- Email: [email protected]
Zinc may be a mediator of leptin production in humans
Life Sciences.,Vol. 66,No. 22, pp. 2143-2149.2000 Copyright 0 2000 Elsevicr Science Inc. Printedin the USA. All ri&s reserved 0024-3205/00&see front matter
PIISOO243205(00)00541-5
ZINC MAY BE A MEDIATOR OF LEPTIN PRODUCTION IN HUMANS Ming-Der Chen, Yuh-Min Song, Pi-Yao Lin Division of Endocrinology and Metabolism, Department of Internal Medicine, Taichung Veterans General Hospital and Departments of Biology and Chemistry, Tunghai University, Taichung, Taiwan, ROC. (Received in fmal form December 23, 1999)
Summary Obese individuals have hyperleptinemia and hypozincemia. Moreover, leptin and zinc have circadian changes in circulating concentrations. We investigated their possible interaction and examined whether a difference existed between obese men and their lean controls. The results indicated the pattern of circadian change in plasma zinc and leptin did not markedly differ between the obese subjects and the lean controls. However, the obese had higher leptin and lower zinc plasma values at each sampling time than did the lean controls. Because an inverse correlation was found in plasma values between zinc and leptin (r=-0.51, p=O.O12), we further determined the role zinc might play in leptin production by human subcutaneous adipose tissue from female donors. The in vitro study showed that zinc treatment (0.2 mmol/L) significantly increased leptin production (142%) however, this increment did not surpass that by insulin (10 nmol/L). The data of this study suggest an interactive connection between zinc and leptin. Key Words: zinc, leptin, circadian change, insulin, subcutaneous adipose tissue, human
Leptin, a hormone secreted by the adipocytes (1) plays a role in mediating satiety and energy expenditure (reviewed in ref. 2). The circulating leptin concentration closely correlates with the amount of body fat mass (3). A blunt central responsiveness to leptin may explain the finding that a higher than normal plasma leptin concentration occurs in obese humans. Furthermore, a variety of physiological factors other than the amount of body fat mass could affect plasma leptin values (2). A physiological nocturnal rise in the circulating leptin concentration in humans has been noted, which may participate in appetite suppression and in body fat gain (4-6). Zinc, an essential trace mineral, also participates in appetite control. Its deficiency induces anorexia and is accompanied by a compensatory increased activity of central Corresponding author: Ming-Der CheK Division of Endocrinology and Metabolism, Taichung Veterans General Hospital, 160 Section 3 Chung-Kang Road, Taichung 40705, Taiwan (ROC). FAX: (666)-4-3593662, E-mail: mdchenBvghtc.vghtc.gov.tw
2144
Zinc and Leptin
Vol. 66, No. 22,zooo
neuropeptide Y, a potent orexigenic agent (7). Obese individuals frequently have hypozincemia (8,9). Furthermore, a diurnal variation in circulating zinc concentration has been well noted (10,ll). According to the data described above, it seems reasonable to speculate an interactive connection may exist between zinc and leptin. Thus, we examined this hypothesis by measuring the circadian changes in plasma concentrations of leptin and zinc. We also evaluated whether an altered rhythm might be shown in obese humans. Moreover, the in vitro effect of zinc on leptin production by human subcutaneous adipose tissue was assessed. Materials and Methods Protocol for experiment 7 Obese male subjects and their sex- and age-matched lean controls were recruited from students in the Scientific College of Tunghai University. The obese men met the criteria that their body weight be 20% higher than their desirable weight and their BMI > 28 kg/m2. Subjects who smoked, who were taking any drug or vitamin-mineral supplement, or who were on a special diet were excluded. All participants were clinically free from endocrine disorders, Five obese subjects (mean BMI=30.1, range=29.7-32.9) and five lean controls (BMI=21.5, range=19.&22.4) were included in this study. This study was approved by the Medical Ethics Committee of this medical center. Written informed consent was also obtained before enrollment. Following an overnight fast, the participants reported to the laboratory at 7 AM. An indwelling intravenous catheter was placed. First, a fasting blood sample was drawn at 8:OO;thereafter, blood samples were taken every 2 hours until the next morning. A total of 12 blood samples were collected from each participant. Standardized hospital meals containing approximately 7000 kJ (1750 kcal/day) with 55% carbohydrates, 30% fat, and 15% protein by weight were offered at 8:30, 12:30, and 18:30 to all participants. The average daily zinc content of this diet calculated from the database of Food Composition and Nutrient Values of Food in Taiwan was 10.8 mg/day. Protocol for experiment 2 The in vitro effect of zinc on leptin production was examined using a method previously reported by Pineiro et al. (12). Abdominal subcutaneous adipose tissue was obtained with permission from three non-obese women (age=28.3*3.5 y) during elective abdominal surgery conducted in the afternoon. None of the tissue donors smoked or were taking any drugs or vitamin-mineral supplements. Excised adipose tissue was immediately transported to the laboratory in ice-cold KrebsRinger HEPES buffer. After dissection to remove blood vessels and connective tissues, the adipose tissue was washed and cut into small pieces. Tissue fragments were then weighed and placed in multi-well culture plates (300 to 500 mg tissue/well) and flushed with 2.5 mL of Dulbecco’s Modified Eagle Medium plus 0.5% fetal calf serum, penicillin and streptomycin sulfate. After pre-incubation (1 hour at 37°C under 95% air-5% CO2), the medium in each well was removed and replaced with 2.5 mL fresh medium with or without the chemicals to be studied. These chemicals included zinc (0.2 mmol/L), sodium orthovanadate (0.1 mmol/L), insulin (10 nmol/l_), and thyroxine (50 nmol/L). The half maximal dosages of zinc and vanadate that have been reported to mimic insulin actions on stimulating glucose uptake and lipogenesis were used in this study. For each tested variable, adipose tissue from a given subject was independently assessed in triplicate. Culture media were collected and refreshed after incubation periods of 2, 12 and 24 hours. The media collected at 2, 12 and 24 hours after the initiation of incubation were used for the determinations of leptin and tumor necrosis factor-alpha. The
2145
Zinc and Leptin
Vol. 66, No. 22,200O
prepared tissue fragments showed >96% viability (assessed by MTT assay) at least 46 hours after adding or not adding the chemicals (data not shown). Analyses and measurements Glucose concentration was measured by the glucose oxidase method with an automated glucose analyzer (GA-03T, A&T, Tokyo, Japan). The measurements of insulin and leptin were performed using an enzymatic immunoassay method (AIA 600 for insulin, Tosoh, Tokyo, Japan; Quantikine human leptin, R&D Systems, Minneapolis, MN, USA) according to the manufacturers’ instructions. Plasma zinc concentration was determined using a flame atomic absorption spectrophotometer (11-551, Instrumentation Laboratory, Wilmington, MA, USA) by the method described elsewhere (13). The tumor necrosis factor-alpha concentration in the medium was measured with a commercial kit (Quantikine human TNF-d, R&D Systems). The data are presented as the mean&D. Statistical analyses of the results were conducted by ANOVA (among circadian changes), Student’s t-test (between the obese subjects and the lean controls), multiple comparison test (among studied chemicals), and simple correlation (between leptin and zinc) using a commercial package, StatWorks 1.2 for Macintosh. The difference was considered to be significant when the p value was c 0.05. Results
Figure 1 shows the rhythm of circadian changes in plasma leptin or plasma zinc did not markedly differ between the obese subjects and their lean controls. Moreover, the obese had higher leptin and lower zinc concentrations at each sampling time than did the lean controls. After all the obtained data were pooled, an inverse correlation was observed between leptin and zinc (r=-0.51, p=O.O12). Insulin and glucose had no significant correlation with leptin and zinc (data not shown). Zinc concentration @mol/L)
Leptin concentration (nmolR) 5
* -
ConReptin Obesefleptin
20.0 1
+ -
Con/Zn Obese/B
4 T
10.0 1
-11
08 10 12 14 16
18 20 22 24 02 04 06
::L
08 10 12 14 16 18 20 22 24 02 04 06
Fig. 1 Comparison of circadian changes in plasma concentrations of leptin and zinc between the obese male subjects (n=5) and their lean controls (n=5). All values are significantly different between the obese subjects and the lean controls (pcO.05).
Vol. 66, No. 22,2000
Zinc and Leptin
2146
The in vitro study indicated that both zinc (142% of control) and insulin (168%) significantly increased leptin production after an incubation period of 12 hours (Figure 2). However, the zinc increment (120%) became nonsignificant and the insulin effect (317%) was preserved at the end of incubation (after 24 hours). Furthermore, zinc did not affect the release of tumor necrosis factor-alpha, a stimulant to leptin production (14) (zinc vs. control=O.84&17 vs. 0.81iO.24 fmol/g tissue after 2 hours; and 7.28&Z.14 vs. 6.53ti.54 fmol/g tissue after 12 hours, respectively) Thyroxine acutely increased (after 2 hours) leptin production (164%, pcO.05); thereafter, it became blunted. Vanadate, as zinc, also having an insulin-like effect (15), did not influence leptin production. Interestingly, the intrinsically nocturnal rise in leptin was likely to be preserved in the incubated adipose tissues.
2.5
1
Leptin concentration (pmol/g tissue) *
Control
0.2 mM Zn
0.1 mM V 10 nM Insulin 50 nM T4
Fig. 2 Effects of various chemicals on leptin production by human abdominal subcutaneous adipose tissue. Leptin production was calculated as the amount of released leptin in the well divided by the weight of tissue fragments in a given incubation time period. * Indicates a significant increase in leptin production compared to that of the control group for the same incubation time (pcO.05). Zn: zinc, V: vanadate, T4: thyroxine. Discussion The present study supported previous findings (4,5) that the rhythm of circadian change in plasma leptin does not differ between obese and lean individuals. We further noted that the diurnal variation in plasma zinc did not differ between the obese male subjects and their lean controls, and that, regardless, the obese had hyperleptinemia and hypozincemia at each sampling time. There was a negative correlation between fasting values of zinc and leptin after the data from all participants were pooled to analyzed. This correlation was still retained even after the impact factors of BMI, glucose, insulin
Vol. 66, No. 22,200O
Zinc and Leptin
2147
were adjusted (data not shown). With the small sample size, it might not be possible to determine the precise association of the circadian changes in plasma zinc altered the response in plasma leptin. However, an inverse correlation in circadian changes in plasma concentrations between zinc and leptin was observed after all available data point were pooled and analyzed. Serious or mild dietary zinc deficiency, as well as increased fat intake, has become prevalent in many countries. Leptin is a satiety factor that reduces appetite, even in obese individuals (16). Feeding behaviors, such as fasting or having a large meal intake, could affect the circulating leptin concentration. Moreover, the development of obesity may be partly attributed to a reduction in leptin production (6,17). On the other hand, the known postprandial fall in plasma zinc values may be due to the redistribution of zinc from the plasma to extravascular pools which participate in the utilization and metabolism of fuels (16). Interestingly, some investigators have recently reported that subjects with zinc deficiency have hypoleptinemia (19,20). Moreover, our previous work indicates that hypozincemia in obese humans is a long-term and not an acute result (13). Insulin, in a large dose, is a potent stimulant to increase leptin production. This effect was also observed in the present study. The contribution of insulin-stimulated glucose uptake into adipose tissue to increase leptin production has been shown by Have1 et al. (21). However, the mechanism by which the metabolism of glucose in adipose tissue may be directly linked to leptin production remains unclear. In this study, zinc also increased leptin production, but this increment did not surpass that caused by insulin. This result was comparable to our previous observation of the adipocyte’s glucose uptake and lipogenesis in the mouse with similar treatment doses of zinc and insulin (22). Sodium orthovanadate had no influence on leptin production in this study, while in a similar dose, vanadate could markedly increase the adipocyte’s glucose uptake and lipogenesis (15). It is possible that the zinc’s effect on increasing leptin production may be associated with altered glucose metabolism but cannot be totally attributable to augmented lipogenesis. Shay et al. (23) recently reported that zinc deficiency may reduce leptin gene expression and leptin secretion from adipose tissue in rats. They also found that insulin-stimulated leptin secretion is augmented by a low rather than high incubation dose of zinc. Unfortunately, their data are not comparable to this study because of the differences in experimental design and species studied. Nevertheless, the dose response curve of this mineral on leptin production, as well as zinc’s effect on stimulating leptin production in obese subjects remain to be established. The physiologic significance of this result is still uncertain, because the zinc dosage (0.2 mM) used in this study is far greater than the levels found in circulation (1 l-23 PM). However, high endogenous zinc content occurs in mammalian tissues (0.5-l .O mM) and exogenously added zinc is able to influence glucose metabolism, which suggests that intracellular zinc may be not easily available and higher extracellular zinc may be needed to produce significant changes in the intracellular concentration over short incubations. Our data also did not support a suggestion that zinc may influence circulating leptin by increasing the production of tumor necrosis factor-alpha (20). This discrepancy may be due to the different experimental designs (in vivo and in vitro). Nevertheless, the possible effect of zinc on increasing cytokine production from extraadipose tissues should be considered for further study. At present, we cannot explain the mechanism for the acute effect of thyroxine on increasing leptin production. There are contradictory data regarding the interaction between thyroid hormones and leptin (24,25). The discrepancy may be associated with the differences in studied subjects, experimental designs, and treatment doses.
2148
Zinc and Leptin
Vol. 66, No. 22,200O
Though limited by a small sample size in this study, based on those data described above, an interactive connection should exist between zinc and leptin which may participate in the development of obesity. As has been suggested by Mantzoros et al. (20), increased appetite and energy intake caused by zinc supplementation in conjunction with simultaneous hyperleptinemia may result in the use of the increased energy intake to build lean body mass and keep fat mass lower. Thus, zinc supplementation may be beneficial for obese subjects who often have a marginal zinc deficiency. In summary, the present study indicated that zinc is a mediator of leptin production, though the detailed mechanism is still unknown and requires further investigation. Acknowledgments This work was supported by grants from the National Science Council (NSC88-2314-B075A-015 for MDC) and Taichung Veterans General Hospital (TCVGH873504A for YMS), Taiwan, ROC. References 1. Y. ZHANG, Ft. PROENCA, M. MAFFEI, M. BARONE, L. LEOPOLD and J.M. FRIEDMAN, Nature 372 425-432 (1994) 2. J.M. FRIEDMAN and J.L. HALAAS, Nature 395 763-770 (1998) 3. R.V. CONSIDINE, M.K. SINHA, M.L. HEIMAN, A. KRIAUCIUNAS, T.W. STEPHENS, MR. NYCE, J.P. OHANNESIAN, C.C. MARCO, L.J. MCKEE, T.L. BAUER and J.F. CARO, N. Engl. J. Med. 33 4 292295 (1996) 4. M.K. SINHA, J.P. OHANNESIAN, M.L. HEIMAN, A. KRIAUCIUNAS, R.W. STEPHENS, S. MAGOSIN, C. MARCO and J.F. CARO, J. Clin. Invest. 9 7 13441347 (1996) 5. J. LICINIO, C. MANTZOROS, A.B. NEGRAO, G. CIZZA, M. WONG, P.B. BONGIORNO, G.P. CHROUSOS, B. KARP, C. ALLEN, J.S. FLIER and P.W. GOLD, Nature Med. 3 575-579 (1997) 6. V. MATKOVIC, J.Z. ILICH, N.E. BADENHOP, M. SKUGOR, A. CLAIRMONT, D. KLISOVIC and J.D. LANDOLL, J. Clin. Endocrinol. Metab. 82 1368-1372 (1997) 7. P.L. SELVAIS, C. LABUCHE, N.X. NINH, J. KETELSLEGERS, J. DENEF and D.M. MAITER, J. Neuroendocrinol. 9 55-62 (1997) 8. M.D. CHEN, P.Y. LIN, W.H. LIN and V. CHENG, Am. J. Clin. Nutr. 4 8 1307-1309 (1988) 9. G. DIMARTINO, M.G. MATERA, B. DIMARTINO, C. VACCA, S. DIMARTINO and F. ROSSI, J. Med. 2 4 177-183 (1993) IO. 0. HETLAND and E. BRUBAKK, Stand. J. Clin. Lab. Invest. 3 2 225226 (1973) 11. J.C. KING, K.M. HAMBIDGE, J.L. WESTCOlT, D.L. KERN and G. MARSHALL, J. Nutr. 12 4 508-516 (1994) 12. V. PINEIRO, X. CASABIELL, R. PEINO, L. GARCIA-VALLEJO, C. DIEGUEZ and F.F. CASANUEVA, Biochem. Biophys. Res. Commun. 252 345-347 (1998) 13. M.D. CHEN, P.Y. LIN and W.H. SHEU, Biol. Trace Elem. Res. 60 123-129 (1997) 14. C. GRUNFELD, C. ZHAO, J. FULLER, A. POLlACK, A. MOSER, J. FRIEDMAN and K.R. FEINGOLD, J. Clin. Invest. 9 7 2152-2157 (1996) 15. Y. SHECHTER, Diabetes 39 l-5 (1990) 16. A.F. HEINI, C. LARA-CASTRO, K.A. KIRK, R.V. CONSIDINE, J.F. CAROand R.L. WEINSIER, Int. J. Obesity22 1084-1087 (1998) 17. J. COOLING, J. BARTH and J. BLUNDELL, Int. J. Obesity 2 2 1132-I 135 (1998) 18. N.M. LOWE, L.R. WOODHOUSE and J.C. KING, Br. J. Nutr. 8 0 363-370 (1998) 19. H.F. MANGIAN, R.G. LEE, G.L. PAUL, J.L. EMMERTand N.F. SHAY, J. Nutr. Biochem. 9 47-51 (1998) 20. C.S. MANTZOROS, A.S. PRASAD, F.W.J. BECK, S. GRABOWSKI, J. KAPLAN, C. ADAIR and G.J. BREWER, J. Am. Coil. Nutr. 1 7 270-275 (1998)
Vol. 66, No. 22,200O
Zinc and Leptin
2149
21. P.J. HAVEL, J.Y. URIU-HARE, T. LIU, K.L. STANHOPE, J.S. STERN, C.L. KEEN and B. AHREN, Am. J. Physiol. 274 R1482-R1491 (1998) 22. M.D. CHEN, S.J. LIOU, P.Y. LIN, V.C. YANG, P.S. ALEXANDER and W.H. LIN, Biol. Trace Elem. Res. 6 1 303-311 (1998) 23. E.S. OTT and N.F. SHAY, FASEB J. 1 3 A521 (1999) 24. G. SESMILO, R. CASAMITJANA, I. HALPERIN, R. GOMIS and E. VILARDELL, Eur. J. Endocrinol. 13 9 428-430 (1998) 25. M. OZATA, G. OZISIK, N. BINGOL, A. CORAKCI and M.A. GUNDOGAN, J. Endocrinol. Invest. 2 1 337-341 (1998)
PIISOO243205(00)00541-5
ZINC MAY BE A MEDIATOR OF LEPTIN PRODUCTION IN HUMANS Ming-Der Chen, Yuh-Min Song, Pi-Yao Lin Division of Endocrinology and Metabolism, Department of Internal Medicine, Taichung Veterans General Hospital and Departments of Biology and Chemistry, Tunghai University, Taichung, Taiwan, ROC. (Received in fmal form December 23, 1999)
Summary Obese individuals have hyperleptinemia and hypozincemia. Moreover, leptin and zinc have circadian changes in circulating concentrations. We investigated their possible interaction and examined whether a difference existed between obese men and their lean controls. The results indicated the pattern of circadian change in plasma zinc and leptin did not markedly differ between the obese subjects and the lean controls. However, the obese had higher leptin and lower zinc plasma values at each sampling time than did the lean controls. Because an inverse correlation was found in plasma values between zinc and leptin (r=-0.51, p=O.O12), we further determined the role zinc might play in leptin production by human subcutaneous adipose tissue from female donors. The in vitro study showed that zinc treatment (0.2 mmol/L) significantly increased leptin production (142%) however, this increment did not surpass that by insulin (10 nmol/L). The data of this study suggest an interactive connection between zinc and leptin. Key Words: zinc, leptin, circadian change, insulin, subcutaneous adipose tissue, human
Leptin, a hormone secreted by the adipocytes (1) plays a role in mediating satiety and energy expenditure (reviewed in ref. 2). The circulating leptin concentration closely correlates with the amount of body fat mass (3). A blunt central responsiveness to leptin may explain the finding that a higher than normal plasma leptin concentration occurs in obese humans. Furthermore, a variety of physiological factors other than the amount of body fat mass could affect plasma leptin values (2). A physiological nocturnal rise in the circulating leptin concentration in humans has been noted, which may participate in appetite suppression and in body fat gain (4-6). Zinc, an essential trace mineral, also participates in appetite control. Its deficiency induces anorexia and is accompanied by a compensatory increased activity of central Corresponding author: Ming-Der CheK Division of Endocrinology and Metabolism, Taichung Veterans General Hospital, 160 Section 3 Chung-Kang Road, Taichung 40705, Taiwan (ROC). FAX: (666)-4-3593662, E-mail: mdchenBvghtc.vghtc.gov.tw
2144
Zinc and Leptin
Vol. 66, No. 22,zooo
neuropeptide Y, a potent orexigenic agent (7). Obese individuals frequently have hypozincemia (8,9). Furthermore, a diurnal variation in circulating zinc concentration has been well noted (10,ll). According to the data described above, it seems reasonable to speculate an interactive connection may exist between zinc and leptin. Thus, we examined this hypothesis by measuring the circadian changes in plasma concentrations of leptin and zinc. We also evaluated whether an altered rhythm might be shown in obese humans. Moreover, the in vitro effect of zinc on leptin production by human subcutaneous adipose tissue was assessed. Materials and Methods Protocol for experiment 7 Obese male subjects and their sex- and age-matched lean controls were recruited from students in the Scientific College of Tunghai University. The obese men met the criteria that their body weight be 20% higher than their desirable weight and their BMI > 28 kg/m2. Subjects who smoked, who were taking any drug or vitamin-mineral supplement, or who were on a special diet were excluded. All participants were clinically free from endocrine disorders, Five obese subjects (mean BMI=30.1, range=29.7-32.9) and five lean controls (BMI=21.5, range=19.&22.4) were included in this study. This study was approved by the Medical Ethics Committee of this medical center. Written informed consent was also obtained before enrollment. Following an overnight fast, the participants reported to the laboratory at 7 AM. An indwelling intravenous catheter was placed. First, a fasting blood sample was drawn at 8:OO;thereafter, blood samples were taken every 2 hours until the next morning. A total of 12 blood samples were collected from each participant. Standardized hospital meals containing approximately 7000 kJ (1750 kcal/day) with 55% carbohydrates, 30% fat, and 15% protein by weight were offered at 8:30, 12:30, and 18:30 to all participants. The average daily zinc content of this diet calculated from the database of Food Composition and Nutrient Values of Food in Taiwan was 10.8 mg/day. Protocol for experiment 2 The in vitro effect of zinc on leptin production was examined using a method previously reported by Pineiro et al. (12). Abdominal subcutaneous adipose tissue was obtained with permission from three non-obese women (age=28.3*3.5 y) during elective abdominal surgery conducted in the afternoon. None of the tissue donors smoked or were taking any drugs or vitamin-mineral supplements. Excised adipose tissue was immediately transported to the laboratory in ice-cold KrebsRinger HEPES buffer. After dissection to remove blood vessels and connective tissues, the adipose tissue was washed and cut into small pieces. Tissue fragments were then weighed and placed in multi-well culture plates (300 to 500 mg tissue/well) and flushed with 2.5 mL of Dulbecco’s Modified Eagle Medium plus 0.5% fetal calf serum, penicillin and streptomycin sulfate. After pre-incubation (1 hour at 37°C under 95% air-5% CO2), the medium in each well was removed and replaced with 2.5 mL fresh medium with or without the chemicals to be studied. These chemicals included zinc (0.2 mmol/L), sodium orthovanadate (0.1 mmol/L), insulin (10 nmol/l_), and thyroxine (50 nmol/L). The half maximal dosages of zinc and vanadate that have been reported to mimic insulin actions on stimulating glucose uptake and lipogenesis were used in this study. For each tested variable, adipose tissue from a given subject was independently assessed in triplicate. Culture media were collected and refreshed after incubation periods of 2, 12 and 24 hours. The media collected at 2, 12 and 24 hours after the initiation of incubation were used for the determinations of leptin and tumor necrosis factor-alpha. The
2145
Zinc and Leptin
Vol. 66, No. 22,200O
prepared tissue fragments showed >96% viability (assessed by MTT assay) at least 46 hours after adding or not adding the chemicals (data not shown). Analyses and measurements Glucose concentration was measured by the glucose oxidase method with an automated glucose analyzer (GA-03T, A&T, Tokyo, Japan). The measurements of insulin and leptin were performed using an enzymatic immunoassay method (AIA 600 for insulin, Tosoh, Tokyo, Japan; Quantikine human leptin, R&D Systems, Minneapolis, MN, USA) according to the manufacturers’ instructions. Plasma zinc concentration was determined using a flame atomic absorption spectrophotometer (11-551, Instrumentation Laboratory, Wilmington, MA, USA) by the method described elsewhere (13). The tumor necrosis factor-alpha concentration in the medium was measured with a commercial kit (Quantikine human TNF-d, R&D Systems). The data are presented as the mean&D. Statistical analyses of the results were conducted by ANOVA (among circadian changes), Student’s t-test (between the obese subjects and the lean controls), multiple comparison test (among studied chemicals), and simple correlation (between leptin and zinc) using a commercial package, StatWorks 1.2 for Macintosh. The difference was considered to be significant when the p value was c 0.05. Results
Figure 1 shows the rhythm of circadian changes in plasma leptin or plasma zinc did not markedly differ between the obese subjects and their lean controls. Moreover, the obese had higher leptin and lower zinc concentrations at each sampling time than did the lean controls. After all the obtained data were pooled, an inverse correlation was observed between leptin and zinc (r=-0.51, p=O.O12). Insulin and glucose had no significant correlation with leptin and zinc (data not shown). Zinc concentration @mol/L)
Leptin concentration (nmolR) 5
* -
ConReptin Obesefleptin
20.0 1
+ -
Con/Zn Obese/B
4 T
10.0 1
-11
08 10 12 14 16
18 20 22 24 02 04 06
::L
08 10 12 14 16 18 20 22 24 02 04 06
Fig. 1 Comparison of circadian changes in plasma concentrations of leptin and zinc between the obese male subjects (n=5) and their lean controls (n=5). All values are significantly different between the obese subjects and the lean controls (pcO.05).
Vol. 66, No. 22,2000
Zinc and Leptin
2146
The in vitro study indicated that both zinc (142% of control) and insulin (168%) significantly increased leptin production after an incubation period of 12 hours (Figure 2). However, the zinc increment (120%) became nonsignificant and the insulin effect (317%) was preserved at the end of incubation (after 24 hours). Furthermore, zinc did not affect the release of tumor necrosis factor-alpha, a stimulant to leptin production (14) (zinc vs. control=O.84&17 vs. 0.81iO.24 fmol/g tissue after 2 hours; and 7.28&Z.14 vs. 6.53ti.54 fmol/g tissue after 12 hours, respectively) Thyroxine acutely increased (after 2 hours) leptin production (164%, pcO.05); thereafter, it became blunted. Vanadate, as zinc, also having an insulin-like effect (15), did not influence leptin production. Interestingly, the intrinsically nocturnal rise in leptin was likely to be preserved in the incubated adipose tissues.
2.5
1
Leptin concentration (pmol/g tissue) *
Control
0.2 mM Zn
0.1 mM V 10 nM Insulin 50 nM T4
Fig. 2 Effects of various chemicals on leptin production by human abdominal subcutaneous adipose tissue. Leptin production was calculated as the amount of released leptin in the well divided by the weight of tissue fragments in a given incubation time period. * Indicates a significant increase in leptin production compared to that of the control group for the same incubation time (pcO.05). Zn: zinc, V: vanadate, T4: thyroxine. Discussion The present study supported previous findings (4,5) that the rhythm of circadian change in plasma leptin does not differ between obese and lean individuals. We further noted that the diurnal variation in plasma zinc did not differ between the obese male subjects and their lean controls, and that, regardless, the obese had hyperleptinemia and hypozincemia at each sampling time. There was a negative correlation between fasting values of zinc and leptin after the data from all participants were pooled to analyzed. This correlation was still retained even after the impact factors of BMI, glucose, insulin
Vol. 66, No. 22,200O
Zinc and Leptin
2147
were adjusted (data not shown). With the small sample size, it might not be possible to determine the precise association of the circadian changes in plasma zinc altered the response in plasma leptin. However, an inverse correlation in circadian changes in plasma concentrations between zinc and leptin was observed after all available data point were pooled and analyzed. Serious or mild dietary zinc deficiency, as well as increased fat intake, has become prevalent in many countries. Leptin is a satiety factor that reduces appetite, even in obese individuals (16). Feeding behaviors, such as fasting or having a large meal intake, could affect the circulating leptin concentration. Moreover, the development of obesity may be partly attributed to a reduction in leptin production (6,17). On the other hand, the known postprandial fall in plasma zinc values may be due to the redistribution of zinc from the plasma to extravascular pools which participate in the utilization and metabolism of fuels (16). Interestingly, some investigators have recently reported that subjects with zinc deficiency have hypoleptinemia (19,20). Moreover, our previous work indicates that hypozincemia in obese humans is a long-term and not an acute result (13). Insulin, in a large dose, is a potent stimulant to increase leptin production. This effect was also observed in the present study. The contribution of insulin-stimulated glucose uptake into adipose tissue to increase leptin production has been shown by Have1 et al. (21). However, the mechanism by which the metabolism of glucose in adipose tissue may be directly linked to leptin production remains unclear. In this study, zinc also increased leptin production, but this increment did not surpass that caused by insulin. This result was comparable to our previous observation of the adipocyte’s glucose uptake and lipogenesis in the mouse with similar treatment doses of zinc and insulin (22). Sodium orthovanadate had no influence on leptin production in this study, while in a similar dose, vanadate could markedly increase the adipocyte’s glucose uptake and lipogenesis (15). It is possible that the zinc’s effect on increasing leptin production may be associated with altered glucose metabolism but cannot be totally attributable to augmented lipogenesis. Shay et al. (23) recently reported that zinc deficiency may reduce leptin gene expression and leptin secretion from adipose tissue in rats. They also found that insulin-stimulated leptin secretion is augmented by a low rather than high incubation dose of zinc. Unfortunately, their data are not comparable to this study because of the differences in experimental design and species studied. Nevertheless, the dose response curve of this mineral on leptin production, as well as zinc’s effect on stimulating leptin production in obese subjects remain to be established. The physiologic significance of this result is still uncertain, because the zinc dosage (0.2 mM) used in this study is far greater than the levels found in circulation (1 l-23 PM). However, high endogenous zinc content occurs in mammalian tissues (0.5-l .O mM) and exogenously added zinc is able to influence glucose metabolism, which suggests that intracellular zinc may be not easily available and higher extracellular zinc may be needed to produce significant changes in the intracellular concentration over short incubations. Our data also did not support a suggestion that zinc may influence circulating leptin by increasing the production of tumor necrosis factor-alpha (20). This discrepancy may be due to the different experimental designs (in vivo and in vitro). Nevertheless, the possible effect of zinc on increasing cytokine production from extraadipose tissues should be considered for further study. At present, we cannot explain the mechanism for the acute effect of thyroxine on increasing leptin production. There are contradictory data regarding the interaction between thyroid hormones and leptin (24,25). The discrepancy may be associated with the differences in studied subjects, experimental designs, and treatment doses.
2148
Zinc and Leptin
Vol. 66, No. 22,200O
Though limited by a small sample size in this study, based on those data described above, an interactive connection should exist between zinc and leptin which may participate in the development of obesity. As has been suggested by Mantzoros et al. (20), increased appetite and energy intake caused by zinc supplementation in conjunction with simultaneous hyperleptinemia may result in the use of the increased energy intake to build lean body mass and keep fat mass lower. Thus, zinc supplementation may be beneficial for obese subjects who often have a marginal zinc deficiency. In summary, the present study indicated that zinc is a mediator of leptin production, though the detailed mechanism is still unknown and requires further investigation. Acknowledgments This work was supported by grants from the National Science Council (NSC88-2314-B075A-015 for MDC) and Taichung Veterans General Hospital (TCVGH873504A for YMS), Taiwan, ROC. References 1. Y. ZHANG, Ft. PROENCA, M. MAFFEI, M. BARONE, L. LEOPOLD and J.M. FRIEDMAN, Nature 372 425-432 (1994) 2. J.M. FRIEDMAN and J.L. HALAAS, Nature 395 763-770 (1998) 3. R.V. CONSIDINE, M.K. SINHA, M.L. HEIMAN, A. KRIAUCIUNAS, T.W. STEPHENS, MR. NYCE, J.P. OHANNESIAN, C.C. MARCO, L.J. MCKEE, T.L. BAUER and J.F. CARO, N. Engl. J. Med. 33 4 292295 (1996) 4. M.K. SINHA, J.P. OHANNESIAN, M.L. HEIMAN, A. KRIAUCIUNAS, R.W. STEPHENS, S. MAGOSIN, C. MARCO and J.F. CARO, J. Clin. Invest. 9 7 13441347 (1996) 5. J. LICINIO, C. MANTZOROS, A.B. NEGRAO, G. CIZZA, M. WONG, P.B. BONGIORNO, G.P. CHROUSOS, B. KARP, C. ALLEN, J.S. FLIER and P.W. GOLD, Nature Med. 3 575-579 (1997) 6. V. MATKOVIC, J.Z. ILICH, N.E. BADENHOP, M. SKUGOR, A. CLAIRMONT, D. KLISOVIC and J.D. LANDOLL, J. Clin. Endocrinol. Metab. 82 1368-1372 (1997) 7. P.L. SELVAIS, C. LABUCHE, N.X. NINH, J. KETELSLEGERS, J. DENEF and D.M. MAITER, J. Neuroendocrinol. 9 55-62 (1997) 8. M.D. CHEN, P.Y. LIN, W.H. LIN and V. CHENG, Am. J. Clin. Nutr. 4 8 1307-1309 (1988) 9. G. DIMARTINO, M.G. MATERA, B. DIMARTINO, C. VACCA, S. DIMARTINO and F. ROSSI, J. Med. 2 4 177-183 (1993) IO. 0. HETLAND and E. BRUBAKK, Stand. J. Clin. Lab. Invest. 3 2 225226 (1973) 11. J.C. KING, K.M. HAMBIDGE, J.L. WESTCOlT, D.L. KERN and G. MARSHALL, J. Nutr. 12 4 508-516 (1994) 12. V. PINEIRO, X. CASABIELL, R. PEINO, L. GARCIA-VALLEJO, C. DIEGUEZ and F.F. CASANUEVA, Biochem. Biophys. Res. Commun. 252 345-347 (1998) 13. M.D. CHEN, P.Y. LIN and W.H. SHEU, Biol. Trace Elem. Res. 60 123-129 (1997) 14. C. GRUNFELD, C. ZHAO, J. FULLER, A. POLlACK, A. MOSER, J. FRIEDMAN and K.R. FEINGOLD, J. Clin. Invest. 9 7 2152-2157 (1996) 15. Y. SHECHTER, Diabetes 39 l-5 (1990) 16. A.F. HEINI, C. LARA-CASTRO, K.A. KIRK, R.V. CONSIDINE, J.F. CAROand R.L. WEINSIER, Int. J. Obesity22 1084-1087 (1998) 17. J. COOLING, J. BARTH and J. BLUNDELL, Int. J. Obesity 2 2 1132-I 135 (1998) 18. N.M. LOWE, L.R. WOODHOUSE and J.C. KING, Br. J. Nutr. 8 0 363-370 (1998) 19. H.F. MANGIAN, R.G. LEE, G.L. PAUL, J.L. EMMERTand N.F. SHAY, J. Nutr. Biochem. 9 47-51 (1998) 20. C.S. MANTZOROS, A.S. PRASAD, F.W.J. BECK, S. GRABOWSKI, J. KAPLAN, C. ADAIR and G.J. BREWER, J. Am. Coil. Nutr. 1 7 270-275 (1998)
Vol. 66, No. 22,200O
Zinc and Leptin
2149
21. P.J. HAVEL, J.Y. URIU-HARE, T. LIU, K.L. STANHOPE, J.S. STERN, C.L. KEEN and B. AHREN, Am. J. Physiol. 274 R1482-R1491 (1998) 22. M.D. CHEN, S.J. LIOU, P.Y. LIN, V.C. YANG, P.S. ALEXANDER and W.H. LIN, Biol. Trace Elem. Res. 6 1 303-311 (1998) 23. E.S. OTT and N.F. SHAY, FASEB J. 1 3 A521 (1999) 24. G. SESMILO, R. CASAMITJANA, I. HALPERIN, R. GOMIS and E. VILARDELL, Eur. J. Endocrinol. 13 9 428-430 (1998) 25. M. OZATA, G. OZISIK, N. BINGOL, A. CORAKCI and M.A. GUNDOGAN, J. Endocrinol. Invest. 2 1 337-341 (1998)