Pathophysiology/Complications ORIGINAL ARTICLE Vitamin E Supplementation Restores Glutathione and Malondialdehyde to Normal Concentrations in Erythrocytes of Type 1 Diabetic Children SUSHIL K. JAIN, PHD ification of intracellular peroxides. Thus, ROBERT MCVIE, MD maintenance of glutathione level is pivotal TINEY SMITH, RN for cellular defense against oxidative injury and for cellular integrity (1,2). Hyperglycemia in diabetes can generate free radicals, hydrogen peroxide, and reac- tive ketoaldehydes by the auto-oxidation of OBJECTIVE — This study examined the relationship between cellular glutathione and vita- glucose or from glycated proteins (3–8). ␣ min E concentrations and the effect of vitamin E ( -tocopherol) supplementation on glutathione Many investigators have reported lower con- and lipid peroxidation product concentrations in the erythrocytes of type 1 diabetic patients. centrations of glutathione in the erythro- RESEARCH DESIGN AND METHODS — We obtained written informed consent to cytes, aorta, and lenses of diabetic patients participate in this study from diabetic patients (n = 29) and their age-matched nondiabetic sib- compared with healthy subjects (9–19), lings (n = 21) according to the guidelines of the Institutional Review Board on Human Exper- except for some studies that reported similar imentation. Diabetic patients were supplemented with a DL-␣-tocopherol (vitamin E) capsule glutathione concentrations (20). In addi- (100 IU/ orally) or placebo for 3 months in a double-blind clinical trial. Fasting blood samples tion, an increase in malondialdehyde con- were collected from each diabetic patient before the start of and after the 3 months of vitamin centration has been reported in erythrocytes E or placebo supplementation. Glutathione, malondialdehyde (which is a product of lipid per- and other tissues of diabetic animals and oxidation), and ␣-tocopherol were determined using high-performance liquid chromatography. patients (21–28). A total of 5 diabetic patients were excluded after randomization from the data analyses. Data Lower glutathione and elevated lipid were analyzed statistically using a paired Student’s t test to compare 12 diabetic patients tak- peroxidation concentrations are risk factors ing vitamin E with 12 diabetic patients receiving placebo supplementation and to compare dia- betic patients with healthy nondiabetic subjects. for the development of pathological states such as retinopathy, neuropathy, cataracts, RESULTS — Erythrocytes of diabetic patients had 21% higher (P Ͻ 0.001) malondialdehyde and atherosclerosis (1–3). Vitamin E has and 15% lower (P Ͻ 0.05) glutathione concentrations than healthy subjects. Vitamin E in ery- been proposed to be the major lipid-soluble throcytes had a significant correlation with the glutathione concentrations in the erythrocytes chain-breaking antioxidant and protects (r = 0.46, P Ͻ 0.02). Vitamin E supplementation increased glutathione concentrations by 9% biologic membranes from lipid peroxida- Ͻ Ͻ (P 0.01) and lowered concentrations of malondialdehyde by 23% (P 0.001) and of HbA1c tion (29). In vitro studies have shown that by 16% (P Ͻ 0.02) in erythrocytes of diabetic patients. No differences were evident in these reduced glutathione can protect against per- parameters before versus after placebo supplementation. oxidation of lipids in cytosolic and particu- late subfraction components of rat liver and CONCLUSIONS — Glutathione level is significantly related to vitamin E level, and sup- other tissues (29). In some experiments, plementation with vitamin E (100 IU/day) significantly increases glutathione and lowers lipid ␣ peroxidation and HbA concentrations in the erythrocytes of type 1 diabetic patients. this protection depends on membrane - 1c tocopherol (30), whereas in others it does Diabetes Care 23:1389–1394, 2000 not (31). Mechanisms that have been pro- posed to explain the glutathione effect include the removal of species that initiate lipid peroxidation, scavenging of radicals by educed glutathione is a major intra- reduced form, the oxidation of which can a glutathione-dependent protein (32), scav- cellular nonprotein sulfhydryl com- otherwise cause altered cellular structure enging of peroxy radicals by glutathione pound. Glutathione has many bio- and function (1,2). Glutathione is also a R (33), maintenance of membrane protein logic functions, including maintenance of cofactor for many enzymes such as glu- thiols by glutathione (34), the protection by membrane protein sulfhydryl groups in the tathione peroxidase, which catalyzes detox- glutathione and ␣-tocopherol of a glu- tathione S-transferase isozyme responsible From the Department of Pediatrics, Louisiana State University Health Sciences Center, Shreveport, Louisiana. for the reduction of lipid hydroperoxides Address correspondence and reprint requests to Sushil K. Jain, PhD, Department of Pediatrics, Louisiana (35), and a glutathione-dependent protein State University Health Sciences Center, 1501 Kings Highway, Shreveport, LA 71130. E-mail: that recycles ␣-tocopherol from the ␣-toco- [email protected]. pheroxy radical (31). Received for publication 4 November 1999 and accepted in revised form 23 May 2000. Abbreviations: HPLC, high-performance liquid chromatography; TBA, thiobarbituric acid. In vivo studies in animal models have A table elsewhere in this issue shows conventional and Système International (SI) units and conversion shown inhibition of lipid peroxidation and factors for many substances. increased glutathione concentrations in the DIABETES CARE, VOLUME 23, NUMBER 9, SEPTEMBER 2000 1389 Vitamin E supplementation in diabetes ␣ Table 1—Age, duration of diabetes, HbA1c, glutathioine, malondialdehyde, and -tocopherol nondiabetic subjects (healthy siblings) were concentrations in erythrocytes of nondiabetic subjects and diabetic patients also enrolled to serve as healthy control subjects, and data from healthy control subjects were used for the cross-sectional Diabetic patients Nondiabetic subjects P aspect of the study. Fasting blood samples n 29 21 — were collected into tubes with and without Age (years) 12.4 ± 0.7 10.9 ± 0.9 NS EDTA before and after the vitamin E or Duration of diabetes (years) 5.3 ± 0.7 — — placebo supplementation from each Ͻ Blood HbA1c (%) 8.44 ± 0.21 5.03 ± 0.14 0.001 patient. Erythrocytes were separated and Glutathione erythrocytes (µmol/g Hb) 5.8 ± 0.2 6.4 ± 0.2 Ͻ0.05 washed from the EDTA-treated blood (21). Malondialdehyde erythrocytes (nmol) All analyses were conducted immediately Per milliliter packed cell volume 1.46 ± 0.04 1.30 ± 0.04 Ͻ0.02 after blood collection. Per gram Hb 4.27 ± 0.13 3.67 ± 0.14 Ͻ0.03 Glutathione was determined using Per micromole total lipid 0.29 ± 0.008 0.24 ± 0.006 Ͻ0.01 high-performance liquid chromatography ␣-Tocopherol erythrocytes (nmol) (HPLC) with a Lichrosphere (EM Separation Per milliliter packed cell volume 4.38 ± 0.13 4.11 ± 0.13 NS Technology, Gibbstown, NJ) 100 NH2 col- Per gram Hb 12.88 ± 0.40 12.00 ± 0.40 NS umn, a gradient system using buffer A (80% Per micromole total lipid 0.87 ± 0.03 0.75 ± 0.03 NS methanol and 20% water) and buffer B (4 mol/l acetate, pH 4.5 in 64% methanol- Data are means ± SEM unless otherwise indicated. Note significant differences in the blood HbA1c and in glu- tathione and malondialdehyde erythrocytes concentrations in erythrocytes of diabetic patients versus healthy water), and a ultraviolet/visible detector set nondiabetic subjects. Details of statistical analyses are discussed in RESEARCH DESIGN AND METHODS. at 365 nm (40). Vitamin E as ␣-tocopherol was measured by HPLC (23). Freshly obtained erythrocytes were stored in 2% liver and heart of guinea pigs after dietary RESEARCH DESIGN AND pyrogallol in ethanol at Ϫ70°C. All samples supplementation (150 mg/kg diet) with vit- METHODS — Informed written con- were analyzed within 1 month of storage amin E (36,37). Similarly, subcutaneous sent of all patients was obtained in accor- using a reverse-phase C-18 column (Waters, treatment with vitamin E (10 mg/100 kg dance with the protocol approved by the Milford, MA), a 95% methanol solvent sys- body wt) increased glutathione concentra- Institutional Review Board on Human tem, and a uv/vis detector set at 292 nm tions and reduced lipid peroxidation in Experimentation. No specific criterion was (23). Lipid peroxidation was determined by renal tissues of adult rats (38). Increases in used for inclusion or exclusion of patients measuring malondialdehyde, which is an glutathione level were reported in the ery- in this study except that patients who had end product of lipid peroxidation. Malondi- throcytes, aqueous humor, and lenses of other disorders such as sickle cell disease or aldehyde can react with thiobarbituric acid rabbits supplemented orally with 30 mg thyroid disorders or those taking other (TBA), and the concentration of malondi- vitamin E и dayϪ1 и kgϪ1 body wt for 10 or medications were excluded from the study. aldehyde-TBA complex was determined by 20 days (39). In humans, increases in glu- All patients agreed to participate in this its separation with an HPLC system (Waters) tathione level were shown in the erythro- study. Diabetic patients who were invited and a uv/vis detector set at 532 nm (41). cytes and aqueous humor after high-dose and agreed to participate in this study were Variation in the assay of the same sample on oral supplementation of 1,000 IU vitamin E asked to come to the clinic after overnight different days was Ͻ7%. for 10 days in 20 subjects hospitalized