Main

It is widely known that oxidative stress may play a relevant role in the pathogenesis of diabetic vascular complications (1–3). Increased production of reactive oxygen metabolites and species is a direct consequence of high glucose concentrations (3,4). Hyperglycemia is able to increase the levels of oxygen radical scavenging enzymes in cultured endothelial cells (5) and in the kidney of rats with diabetes induced by streptozotocin (6,7). Finally, hyperglycemia can induce formation of free radicals and activation of oxidative stress through nonenzymatic glycation of proteins (8,9), auto-oxidative glycation (10), activation of protein kinase C (11), and increased polyol pathway (12). In normal individuals, exposure to high glucose concentrations induces an antioxidant defensive mechanism in skin fibroblasts; in adults with type 1 diabetes with macroalbuminuria and overt nephropathy, this defensive mechanism is absent (13).

Recently, we demonstrated that fluorescent products of lipid peroxidation and malondialdehyde both are increased in adolescents and young adults with early nephropathy (14). Concurrently, vitamin E levels were markedly reduced in these individuals. In the present study we evaluated intracellular antioxidant enzyme production in skin fibroblasts of young patients with persistent microalbuminuria and early diabetic nephropathy; we also investigated whether administration of vitamin E (600 mg twice daily for 3 mo) is able to modify this cellular antioxidant mechanism.

METHODS

Participants

All patients gave their informed consent to the study, which was approved by the Ethics Committee of the School of Medicine, University of Chieti, Italy. Twelve adolescents with type 1 diabetes agreed to participate; their age ranged from 11 to 21 y, and duration of diabetes ranged from 10 to 18 y. Eight of these patients had retinopathy [background (four), preproliferative (three), or proliferative (one)], four had persistent microalbuminuria [defined as an albumin excretion rate >50 μg/min in two of three overnight urinary collections), and seven had both. Skin fibroblasts obtained by skin biopsies were taken by excision under local anesthetic from the anterior surface of the forearm and cultured in Dulbecco's modified Eagle's medium (DMEM). CuZn superoxide dismutase (SOD), MnSOD, catalase (CAT), and glutathione-peroxidase (GPX) activity and mRNA expression were measured before and after 3 mo of vitamin E supplementation (600 mg twice daily); on both occasions, antioxidant enzyme activity was evaluated ex vivo at different glucose concentrations (5 and 22 mM). Ten adolescents (aged 12–20 y) without diabetic angiopathy and eight healthy volunteers (aged 15–22 y) participated in the study as control groups. Clinical characteristics of participants enrolled in the study are summarized in Table 1.

Table 1 Characteristics of adolescents with diabetic angiopathy (DA), adolescents without diabetic angiopathy (NDA), and nondiabetic control subjects (CS)
Full size table

Arterial blood pressure was measured in all patients and control subjects following the recommendations of the American Heart Association and the American Academy of Pediatrics (15,16). Glomerular filtration rate (GFR) was measured as previously described (17). Vitamin E serum levels were evaluated every 2 wk and measured as previously described (as α-tocopherol by HPLC) (18).

Cell Culture

Fibroblasts were cultured in DMEM (ICN Biochemicals, Thame, UK) supplemented with 20% FCS (Life Technologies, Paisley, Scotland, UK), 2 mM glutamine (Sigma Chemical Co., Dorset, UK), 50 U/mL of penicillin (Life Technologies), and 50 μg/mL of streptomycin (Life Technologies). At the fourth passage, cells were cooled gradually and then frozen at −180°C in 10% DMSO in DMEM until used for the experiments. It is well recognized that even long-term cryopreservation does not affect fibroblasts' functional activities (19).

Experiments

All experiments were conducted between the sixth and eighth passages, using the same batches of medium and FCS. The purchased medium contained 5 mM of glucose, to which mannitol or glucose was added to obtain iso-osmolal experimental media; in other words, mannitol was added to the medium to ensure that the high glucose culture media had the same osmolarity. Cells were cultured in iso-osmolal normal (5 mM) ex vivo glucose and in high ex vivo glucose concentrations (22 mM).

Each sample of cells was grown for 12 wk, with renewal of the medium every second day. For each culture condition (normal or high glucose), 12 80-cm2 plastic tissue culture flasks were used. Three flasks were used for RNA extraction, three flasks for enzyme activity measurement, three flasks for the evaluation of cell membrane lipid peroxidation, and three flasks to determine cell number.

Cell Counting

The medium was aspirated, and the monolayers were washed twice with PBS and detached by treatment with 2.5 mL trypsin-EDTA (Life Technologies) for 4–6 min at 37°C. Trypsin activity was stopped by the addition of 7 mL of medium that contained serum, after verification under the microscope of the complete detachment of the cells. The cell suspension was passed several times through a fine Pasteur pipette to disaggregate cell clumps, and 1 mL was counted in an electronic Coulter counter (ZBI model; Coulter Electronics, Beds, UK) equipped with a 100-μm aperture.

Antioxidant Enzyme Activity

CAT and GPX activities.

The monolayers were rinsed twice with ice-cold PBS, and the cells were harvested with a sterile rubber cell scraper. The cells were sedimented for 4 min at 1600 × g and processed either for enzyme/protein or for mRNA analyses. For enzyme/protein lysates, cells were resuspended in 50 mM of potassium-phosphate buffer that contained 0.5% Triton X-100 and sonicated (in an ice-water bath) for two 30-s bursts on a Branson sonicator B15 (position 2, continuous setting; Branson Ultrasonic, Danbury, CT) with a 30-s cooling interval. Total protein concentration was determined according to the procedure of Bradford (20). For CAT and GPX activities, sonicates were first spun 5 min at 800 × g (4°C). The supernatants were assayed according to the procedure of Clairborne (21) for CAT activity and Gunzler and Flohè (22) for GPX activity.

SOD measurements.

Cells were suspended in 100 mM of triethanolamine-diethanolamine buffer and homogenized with a Teflon glass Dounce homogenizer. The homogenate was centrifuged at 105,000 × g for 1 h (4°C), and the supernatant was passed through a small Sephadex G25 (coarse) column to remove low-molecular-weight substances that interfere with the enzyme assay (23). An aliquot of the eluate was applied onto a 5.5% polyacrylamide gel to localize SOD activity (24), with the exception that no tetramethyl-ethylenediamine was used for staining.

MnSOD activity.

MnSOD activity was determined in mitochondrial fractions that were prepared by differential centrifugation, as previously described (25). Mitochondria were disrupted by freezing-thawing in a high ionic strength buffer [0.25 mM of sucrose, 0.12 M of KCl, and 10 mM of Tris-HCL (pH 7.4)]. Mitochondrial membranes were removed by sedimentation at 105,000 × g for 1 h (4°C), and enzyme activity was measured in the supernatant.

Northern blot analysis.

Total RNA was prepared according to the procedure of Chirgwin et al. (26). Briefly, 10 μg of total RNA was electrophoresed on a 1.4% agarose-formaldehyde gel and then transferred to gene screen membranes. The filters were prehybridized in 50 mM of Tris-HCl (pH 7.5), 0.1% sodium pyrophosphate, 0.2% Ficoll, 5 mM of EDTA, 1% SDS, 2.2% poly(vinylpyrrolidone), 50% formamide, 0.2% BSA, 1× standard sodium citrate (SSC), and 150 μg/mL of denatured salmon sperm DNA at 65°C for 6 h. Blots were hybridized with 32P-labeled probes for human CuZnSOD (27), human CAT (27), human MnSOD (28), and bovine GPX (29), to a specific activity of 1 × 106 cpm/mL in hybridization fluid at 65°C overnight. The filters were washed at 65°C twice for 15 min with 2× SSC-0.1% and twice for 15 min with 0.1× SSC-0.1% SDS and then subjected to autoradiography using an intensifying screen at −85°C. Densitometry was performed on an LKB laser scanning densitometer. Hybridization to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA was used as internal control to correct for loading inequalities.

The filters were probed for the four antioxidant enzymes separately, and GAPDH was also used separately. The results were normalized against an ideal reference value obtained from healthy individuals at 5 mmol glucose/L ex vivo.

Lipid peroxidation.

Cells were trypsinized and centrifuged at 250 × g for 10 min at 4°C. Cell pellets were resuspended in 1 mL of cold PBS for assay of thiobarbituric acid–reactive substances and conjugated dienes, as previously described (30).

Statistical Analysis

ANOVA was used to test differences among the three groups. Paired t test was used to compare, for each group of fibroblasts, the results under conditions of normal versus high ex vivo glucose concentration, whereas Fisher least significant differences test was used to evaluate the difference among the three different groups in either normal or high glucose condition. A p < 0.05 was considered significant. Data are expressed as means ± SD or as median and range.

RESULTS

CuZnSOD.

In normal ex vivo glucose concentration, CuZnSOD activity and mRNA expression were not different among the four groups. In high ex vivo glucose conditions, CuZnSOD mRNA and activity increased similarly in all groups (p = NS by ANOVA).

MnSOD.

In normal ex vivo glucose concentration, MnSOD activity and mRNA expression were not different among the four groups. In high ex vivo glucose conditions, MnSOD did not change in any group.

CAT and GPX activity.

In normal ex vivo glucose concentration, CAT and GPX activity and mRNA expression were not different among the four groups. In high ex vivo glucose conditions, CAT and GPX mRNA (p < 0.001) and activity (p < 0.001) were significantly different between the groups by ANOVA (Figs. 1 and 2). Comparing the groups in high glucose conditions, CAT and GPX mRNA expression and CAT and GPX protein activity were significantly higher in control subjects and diabetic subjects without angiopathy versus angiopathic diabetic subjects, with no difference between adolescents without diabetic angiopathy and control subjects (Figs. 1 and 2).

Figure 1
figure 1

Antioxidant enzyme activity in skin fibroblasts from adolescents with diabetic angiopathy (n = 12), adolescents without diabetic angiopathy (n = 10), and healthy control subjects (n = 8). Enzyme activity was measured in normal glucose concentration (5 mmol/L; □) and in high glucose condition (22 mmol/L; ▪); *p < 0.001.

Full size image
Figure 2
figure 2

mRNA expression of antioxidant enzymes in skin fibroblasts from adolescents with diabetic angiopathy (n = 12), adolescents without diabetic angiopathy (n = 10), and healthy control subjects (n = 8). mRNA expression was measured in normal glucose concentration (5 mmol/L; □) and in high glucose condition (22 mmol/L; ▪); *p < 0.001.

Full size image

Lipid peroxidation.

High ex vivo glucose concentrations significantly increased lipid peroxidation in every group of cells. Higher levels were found in cells of adolescents and young adults with diabetic angiopathy (p < 0.001).

Vitamin E supplementation.

Vitamin E serum levels increased 2–3 wk after the administration and remained high throughout the study. No adverse event was evident in any patient. Vitamin E supplementation (600 mg twice daily for 3 mo) did not change significantly any of the enzymatic activity in both normal and hyperglycemic conditions (Table 2). With regard to mRNA expression, vitamin E was not able to modify the mRNA/GAPDH ratio for CuZnSOD at 5 and 22 mM ex vivo glucose concentrations [diabetics with angiopathy (DA): 4.6 ± 1.6, 10.1 ± 2.9; diabetics without angiopathy (NDA): 4.9 ± 1.6, 10.3 ± 3.1], for MnSOD (DA: 0.9 ± 0.4, 1.1 ± 0.2; NDA: 0.9 ± 0.5, 1.2 ± 0.4), for catalase (DA: 4.3 ± 1.3, 4.5 ± 1.4; NDA: 4.4 ± 1.5, 8.4 ± 2.7), and for GPX (DA: 2.3 ± 1.0, 2.5 ± 0.9; NDA: 2.4 ± 1.1, 4.3 ± 1.1).

Table 2 Antioxidant enzyme activity after 3 months of vitamin E administration in skin fibroblasts from adolescents with (DA) and without (NDA) diabetic angiopathy
Full size table

DISCUSSION

The present study indicates that exposure to high ex vivo glucose concentrations induces an increase in mRNA levels and biologic activity of CuZnSOD, CAT, and GPX in fibroblasts from control subjects and adolescents without diabetic angiopathy; by contrast, in fibroblasts from diabetic adolescents with angiopathy, only CuZnSOD is increased. This finding may have important consequences concerning glucose-induced oxidative stress damage to the cell; in fact, glucose-induced oxidative stress has been demonstrated to damage several cells, including endothelial cells (2,5).

Both CuZnSOD, which is located primarily in the cytoplasm, and MnSOD, a structurally distinct protein located in the mitochondria, catalyze the reaction O2− + O2− + 2H+ = O2 + H2O2 (28). H2O2 is converted to H2O in peroxisomes by the antioxidant enzyme CAT and in the cytoplasm by GPX (31). These antioxidant enzymes protect the cell from oxidative stress, but the threshold of protection can vary dramatically as a function of their activity and balance (32). CAT and GPX are far more efficient than CuZnSOD in protecting fibroblasts against oxidative stress (32,33). However, in several instances, cells with increased levels of CuZnSOD are hypersensitive to oxidative stress rather than protected from it (32). This happens because CuZnSOD increases the formation of H2O2, which, if not efficiently converted to H2O by an adequate level of CAT and GPX, may be detrimental to the cell (32). It is therefore not surprising that generally an increase in CuZnSOD is accompanied by a concomitant increase in CAT and GPX (32). In the presence of high ex vivo glucose concentrations, we confirmed this phenomenon in the fibroblasts derived from control subjects and diabetic young patients without microvascular complications. In the fibroblasts of young patients with childhood-onset diabetes and angiopathy, however, high glucose induced a significant increase only in CuZnSOD but no change in the activity of CAT and GPX. These results are largely confirmatory of previous results obtained in adult diabetic patients with macroproteinuria and overt nephropathy (13) and suggest that cells of youths with type 1 diabetes and incipient angiopathy are not able to adjust their antioxidant defenses when high ex vivo glucose concentration–induced oxidative stress is produced, so they are more susceptible to oxidative stress. Alternatively, one could argue that in the absence of the ability to increase CAT and GPX, the cells may “decide” not to enhance CuZnSOD and MnSOD, in that the mechanism could simply be switched-off. However, this event should be operative also in individuals with diabetes without angiopathy and in control subjects.

High glucose concentrations in vitro and hyperglycemia in vivo are well-known stimuli for the production of free radicals and the generation of oxidative stress, with a consequent increase in the expression and activity of antioxidant enzymes (1–3), which act as a defense system against cell damage (33). Hyperglycemia is also a necessary factor for the development of the glomerular lesions of diabetes. The observation that, despite hyperglycemia, only a portion of the population of patients with type 1 diabetes will progress to diabetic microangiopathy indicates that there is individual diversity in cell response to high ex vivo glucose concentrations. It is therefore of great relevance that a disturbance in the mechanisms of protection from oxidative stress was found only in the cells of adolescents with angiopathy. By contrast, in adolescents and young adults with long-term type 1 diabetes without angiopathy, a group that seems to be protected from vascular complications, the defense mechanisms against high glucose–induced oxidative stress were intact or similar to those of nondiabetic individuals.

The novel finding of this study is that vitamin E supplementation (600 mg twice daily for 3 mo) was unable to substantially modify the antioxidant enzyme production and activity in young patients with early signs of angiopathy. Contrasting results have been obtained on the effects of vitamin E on markers of oxidative stress: recently, we were able to demonstrate that administration of vitamin E (at the same dosage used in the present study) was able to reduce plasma concentrations of MCP-1 (an inflammatory chemokine possibly involved in the pathogenesis of diabetic angiopathy) (14); fluorescent products of lipid peroxidation and malondialdehyde were also reduced after treatment and vitamin E levels increased.

In a double-blinded, placebo-controlled study, high-dose vitamin E supplementation (∼1230 mg/d) was able to normalize retinal blood flow and creatinine clearance in patients with type 1 diabetes (34). In patients with type 2 diabetes, supplementation with vitamin E (∼550 mg/d) induced a significant reduction of risk factors for macrovascular complications (35). Even low doses of vitamin E (∼70 mg/d) were able to reduce glutathione and lower lipid peroxidation and HbA1c concentrations in the erythrocytes of patients with type 1 diabetes (36). At variance, in a recent study, a lower dose of vitamin E (∼270 mg/d) taken orally for 8 wk had no significant effect on oxidatively induced LDL or DNA damage in patients with type 1 diabetes, but the same dosage regimen did reduce susceptibility to LDL oxidative change in control subjects (37). These studies are difficult to compare, because they examine patients with type 1 and type 2 diabetes; in some, natural vitamin E is used; in others, synthetic vitamin is used, different doses are given, and the duration of treatment is also different. These factors must be taken into consideration when comparing studies. Furthermore, clinical trials with vitamin E failed to demonstrate any beneficial effect on the development of diabetic complications (38). On this matter, it was suggested recently that antioxidant therapy with vitamin E or other antioxidants is limited to scavenging already-formed antioxidants and therefore may be considered a more “symptomatic” rather than a “causal” treatment for vascular oxidative stress (39,40). Some studies have documented that vitamin E is able to inhibit protein kinase C activation and consequently to induce a beneficial effect on endothelial cell dysfunction and diabetic angiopathy (41,42); however, the lack of effect of vitamin E in the present study was evident in both youths with diabetes and angiopathy and in those with no signs of diabetic vascular disease. Therefore, at least at the doses and for the time used in the present study, vitamin E is not effective in modifying the defective intracellular (in skin fibroblasts) antioxidant enzyme production in young adults with childhood-onset diabetes and signs of incipient retinopathy and nephropathy.

CONCLUSION

In conclusion, this study confirms that exposure to high ex vivo glucose concentrations induces an antioxidant defense mechanism in skin fibroblasts of normal young subjects and that a failure of this defensive mechanism is present in fibroblasts obtained from young patients with childhood-onset diabetes and early signs of diabetic retinopathy and nephropathy. Vitamin E supplementation (at least at the dose of 600 mg twice daily for 3 mo) is unable to significantly modify these cellular antioxidant mechanisms. Consequently, treatment with these doses of vitamin E should not be used routinely as an adjunct treatment for secondary prevention of angiopathy in patients with childhood-onset type 1 diabetes. Similar to the Diabetes Control and Complications Trial, researchers need to do a long-term clinical trial with a large patient population to assess whether vitamin E supplementation (in different doses and for longer periods) may help to lower the incidence of development and progression of microvascular complications in patients with diabetes (43).