Early Increase of Oxidative Stress and Reduced Antioxidant Defenses in Patients With Uncomplicated Type 1 Diabetes: A case for gender difference

We studied 37 patients with type 1 diabetes of short duration (19 men, 18 women) who were attending the Diabetic Outpatient Clinic of the Catholic University School of Medicine, and 29 control subjects (15 men, 14 women) who were comparable for age, gender, smoking habit, diet, and physical activity. Patients were included in the study only if their duration of disease was <10 years and if they had no history of diabetic complications, such as nephropathy, neuropathy, hypertension, cardiac ischemic disease, and peripheral vasculopathy, according to the criteria of the EURODIAB IDDM Complications Study (16), as reported in detail in a previous study (15). Further exclusion criteria included administration of drugs other than insulin (such as antihypertensive agents, aspirin, or vitamin supplements), history of smoking, pregnancy, and current illness (such as hepatic, cardiac, or renal disease). Neither diabetic nor control women were on estroprogestinic therapy. The characteristics of the study population are shown in Table 1.

Venous blood samples were collected after 12-h overnight fasting only from patients in whom values of glycemia were <125 mg/dl that morning. The samples were collected in sampling vials containing heparin for preparation of plasma. Plasma samples were stored under nitrogen at −80°C and used within 1 month for analysis of TRAP and measurement of CD and ROOH.

As previously reported (15,17), the assay is based on the quenching of the 2,2′-azinobis-(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) radical cation (ABTS · ⁺) by antioxidants contained in the sample. The ABTS · ⁺ formation was monitored continuously as an increased absorbance at 734 nm, at 20°C. The delay or inhibition time between the addition of H₂O₂ (time 0) and the onset of increased absorbance (ABTS · ⁺ formation) was measured. The assay was standardized using a water-soluble vitamin E analog (Trolox). Intra and interassay coefficients of variation for this method were 6.2 and 8.5%, respectively.

Plasma lipids were extracted by a modification of the method of Folch, as previously described (15,18).

As previously reported, CDs were measured by second-derivative spectrophotometry (15,19), which allows great sensitivity, because the CD shoulder that appears in the ordinary spectrum translates into sharp minima peaks, which are more easily measurable. Furthermore, this technique permits a discrimination between the different CD structures (e.g., cis-trans and trans-trans conformation) present in the sample (20). Minima at 232 and 246 nm were ascribed to the trans-trans and cis-trans CD isomers, respectively (21,22). These minima were quantified in arbitrary units as d²A/dλ². The CD ratio measurement at 246- and 232-nm minima of the derivative spectra, which can provide an indirect evaluation of plasma antioxidant capacity, was then calculated (19,20). Intra- and interassay coefficients of variation for this method were 8.2 and 11.4%, respectively.

As reported elsewhere (15), plasma ROOH content was determined with the FOX 2 assay for lipids (23,24). ROOH content in plasma samples was determined as a function of the mean absorbance difference of samples with and without elimination of ROOH by triphenylphosphine (TPP). Intra- and interassay coefficients of variation for this method were 6.5 and 8.3%, respectively.

Statistical calculations were performed using StatView software (version 4.01; SAS Institute, Cary, NC). Values are given as means ± SD. Comparisons between diabetic and control groups and within individual groups were made using ANOVA and Scheffe’s F test for multiple comparisons. Simple linear regression analysis was used to detect relationships among TRAP values, 246-nm CD, ROOH, and plasma HbA_1c as the independent variable.

As shown in Table 2, compared with control subjects, type 1 diabetic patients had significantly lower TRAP values, significantly higher ROOH levels, increased total CD levels (only in men), significantly lower 246-nm CD levels, and significantly augmented 232-nm CD levels. In type 1 diabetic patients of both sexes, the CD isomer ratio was directly correlated with TRAP (R = 0.468, P < 0.02) and inversely correlated with ROOH (R = −0.486, P = 0.003) (Fig. 1). No significant differences were found between patients with type 1 diabetes and control subjects in the concentration of total, HDL, and LDL cholesterol and triglycerides; in age; and in BMI.

As shown in Table 2, compared with diabetic men, women with type 1 diabetes had significantly higher ROOH levels, significant lower TRAP values, and significantly lower 246-nm CD levels, but there were no significant differences in total and 232-nm CD levels. No significant differences were found between men and women with type 1 diabetes in concentration of total, HDL, and LDL cholesterol and triglycerides, age, BMI, and insulin dosage.

As shown in Table 1, uric acid levels were significantly lower in type 1 diabetic patients than in control subjects, both in men with type 1 diabetes compared with male control subjects and in women with type 1 diabetes compared with female control subjects. Moreover, there was a significant difference between women and men with type 1 diabetes.

In type 1 diabetic patients, uric acid levels were directly correlated with TRAP (R = 0.484, P = 0.02) and inversely correlated with ROOH (R = −0.449, P = 0.03).

In a simple linear regression analysis between HbA_1c as the independent variable and TRAP, ROOH, and 246-nm CD, we did not find any significant relationship in type 1 diabetic patients.

In this study, we investigated whether oxidative stress occurs early during the course of disease in type 1 diabetic patients without diabetic complications, and whether any difference exists between diabetic men and women in this regard.

Compared with control subjects, our type 1 diabetic patients had significantly lower TRAP and 246-nm CD levels and significantly higher ROOH, 232-nm CD, and total CD levels, at least in men (Table 2). Therefore, diabetic patients have an early impairment of plasma antioxidant capacity and increased oxidative stress, even in the early stages of the disease. These data are in agreement with those of other authors (14), although they studied the oxidative stress by evaluating the enzymatic activity of superoxide dismutase and glutathione peroxidase in children and adolescents.

The decrease in 246-nm CD and the increase in 232-nm CD, expression of a decreased CD isomer ratio, are indicative of a large depletion of antioxidants in type 1 diabetic patients, because in physiological conditions, oxidative reactions are directed toward the formation of cis-trans CDs. In type 1 diabetes, the hydrogen-donating ability of plasma, as measured by TRAP, is decreased because of the consumption of antioxidants and because the kinetic equilibrium of the oxidation process is shifted toward trans-trans products, with an increase of the 232-nm CD. Interestingly, in our patients, the CD isomer ratio was positively correlated with TRAP and inversely correlated with ROOH, as shown in Fig. 1.

Our reported increase in plasma lipid ROOHs and CDs was not due to an increase in the content of circulating lipids, because there were no significant differences in the plasma lipid levels between normal and diabetic subjects.

We did not find any relationship between metabolic control, as evaluated by HbA_1c, antioxidant status, and markers of oxidative stress. This can be explained by the fact that HbA_1c is a “mean” evaluation of glycemic control and reflects, only in part, glycemic fluctuations, such as postprandial hyperglycemia or disglycemia states, which may induce a pro-oxidative status and may play a significant role in the pathogenesis of diabetic complications (10,11,25). In fact, a recent prospective study by Berg et al. (26) showed that intensive insulin treatment can reduce the increased ROOH levels in type 1 diabetic patients. Moreover, recent epidemiological studies (27) suggest that HbA_1c is related to total mortality in the general population, determining an increased risk even for values >5%.

Data from epidemiological surveys have shown a significant decrease of uric acid levels after the onset of type 1 diabetes (28). Uric acid has been proposed to play a pivotal role in the antioxidant defense systems in humans (29–31). The intriguing question of a role for uric acid in the free radical–scavenging processes in diabetes has received little attention. Elevated glucose levels seem to be the principal mechanism, because hyperglycemia has, at least in part, an osmotic diuretic effect, leading to an excessive excretion of uric acid (32). Some authors reported that even in type 1 diabetic patients with good metabolic control, plasma uric acid levels are not completely normalized (33), possibly because of the glomerular hyperfiltration that characterizes the first years of type 1 diabetes, and can contribute to the hyperuricosuria and hypouricemia found in type 1 diabetic patients (34).

In our study, we report not only decreased levels of uric acid in type 1 diabetic patients compared with control subjects but also, for the first time, a more evident decrease in uric acid levels in diabetic women compared with diabetic men. This fact, together with the positive relationship found between uric acid levels and antioxidant capacity, as shown in Fig. 1, confirms the importance of uric acid for the antioxidant defenses (35).

Many studies have shown that the protective effect of female gender against cardiovascular risk is lost in the presence of diabetes (5) and that the increased relative risk for CHD in diabetic women is independent from the conventional risk factors (3–4). Moreover, it has been reported recently that in diabetic women, coronary artery calcification, a measure of atherosclerotic plaque burden, is greatly increased and the gender difference is lost (36). Because little of this can be explained by known coronary risk factors, the observed increase in lipid ROOH along with the decrease in TRAP and uric acid levels in diabetic women compared with diabetic men could play an important role in the development of the atherosclerotic plaque and may represent one of the pathophysiological explanations for the higher cardiovascular risk of diabetic versus nondiabetic women (3–6,37).

Many studies support the role of estrogens in the primary and secondary prevention of CHD among women (38,39), particularly in normalizing blood lipids or inducing endothelium-dependent vasodilation stimulating nitric oxide synthetase (40–42). It has recently been demonstrated in premenopausal diabetic women that there is a decreased estrogen-inducing endothelium-dependent vasodilation (43), which can explain the increased risk of cardiovascular mortality in diabetic women. It could be possible that the augmented oxidative stress in diabetic women induced by hyperglycemic events, not detectable by HbA_1c values, could impair the estrogen protective effect on lipoprotein oxidizability (44,45) and/or on endothelium-dependent vasodilation (46).

In conclusion, our data suggest that type 1 diabetic patients with a short duration of disease and good metabolic control show an early imbalance in their antioxidant capacity and augmented levels of lipid ROOH and CD, even in the absence of complications. Reduced uric acid levels seem to contribute to this phenomenon, especially in diabetic women. Diabetic women show, independently from other factors, a decreased antioxidant capacity and an increased rate of lipoperoxidation compared with diabetic men. The severe alteration of the oxidative pattern in diabetic women may offer one possible pathogenetic explanation for the higher incidence of cardiovascular complications observed in diabetic versus nondiabetic women.

Further studies are needed to investigate a possible effect of hyperglycemia and oxidative stress on the hormonal pattern and/or activity in women with type 1 diabetes.

This work was supported by a MURST ‘97 grant “Prediction, prevention and intervention in diabetes mellitus and its complications.”

We thank Dr. Paolo Magnani for special revision of the English language in the manuscript. This paper is dedicated to Patrizia Cotroneo.