Race-Specific Differences in Antioxidant Enzyme Activity in Patients With Type 2 Diabetes: A potential association with the risk of developing nephropathy

Patients with type 2 diabetes attended the outpatient department of the Whittington Hospital, which serves a multiracial population of 155,000 in North London. In the assignment of racial origin, patients were considered to be of black African origin (Caribbean and African) if both parents were native to either African or Caribbean countries. Caucasian patients were confirmed as such if both parents originated from Western European countries. A diagnosis of type 2 diabetes was based on an absence of ketosis at, or need for insulin within, 1 year of diagnosis. Microalbuminuria was diagnosed if albumin-to-creatinine ratios were ≥3 mg/mmol in at least two sequential, sterile, early-morning urine samples and if the urinary albumin excretion rate was between 30 and 300 mg/day in a 24-h urine collection. A history of cardiovascular disease was determined using the Rose questionnaire (12).

Nondiabetic control plasma samples were obtained from subjects who took part in the Wandsworth Heart and Stroke Study, a cross-sectional study of cardiovascular risk factors in racial minority groups carried out in South London. In this study, individuals of African and Caucasian racial origin (assigned as above) were randomly sampled from a population of 190,000 in the Wandsworth Health Authority, as previously described (13). A 75-g oral glucose tolerance test was performed to exclude glucose intolerance in those not known to have diabetes or glycosuria (13,14). Clinical and biochemical data were collected from two sex-, age-band (within 5 years)-, and race-matched control subjects without glucose intolerance for each patient with type 2 diabetes. The study was approved by the ethics committees of the Whittington Hospital and the Wandsworth Health Authority.

BMI was calculated from weight in kilograms divided by height in meters squared. Mean systolic and diastolic blood pressure were calculated according to a standardized protocol after at least 10 min rest using a validated automated machine and appropriate cuff. In patients with diabetes, direct fundoscopy was performed after the pupils were dilated with tropicamide. Retinopathy status was recorded as either present (background, preproliferative, or proliferative) or absent.

Fasting venous blood was taken from an antecubital vein. HbA_1c (A1C) was measured by a high-performance liquid chromatography system (Menarini 8140; Menarini Diagnostics, Wokingham, U.K.). Total and HDL cholesterol and total triglycerides were estimated using enzymatic methods (Boehringer-Mannheim, Mannheim, Germany). Urinary albumin and creatinine were measured by immunoturbidimetry (Cobas Fara; Roche) and the Jaffe rate reaction methods, respectively. The samples were coded in order to blind the laboratory staff to the patients’ sex, age, racial origin, diabetes, and comorbidity status. Renal function was determined from the Cockcroft and Gault formula and termed the estimated glomerular filtration rate corrected for a body surface area of 1.73 m⁻².

GPx activity was measured using a coupled assay system. The oxidation of reduced glutathione was coupled to NADPH oxidation in a reaction catalyzed by glutathione reductase. A total of 500 μl Tris-HCl (0.2 mmol/l, pH 8) was mixed with 100 μl NADPH (2 mmol/l) followed by 100 μl EDTA (5 mmol/l) and 100 μl glutathione (20 mmol/l). Plasma samples (100 μl) were then mixed with glutathione reductase (0.002 units) and incubated for 5 min at 37°C and for a further 5 min after the addition of 100 μl of tertiar-butyl hydroperoxide (0.7 mmol/l). Absorbances were read in a spectrophotometer (Shimadzu UV 240; Shimadzu, Kyoto, Japan) at 340 nm. A unit of GPx activity was defined as being equivalent to the oxidation of 1 μmol of NADPH per second at 37°C (15).

Plasma SOD activity was measured in 500-μl aliquots of plasma treated with 300 μl chloroform and 500 μl ethanol. The samples were then centrifuged at 18,000g for 30 min, and 50 μl of the supernatant were removed and mixed with 900 μl of an SOD reagent (0.1 mmol/l xanthine, 0.1 mmol/l EDTA, 50 mg BSA, 25 mmol/l NBT (nitro blue tetrazolium), and 40 mmol/l Na₂CO₃ (pH 10.2). Xanthine oxidase (10 μl of solution containing 25 units in 0.8 ml of 2 mol ammonium sulfate) was then added, and the samples were incubated for 20 min at 25°C. The reaction was stopped by the addition of 1 ml CuCl₂ (0.8 mmol/l), and the absorbance of the samples was measured at 560 nm (16).

Plasma samples (100 μl) were placed in glass tubes with 100 ηg γ-tocopherol in 500 μl ethanol as internal standard and mixed with 500 μl water and 1 ml n-hexane. The upper (hexane) layer was aspirated after centrifugation, and lipid extraction was performed twice. The solvent was evaporated to dryness under a stream of nitrogen. The residue was dissolved into 100 μl acetonitrile, and 20 μl was then injected on to an high-performance liquid chromatography system (Gilson, Anachem, Luton, Scotland) composed of a model 305 pump, Mixer 805 Monometric module, and Fluorometer model 121. Tocopherols were separated by reverse-phase chromatography on an ODS column (10 cm × 5 mm; particle size 5 μm; Chrompack, Middelburg, the Netherlands) using acetonitrile/water (80/20 [vol/vol]) at a flow rate of 0.7 ml/min as mobile phase. α-Tocopherol was detected using excitation and emission wavelengths at 295 nm and 340 nm, respectively (4). α-Tocopherol levels were corrected for total cholesterol concentrations.

Continuous data were analyzed with parametric or nonparametric tests according to their distribution and categorical data with the χ² test, using SPSS version 10.1 for Windows (Chicago, IL). Skewed data were log transformed before analyses. Regression analyses were carried out with antioxidant enzyme activity as the dependent variable. All analyses were two tailed, and a P value <0.05 was accepted as being statistically significant. Data are presented as means ± SD, unless otherwise stated.

Clinical and biochemical data for the study population are shown in Table 1. Patients with type 2 diabetes were older, had a greater BMI, and had higher systolic blood pressure compared with the control population. Men and women without diabetes had significantly higher total cholesterol but lower triglycerides compared with patients with type 2 diabetes. Differences in antioxidant enzyme activity between the patients with and without type 2 diabetes are shown in Fig. 1.

The clinical and biochemical characteristics of the patients with type 2 diabetes according to racial origin are shown in Table 2. The patients were well matched for age and had equivalent BMI and glycemic control. Similar numbers of patients were receiving either diet alone/oral hypoglycemic agents or insulin-based regimens for treatment of their diabetes in the Caucasian and African groups (4/27/13/1 and 2/13/12/3, respectively, P = 0.30). The frequency of cardiovascular disease (stroke, ischemic heart disease) and treatment for hypertension was similar in the Caucasian and African groups (13 vs. 5, P = 0.23 and 25 vs. 19, P = 0.14). Peripheral vascular disease was reported in 15% (n = 6) of the Caucasian group and none of the African patients (P = 0.04).

Patients of African origin with type 2 diabetes had significantly higher diastolic blood pressure, HDL cholesterol, and plasma creatinine but lower fasting triglycerides compared with the Caucasian group. The concentration of α-tocopherol unadjusted and adjusted for total cholesterol was lower in the African patients (31.7 ± 9.7 vs. 37.8 ± 13.7 μmol/l, P = 0.05 and 6.2 ± 1.64 vs. 7.81 ± 2.78 (μmol/l)/(mmol/l) P = 0.01, respectively). SOD activity was higher and GPx activity lower in African compared with Caucasian patients (Fig. 2).

Patients of African origin with normal urinary albumin excretion had significantly higher plasma creatinine concentrations (100.7 ± 14.2 vs. 88.1 ± 14.9 μmol/l, P = 0.007) and lower GPx activity (99.0 ± 72.4 vs. 173.7 ± 107.4 units/l, P = 0.02) compared with those of Caucasian origin. Patients in these respective racial groups with microalbuminuria had similar concentrations of plasma creatinine and levels of GPx activity (123.9 ± 42.4 vs. 104.9 ± 27.3 μmol/l, P = 0.15 and 160.9 ± 82.5 vs. 152.8 ± 80.5 units/l, P = 0.81). Urinary albumin excretion correlated with GPx activity in patients of African origin (Fig. 3).

Regression analyses were performed for SOD and GPx activity separately as independent variables. For patients with diabetes, the dependent variables of each model entered were age, sex, systolic blood pressure, estimated glomerular filtration rate, A1C, and race. African race remained a significant and independent determinant of antioxidant enzyme activity in the patients with diabetes (Tables 3 and 4). In a model including the whole cohort of patients with and without type 2 diabetes, race remained an independent predictor of SOD activity (P = 0.015).

Superoxide is considered an important free radical contributing to oxidative stress (17). It is dismuted to hydrogen peroxide, a much less harmful product, by the family of SOD enzymes. The subsequent conversion of hydrogen peroxide to water is dependent on GPx activity. Recent experimental data have shown that increased superoxide production due to increased NADPH oxidase activity can be effectively suppressed by α-tocopherol supplementation (18). Relatively lower α-tocopherol concentrations and GPx activity could therefore explain the higher rates of lipid hydroperoxides in patients of African origin reported earlier (11).

Plasma GPx activity is largely derived from the renal tubular epithelium (19). Both plasma and erythrocyte GPx activity change in relation to the evolution of renal disease (20–22). Activity progressively rises in the phase of incipient to established renal failure and then declines with the loss of functional renal mass to end-stage renal disease (23). Our data suggest that patients of African origin with reduced GPx activity already have a lower functional renal mass compared with Caucasian patients before the development of incipient nephropathy. Of note, GPx activity was higher in patients with microalbuminuria in this racial group, suggesting a preserved response to the development of incipient renal disease or microvascular damage. These data suggest that the differential regulation of reduction/oxidation (redox) pathways could play a role in renal disease susceptibility.

Greater deposition of the products of oxidative stress in the kidney is linked to faster rates of functional decline (24–27). Oxidative stress also contributes to hemodynamic alterations that occur with and promote diabetic nephropathy (28,29). An elevation in circulating lipid hydroperoxide activates the p38 mitogen-activated kinase in platelets, promoting their activation and aggregation and thereby predisposing to microvascular ischemia (30). Superoxide neutralizes nitric oxide and increases formation of the toxic free radical peroxynitrite and impairs vascular dilatation (30). This could explain the blunted renal vasodilatory responses in healthy African Americans and patients of African origin with type 2 diabetes and microalbuminuria compared with matched Caucasian control subjects (31,32).

Central to the origin and development of vascular complications in diabetes is impaired endothelial function and reduced insulin sensitivity. This syndrome, like aging, could be dependent upon a progressive reduction in antioxidant enzyme activity (33,34). In our study, patients with type 2 diabetes of African origin had higher SOD and lower GPx activity, which, together with reduced total triglycerides and raised HDL cholesterol, suggests relatively higher insulin sensitivity than in Caucasians. Of note, treatment that improves insulin sensitivity also increases SOD activity in patients with diabetes (35). Paradoxically, overexpression of GPx activity could promote the development of insulin resistance through impaired insulin receptor signaling due to faulty lipid hydroperoxide–dependent activation of serine/threonine kinase Akt (36). We have not determined whether the patients of African origin have a relatively higher insulin sensitivity. However, together these data suggest that differences in the magnitude of antioxidant defense (and possibly insulin resistance) could be relevant to the development, or not, of the vascular complications of diabetes.

It remains unclear if these differences in antioxidant enzyme activity are inherent or due to environmental factors. The racial differences in lipid hydroperoxide that we reported earlier could also be dependent on free fatty acid constitution. In a preliminary study, we found higher levels of the polyunsaturated fat eicosapentanoic acid in patients of African origin with type 2 diabetes than in Caucasians (data not shown). This could contribute to higher levels of lipid hydroperoxide, affect the consumption of vitamin E, and thereby modulate GPx activity. In addition, treatment of hypercholesterolemia with statins has variable effects on antioxidant enzyme activity (37,38). We have no evidence to suggest that differences in the usage of these or other drugs that can modify antioxidant enzymes or dietary factors account for our observations. Data showing that cultured endothelial cells derived from African patients release greater quantities of superoxide and peroxynitrite radicals compared with Caucasians are consistent with our findings and imply an inherent basis for the racial differences in redox metabolism (39). Further studies are needed to explore whether treatments that can potentially alter these antioxidant enzymes will modify the susceptibility to or occurrence of diabetes complications.

This work was supported by grants awarded to K.A.E. by the St. George’s Charitable Trust and The Sir Jules Thorn Foundation. A list of the Wandsworth Heart and Stroke Study group is given elsewhere (12). The Wandsworth Heart and Stroke Study has received support from the former Wandsworth and South Thames Regional Health Authorities, the National Health Service Research and Development Directorate, the British Heart Foundation, the former British Diabetic Association, and the Stroke Association. F.P.C. and K.A.E. are members of the St. George’s Cardiovascular Research group. J.N.-Z. is supported by the British Heart Foundation.