�9 1998 by Humana Press Inc. All rights of any nature, whatsoew?r, reserved. 0163-4984/98/6103-0303 $10.25
Effects of Zinc Supplementation on the Plasma Glucose Level
and Insulin Activity in Genetically Obese (ob/ob) Mice
MING-DER CHEN, *'1'3 SHY-JANE Llou, 1 PI-YAo LIN, 2 VIVlAN C. YANG, 1 PAUL S. ALEXANDER, 1 AND WEN-HAN LIN 3
~Graduate Institute of Biology and 2Department of Chemistry, Tunghai University; and 3Division of Endocrinology and Metabolism, Department of Medicine, Taichung Veterans General Hospital, Taichung, Taiwan, ROC
Received May 3, 1996; Accepted October 24, 1996
A B S T R A C T
The effects of zinc supplementation (20 mM ZnC12 from the drinking water for eight weeks) on plasma glucose and insulin levels, as well as its in vitro effect on lipogenesis and lipolysis in adipocytes were studied in genetically obese (ob/ob) mice and their lean controls (+/?). Zinc supplementation reduced the fasting plasma glucose lev- els in both obese and lean mice by 21 and 25%, respectively (p < 0.05). Fasting plasma insulin levels were significantly decreased by 42% in obese mice after zinc treatment. In obese mice, zinc supplementation also attenuated the glycemic response by 34% after the glucose load. The insulin-like effect of zinc on lipogenesis in adipocytes was sig- nificantly increased by 80% in lean mice. However, the increment of 74% on lipogenesis in obese mice was observed only when the zinc plus insulin treatment was given. This study reveals that zinc sup- plementation alleviated the hyperglycemia of ob/ob mice, which may be related to its effect on the enhancement of insulin activity.
Index Entries: Zinc; obesity; insulin; glucose tolerance test; lipogenesis; lipolysis.
*Author to whom all correspondence and reprint requests should be addressed.
Biological Trace Element Research 303 Vol. 61, 1998
304 Chen et al.
INTRODUCTION
Obese individuals have low serum and hair zinc levels, and this decrement is well correlated to their BMI (kg/m 2) values (1-3). Zinc administration plus dietary restriction decreases body weight gain in obese children (1), but this effect is not significant in obese adults receiv- ing zinc treatment alone (4). Genetically obese (ob/ob) mice also have hypozincemia, hyperzincuria, and a maldistribution of zinc in their tis- sues (5-7). Zinc supplementation attenuates hyperglycemia and pancre- atic insulin secretion (5), as well as increases body fat deposition in this obese mutant (8).
Obesity is becoming prevalent in developed and developing coun- tries, and is a factor in the pathological development of noninsulin- dependent diabetes mellitus (NIDDM) (9). Zinc deficiency found in diabetes mellitus may be related to the insulin resistance of the disease. Moreover, the physiological or pharmacological supplementation of zinc is found to be beneficial to alleviate the hyperglycemia of diabetic patients (10,11).
There is a strong relationship between zinc and insulin. This essen- tial trace element is well known to take part in the synthesis, secretion, and activity of insulin (12,13). Furthermore, zinc also mimics some actions of insulin (14-17). It is unknown whether zinc exerts an insulin- like action in obese-diabetics the same as observed in nonobese diabetics (17), this study was undertaken to examine the effect of zinc supple- mentation on the obese-diabetic (ob/ob) mouse, which exhibits obesity, hyperglycemia, hyperinsulinemia, and insulin resistance (18).
METHODS
Animals
Male obese (C57BL/6J-ob/ob) mice and their lean controls (+/?) were obtained from the Jackson Laboratory (Bar Harbor, ME). Through- out the study, all animals were kept in a temperature-controlled room (25 +_ 2~ with a 12-12 dark-light cycle, and were housed in individual cages. Purina chow (zinc content = 44 _+ 1 mg/kg diet, #5001, St. Louis, MO) was provided ad libitum.
Glucose Tolerance Test
Mice at eight weeks of age were separated into four groups (N-control; ln-Zn; ob-control; ob-Zn) according to the absence of zinc (deionized water), or with zinc supplementation (20 mM ZnC12) in the drinking water. Each group contained six mice. After eight weeks of zinc administration, mice (at 16 wk of age) were prefasted over 12 h and subject to the glucose tolerance test (1 g glucose/kg body wt, ip). Blood was obtained from the retro-orbital
Biological Trace Element Research Vol. 61, 1998
Zinc and Obesity 305
sinus of the conscious mice and the plasma levels of glucose (glucose- oxidase method) and insulin (RIA) determined.
Zinc Determination
Tissue's zinc concentrations were determined by the method of atomic absorption spectrophotometry (19). The compatible standard ref- erence materials obtained from NIST (Gaithersburg, MA) were also assayed to measure the zinc recovery (>97%).
Lipogenesis and Lipolysis in Adipocytes
An additional 48 male mice of both phenotypes at 20 wk of age were used, and killed by dislocation. The epididymal adipose tissues were taken, dissected free of contaminating muscle and connective tissues, and weighed. Adipocytes were prepared by collagenase digestion (20). The cell preparation showed >97% viability at least 3 h after digestion and with or without 0.5 mM ZnC12 addition, which was determined by the 3-(4,5-dimethyl-thiazol-z-yl)-2,5-diphenyl-tetra-zolium bromide (MTT) assay (21). In a preliminary study, the half maximum stimulation of lipogene- sis by zinc and insulin occurred respectively at 0.2 mM and 10 nM (data not shown). All assays were in duplicate and as described in the figure legends. Lipogenesis was performed with [U-14C]-glucose, and its con- version to lipids was examined as previously reported (22). Lipolysis was determined by measuring glycerol released following stimulation with 1 ~tM isoproterenol (23).
Data Analysis
Statistical analyses of the data were conducted by analysis of vari- ance (ANOVA) and unpaired Student's t-test using statistical analysis system (SAS) commercial program (24). The difference was considered to be significant when the p value <0.05.
RESULTS
Table 1 shows that zinc supplementation significantly increases body fat and reduces food intake in ob/ob mice, but this effect was not sig- nificant in lean mice.
Zinc-supplemented mice had lower fasting plasma glucose levels than the controls (ln-Zn vs In-control = 5.2 vs 7.0; ob-Zn vs ob-control = 10.2 vs 12.9 mM).
Fasting plasma insulin was significantly decreased by 42% in ob-Zn mice, but this value was still higher than that of lean mice (Table 2).
After the administration of the glucose load, plasma glucose values of the ob-Zn group rose to levels that were significantly different from
Biological Trace Element Research Vol. 61, 1998
306 Chen et al.
Table 1 Body Weight, Body Fat, Food and Water Intake in Control
and Zinc-Supplemented Mice a,b,c,d
Initial wt Final wt Body fat Food Water (g) (g) (%) (g/day) (mL/day)
Ln-control 21.1 (0.9) 25.8 (1.1) 12.4 (1.7) 4.0 (0.2) 7.2 (0.3) Ln-Zn 21.0 (0.5) 25.3 (0.5) 13.2 (2.0) 3.8 (0.2) 7.3 (0.4) Ob-control 41.7 (0.3) 55.7 (1.1) 45.9 (0.9) 7.4 (0.2) 7.1 (0.2) Ob-Zn 40.8 (0.5) 53.5 (1.1) 53.4 (1.1)* 6.3 (0.2)* 7.2 (0.6)
aMean (SEM), each group contained six mice. bThe mice were fed their respective drinking water (deionized or 20 mM ZnC12 zinc-
supplemented) for a period of eight weeks. CThe body fat composition was measured gravimetrically after chloroform/methanol
extraction from the hom*ogenate of mouse carcass. aAll values were significantly different between lean and obese mice, except the data
of water intake. *Significant difference (p < 0.05) between control and Zn group.
Table 2 Plasma Insulin and Zinc Levels after the Administration of the Glucose Load
(1 g glucose/kg of body wt, ip) in Control and Zinc-Supplemented Mice a
0 min 30 min 60 min 90 min
Insulin (pM) Ln-control 144 (15) 102 (14) 118 (6) 154 (14) Ln-Zn 127 (6) 152 (19) 122 (12) 128 (19) Ob-control 1090 (92) 1375 (70) 1287 (79) 1335 (77) Ob-Zn 630 (112)* 672 (120)* 880 (96)* 603 (111)* Zinc (~tM) Ln-control 15.0 (1.0) 12.1 (3.3) 10.3 (1.0) 10.2 (0.8) Ln-Zn 17.0 (2.1) 10.9 (1.4) 11.3 (2.7) 8.6 (1.2) Oh-control 12.4 (1.7) 16.0 (0.5) 10.6 (1.4) 15.4 (1.1) Ob-Zn 14.4 (1.3) 10.8 (1.5)* 10.0 (2.1) 12.6 (1.8)*
aMean (SEM), each group contained six mice. *Significant difference (p < 0.05) between control and Zn group.
the controls. In other words, zinc supplementation attenuated the glycemic response by 34% (area under the glycemia curve) in ob/ob mice (Fig. 1). In brief, plasma insulin values in all sampling times, and plasma zinc levels at 30 and 90 min were lower in ob-Zn mice as compared with their controls after the glucose load (Table 2). There was no significant correlation between these determined variates in the glucose tolerance test, except for a positive correlation (r -- 0.994) between glucose and insulin found in ob-control mice.
Biological Trace Element Research Vol. 61, 1998
Zinc and Obesity 307
v
o _=
30
20 �84
10,
0 -I
- - " o - - ' In -con t ro l .
.... -l- .... tn-Zn .
ob-control I ob-Zn
,#I ~
I'I ./-"
1'5 3'0 4'5 G.O 7'5 9.O Time (min)
Fig. 1. Effect of zinc supplementation on plasma glucose levels after the administration of the glucose load (1 g glucose/kg of body wt, ip) in obese and lean mice. The data were mean +SEM from six mice in each group. *Indicates a significant difference between control and Zn group (p < 0.05).
Table 3 Zinc Concentrations of Tissues in Control and Zinc-Supplemented Micea,b, c
Adipose Liver Urine Stool (~tmol/g) (~mol/g) (nmol/day) (nmol/day)
Ln-control 9.7 (1.1) 1.9 (0.1) 67 (6) 336 (21) Ln-Zn 10.9 (1.3) 2.1 (0.1) 78 (8) 413 (26)* Ob-control 16.3 (1.9) 2.7 (0.2) 165 (18) 612 (37) Ob-Zn 21.2 (2.1)* 3.0 (0.2)* 202 (21) 719 (72)*
aMean (SEM), each group contained six mice. bZinc values for epididymal adipose tissue and liver, as well as the stool were based
on the dry weight of the samples. cAll values were significantly different between lean and obese mice (p < 0.05). *Significant difference (p < 0.05) between control and Zn group.
Tissue zinc status in this s tudy was consistent with previous reports (5-8). The ob /ob mouse had high zinc in excretion from urine and stool, as well as tissue-specific accumulation of zinc in liver and epididymal adipose tissue (Table 3).
Figure 2 shows that lipogenesis in adipocytes derived from lean mice was increased by 80% after zinc treatment (Zn vs Basal = 3756 vs
Biological Trace Element Research Vol. 61, 1998
308 Chen et al.
8 0 0 0 -
6 0 0 0 �9
4 0 0 0 '
2000 -
Bas a l Zn
[ ] Lean
B "-F-
A " - - T - -
I n s u l i n Z n + l n s u l i n
C
c T - 1 -
o 8
2 ? ,r '7,
.o
8
[ ] Obese 12000
8000 ] A
4 0 0 0 -t
//////
0 / / / / / /
B a s a l
A
/ / / / / . , z / z / z /
, / / / / , , / / / / / /
Zn I n s u l i n
B
T Y / / / X
,z/zz, 1 / / / / ~ "/////
1 z / l / ,
, 1 1 1 1 , I///I~ / / / / / ,
"///// V//// r / i l l ~
z'/f//~
Z n + I n s u l i n
Fig. 2. Adipocytes (5 X 104 cells/mL) derived either from obese or lean mice that received the zinc treatment (groups of Zn and Zn+Insulin) were pre- incubated for 30 min at 37~ with 0.2 mM ZnC12. Lipogenesis was then carried out for additional 1 h at 37~ with [U-14C]-glucose (specific activity = 10.6 mCi/ mmol, final concentration -- 0.15 mM) after the addition of nothing (groups of Basal and Zn) or 20 nM insulin (groups of Insulin and Zn+Insulin). The differ- ent letters which are given above the columns indicate that there were signifi- cant differences among treated groups (p < 0.05).
2073 dpm). However, the increment of 74% in ob /ob mice was only sig- nificant when a zinc plus insulin treatment was given (Zn+Insulin vs Basal = 9929 vs 5713 dpm). Zinc treatment alone had no significant effect on lipolysis induced by isoproterenol (Fig. 3). However, the zinc plus insulin treatment significantly attenuated the inhibitive effect of insulin on the isoproterenol-mediated lipolysis in lean mice.
DISCUSSION
In vivo, zinc supplementat ion has been found to reduce the plasma glucose levels in ob /ob mice and STZ-induced diabetic rats (5,17), which appear because the insulin-like action of zinc. Hyperinsulinemia, a characteristic of obese animals, is accompanied by hyperglycemia, and also appear implicated in the establishment of hyperphagia. High rates of lipogenesis are apparent in tissues of obese mutants after weaning, and hyperinsul inemia is central to this (18). In this study, zinc supple- mentat ion significantly reduced the fasting plasma levels of glucose and insulin, as well as their values after the glucose load in ob /ob mice. Zinc also reduced the fasting plasma glucose level in lean mice. This indi- cated that zinc might, through its insulin-like action or a regulatory
Biological Trace Element Research Vol. 61, 1998
Zinc and Obesity 309
30
[ ] Lean
A
"-I-- 20"
i
0 lso
A
T
lso+Zn
C
B
lso+lns Iso+Zn+Ins
12 [ ] Obese
AC
/ t / / t / I
| so
A
Y//,,h lso+Zn
BC
B
/ / / / / / I / / / / t
~/" //////
/////Z Iso+tns Iso+Zn+lns
Fig. 3. Adipocytes (5 • 104 cells/mL) derived either from obese or lean mice were added without any additive, or with 0.2 mM ZnC12, or/and with 20 nM insulin, respectively. Lipolysis was then carried out for 2 h at 37~ with a stimulant of isoproterenol (1 gM). Amount of glycerol released was deter- mined. The different letters which are given above the columns indicate that there are significant differences among treated groups (p < 0.05).
effect on insulin activity, respond to affect the plasma glucose status. The zinc effect on reducing the plasma insulin level in ob /ob mice might either be related to the direct effect of zinc on attenuating pan- creatic insulin secretion (5), or an indirect effect of zinc on the enhance- ment of peripheral response and utilization of insulin. Chronic hyperglycemia, which has been referred to as glucose toxicity, inhibits both insulin secretion and glucose utilization (25). In this study, zinc supplementation reduced the food intake in ob /ob mice as previously reported (5). However, the possibility that zinc might participate in relieving the glucose toxicity by affecting food intake should also be considered.
In vitro, zinc mimics some actions of insulin to increase glucose uti- lization and lipogenesis in adipocytes from normal and streptozotocin (STZ)qnduced diabetic rats (14-17). In this study, zinc addition increased the lipogenesis in adipocytes of lean mice, but this effect did not exceed the maximum stimulation by insulin. Adipocytes of ob/ob mice were insulin resistant; insulin or zinc treatment showed no effect on lipogenesis. However, the zinc plus insulin treatment caused enhanced lipogenesis in ob/ob mice. The results indicate that insulin-like action of zinc on lipoge- nesis is not manifested in the adipocytes of ob/ob mice, contrary to the findings in lean mice and in rats described above (14-17). This also sug- gests that zinc acts differently in the states of hyperinsulinemia (ob/ob mice) and insulin deficiency (STZ-induced diabetic rats).
Biological Trace Element Research Vol. 61, 1998
310 Chen et al.
Zinc is known to have the antilipolytic effect in rats' adipocytes (15). A weak lipolytic activity of zinc is also found in rats adipocytes, yet this effect does not exceed 5% of the maximum lipolytic rates obtained in the presence of isoproterenol (17). Zinc addition alone had no significant effect on the isoproterenol-mediated lipolysis in this study. However, an increase of lipolysis was found in the adipocytes of lean mice that received the zinc plus insulin treatment. This showed that zinc attenu- ated the insulin inhibition on the isoproterenol-mediated lipolysis. It is still unknown how zinc might act on the signals derived from insulin and adrenaline, which affect lipolysis.
The in vivo results of this study were consistent with a previous report (5). Some differences in the improvement of glucose intolerance in ob/ob mice and the fasting plasma value of insulin in lean mice might be attributed to the added dosage (1000 mg Zn/kg diet in that study, approx 20 times the current study), and the duration of supple- mentation. Zinc is relatively nontoxic to mammals (26). It is still unclear whether the long-term or over-dosage of zinc supplementation might act differently on its regulatory function. In this study, the adaptive increase of zinc loss was apparent, but ob/ob mice specifically retained more zinc in the liver and adipose tissues. Although dietary-obese rats have been found to adapt their absorbing and retaining strategies to match the dietary availability of essential minerals (27), the appearance of tissue-specific real-distribution of zinc in the ob/ob mice might play a more important role in the development of obesity. Zinc supplemen- tation contributed to the tissue-specific maldistribution of zinc in ob/ob mice, and may have affected peripheral glucose utilization and insulin activity. Increases of lipogenesis in adipocytes and body fat accumula- tion have been reported elsewhere (8).
In summary, zinc supplementation reduced the fasting plasma levels of glucose and insulin in ob/ob mice. The zinc effect on increasing lipo- genesis in the adipocytes of ob/ob mice was mainly by its enhancement on insulin activity, but not an insulin-like action.
REFERENCES
1. P.J. Collipp, New development in medical therapy of obesity: thyroid and zinc, Pedi- atr. Ann. 13, 465-472 (1984).
2. M. D. Chen, P. Y. Lin, W. H. Lin, and V. Cheng, Zinc in hair and serum of obese indi- viduals in Taiwan, Am. J. Clin. Nutr. 48, 1307-1309 (1988).
3. G. Di-Martino, M. G. Matera, B. De-Martino, C. Vacca, S. Di-Martino, and F. Rossi, Relationship between zinc and obesity, J. Med. 24, 177-183 (1993).
4. M. D. Chen, W. H. Lin, and P. Y. Lin, Zinc sulfate and thyroxine treatment on the obese patients, Chin. Med. J. 48, 210-216 (1991).
5. N. Begin-Heick, M. Daple-Scott, J. Rowe, and H. M. C. Heick, Zinc supplementation attenuates insulin secretory activity in pancreatic islet of the ob/ob mouse, Diabetes 34, 179-184 (1985).
6. M. L. Kennedy and M. L. Failla, Zinc metabolism in genetically obese (ob/ob) mice, J. Nutr. 117, 886-893 (1987).
Biological Trace Element Research Vol. 61, 1998
Zinc and Obesity 311
7. W. H. Lin, M. D. Chen, and P. Y. Lin, Investigation of the profile of selected trace met- als in genetically obese (ob/ob) and lean (+/?) mice, J. Formosan. Med. Assoc. 91, s27-s33 (1992).
8. M. D. Chen, P. Y. Lin, V. Cheng, and W. H. Lin, Zinc supplementation aggravates body fat accumulation in genetically obese (ob/ob) mice and dietary-obese mice, Biol. Trace. Elem. Res. 52, 125-132 (1996).
9. D. A. Leaf, Overweight: assessment and management issues, Am. Fam. Physician. 42, 653-660 (1990).
10. J. F. Burn, R. Guintrand-Hugret, C. Fons, J. Carvajal, C. Fedou, M. Fussellier, L. Bardet, and A. Orsetti, Effects of oral zinc gluconate on glucose effectiveness and insulin sensitivity in humans, Biol. Trace. Elem. Res. 47, 385-391 (1995).
11. J. E. Sprietsma and G. E. Schuitemaker, Diabetes can be prevented by reducing insulin production, Med. Hypotheses 42, 15-23 (1994).
12. H. P. Roth and M. Kirchgessner, Zinc and insulin metabolism, Biol. Trace. Elem. Res. 3, 13-32 (1981).
13. P. Faure, A. Roussel, C. Coudray, M. J. Richard, S. Halimi, and A. Favier, Zinc and insulin sensitivity, Biol. Trace. Elem. Res. 32, 305-310 (1992).
14. L. Coulston and P. Nandona, Insulin-like effect of zinc on adipocytes, Diabetes 29, 665-667 (1980).
15. J. M. May and C. S. Contoreggi, The mechanism of the insulin-like effects of ionic zinc, ]. Biol. Chem. 257, 4362-4368 (1982).
16. O. Ezaki, lib group metal ions (Zn 2+, Cd 2+, Hg 2+) stimulate glucose transport activity by post-insulin receptor kinase mechanism in rat adipocytes, J. Biol. Chem. 264, 16,118-16,122 (1989).
17. A. Shisheva, D. Gefel, and Y. Shechter, Insulinlike effects of zinc ion in vitro and in vivo: preferential effects on desensitized adipocytes and induction of normoglycemia in streptozocin-induced rats, Diabetes 41, 982-988 (1992).
18. G. A. Bray and D. A. York, Hypothalamic and genetic obesity in experimental ani- mals: an autonomic and endocrine hypothesis, Physiol. Rev. 59, 719-809 (1979).
19. K. H. Falchuk, K. L. Hilt, and B. L. Vallee, Determination of zinc in biological sam- ples by atomic absorption spectrometry, Method Enzymol. 158, 422-434 (1988).
20. M. Rodbell, Metabolism of isolated fat cells, I. Effects of hormones on glucose metab- olism and lipolysis, J. Biol. Chem. 239, 375-380 (1964).
21. F. Denizot and R. Lang, Rapid colorimetric assay for cell growth survival, J. lmmunol. Method. 89, 271-277 (1986).
22. J. A. Carnie, D. G. Smith, and M. Mavris-Vavayannis, Effects of insulin on lipolysis and lipogenesis in adipocytes from genetically obese (ob/ob) mice, Biochem. J. 184, 107-112 (1979).
23. S. R. Jooly, J. J. Lech, and L. A. Menahan, Influence of genetic obesity in mice on the lipolytic response of isolated adipocytes to isoproterenol an ACTH-(1-24), Horm. Metab. Res. 10, 223-227 (1978).
24. Statistical Analysis System Institute Inc, SAS users guide: statistics, SAS Institute, Cary, NC (1985).
25. H. Yki-Jarvinen, Glucose toxicity, Endocrine Rev. 13, 415-431 (1992). 26. M. K. Hambridge, C. E. Casey, and N. F. Krebs, Zinc, in Trace Elements in Human and
Animal Nutrition-Fifth Edition, W. Mertz, ed., Academic, New York, pp. 1-137 (1986). 27. J. A. Fernandez-Lopez, M. Esteve, I. Rafecas, X. Remesar, and M. Alemany, Management
of dietary essential metals (iron, copper, zinc, chromium and manganese) by Wistar and Zucker obese rats fed a self-selected high-energy diet, Biometals 7, 117-129 (1994).
Biological Trace Element Research Vol. 61, 1998
