Diabetic retinopathy remains one of the main causes of vision loss and blindness in many countries. There have been recent advancements in treatments for preproliferative or proliferative diabetic retinopathy, laser photocoagulation and intra-ocular surgery, such as vitrectomy; however, the prognosis of difficult cases with maculopathy, rubeotic glaucoma, retinal detachment, and so on, is still poor. Therefore, it is important to prevent the progression of retinopathy at an early stage, that is, when it is background retinopathy. In background retinopathy, before dot hemorrhages and microaneurysms are seen with funduscopy, microscopic changes, such as thickening of basement membranes or pericyte ghosts, have already occurred in microvascular walls. Pericyte ghosts, which are the pockets in basement membranes marking the space from which pericytes have disappeared, are well-known features in the histology of background retinopathy 1, 2 and indicate the demise of pericytes. Recently, we reported that the death of RMCs, that is, pericytes and endothelial cells, by apoptosis was accelerated in diabetic patients and in two animal models that are known to develop the early stages of retinopathy (the alloxan-diabetic rat and the galactose-fed rat) 3. We concluded that the increased apoptosis in these cells indicated the progression of diabetic retinopathy. Furthermore, we speculated that the progression of early diabetic retinopathy could be quantitatively evaluated by determining the number of apoptotic cells in diabetic animal models.
Goto-Kakizaki (GK) rats are considered to be a spontaneous, non-obese model of type 2 diabetes mellitus with glucose intolerance due to impaired insulin secretion 4–6. Mild hyperglycemia has been demonstrated already in 8-day-old GK rats 5, 7. Regarding diabetic complications, thickening of the glomerular capillary basement membrane 8, slowing of the conduction velocity of impulse in the tail motor nerve 9, and increased endothelial/pericyte ratio 10 were confirmed. However, there has been no study of apoptosis of RMCs.
Oxidative stress, accumulation of advanced glycation end-products (AGE), elevation of protein kinase C (PKC) activity, activation of a polyol pathway 11, and so on, were previously reported as pathogenesis of diabetic retinopathy, but the detailed mechanisms are unknown and prevention strategies have not been established except improving blood glucose levels and blood pressure. We were interested in oxidative stress and AGE, in particular, and speculated that an antioxidant or an inhibitor of advanced glycation might reduce apoptosis of RMCs. There is no report about the effect of antioxidant on apoptosis of RMCs in diabetic animal models.
Currently, many types of antioxidants are known and available. Vitamins C (ascorbic acid) and E (alpha-tocopherol) are well-known dietary antioxidants. Vitamin E is lipophilic and inhibits lipid peroxidation, scavenging lipid peroxyl radicals to yield lipid hydroperoxides and the tocopheroxyl radical 12. Vitamin C, a water-soluble vitamin, functions cooperatively with vitamin E by regenerating tocopherol from tocopheroxl radical. We used a combination of ascorbic acid and tocopherol as antioxidants.
OPB-9195, (±)-2-Isopropylidenehydrazono-4-oxo-thiazolidin-5-ylacetanilide, is a novel inhibitor of advanced glycation that was developed by Otsuka Pharmaceutical Co., Japan. In diabetic rats, this drug inhibits the formation of AGE, but has no effect in lowering blood glucose, which prevents the development of diabetic nephropathy 13, 14. Recently, it was reported that OPB-9195 inhibited formation of both AGE and advanced lipoxidation end-products (ALE), such as malondialdehyde (MDA)-lysine and 4-hydroxynonenal (HNE)-protein adduct, through its ability to trap reactive carbonyl compounds (RCO) 15, 16.
We studied the effects of antioxidants (using a combination of ascorbic acid and tocopherol) or a novel inhibitor of advanced glycation, OPB-9195, on apoptosis of RMCs in diabetic GK rats.
- Top of page
One of the novel findings of our study is that apoptosis of RMCs actually occurred in moderately diabetic GK rats. Apoptosis of RMCs has only been confirmed in diabetic patients, alloxan-diabetic rats, and galactose-fed rats 3, 24. It is interesting that apoptosis of RMCs was found in 1-year-old GK rats with much milder hyperglycemia than streptozotocin-induced or alloxan-induced diabetic rats. This fact disproves the assertion that only severe hyperglycemia or hypergalactosemia might lead to apoptosis of RMCs and shows that it was not caused by streptozotocin or alloxan.
Our data shows that the evaluation of apoptosis of RMCs is a good marker of diabetic retinopathy in animal models because the amount of apoptosis, as well as the number of acellular capillaries, increased in the diabetic rats more than in the control, and the difference in the amount of apoptosis between the diabetic GK rats and the controls was large. A cellular capillaries and pericyte ghosts are well known as early histological changes of diabetic retinopathy, and these changes mean that the nuclei of RMCs have disappeared from capillaries. An acellular capillary is a capillary from which all endothelial cells and all pericytes have disappeared, and a pericyte ghost is an empty pocket in the basement membrane from which a pericyte nucleus has disappeared. Because cells in which apoptosis has started will soon die and disappear, we presume that these two changes appear as a result of apoptosis of RMCs. The evaluation of apoptosis shows how many RMCs are dying now, and the evaluation of the two morphological changes show how many RMCs have died so far. Because the evaluation of apoptosis predicts prospective histological changes, it is useful as well as the evaluation of morphological changes.
Because it has been known that TUNEL labeling is not an apoptosis-exclusive assay, our data cannot suggest that apoptosis is the only way in which vascular cells die during diabetes. However, it suggested that it was a useful evaluation method of diabetic retinopathy to count TUNEL-positive cells, which are dying mainly through apoptosis presumably.
Another of the novel findings in our study is that antioxidants (Vitamin C and E) and an AGE inhibitor (OPB-9195) ameliorated the increase of apoptosis of RMCs and acellular capillaries in diabetic GK rats. Vitamin C is water soluble and one of the most powerful natural antioxidants 12. It scavenges reactive oxygen species in aqueous phase (plasma, cytoplasm, and so on). Moreover, there is evidence from in vitro studies that it is capable of regenerating tocopherol from the tocopheroxyl radical that is formed upon the inhibition of lipid peroxidation by vitamin E. Vitamin E is lipophilic, operating in membranes or lipoprotein particles. It scavenges lipid peroxyl radicals and inhibits lipid peroxidation. It was reported that, in the lipid phase, it might be the most efficient of lipophilic antioxidants 25. Therefore, a combination of vitamins C and E is thought to work as a powerful antioxidant in aqueous and lipid phases.
Our data demonstrated that the combination of vitamins C and E reduced apoptosis of RMCs and acellular capillaries in diabetic rats and suggested that the combination of vitamins C and E might inhibit the progression of early retinopathy in diabetic patients.
Brownlee M et al. reported 26 that high concentrations of glucose increased the production of reactive oxygen species (ROS) in endothelial cells, and inhibition of mitochondrial ROS production prevented high glucose-induced activation of PKC, formation of AGE, sorbitol accumulation, and NFκB activation. This means oxidative stress though mitochondrial ROS production should be the most important cause of diabetic endothelial dysfunction and inhibition of mitochondrial ROS production or elimination of intracellular excess ROS will be a effective and reasonable therapeutic approach to prevent diabetic complications.
A couple of recent articles studied whether a combination of vitamins C and E can improve diabetic retinopathy in diabetic rats. One of them 27 reported that a combination of vitamins C (10 g/kg diet) and E (1 g/kg diet) inhibited 50% of acellular capillaries significantly and tended to inhibit pericyte ghosts in alloxan-induced diabetic Sprague-Dawley rats. Moreover, multiple-antioxidant diet including vitamins C and E improved both histological changes significantly. The other article 28 reported a combination of vitamins C (1 g/L drinking water) and E (10 IU/kg BW) inhibited retinal superoxide production significantly and tended to inhibit pericyte ghosts or acellular capillaries (but the differences were not significant) in streptozotocin-induced diabetic Lewis rats. Interestingly, green tea with antioxidant property inhibited 23% of acellular capillaries significantly.
It is difficult to compare our data with theirs because of the levels of hyperglycemia, amount of administered vitamins C and E, and strains of rats. Hyperglycemia was milder in our GK rats than in diabetic rats used in the above studies. Fortunately, because glucose levels of GK rats were adequate for a treatment of vitamins C and E to work well, the difference between the Diabetic and the Vit. C + E groups might have become significant even if the number of experimental animals was low.
It was previously reported that aminoguanidine ameliorated the increase of apoptosis of RMCs and acellular capillaries in alloxan-induced diabetic rats 24. We showed that OPB-9195 ameliorated it in diabetic GK rats. These results demonstrate that advanced glycation is also cross related to the pathogenesis of diabetic retinopathy and suggest that AGE inhibitors might inhibit the progression of early retinopathy in diabetic patients. However, the antioxidative effects of aminoguanidine 29–32 and OPB-9195 15, 16 have recently been reported. Therefore, these AGE inhibitors may improve diabetic retinopathy through not only the inhibition of advanced glycation but also through an antioxidative effect.
Immunohistochemical analysis of AGE demonstrated that the antioxidants combination and OPB-9195 reduced AGE-like immunoreactivity, which increased in retinal vascular walls and the inner segment of the photoreceptor layer in a subject with diabetes. The epitope of anti AGE antibody we used was N-(carboxymethyl) lysine (CML). Oxidative stress is known to be related to the formation of CML, and it has been reported that CML is a biomarker of not only advanced glycation but also oxidative stress 33–38. We speculate that the antioxidants combination might reduce the accumulation of AGE through the inhibition of oxidative stress.
In conclusion, we showed that retinal pathogenic changes were suppressed by antioxidants and AGE inhibitors in diabetic GK rats and the combination of Vitamin C + E, and OPB-9195 had beneficial effects.
- Top of page
The authors thank Dr Giulio Romeo and Dr Mara Lorenzi of Schepens Eye Research Institute and the Department of Ophthalmology, Harvard Medical School, Boston, Massachusetts, and Sayon Roy of the Department of Ophthalmology, School of Medicine, Boston University, for information about the trypsin digest method of rat retina.
This study was supported by the Promotion of Fundamental Studies in Health Science of the Organization for Pharmaceutical Safety and Research (OPSR) and by a Health Sciences Research Grant (Research on Human Genome and Gene Therapy) from the Ministry of Health and Welfare. This study was also supported by a grant from the Japanese Ministry of Education, Science, Sports and Culture and by a grant for Project Research from the University of Tsukuba.