This study analyzes the effect of chronic treatment with different antioxidants (N-acetyl-cysteine [NAC], taurine, a combination of NAC and taurine, and oxerutin) on long-term experimental diabetes induced by streptozotocin in rats. Glycoxidative damage was evaluated in the skin; glomerular structural changes were studied with morphometry and immunohistochemistry. Oxerutin treatment and the combined NAC plus taurine treatment resulted in reduced accumulation of collagenlinked fluorescence in skin in comparison with untreated diabetic rats. All treatments except taurine reduced glomerular accumulation of N ⑀ -(carboxymethyl)lysine and protected against the increase in glomerular volume typical of diabetes; furthermore, the apoptosis rate was significantly decreased and the glomerular cell density was better preserved. Glycoxidative markers in the skin turned out to be good indicators of the glomerular condition. The findings that emerged from our study support the hypothesis that glomerular damage in diabetes can be prevented or at least attenuated by supplementation with specific antioxidants. Treatment with oxerutin and combined treatment with NAC plus taurine gave the most encouraging results, whereas the results of taurine-only treatment were either negligible or negative and therefore suggest caution in the use of this molecule in single-drug treatment courses. Diabetes 52: 499 -505, 2003

Background The main functional property of collagen is to provide a supporting framework to almost all tissues: the effects of non‐enzymatic glycation on this protein are deleterious and in diabetes mellitus contribute to the mechanism of late complications. The aim of this work is to provide evidence by scanning force microscopy of modifications in collagen structure caused by high glucose concentration, in vivo and in vitro, and to correlate the data with markers of non‐enzymatic glycation. Methods Tendon fibrils were obtained from the tails of 8‐month‐old rats (BB/WOR/MOL\BB) which developed diabetes spontaneously at least 12 weeks before they were killed, and from diabetes‐resistant rats of the same strain (BB/WOR/MOL\WB). A scanning force microscope (SFM; Nanoscope III) equipped with a Contact Mode Head was used for imaging. Band interval, diameter and depth of D‐band gap were measured in non‐diabetic and diabetic tail tendon fibrils and in fibrils incubated with glucose (0.5 M for 2 weeks). Fructosamine was determined in the tendon fibrils by a colorimetric method and pentosidine was evaluated in acid‐hydrolyzed samples by coupled reverse phase‐ionic exchange column HPLC. Results Incubated fibrils revealed modifications in radius (228±5 nm) and gap depth (3.65±0.10 nm) that closely reproduce diabetes‐induced damage (236±3 and 3.20±0.04 nm respectively) and were significantly different from the pattern seen in non‐diabetic fibrils (151±1 and 2.06±0.03 nm; p<0.001). Both fructosamine and pentosidine were higher in diabetic (3.82±1.43 nmol/mg and 2.23±0.24 pmol/mg collagen respectively) and in glucose‐incubated fibrils (9.27±0.55 nmol/mg and 5.15±0.12 pmol/mg collagen respectively) vs non‐diabetic tendons (1.29±0.08 nmol/mg and 0.88±0.11 pmol/mg collagen respectively; p<0.01); during the time course of incubation, an early increase in fructosamine was seen, whereas pentosidine increased later. The D‐band parameter was similar in all three groups, indicating that axial organization is not modified by non‐enzymatic glycation. Conclusion This is the first description obtained with SFM of diabetes‐induced ultrastructural changes in collagen fibrils. Moreover, the data presented are consistent with the concept that chronic exposure of collagen to glucose in vivo or in vitro leads to similar structural modifications in collagen fibrils, probably through crosslinks. The correlation between morphologic parameters and both markers of glycation provides strong evidence for a crucial role of this non‐enzymatic modification. Copyright © 2000 John Wiley & Sons, Ltd.

Blood glucose control plays a prominent role in the aetiology of diabetic complications. Recent data support the hypothesis that non-enzymatic pathways (glycation and oxidation) are involved in the pathogenesis of tissue damage in diabetes mellitus. In this study the level of pentosidine, a marker of glycation, and the intensity of collagen-linked fluorescence glycation (370/440 and 335/385 nm) and oxidation-related (356/460 and 390/460 nm), have been examined in spontaneously diabetic rats with good and poor glycaemic control. Pentosidine increased dramatically in rats with poor control, and slightly in those with good control. At the end of the study, after 6 months of diabetes, pentosidine levels were 13 +/- 5 and 2.1 +/- 0.5 pmol/mg collagen, respectively (control rats: 1.1 +/- 0.1 pmol/mg collagen). A similar pattern was observed for both glycation or oxidation-related fluorescence. The group of rats with poor control always showed elevated average values when compared to rats with good control, with a relative increase of over 200%. The results emphasize the role of good glycaemic control in preventing the growth of glycation or oxidation end-products in collagen. On comparison between the general mean level of all glycated haemoglobin and the mean pentosidine level of the three groups, a very good exponential correlation was found (r = 0.993, p < 0.001). The fluorescence values presented a less strong relationship, but a correlation with glycaemic control was still present. If the post-translational modifications of proteins play a leading role in the pathogenesis of complications it is possible to conclude that strict glycaemic control, obtained by accurate insulin therapy can prevent them by inhibiting the non-enzymatic modification of proteins and delaying their accumulation in collagen. The therapeutic implications are obvious.

The role of oxidative stress in aging and diabetes mellitus is currently under discussion. We previously showed age-dependent accumulations of fluorescent protein adducts with lipoperoxidative aldehydes, (malondialdehyde (MDA), and hydroxynonenal (HNE)) in rat skin collagen with diabetic BB rats exhibiting faster accumulation. Modified proteins have been shown to be immunogenic: antibody titres against rat serum albumin modified by MDA and HNE (MDA-RSA and HNE-RSA) or oxidized by reactive oxygen species were measured by ELISA as markers of oxidative damage in BB diabetic and non-diabetic rats. Each tested antibody titre was significantly higher in the diabetic than in the non-diabetic rats. A significant correlation existed between anti-MDA-RSA and anti-HNE-RSA antibody titers. Only the anti-HNE-RSA antibody titre increased significantly with age (p=0.052) in diabetic animals, while no titres increased significantly in non-diabetic animals. A major factor which correlated with the development of these antibodies was diabetes duration: this was significant (p=0.032) for anti-HNE-RSA antibody titre and slightly significant (p=0.05) for anti-MDA-RSA antibody titre. Thus, chronic hyperglycaemia is probably fundamental in the increase of oxidative stress. There is correlation between anti-aldehyde-RSA antibody titres and the corresponding aldehyde-related collagen-linked fluorescence: modified collagen may play a part in the observed immune response. Our data indicate a stronger immune response of diabetic rats against proteins modified by lipoperoxidative aldehydes and oxygen free radicals, and they support the hypothesis of increased oxidative damage in diabetes.

Heme oxygenase 1 (HO-1) expression is recognized as a marker of cellular response to oxidative stress; since ageing is believed to be related to oxidative "wear and tear", HO-1 may represent a candidate biomarker of ageing. In our study, the hepatic expression of HO-1 mRNA, evaluated by RT-PCR in 2.5-24 month-old rats, was higher at 6 months than at 2.5 months of age, but thereafter increased no further: on the contrary, a declining trend was observed. However, while 2.5 month-old rats responded to acute ethanol intoxication by displaying increased expression of liver HO-1 mRNA, and 6 month-old rats exhibited a mild response, 18 month-old rats did not show any response; this phenomenon suggests that during development and ageing the transcriptional response to oxidative stress decreases. In our view, the finding that HO-1 expression did not increase progressively during ageing may be explained by a decreased transcriptional ability to respond to stress in older animals, rather than by a reduction in oxidative stress.