- Email: [email protected]
Influence of alloxan-induced diabetes and selenite treatment on blood glucose and glutathione levels in mice
ARTICLE IN PRESS Journal of
Trace Elements in Medicine and Biology Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267 www.elsevier.de/jtemb
PATHOBIOCHEMISTRY
Influence of alloxan-induced diabetes and selenite treatment on blood glucose and glutathione levels in mice Xi-Qun Sheng, Kai-Xun Huang, Hui-Bi Xu Institute of Materia Medica, Huazhong University of Science and Technology, Wuhan 430074, P.R. China Received 9 June 2004; accepted 17 January 2005
Abstract Many clinical studies reported that diabetic patients had lower glutathione contents in erythrocytes or plasma. Recently, selenium, an essential trace element with well-known antioxidant characteristics, has been found to have insulin-mimetic properties. But seldom information is available about the influence of selenium on glutathione changes induced by diabetes mellitus in animals. Therefore, this study was designed to compare the impacts of selenite treatment on glutathione (GSH) levels of blood and tissues such as brain, kidney, liver, spleen and testis in mice. Four groups were used in this study: a control group, a diabetic group, a selenite-treated normal group and a selenite-treated diabetic group. Selenite was administered to the mice for 4 weeks with an oral dose of 2 mg kg1 day1 by gavage. The blood glucose level, and GSH level in blood and tissues were determined. The results show that the selenite-treated diabetic group had significantly lower blood glucose levels than the diabetic group. Moreover, alloxan-induced diabetes significantly decreased GSH levels in blood, kidney, liver and testis compared to the controls. Selenite treatment of the diabetic mice only improved the GSH levels in liver and brain. On the other hand, selenite administered to the normal mice reduced GSH levels in the liver compared to the controls. In conclusion, this study suggests that selenite treatment of diabetic mice with an effective dose would be beneficial for the antioxidant system of liver and brain although it exerts a toxic effect on the liver of normal mice. r 2005 Elsevier GmbH. All rights reserved. Keywords: Selenite; Alloxan; Diabetes mellitus; Glutathione; Glucose
Introduction Reduced glutathione (GSH) is the prevalent nonprotein thiol in mammalian cells. GSH has many biological functions, such as maintaining membrane protein sulfydryl groups in the reduced form, acting as a substrate for GSH peroxidase and detoxification of xenobiotics. Therefore, the maintenance of the GSH Corresponding author. Tel.: +86 27 87543532; fax: +86 27 87543632. E-mail addresses: [email protected] (X.-Q. Sheng), [email protected] (H.-B. Xu).
0946-672X/$ - see front matter r 2005 Elsevier GmbH. All rights reserved. doi:10.1016/j.jtemb.2005.01.001
level is crucial for cellular defense against oxidative injury and cellular integrity [1,2]. It has been reported that diabetic patients had lower GSH concentrations in erythrocytes and plasma [3–5]. In diabetic animals, a reduction of the GSH level was also observed both in erythrocytes [6,7] and in aortic tissue [8]. Selenium has been proved to have insulin-mimetic properties in vitro [9] and in vivo [10] and exert antioxidant characteristics in diabetic animal models [11]. However, seldom information is available about the influence of selenium treatment on GSH levels of tissues in diabetic animals. Therefore, this study was
ARTICLE IN PRESS 262
X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
designed to systematically compare the influence of selenite treatment on GSH levels in blood and tissues such as brain, kidney, liver, spleen and testis in KunMing mice with diabetes induced by alloxan. We found that mice with alloxan-induced diabetes had lower GSH levels in blood, kidney, liver and testis compared to the controls, and that selenite treatment of the diabetic mice with an oral dose of 2 mg kg1 day1 only improved GSH levels in liver and brain.
volume of redistilled water for the GSH assay. The remaining blood was centrifuged 15 min at 1500g for isolation of plasma. After the blood was collected, tissues of brain, liver, kidney, spleen and testis were removed, weighed and rinsed three times with cold physiological saline, and stored at 75 1C for the GSH assay.
GSH assay
Materials and methods Materials Alloxan, o-phthalaldehyde (OPA), diethylenetriaminepentaacetic acid (DTPA) and N-ethylmaleimide (NEM) were purchased from Sigma Co. GSH was obtained from Amresco Co. All other chemicals were of the highest commercial grade available in the domestic market. The water used was freshly prepared redistilled water.
Animals, induction of diabetes and sodium selenite treatment Fifty male Kun-Ming mice obtained from Hubei Research Animal Center, weighing 2072 g, were kept in an air-conditioned animal house with a normal day/ night cycle. The mice were fed with mouse chow (also purchased from Hubei Research Animal Center) and tap water ad libitum. Twenty mice were used as controls; they were again randomly divided into two subgroups: the control group and the selenite-treated normal group. Thirty mice, after 48 h of fasting, were intraperitoneally injected with 200 mg kg1 alloxan dissolved in 50 mmol l1 citrate buffer (pH 3.0). Seventy-two hours after alloxan injection, whole blood samples were obtained from the tail vein of the overnight fasted mice and their glucose levels were tested by glucose test strips (Roche Diagnostic Corp., Indiannapolis, IN, USA). Twenty mice with glucose levels over 15 mmol l1 were included in the diabetic group and then randomly assigned to two subgroups: the selenite-treated diabetic group and the diabetic group without any treatment. The mice in the selenite-treated diabetic and normal group were treated with a dose of 2 mg kg1 selenite dissolved in redistilled water by gavage. The body weight and blood glucose level were measured once a week when the mice were fasted overnight. After 4 weeks of selenite treatment, the mice were anesthetized with ether, and blood samples were drawn from postcava into heparinized tubes. An aliquot of whole blood was removed and hemolyzed with a ninefold
The content of reduced GSH in whole blood samples was measured with an assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) using 5,50 dithio-bis (2-nitrobenzoic acid) (DTNB). Levels of reduced GSH in brain, liver, spleen, kidney and testis where determined using a fluorescence method [12]. A 10% homogenate of the tissue samples was prepared in an ice-cold homogenization solution (20 mmol l1 HCl, 5 mmol l1 DTPA, 10 mmol l1 ascorbic acid, and 5% trichloroacetic acid (TCA)) using a glass piston homogenizer for 1 min. The suspension was centrifuged at 14,000g and the resulting supernatant solution was centrifuged through a 0.45-mm microcentrifuge filter (Millipore Corp., USA), yielding a 5% deproteinized homogenate that was used for the OPA assay procedure. The following solutions were required to perform the OPA assay: redox quenching buffer (RQB) (20 mmol l1 HCl, 5 mmol l1 DTPA, 10 mmol l1 ascorbic acid); 5% TCA in RQB (TCARQB); 7.5 mmol l1 NEM in RQB; 1.0 mol l1 potassium phosphate (KPi) buffer (pH 7.0); 0.1 mol l1 KPi buffer (pH 6.9); 5.0 mg ml1 OPA in methanol. The OPA solution was prepared just before use. 0.1 mmol l1 GSH was prepared as a standard. In brief, a mixture of sample (or standard), TCA-RQB, NEM and 1 mol l1 KPi was incubated for 5 min at room temperature (20–21 1C), 0.1 mol l1 KPi and OPA were added and incubated for 30 min at room temperature before measuring the OPA-derived fluorescence at 365 nm excitation and 430 nm emission [12].
Blood glucose assay Blood glucose was determined by the glucose oxidase method (blood glucose assay kit, Shanghai Rongsheng Biotech Co., Shanghai, China).
Plasma insulin determination The insulin level in plasma was estimated using an insulin radioimmunoassy kit (Huaxi Institute of diabetic technology, Chengdu, Sichuan Province, China).
ARTICLE IN PRESS X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
Statistical analysis
263
Data were expressed as mean7standard deviation (SD). Statistical analysis was performed using the statistical package ‘‘SPSS 12.0 for Windows’’. A oneway analysis of variation (ANOVA) was performed after ascertainment of the normality of distribution (Kolmogorov–Smirnov test) and the homogeneity of variance (Levene test) of the experimental data. If both conditions were fulfilled, differences between means were examined by Tukey’s test. If homogeneity of variance could not be ensured, differences between means were evaluated by Dunnett-T3 test. po0:05 was considered significantly different.
Blood glucose level (mmol/L)
35
30
25 a
20 b
15
10
5
0 0
1
2
3
4
Week
Results Changes in the mean body weight and the mean blood glucose level of the four groups are shown in Figs. 1 and 2, respectively. The body weight of the selenite-treated diabetic group started to be higher than that of the diabetic control group from week 3 on, but the difference only reached significance at week 4 (po0:05) (Fig. 1). Accordingly, from week 3 on, the glucose level of the selenite-treated diabetic group was significantly lower compared to the untreated diabetic group, suggesting that an oral administration of 2 mg kg1 selenite effectively decreases the hyperglycemia of alloxan-induced diabetic mice. However, after 4 weeks,
44 42 40 a
36
Body weight (g)
a
a
38
a a
34
a
a
32 b
30 b
a
28 26
b
24
b
c
b
b b
22 20 18 16 0
1
2
3
4
Week
Fig. 1. Body weight changes of the normal and diabetic mice treated and untreated with selenite for 4 weeks. ’, normal control; K, diabetic control; ., selenite-treated normal group; ,, selenite-treated diabetic group; n ¼ 10 in all groups. Values of the same week marked with different letters (a, b, c) were significantly different from each other as determined by Tukey’s test (po0:05).
Fig. 2. Blood glucose level changes of the normal and diabetic mice treated and untreated with selenite for 4 weeks. ’, normal control; K, diabetic control; ., selenite-treated normal group; ,, selenite-treated diabetic group; n ¼ 10 in all groups. In the selenite-treated diabetic group, at week 3 and week 4, the blood glucose level (marked with a and b) was significantly lower than in the diabetic control group at the same time (a: po0:05; b: po0:01).
the selenite treatment could not restore the hyperglycemia of diabetic mice to a normal level (Fig. 2). Results of the plasma insulin assay show that the hypoglycemic effect of selenite treatment on the diabetic mice was not due to an increase of pancreatic insulin release stimulated by selenite, because there were no differences in the plasma insulin levels between the selenite-treated diabetic group and the diabetic group (Fig. 3). On the contrary, the plasma insulin level of the selenite-treated normal group was lower than that of the control group. Fig. 4 shows GSH levels in whole blood of the four groups. In the diabetic group, the blood GSH concentration was markedly lower than that in the control group. However, the blood GSH content of the selenitetreated diabetic group and the diabetic group was at the same level, and so was the GSH level in the selenitetreated normal group and the control group. This demonstrates that selenite treatment had no influence on the GSH level in diabetic or normal mice. GSH levels in tissues of the four groups are shown in Fig. 5. In the diabetic group, GSH contents were generally lower than in the control group, but significant decreases were observed only in kidney, liver and testis (Figs. 5B, C and E). Selenite treatment of diabetic mice increased GSH levels in liver and brain: there were significant differences between the GSH content in brain and liver of the selenite-treated diabetic group and the diabetic group (Figs. 5A and C). Selenite treatment had no impact on spleen both for the diabetic group and the normal group (Fig. 5D).
ARTICLE IN PRESS 264
X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
Plasma insulin level (microIU/ml)
16
b
14 12
a a
a
10 8 6 4 2 0
DM+Se
Normal+Se
DM
Normal
Groups
GSH levels in the whole blood (micromol/ml)
Fig. 3. Plasma insulin levels of the normal and diabetic mice (DM) treated and untreated with selenite for 4 weeks; n ¼ 10 in all groups. Values marked with different letters (a, b) were significantly different from each other as determined by Tukey’s test (po0:05).
40
b
b
a
a
35 30 25 20 15 10 5 0
DM+Se
Normal+Se
DM
Normal
Groups
Fig. 4. GSH levels in whole blood of the normal and diabetic mice (DM) treated and untreated with selenite for 4 weeks; n ¼ 10 in all groups. Values marked with different letters (a, b) were significantly different from each other as determined by Tukey’s test (po0:05).
Discussion It has been proved that hyperglycemia generates oxidative stress leading to the development of diabetic complications [13]. However, the reported results about the relationship between diabetes and GSH level are
inconsistent. Several investigators have found lower GSH concentrations in erythrocytes, plasma, aorta and lenses in diabetic compared to normal subjects [3–8,14], while others have reported no changes of GSH levels in human erythrocytes [15], or in diabetic rats’ erythrocytes, liver, muscle, testis and lense [16–18]. The present study shows that alloxan-induced diabetes can decrease GSH levels in blood and some tissues such as kidney, liver and testis (Figs. 4 and 5B, C, E). A possible explanation for the discrepancy might be the different degrees and durations of hyperglycemia, some cases showing ketonemia resulting in a low glutahione level [3]. In our study, we did not measure the levels of the ketone body. However, all diabetic mice had, on an average, high blood glucose levels (30.674.1 mmol l1 at the beginning; 24.174.2 mmol l1 at the end of the experiment), in such a way that hyperglycemia would probably cause ketonemia in 4 weeks. A further study needs to confirm this. It is well established that selenium has biological functions as an antioxidant [19], and its insulin-like properties have also been documented in vitro and in vivo [9,10,20]. Therefore, we deduced that selenium might have some impact on GSH levels in diabetic mice. Indeed, in this study, we found that selenite treatment of diabetic mice significantly increased GSH levels in liver and brain, but not in blood, kidney, spleen and testis. However, selenite treatment of normal mice markedly decreased GSH levels in the liver compared to the controls, suggesting that 2 mg kg1 selenite oral administration had a toxic effect on normal mice. A similar beneficial effect of selenite treatment in the liver of diabetic rats was also observed by another group [16], who, on the other hand, reported a liver GSH content at the same level for the normal and diabetic animals. On considering the restoration of GSH levels in the liver of diabetic mice by selenite treatment, it seems that diabetes decreased the toxicity of a high dose of selenite in mice. The mechanism underlying this phenomenon needs further study. Regarding the glucose lowering effect of selenium on type II diabetic animals, the dose of selenite used in the present study was 2 mg kg1 day1 (11.6 mmol kg1 day1), which was in accord with McNeill et al.’s studies (10–15 mmol kg1 day1) [10,21], but in conflict with another report in which oral administration of 80 mg kg1 day1 (0.46 mmol kg1 day1) selenite was found to be effective in 2 weeks in lowering hyperglycemia of Kun-Ming mice with diabetes induced by alloxan [22]. In addition, we observed that the hypoglycemic effect of selenium on diabetic mice was not due to an increase of insulin release from pancreatic islets stimulated by selenite (Fig. 3). On the contrary, selenite reduced insulin release of pancreatic islets in normal mice: the plasma insulin level in the selenite-treated normal group was markedly lower than in the controls (Fig. 3).
ARTICLE IN PRESS X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
265
3.5
A
5
2.5 ac
2.0 1.5
bc
bc
1.0
GSH level in liver (micromol/g)
GSH level in brain (micromol/g)
3.0 a
C
a
a b
4
3 c 2
1
0.5 0
0.0 DM+Se
DM
Normal+Se
DM+Se
Normal
DM
Normal+Se
Groups
Normal
Groups 4.5
5.5
b
4.5 b
4.0 3.5
a
3.0 2.5
a
2.0 1.5 1.0
a
D
4.0
B
a
3.5
GSH level in spleen (micromol/g)
GSH level in kidney (micromol/g)
5.0
3.0
a
a
2.5 2.0 1.5 1.0 0.5
0.5
0.0
0.0 DM+Se
DM
Normal+Se
DM+Se
Normal
Normal+Se
6.5
DM
Normal
Groups
Groups E
6.0
GSH lelvel in testis (mciromol/g)
5.5 5.0 bc
4.5 4.0 3.5 ac
3.0 2.5 2.0
aa
a
1.5 1.0 0.5 0.0 DM+Se
Normal+Se
DM
Normal
Groups
Fig. 5. GSH levels in brain, kidney, liver, spleen and testis of the normal and diabetic mice treated and untreated with selenite for 4 weeks; n ¼ 10 in all groups. (A) brain, (B) kidney, (C) liver, (D) spleen, (E) testis. Values marked with different letters (a, b, c) were significantly different from each other as determined by Tukey’s test or Dunnett-T3 test (po0:05).
ARTICLE IN PRESS 266
X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
Recently, Mueller et al. [23] have found that selenate corrected glucose tolerance and lowered insulin resistance in type II diabetic dbdb mice, but selenite did not, therefore assuming that selenite might possess no peripheral insulin mimicking properties. However, our study has shown that selenite had a hypoglycemic effect on type II diabetic mice not due to a pancreatic insulin secretion increased by selenite treatment (Fig. 3), suggesting that selenite has peripheral insulin mimicking properties. To clarify this issue, more studies are needed, especially in vitro. In our previous study [24], a higher dose of selenite (4 mg kg1 day1) was used to trigger the glucose-lowering effect quickly (within a week), but failed to maintain this insulin-like effect on the alloxan-induced diabetic mice more than a week due to the toxicity of selenite. We also observed that more selenium was deposited in the liver and brain of the diabetic mice. In the present study, we found that a selenite treatment of 2 mg kg1 day1 improved the GSH concentration only in the liver and brain of diabetic rats, implying that selenium exerts its antioxidant properties beneficial to diabetic mice in liver and brain, in addition to its insulinlike properties. The question whether selenium has the same insulin-mimetic properties in the brain or the central nervous system needs further investigation. In summary, alloxan-induced diabetes decreased GSH levels in blood, kidney, liver and testis in KunMing mice, and oral administration of 2 mg kg1 day1 selenite improved the liver and brain GSH levels in the diabetic mice. Considering a lack of effect or toxicity of selenite in blood, kidney, spleen and testis on the GSH levels, more studies are needed to assess the beneficial effect of selenium in diabetes.
Acknowledgements We would like to give our thanks to associate professor Qiong Liu for revising the English language.
References [1] Will Y. Overview of glutathione function and metabolism. In: Maines MD, editor. Current protocols in toxicology. New York: Wiley; 1999. p. 6.1.1–6.1.18. [2] Dickinson DA, Forman HJ. Cellular glutahione and thiol metabolism. Biochem Pharmacol 2002;64:1019–26. [3] Jain SK, McVie R. Hyperketonemia can increase lipid peroxidation and lower glutathione levels in human erythrocytes in vitro and in type 1 diabetic patients. Diabetes 1999;48:1850–5. [4] Jain SK, McVie R, Smith T. Vitamin E supplementation restores glutathione and malondialdehyde to normal concentrations in erythrocytes of type 1 diabetic children. Diabetes Care 2000;23:1389–94.
[5] Samiec PS, Drews-Botsch C, Flagg EW, Kurtz JC, Sternberg Jr P, Reed RL, Jones DP. Glutathione in human plasma: decline in association with aging, agerelated macular degeneration, and diabetes. Free Radic Biol Med 1998;24:699–704. [6] Mohan IK, Das UN. Effect of L-arginine-nitric oxide system on chemical-induced diabetes mellitus. Free Radic Biol Med 1998;25:757–65. [7] Yilmaz O¨, O¨zkan Y, Yildirim M, O¨ztu¨rk AI, Ersan Y. Effect of alpha lipoic acid, ascorbic acid-6-palmitate, and fish oil on the glutathione, malonaldehyde, and fatty acids levels in erythrocytes of streptozotocin induced diabetic male rats. J Cell Biochem 2002;86:530–9. [8] Tachi Y, Okuda Y, Bannai C, Bannai S, Shinohara M, Shimpuku H, Yamashita K, Ohura K. Hyperglycemia in diabetic rats reduces the glutathione content in the aortic tissue. Life Sci 2001;69:1039–47. [9] Ezaki O. The insulin-like effects of sodium selenate in rat adipocytes. J Biol Chem 1990;265:1124–8. [10] McNeill JH, Delgatty HLM, Battell ML. Insulinlike effects of sodium selenate in streptozotocin diabetic rats. Diabetes 1991;40:1675–8. [11] Mukherjee B, Anbazhagan S, Roy A, Ghosh R, Chatterjee M. Novel implications of the potential role of selenium on antioxidant status in streptozotocininduced diabetic mice. Biomed Pharmacother 1998; 52:89–95. [12] Senft AP, Dalton TP, Shertzer HG. Determining glutathione and glutathione disulfide using the fluorescence probe o-phthalaldehyde. Anal Biochem 2000;280:80–6. [13] Baynes JW. Role of oxidative stress in development of complications in diabetes. Diabetes 1991;40:405–11. [14] Mitton KP, Dean PAW, Dzialoszynski T, Xiong H, Sanford SE, Trevithick JR. Modelling cortical cataractogenesis 13. Early effects on lense ATP/ADP and glutathione in the streptozotocin rat model of the diabetic cataract. Exp Eye Res 1993;56:187–98. [15] Sinclair AJ, Girling AJ, Gray L, Lunec J, Barnett AH. An investigation of the relationship between free radical activity and vitamin C metabolism in elderly diabetic subjects with retinopathy. Gerontology 1992;38:268–74. [16] Naziroglu M, Cay M. Protective role of intraperitoneally administered vitamin E and selenium on the antioxidative defense mechanisms in rats with diabetes induced by streptozotocin. Biol Trace Elem Res 2001;79:149–59. [17] Unlucerci Y, Bekpinar S, Kocak H. Testis glutathione peroxidase and phospholipid hydroperoxide glutathione peroxidase activities in aminoguanide-treated diabetic rats. Arch Biochem Biophys 2000;379:217–20. [18] Naziroglu M, Dilsiz N, Cay M. Protective role of introperitoneally administered vitamine C and E and selenium on the levels of lipid peroxidation in the lens of rats made diabetic with streptozotocin. Biol Trace Elem Res 1999;70:223–32. [19] Xu H, Huang K. Selenium: its chemistry, biochemistry and application in life science. China: The publishing press of HUST; 1994. p. 152–167. [20] Becker DJ, Reul B, Ozcelikay AT, Buchet JP, Henquin JC, Brichard SM. Oral selenate improves glucose homeostasis and partly reverses abnormal expression of liver
ARTICLE IN PRESS X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
glycolytic and gluconeogenic enzymes in diabetic rats. Diabetologia 1996;39:3–11. [21] Battell ML, Delgatty HLM, McNeill JH. Sodium selenate corrects glucose tolerance and heart function in STZ diabetic rats. Mol Cell Biochem 1998;179:27–34. [22] Ding H, Pen R, Chen J, Li Q, Wang H, Kong R. Effect of selenite and allitridi on the diabetic mice. Acta Nutr Sin 1997;19:384–7.
267
[23] Mueller AS, Pallauf J, Rafael J. The chemical form of selenium affects insulinomimetic properties of the trace element: investigations in type II diabetic dbdb mice. J Nutr Biochem 2003;14:637–47. [24] Sheng X-Q, Huang K-X, Xu H-B. New experimental observation on the relationship of selenium and diabetes mellitus. Biol Trace Elem Res 2004;99: 241–53.
Trace Elements in Medicine and Biology Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267 www.elsevier.de/jtemb
PATHOBIOCHEMISTRY
Influence of alloxan-induced diabetes and selenite treatment on blood glucose and glutathione levels in mice Xi-Qun Sheng, Kai-Xun Huang, Hui-Bi Xu Institute of Materia Medica, Huazhong University of Science and Technology, Wuhan 430074, P.R. China Received 9 June 2004; accepted 17 January 2005
Abstract Many clinical studies reported that diabetic patients had lower glutathione contents in erythrocytes or plasma. Recently, selenium, an essential trace element with well-known antioxidant characteristics, has been found to have insulin-mimetic properties. But seldom information is available about the influence of selenium on glutathione changes induced by diabetes mellitus in animals. Therefore, this study was designed to compare the impacts of selenite treatment on glutathione (GSH) levels of blood and tissues such as brain, kidney, liver, spleen and testis in mice. Four groups were used in this study: a control group, a diabetic group, a selenite-treated normal group and a selenite-treated diabetic group. Selenite was administered to the mice for 4 weeks with an oral dose of 2 mg kg1 day1 by gavage. The blood glucose level, and GSH level in blood and tissues were determined. The results show that the selenite-treated diabetic group had significantly lower blood glucose levels than the diabetic group. Moreover, alloxan-induced diabetes significantly decreased GSH levels in blood, kidney, liver and testis compared to the controls. Selenite treatment of the diabetic mice only improved the GSH levels in liver and brain. On the other hand, selenite administered to the normal mice reduced GSH levels in the liver compared to the controls. In conclusion, this study suggests that selenite treatment of diabetic mice with an effective dose would be beneficial for the antioxidant system of liver and brain although it exerts a toxic effect on the liver of normal mice. r 2005 Elsevier GmbH. All rights reserved. Keywords: Selenite; Alloxan; Diabetes mellitus; Glutathione; Glucose
Introduction Reduced glutathione (GSH) is the prevalent nonprotein thiol in mammalian cells. GSH has many biological functions, such as maintaining membrane protein sulfydryl groups in the reduced form, acting as a substrate for GSH peroxidase and detoxification of xenobiotics. Therefore, the maintenance of the GSH Corresponding author. Tel.: +86 27 87543532; fax: +86 27 87543632. E-mail addresses: [email protected] (X.-Q. Sheng), [email protected] (H.-B. Xu).
0946-672X/$ - see front matter r 2005 Elsevier GmbH. All rights reserved. doi:10.1016/j.jtemb.2005.01.001
level is crucial for cellular defense against oxidative injury and cellular integrity [1,2]. It has been reported that diabetic patients had lower GSH concentrations in erythrocytes and plasma [3–5]. In diabetic animals, a reduction of the GSH level was also observed both in erythrocytes [6,7] and in aortic tissue [8]. Selenium has been proved to have insulin-mimetic properties in vitro [9] and in vivo [10] and exert antioxidant characteristics in diabetic animal models [11]. However, seldom information is available about the influence of selenium treatment on GSH levels of tissues in diabetic animals. Therefore, this study was
ARTICLE IN PRESS 262
X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
designed to systematically compare the influence of selenite treatment on GSH levels in blood and tissues such as brain, kidney, liver, spleen and testis in KunMing mice with diabetes induced by alloxan. We found that mice with alloxan-induced diabetes had lower GSH levels in blood, kidney, liver and testis compared to the controls, and that selenite treatment of the diabetic mice with an oral dose of 2 mg kg1 day1 only improved GSH levels in liver and brain.
volume of redistilled water for the GSH assay. The remaining blood was centrifuged 15 min at 1500g for isolation of plasma. After the blood was collected, tissues of brain, liver, kidney, spleen and testis were removed, weighed and rinsed three times with cold physiological saline, and stored at 75 1C for the GSH assay.
GSH assay
Materials and methods Materials Alloxan, o-phthalaldehyde (OPA), diethylenetriaminepentaacetic acid (DTPA) and N-ethylmaleimide (NEM) were purchased from Sigma Co. GSH was obtained from Amresco Co. All other chemicals were of the highest commercial grade available in the domestic market. The water used was freshly prepared redistilled water.
Animals, induction of diabetes and sodium selenite treatment Fifty male Kun-Ming mice obtained from Hubei Research Animal Center, weighing 2072 g, were kept in an air-conditioned animal house with a normal day/ night cycle. The mice were fed with mouse chow (also purchased from Hubei Research Animal Center) and tap water ad libitum. Twenty mice were used as controls; they were again randomly divided into two subgroups: the control group and the selenite-treated normal group. Thirty mice, after 48 h of fasting, were intraperitoneally injected with 200 mg kg1 alloxan dissolved in 50 mmol l1 citrate buffer (pH 3.0). Seventy-two hours after alloxan injection, whole blood samples were obtained from the tail vein of the overnight fasted mice and their glucose levels were tested by glucose test strips (Roche Diagnostic Corp., Indiannapolis, IN, USA). Twenty mice with glucose levels over 15 mmol l1 were included in the diabetic group and then randomly assigned to two subgroups: the selenite-treated diabetic group and the diabetic group without any treatment. The mice in the selenite-treated diabetic and normal group were treated with a dose of 2 mg kg1 selenite dissolved in redistilled water by gavage. The body weight and blood glucose level were measured once a week when the mice were fasted overnight. After 4 weeks of selenite treatment, the mice were anesthetized with ether, and blood samples were drawn from postcava into heparinized tubes. An aliquot of whole blood was removed and hemolyzed with a ninefold
The content of reduced GSH in whole blood samples was measured with an assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) using 5,50 dithio-bis (2-nitrobenzoic acid) (DTNB). Levels of reduced GSH in brain, liver, spleen, kidney and testis where determined using a fluorescence method [12]. A 10% homogenate of the tissue samples was prepared in an ice-cold homogenization solution (20 mmol l1 HCl, 5 mmol l1 DTPA, 10 mmol l1 ascorbic acid, and 5% trichloroacetic acid (TCA)) using a glass piston homogenizer for 1 min. The suspension was centrifuged at 14,000g and the resulting supernatant solution was centrifuged through a 0.45-mm microcentrifuge filter (Millipore Corp., USA), yielding a 5% deproteinized homogenate that was used for the OPA assay procedure. The following solutions were required to perform the OPA assay: redox quenching buffer (RQB) (20 mmol l1 HCl, 5 mmol l1 DTPA, 10 mmol l1 ascorbic acid); 5% TCA in RQB (TCARQB); 7.5 mmol l1 NEM in RQB; 1.0 mol l1 potassium phosphate (KPi) buffer (pH 7.0); 0.1 mol l1 KPi buffer (pH 6.9); 5.0 mg ml1 OPA in methanol. The OPA solution was prepared just before use. 0.1 mmol l1 GSH was prepared as a standard. In brief, a mixture of sample (or standard), TCA-RQB, NEM and 1 mol l1 KPi was incubated for 5 min at room temperature (20–21 1C), 0.1 mol l1 KPi and OPA were added and incubated for 30 min at room temperature before measuring the OPA-derived fluorescence at 365 nm excitation and 430 nm emission [12].
Blood glucose assay Blood glucose was determined by the glucose oxidase method (blood glucose assay kit, Shanghai Rongsheng Biotech Co., Shanghai, China).
Plasma insulin determination The insulin level in plasma was estimated using an insulin radioimmunoassy kit (Huaxi Institute of diabetic technology, Chengdu, Sichuan Province, China).
ARTICLE IN PRESS X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
Statistical analysis
263
Data were expressed as mean7standard deviation (SD). Statistical analysis was performed using the statistical package ‘‘SPSS 12.0 for Windows’’. A oneway analysis of variation (ANOVA) was performed after ascertainment of the normality of distribution (Kolmogorov–Smirnov test) and the homogeneity of variance (Levene test) of the experimental data. If both conditions were fulfilled, differences between means were examined by Tukey’s test. If homogeneity of variance could not be ensured, differences between means were evaluated by Dunnett-T3 test. po0:05 was considered significantly different.
Blood glucose level (mmol/L)
35
30
25 a
20 b
15
10
5
0 0
1
2
3
4
Week
Results Changes in the mean body weight and the mean blood glucose level of the four groups are shown in Figs. 1 and 2, respectively. The body weight of the selenite-treated diabetic group started to be higher than that of the diabetic control group from week 3 on, but the difference only reached significance at week 4 (po0:05) (Fig. 1). Accordingly, from week 3 on, the glucose level of the selenite-treated diabetic group was significantly lower compared to the untreated diabetic group, suggesting that an oral administration of 2 mg kg1 selenite effectively decreases the hyperglycemia of alloxan-induced diabetic mice. However, after 4 weeks,
44 42 40 a
36
Body weight (g)
a
a
38
a a
34
a
a
32 b
30 b
a
28 26
b
24
b
c
b
b b
22 20 18 16 0
1
2
3
4
Week
Fig. 1. Body weight changes of the normal and diabetic mice treated and untreated with selenite for 4 weeks. ’, normal control; K, diabetic control; ., selenite-treated normal group; ,, selenite-treated diabetic group; n ¼ 10 in all groups. Values of the same week marked with different letters (a, b, c) were significantly different from each other as determined by Tukey’s test (po0:05).
Fig. 2. Blood glucose level changes of the normal and diabetic mice treated and untreated with selenite for 4 weeks. ’, normal control; K, diabetic control; ., selenite-treated normal group; ,, selenite-treated diabetic group; n ¼ 10 in all groups. In the selenite-treated diabetic group, at week 3 and week 4, the blood glucose level (marked with a and b) was significantly lower than in the diabetic control group at the same time (a: po0:05; b: po0:01).
the selenite treatment could not restore the hyperglycemia of diabetic mice to a normal level (Fig. 2). Results of the plasma insulin assay show that the hypoglycemic effect of selenite treatment on the diabetic mice was not due to an increase of pancreatic insulin release stimulated by selenite, because there were no differences in the plasma insulin levels between the selenite-treated diabetic group and the diabetic group (Fig. 3). On the contrary, the plasma insulin level of the selenite-treated normal group was lower than that of the control group. Fig. 4 shows GSH levels in whole blood of the four groups. In the diabetic group, the blood GSH concentration was markedly lower than that in the control group. However, the blood GSH content of the selenitetreated diabetic group and the diabetic group was at the same level, and so was the GSH level in the selenitetreated normal group and the control group. This demonstrates that selenite treatment had no influence on the GSH level in diabetic or normal mice. GSH levels in tissues of the four groups are shown in Fig. 5. In the diabetic group, GSH contents were generally lower than in the control group, but significant decreases were observed only in kidney, liver and testis (Figs. 5B, C and E). Selenite treatment of diabetic mice increased GSH levels in liver and brain: there were significant differences between the GSH content in brain and liver of the selenite-treated diabetic group and the diabetic group (Figs. 5A and C). Selenite treatment had no impact on spleen both for the diabetic group and the normal group (Fig. 5D).
ARTICLE IN PRESS 264
X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
Plasma insulin level (microIU/ml)
16
b
14 12
a a
a
10 8 6 4 2 0
DM+Se
Normal+Se
DM
Normal
Groups
GSH levels in the whole blood (micromol/ml)
Fig. 3. Plasma insulin levels of the normal and diabetic mice (DM) treated and untreated with selenite for 4 weeks; n ¼ 10 in all groups. Values marked with different letters (a, b) were significantly different from each other as determined by Tukey’s test (po0:05).
40
b
b
a
a
35 30 25 20 15 10 5 0
DM+Se
Normal+Se
DM
Normal
Groups
Fig. 4. GSH levels in whole blood of the normal and diabetic mice (DM) treated and untreated with selenite for 4 weeks; n ¼ 10 in all groups. Values marked with different letters (a, b) were significantly different from each other as determined by Tukey’s test (po0:05).
Discussion It has been proved that hyperglycemia generates oxidative stress leading to the development of diabetic complications [13]. However, the reported results about the relationship between diabetes and GSH level are
inconsistent. Several investigators have found lower GSH concentrations in erythrocytes, plasma, aorta and lenses in diabetic compared to normal subjects [3–8,14], while others have reported no changes of GSH levels in human erythrocytes [15], or in diabetic rats’ erythrocytes, liver, muscle, testis and lense [16–18]. The present study shows that alloxan-induced diabetes can decrease GSH levels in blood and some tissues such as kidney, liver and testis (Figs. 4 and 5B, C, E). A possible explanation for the discrepancy might be the different degrees and durations of hyperglycemia, some cases showing ketonemia resulting in a low glutahione level [3]. In our study, we did not measure the levels of the ketone body. However, all diabetic mice had, on an average, high blood glucose levels (30.674.1 mmol l1 at the beginning; 24.174.2 mmol l1 at the end of the experiment), in such a way that hyperglycemia would probably cause ketonemia in 4 weeks. A further study needs to confirm this. It is well established that selenium has biological functions as an antioxidant [19], and its insulin-like properties have also been documented in vitro and in vivo [9,10,20]. Therefore, we deduced that selenium might have some impact on GSH levels in diabetic mice. Indeed, in this study, we found that selenite treatment of diabetic mice significantly increased GSH levels in liver and brain, but not in blood, kidney, spleen and testis. However, selenite treatment of normal mice markedly decreased GSH levels in the liver compared to the controls, suggesting that 2 mg kg1 selenite oral administration had a toxic effect on normal mice. A similar beneficial effect of selenite treatment in the liver of diabetic rats was also observed by another group [16], who, on the other hand, reported a liver GSH content at the same level for the normal and diabetic animals. On considering the restoration of GSH levels in the liver of diabetic mice by selenite treatment, it seems that diabetes decreased the toxicity of a high dose of selenite in mice. The mechanism underlying this phenomenon needs further study. Regarding the glucose lowering effect of selenium on type II diabetic animals, the dose of selenite used in the present study was 2 mg kg1 day1 (11.6 mmol kg1 day1), which was in accord with McNeill et al.’s studies (10–15 mmol kg1 day1) [10,21], but in conflict with another report in which oral administration of 80 mg kg1 day1 (0.46 mmol kg1 day1) selenite was found to be effective in 2 weeks in lowering hyperglycemia of Kun-Ming mice with diabetes induced by alloxan [22]. In addition, we observed that the hypoglycemic effect of selenium on diabetic mice was not due to an increase of insulin release from pancreatic islets stimulated by selenite (Fig. 3). On the contrary, selenite reduced insulin release of pancreatic islets in normal mice: the plasma insulin level in the selenite-treated normal group was markedly lower than in the controls (Fig. 3).
ARTICLE IN PRESS X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
265
3.5
A
5
2.5 ac
2.0 1.5
bc
bc
1.0
GSH level in liver (micromol/g)
GSH level in brain (micromol/g)
3.0 a
C
a
a b
4
3 c 2
1
0.5 0
0.0 DM+Se
DM
Normal+Se
DM+Se
Normal
DM
Normal+Se
Groups
Normal
Groups 4.5
5.5
b
4.5 b
4.0 3.5
a
3.0 2.5
a
2.0 1.5 1.0
a
D
4.0
B
a
3.5
GSH level in spleen (micromol/g)
GSH level in kidney (micromol/g)
5.0
3.0
a
a
2.5 2.0 1.5 1.0 0.5
0.5
0.0
0.0 DM+Se
DM
Normal+Se
DM+Se
Normal
Normal+Se
6.5
DM
Normal
Groups
Groups E
6.0
GSH lelvel in testis (mciromol/g)
5.5 5.0 bc
4.5 4.0 3.5 ac
3.0 2.5 2.0
aa
a
1.5 1.0 0.5 0.0 DM+Se
Normal+Se
DM
Normal
Groups
Fig. 5. GSH levels in brain, kidney, liver, spleen and testis of the normal and diabetic mice treated and untreated with selenite for 4 weeks; n ¼ 10 in all groups. (A) brain, (B) kidney, (C) liver, (D) spleen, (E) testis. Values marked with different letters (a, b, c) were significantly different from each other as determined by Tukey’s test or Dunnett-T3 test (po0:05).
ARTICLE IN PRESS 266
X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
Recently, Mueller et al. [23] have found that selenate corrected glucose tolerance and lowered insulin resistance in type II diabetic dbdb mice, but selenite did not, therefore assuming that selenite might possess no peripheral insulin mimicking properties. However, our study has shown that selenite had a hypoglycemic effect on type II diabetic mice not due to a pancreatic insulin secretion increased by selenite treatment (Fig. 3), suggesting that selenite has peripheral insulin mimicking properties. To clarify this issue, more studies are needed, especially in vitro. In our previous study [24], a higher dose of selenite (4 mg kg1 day1) was used to trigger the glucose-lowering effect quickly (within a week), but failed to maintain this insulin-like effect on the alloxan-induced diabetic mice more than a week due to the toxicity of selenite. We also observed that more selenium was deposited in the liver and brain of the diabetic mice. In the present study, we found that a selenite treatment of 2 mg kg1 day1 improved the GSH concentration only in the liver and brain of diabetic rats, implying that selenium exerts its antioxidant properties beneficial to diabetic mice in liver and brain, in addition to its insulinlike properties. The question whether selenium has the same insulin-mimetic properties in the brain or the central nervous system needs further investigation. In summary, alloxan-induced diabetes decreased GSH levels in blood, kidney, liver and testis in KunMing mice, and oral administration of 2 mg kg1 day1 selenite improved the liver and brain GSH levels in the diabetic mice. Considering a lack of effect or toxicity of selenite in blood, kidney, spleen and testis on the GSH levels, more studies are needed to assess the beneficial effect of selenium in diabetes.
Acknowledgements We would like to give our thanks to associate professor Qiong Liu for revising the English language.
References [1] Will Y. Overview of glutathione function and metabolism. In: Maines MD, editor. Current protocols in toxicology. New York: Wiley; 1999. p. 6.1.1–6.1.18. [2] Dickinson DA, Forman HJ. Cellular glutahione and thiol metabolism. Biochem Pharmacol 2002;64:1019–26. [3] Jain SK, McVie R. Hyperketonemia can increase lipid peroxidation and lower glutathione levels in human erythrocytes in vitro and in type 1 diabetic patients. Diabetes 1999;48:1850–5. [4] Jain SK, McVie R, Smith T. Vitamin E supplementation restores glutathione and malondialdehyde to normal concentrations in erythrocytes of type 1 diabetic children. Diabetes Care 2000;23:1389–94.
[5] Samiec PS, Drews-Botsch C, Flagg EW, Kurtz JC, Sternberg Jr P, Reed RL, Jones DP. Glutathione in human plasma: decline in association with aging, agerelated macular degeneration, and diabetes. Free Radic Biol Med 1998;24:699–704. [6] Mohan IK, Das UN. Effect of L-arginine-nitric oxide system on chemical-induced diabetes mellitus. Free Radic Biol Med 1998;25:757–65. [7] Yilmaz O¨, O¨zkan Y, Yildirim M, O¨ztu¨rk AI, Ersan Y. Effect of alpha lipoic acid, ascorbic acid-6-palmitate, and fish oil on the glutathione, malonaldehyde, and fatty acids levels in erythrocytes of streptozotocin induced diabetic male rats. J Cell Biochem 2002;86:530–9. [8] Tachi Y, Okuda Y, Bannai C, Bannai S, Shinohara M, Shimpuku H, Yamashita K, Ohura K. Hyperglycemia in diabetic rats reduces the glutathione content in the aortic tissue. Life Sci 2001;69:1039–47. [9] Ezaki O. The insulin-like effects of sodium selenate in rat adipocytes. J Biol Chem 1990;265:1124–8. [10] McNeill JH, Delgatty HLM, Battell ML. Insulinlike effects of sodium selenate in streptozotocin diabetic rats. Diabetes 1991;40:1675–8. [11] Mukherjee B, Anbazhagan S, Roy A, Ghosh R, Chatterjee M. Novel implications of the potential role of selenium on antioxidant status in streptozotocininduced diabetic mice. Biomed Pharmacother 1998; 52:89–95. [12] Senft AP, Dalton TP, Shertzer HG. Determining glutathione and glutathione disulfide using the fluorescence probe o-phthalaldehyde. Anal Biochem 2000;280:80–6. [13] Baynes JW. Role of oxidative stress in development of complications in diabetes. Diabetes 1991;40:405–11. [14] Mitton KP, Dean PAW, Dzialoszynski T, Xiong H, Sanford SE, Trevithick JR. Modelling cortical cataractogenesis 13. Early effects on lense ATP/ADP and glutathione in the streptozotocin rat model of the diabetic cataract. Exp Eye Res 1993;56:187–98. [15] Sinclair AJ, Girling AJ, Gray L, Lunec J, Barnett AH. An investigation of the relationship between free radical activity and vitamin C metabolism in elderly diabetic subjects with retinopathy. Gerontology 1992;38:268–74. [16] Naziroglu M, Cay M. Protective role of intraperitoneally administered vitamin E and selenium on the antioxidative defense mechanisms in rats with diabetes induced by streptozotocin. Biol Trace Elem Res 2001;79:149–59. [17] Unlucerci Y, Bekpinar S, Kocak H. Testis glutathione peroxidase and phospholipid hydroperoxide glutathione peroxidase activities in aminoguanide-treated diabetic rats. Arch Biochem Biophys 2000;379:217–20. [18] Naziroglu M, Dilsiz N, Cay M. Protective role of introperitoneally administered vitamine C and E and selenium on the levels of lipid peroxidation in the lens of rats made diabetic with streptozotocin. Biol Trace Elem Res 1999;70:223–32. [19] Xu H, Huang K. Selenium: its chemistry, biochemistry and application in life science. China: The publishing press of HUST; 1994. p. 152–167. [20] Becker DJ, Reul B, Ozcelikay AT, Buchet JP, Henquin JC, Brichard SM. Oral selenate improves glucose homeostasis and partly reverses abnormal expression of liver
ARTICLE IN PRESS X.-Q. Sheng et al. / Journal of Trace Elements in Medicine and Biology 18 (2005) 261–267
glycolytic and gluconeogenic enzymes in diabetic rats. Diabetologia 1996;39:3–11. [21] Battell ML, Delgatty HLM, McNeill JH. Sodium selenate corrects glucose tolerance and heart function in STZ diabetic rats. Mol Cell Biochem 1998;179:27–34. [22] Ding H, Pen R, Chen J, Li Q, Wang H, Kong R. Effect of selenite and allitridi on the diabetic mice. Acta Nutr Sin 1997;19:384–7.
267
[23] Mueller AS, Pallauf J, Rafael J. The chemical form of selenium affects insulinomimetic properties of the trace element: investigations in type II diabetic dbdb mice. J Nutr Biochem 2003;14:637–47. [24] Sheng X-Q, Huang K-X, Xu H-B. New experimental observation on the relationship of selenium and diabetes mellitus. Biol Trace Elem Res 2004;99: 241–53.