Hitachi, Tokyo, Japan). Body fat content was measured with EM-SCAN model SA-2 (EMScan, IL, USA) which is a noninvasive analysis device using electromagnetic scanning technology for measurement of the body composition of small animals, and the values are presented as the percentages of body weight.
2.6. Statistics Log-transformation of serum leptin and triglyceride concentration were performed to normalize the distribution for subsequent analysis. Quantitative variables were reported as mean6S.D. The statistical significance of difference was evaluated using the paired Student’s t-test and Scheffe’s method for multiple comparison, if the three groups were evaluated by analysis of variance. All tests were two-tailed and the level of significance was set at P , 0.05. Pearson’s correlation and partial correlation analysis were performed using the software package SPSS 6.1J for Macintosh.
3. Results The serum leptin levels in the 100%, 85% and 70% groups were 30.262.7 ng / ml, 28.067.5 ng / ml and 17.065.2 ng / ml, respectively (Table 1). The value in the 70% group was significantly lower than those in the others. Similar trends were observed in cases of body weight and body fat content among the three groups (Table 1, Fig. 2a,b,c). Insulin-stimulated glucose disposal in vivo, presented as GIR, improved with decrease in body weight and body fat content. The GIR in the 70% group was twice that in the 100%, and the GIR in the 85% group was intermediate between those in the 70% and 100% groups (Fig. 2d). However, serum insulin levels and fasting plasma glucose (FPG) were not significantly different among the three groups. Table 2 also shows correlations between leptin and several anthropometric and metabolic variables. There were statistically significant correlations between leptin level and body weight (r 5 0.626, P 5 0.005), body fat content (r 5 0.728, P 5 0.001), serum triglyceride (r 5 0.614, P 5 0.007) and FPG (r 5 0.515, P 5 0.029). The highest correlation was observed between serum leptin and body fat content. Cholesterol and insulin levels correlated positively and GIR negatively with the serum leptin level, but the correlation coefficients did not reach a statistically significant level. The significance of the correlation to leptin was abolished when the correlations were adjusted for body fat content as a confounding factor by partial correlation analysis. Thus, these significant correlations to leptin were due to dependence on differences in the amounts of body fat content. An acute insulin administra-
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146
144 Table 1 Characteristics of three groups of OLETF rat
BW (g) Body fat (%) GIRc FPG (mmol / l) Insulin (pmol / l) TG (mmol / l) T-cho (mmol / l) Leptin (ng / ml)
100% (n 5 6)
85% (n 5 6)
70% (n 5 6)
598623.3 15.462.12 36.1611.1 6.4961.97 4136135 1.1560.61 2.3160.31 30.262.7
515618.1 a 15.061.54 63.2614.3 5.6860.87 3906107 0.6260.29 2.1860.15 28.067.5
431613.4 a,b 10.763.62 a,b 76.4629.9 a 4.8061.22 3956147 0.2960.15 a 2.1960.13 17.065.2 a,b
Statistical significance was evaluated by Scheffe’s method. a P , 0.05 compared to 100% group, b P , 0.05 compared to 85% group. c Values are presented as mmol glucose / kg / min. BW, body weight at 22 weeks of age; FPG, fasting plasma glucose; TG, triglyceride; T-cho, total cholesterol.
Fig. 2. Effects of caloric restriction on body weight (a), body fat content (b), serum leptin concentration (c) and GIR (d). Statistical significance was evaluated by Scheffe’s method for multiple comparison and data are presented as mean6S.D. ** P , 0.01; NS, not significant.
Table 2 Correlation coefficients and partial correlation coefficients between variables and serum leptin concentration Variables
Correlation coefficient (n 5 18)
Partial correlation coefficient controlled for body fat content (n 5 18)
BW (g) Body fat (%) GIRa FPG (mmol / l) Insulin (pmol / l) TG (mmol / l) T-cho (mmol / l)
0.626 0.728 2 0.378 0.515 0.367 0.614 0.366
0.356 – 2 0.225 0.458 0.037 0.474 0.356
(0.005) (0.001) (0.149) (0.029) (0.148) (0.007) (0.135)
(0.161) (0.419) (0.065) (0.892) (0.055) (0.161)
Correlation and partial correlation coefficient were calculated using the values together with three groups (n 5 18) against the log-transformed leptin concentration; data are presented as r (P-value). a Values are presented as mmol glucose / kg / min. BW, body weight at 22 weeks of age; FPG, fasting plasma glucose; TG, triglyceride; T-cho, total cholesterol.
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146
Fig. 3. Serum leptin concentration before and after a hyperinsulinemic euglycemic clamp test. Statistical significance was evaluated by paired Student’s t-test using log-transformed values and data are presented as mean6S.D. NS, not significant.
tion had no effect on serum leptin levels as shown in Fig. 3 (28.064.8 ng / ml before vs. 27.364.2 ng / ml after a 1 h hyperinsulinemic clamp).
4. Discussion In this study, we developed a RIA for determining circulating rat leptin concentrations, which performed adequately in evaluating circulating leptin levels from the standpoint of both accuracy and sensitivity. The relative crossreactivities of the antiserum used in this RIA system to mouse and human leptin were very low compared with that to rat leptin. Judging from amino acid homology between the leptin from these species, it is conceivable that the epitopes of this antiserum are located in the peptide including amino acid at either the 25th, the 54th, the 99th and the 123rd position (or combination thereof) in the leptin molecule, since these amino acids are identical between mouse and human leptin but different from those of rat leptin [2]. OLETF rats are reported to manifest hyperinsulinemia with insulin resistance in accordance with an increase in body fat deposition, and these metabolic derangements ameliorate in parallel with weight loss by either caloric restriction or exercise [21,23]. The caloric restriction study makes it possible to investigate the animals in different physical and metabolic states, and also unnecessary to use a large number of animals to get satisfactory correlation results. Indeed, we observed that five of the six rats in the free eating group (100% group), one of the six in the 85% group (15% food restricted) and none of the six in the 70% group (30% food restricted) were diabetic, as judged from the results of an oral glucose tolerance test performed at 20 weeks of age (data not shown). Thus, this model animal is suitable for investigating the relationship between circulat-
145
ing leptin and metabolic derangements related to the development of diabetes. It is well known that obesity causes hyperleptinemia [7,13,14] and is accompanied by insulin insensitivity [11,12]. The latter is a stimulus for insulin secretion, resulting in hyperinsulinemia in cases where the pancreatic b-cells are capable of responding to this stimulus. Furthermore, leptin may cause insulin resistance through modulating insulin signal transduction, such as the attenuation of tyrosine phosphorylation of the insulin receptor substrate-1 [19]. Leptin receptors are also expressed in pancreatic b-cells and leptin modulates the secretion of insulin from the isolated islet [24,25]. However, the fact that one modulates the secretion of the other remains controversial. Our study showed that circulating leptin levels closely mirror body fat content in OLETF rats and that serum leptin levels were not altered before and after hyperinsulinemic euglycemic clamp, as described in humans [13]. Furthermore, the partial correlation analysis revealed that there were no significant correlations between circulating leptin levels and various metabolic variables, including insulin and insulin sensitivity, after controlling for body fat content, suggesting that leptin per se did not associate with the development of the hyperglycemia with insulin insensitivity in the OLETF rats. In our study, serum leptin levels were reduced in the 70% food consumption group whose body fat content was also reduced, but the 85% chronic food consumption group did not result in a significant reduction of body fat content or plasma leptin level compared to the free eating group. These results suggest that chronic food restriction per se has no direct effect on plasma leptin levels but reduces them indirectly through its effect on decrease in body fat content, although acute restriction of food intake has been reported to reduce plasma leptin levels [13,14,26]. In any case, as reported by Collins and Surwit [7], there may not be an obligate relationship between plasma leptin levels and food intake.
Acknowledgements We thank Ms. Noriko Okauchi for her excellent work in maintaining the OLETF rats and for valuable advice during this study.
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;372:425–32. [2] Murakami T, Shima K. Cloning of rat obese cDNA and its expression in obese rats. Biochem Biophy Res Commun 1995;209:944–52. [3] Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards JG, Campfield LA, Clark FT, Deeds J, Muir C, Sanker S,
146
[4]
[5]
[6]
[7] [8]
[9]
[10]
[11]
[12]
[13]
[14]
M. Iida et al. / Regulatory Peptides 77 (1998) 141 – 146 Moriarty A, Moore KJ, Smutko JS, Mays GG, WoolF EA, Monroe CA, Tepper RI. Identification and expression cloning of a leptin receptor (OB-R). Cell 1995;83:1236–71. Chen H, Charlat O, Tartaglia LA, Woolf EA, Weng X, Ellis SJ, Lakey ND, Culpepper J, Moore KJ, Breitbart RF, Duyk GM, Tepper RI, Morgenstern JP. Evidence that the diabetes gene encodes the leptin receptor: Identification of a mutation in the leptin receptor gene in db /db mice. Cell 1996;84:491–5. Naggert J, Fricker L, Varlamov D, Nishina P, Roaille Y, Steiner D, Carrol R, Paigen B, Leiter E. Hyperproinsulinemia in obese fat /fat mice associated with carboxypeptidase E mutation which reduce enzyme activity. Nat Genet 1995;10:134–42. Iida M, Murakami T, Ishida K, Mizuno A, Kuwajima M, Shima K. Phenotype-linked amino acid alteration in leptin receptor cDNA from Zucker fatty ( fa /fa) rat. Biochem Biophy Res Commun 1996;222:19–26. Collins S, Surwit RS. Pharmacological manipulation of ob expression in a dietary model of obesity. J Biol Chem 1996;271:9437–40. Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Robinowitz D, Lallone RL, Burley SK, Friedman JM. Weight-reducing effect of the plasma protein encoded by the obese gene. Science 1995;269:543– 6. Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R, Ranganathan S, Kern PA, Friedman JM. Leptin levels in human and rodent: measurement of plasma leptin and ob mRNA in obese and weight-reduced subjects. Nat Med 1995;1:1156–61. ¨ Frederich RC, Hamman A, Andersson S, Lollmann B, Lowell BB, Flier JS. Leptin levels reflect body lipid content in mice: evidence for diet-induced resistance to leptin action. Nat Med 1995;1:1311–4. Bergman RN, Phillips LS, Cobelli C. Physiologic evaluation of factors controlling glucose tolerance in man: measurement of insulin sensitivity and b-cell glucose sensitivity from the response to intravenous glucose. J Clin Invest 1981;68:1456–67. Zunigo-Guajardo S, Jimenez J, Angel A, Zinman B. Effect of massive obesity on insulin sensitivity and insulin clearance and metabolic response to insulin as assessed by the euglycemic clamp technique. Metabolism 1986;35:278–82. ¨ Frederich RC, Lollmann B, Hamann A, Napolitano-Rosen A, Kahn BB, Lowell BB, Flier JS. Expression of ob mRNA and its encoded protein in rodents: impact of nutrition and obesity. J Clin Invest 1995;96:1658–63. Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephans TW, Nyce MR, Ohannesian JP, Marco CC, Mckee LJ, Bauer TL, Caro JF. Serum immunoreactive-leptin concentrations in normalweight and obese humans. New Engl J Med 1996;334:292–5.
[15] Cusin I, Sainsbury A, Doyle P, Rohner-Jeanrenaud F, Jeanrenand B. The ob gene and insulin: a relationship leading to clues to the understanding of obesity. Diabetes 1995;44:1467–70. [16] Saladin R, De Vos P, Guerre-Millo M, Leturque A, Girard J, Stales B, Auwerx J. Transient increase in obese gene expression after food intake or insulin administration. Nature 1995;377:527–9. [17] Kolaczynski JW, Nyce MR, Considine RV, Boden G, Nolan JJ, Henry R, Mudaliar SR, Olefsky J, Caro JF. Acute and chronic effect of insulin on leptin production in humans: studies in vivo and vitro. Diabetes 1996;45:699–701. [18] 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;269:540–3. [19] Choen B, Novick D, Rubinstein M. Modulation of insulin activities by leptin. Science 1996;274:1185–8. [20] Kawano K, Hirashima T, Mori S, Saitho Y, Kurosumi M, Natori T. Spontaneous long-term hyperglycemic rat with diabetic complications: Otsuka Long–Evans Tokushima Fatty (OLETF) strain. Diabetes 1992;41:1422–8. [21] Ishida K, Mizuno A, Zhu M, Sano T, Shima K. Which is the primary etiologic event in Otsuka Long–Evans Tokushima Fatty rats, a model of spontaneous, non-insulin-dependent diabetes mellitus, insulin resistance or impaired insulin secretion? Metabolism 1995;44:940–5. [22] Iida M, Murakami T, Yamada M, Sei M, Kuwajima M, Mizuno A, Noma Y, Aono T, Shima K. Hyperleptinemia in chronic renal failure. Horm Metab Res 1996;28:724–7. [23] Okauchi N, Mizuno A, Zhu M, Ishida K, Sano T, Noma Y, Shima K. Effect of obesity and inheritance on the development of non-insulindependent diabetes mellitus in Otsuka Long–Evans Tokushima Fatty rats. Diabetes Res Clin Pract 1995;29:1–10. [24] Emilsson V, Lin YL, Cawthorne MA, Morton NM, Davenport M. Expression of the functional leptin receptor mRNA in pancreatic islets and direct inhibitory action of leptin on insulin secretion. Diabetes 1997;46:313–6. [25] Ishida K, Murakami T, Mizuno A, Iida M, Kuwajima M, Shima K. Leptin suppresses insulin secretion from rat pancreatic islet. Regul Pept 1997;70:179–82. [26] Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R, Ranganathan S, Kern PA, Friedman JM. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med 1995;1:1155–61.