Free Radicals and Aging ed. by I. Emerit ,. B. Chance
©1992 Birkhiuser Verlag Basel/Switze~and
Oxidative stress in diabetic retina Michel Dol}"" Marie-Therese Droy-Lefaixb and Pierre Braquetb aLaboratoire de Biophysique (lnserm U. 71), Facultes de Medecine et de Pharmacie, BP 38, F-63001 Clermont-Ferrand Cedex, France, and bIBB/IPSEN Laboratoires de Recherche, 17, Avenue Descartes, F-92350 Le Plessis-Robinson, France Summary. The authors describe the alterations usually associated with diabetic retinopathy. They concern the classical thickening of the basal membrane of retinal capillaries and the associated modification of retinal vessel permeability. These alterations correspond to the blood-retinal barrier disruption. The authors then discuss the participation of oxygenated free radicals in the pathogenesis of diabetic retinopathy. They report several experimental studies establishing such a participation and finally describe their own results obtained on a model of retinas isolated from alloxan-induced diabetic rats. After one month of evolution, the electroretinograms (ERG) recorded on isolated retinas from diabetic rats had an amplitude about 20% lower than the controls, whereas after two months of diabetes, this decrease was about 60%. Under these conditons, the authors tested the protective properties of Ginkgo bi/oba extract (EGb 761) on their model. They observed that in EGb-treated animals (100 mg/kg/day), the ERG had a significantly (p < 0.001) greater amplitude than untreated animals after two months of diabetes evolution. In conclusion, the authors discuss the possible utilization of a free radical scavenger, such as EGb 761, in the prevention of the retinal impairment in diabetes.
Introduction
Retinopathy is one of the most serious consequences of diabetes in man. Its gravity is directly linked to irreversible impairment of visual function. At the present time, there is only partial comprehension of the mechanism involved in diabetic retinopathy. Several types of histological impairment have been described. Most notably, a thickening of the basal membrane of the capillaries is observed. More recently, several authors have highlighted a disruption in the blood-retinal barrier both in man and in experimental animals. This disruption is accompanied by an increase in capillary permeability that could play a role in the pathogenesis of diabetic retinopathy. Several recent studies have reported a clear increase in the concentration of oxygenated free radicals in the retinas of diabetic animals (Murata et aI., 1981; Cohen, 1984; Nishimura and Kuriyama, 1985a). Moreover, this increase is correlated with an elevation in the rate of membrane lipid peroxidation. Furthermore, the retina which is a tissue very rich in polyunsaturated fatty acids, is particularly sensitive to the reactivity of oxygenated free radicals (Anderson and Sperling, 1971;
300 Doly et aI., 1984). Excessive production of oxygenated radicals may thus explain the specificity of the retinal impairment involved in diabetes. Retinal alterations in diabetes Diabetic retinopathy has been an important subject in histological studies on animal models. For some time now (Bloodworth and Moditor, 1985; Tanigushi and Nomura, 1968; Papachristodoulou and Health, 1977), it has been known that the characteristic histological finding in retinopathy is the thickening of the basal membrane of the retinal capillaries. More recent studies, in particular those of Ishibashi et ai. (1980), demonstrate that these histological modifications accompany an increase in permeability of certain retinal vessels. Indeed, by injecting horseradish peroxidase (HRP) 30 min before sacrificing animals, a high increase in vesicular transport and the existence of junctional transport between the capillary endothelial cells can be established in rats maintained in a diabetic state for 2 to 6 months. This increase in permeability of the endothelial cells apparently precedes the thickening of the basal membrane. In 1975, Cunha-Vaz et ai. introduced the concept of the blood-retinal barrier by analogy with the blood-brain barrier. They noted that one of the early consequences of diabetes is a disruption of the blood-retinal barrier. These results were subsequently confirmed by several workers, such as Ishibashi et ai. (1980) on streptozotocin-diabetic rats and by Blair et ai. (1984) on a spontaneously diabetic variety of rats. The disruption in the blood-retinal barrier appears to be due to an increase in vesicular transport and a junctional insufficiency between the endothelial cells (Ishibashi et aI., 1980). Its mechanism is still unknown, but hyperglycemia and the related accumulation of sorbitol probably playa primary role (Palmberg, 1977). Aldose reductase, an enzyme responsible for the conversion of glucose to sorbitol, is present in the retina and a significant accumulation of sorbitol is observed in diabetic rat retinas. Thus, aldose reductase inhibitors could delay the disruption of the blood-retinal barrier in experimental animals (Palmberg, 1977). There are fewer studies on the electrophysiological consequences of diabetic retinopathy than there are histological studies on the condition. The first ERG alterations related to diabetes were described at the level of the oscillatory potential (Yonemura et aI., 1962). In humans, these wavelets, superimposed on the ascending branch of the b-wave, decrease in amplitude with the evolution of diabetes and finally disappear. Since this oscillatory potential is directly related to the activity of the amacrin cells, functional impairment at this level has been proposed as the consequences of diabetic retinopathy (Yonemura, 1977; Nishimura and Kuriyama, 1985b).
301 Tamai and Tanaka (1973) have given an account of an electroretinographic study in vivo, performed on streptozotocin-diabetic albino rats. They observed that 4 months after the induction of diabetes, there was an average fall in a-wave amplitude of 40% and 37% for b-wave, which they interpreted as being a moderate impairment of the retinal function. More recently, and still concerning streptozotocin-diabetic rats, Tanaka (1981) demonstrated that injection of dextran sulfate ester sodium had the effect of partially preventing the fall in amplitude of a- and b-waves. Free radicals and diabetic retinopathy The pathogenesis of diabetic retinopathy is still quite obscure but seems to be closely linked with the formation of oxygenated free radicals. The studies of Murata et ai. (1981) performed on alloxan-induced diabetic rats, and by Nishimura and Kuriyama (1985a) on streptozotocin-induced diabetic rats, clearly demonstrate that diabetic retinopathy is accompanied by a high increase in the concentration of oxygenated free radicals in the retina. These radicals react preferentially with unsaturated lipids leading to the formation of lipoperoxide radicals (LOi) inducing an oxidative process which results in membrane lysis (Doly et aI., 1984, 1985). Thus Murata et ai. (1981) reported a 40% increase in the lipid peroxidation rate in the retina of diabetic rats in comparison with control rats. This attack on the membrane has particularly harmful effects on retinal tissue and might be related to the important decrease of lipid replacement observed on streptozotocin-induced diabetic rat retina (Careaga and Bazan, 1981; Bazan et aI., 1985). Indeed, two membranes are implicated in the phototransduction mechanism which generates the electroretinogram (ERG): the plasma membrane and the saccular membrane of the photoreceptor. These two membranes are particularly rich in polyunsaturated fatty acids (Kagan et aI., 1973; Oakley and Pinto, 1981) and so very sensitive to lipoperoxidation. The first one to be impaired and denatured by oxidation see~s to be the saccular membrane. In fact, after lipoperoxidation, rhodopsin is more easily extracted (Novikov et aI., 1975) which may correspond to a weakening of protein-lipid interaction; moreover, regeneration capacity of rhodopsin decreases after peroxidation (Franswooth and Dratz, 1976). Finally, intracellular records performed in the rod outer segment indicate that lipoperoxidative substances interact essentially on the saccular membrane (Shvedova et aI., 1979). Free radical scavengers and diabetic retinopathy In order to demonstrate the participation of free radicals in diabetic retinopathy, experiments have been performed on isolated retina of
302
Ag STIMULATION - - .
Circulation of perfusion solution
Figure 1. Experimental set-up used to record ERG on retina isolated from albino rats.
alloxan-induced diabetic rats (Doly et aI., 1986, 1988). The experimental set-up is shown in Figure 1 and was previously published (Doly et aI., 1980). Experimental diabetes was induced by alloxan injection. The retinal function was assessed by ERG records during the survival of the isolated retina. Alongside the control group, three other groups of rats were examined: the first one comprised animals sacrificed one month after diabetes evolution; the second one was made up of animals sacrificed after two months of diabetes evolution and the third one contained treated animals after two months of diabetes. The treatl)1ent consisted in the daily administration of 100 mg/kg per os Ginkgo bi/oba extract (EGb 761) in aqueous solution. This extract is a potent free radical scavenger and exerts a protective effect on the retina against membrane lipid peroxidation (Braquet et aI., 1982). An example of ERG is shown in Figure 2, so as the evolution of ERG amplitude during the survival of normal retinas in standard conditions. We chose to quantify the ERG by measuring the b-wave amplitude; indeed since the latter is generated at the level of the Muller cells (Tomita and Ynagida, 1981), its development involves all the intraretinal nerve connections, and its amplitude thus reflects the quality of the retinal metabolism. The average survival curve obtained in diabetic animals 1 month after the evolution of diabetes is shown in Figure 3. It can be observed that during the first hour of survival, the ERG amplitudes recorded in I-month diabetic rats are comparable to those of the reference curve, whereas during the remainder of the survival period, the ERGs of the diabetic rats are systematically less in amplitude than the reference ERGs. The amplitude of the ERGs obtained on
303
c.....
o,~----~----~----~----~----~----~ 2 3 4 6 5
Time (hours)
Figure 2. Normal evolution of ERG b-wave amplitude during retina survival in standard conditions. On the upper right an example of electroretinogram (ERG) recorded on isolated retina.
• Normal rats .. Diabetic rats (1 month) o Diabetic rats (2 months)
400
~3oo
..:; Q)
"0 :l
~200 E
as
(!:J
II: W 100
o
------r------rl------TI------TI----------234
~I
o
Times (h)
Figure 3. Evolution of ERG amplitude recorded on retina isolated from diabetic rats (1 month and 2 months) compared to ERG recorded on normal rat retina.
2-month diabetic rat retinas are systematically lower than the reference values over the entire survival period (Fig. 3). It can be observed that the decrease in the ERG amplitude is directly correhlted with the duration of the evolution of diabetes. At the point of maximum amplitude in the survival curve of the control rats, the decrease in the ERG amplitude is around 20% for I-month diabetic rats, whereas it is around 60% for 2-month diabetic rats.
304 The animals that had been treated with Ginkgo bi/oba extract were sacrificed after 2-month evolution of diabetes. In comparison with the non-treated animals, it can be observed that the ERG amplitudes are increased, especially at beginning of the survival period, the second half of the survival curves being superimposable both for treated and nontreated animals (Fig. 4). Figure 5 represents the average survival curves obtained with treated and non-treated rats after computer adjustment; this representation points out graphically the difference between the two curves. On such an alloxanic diabetes model, the increase of capillary permeability responsible for the blood retinal barrier disruption may be correlated with oxygenated free radical release (Murata et aI., 1981). Moreover, the elevated rate of membrane lipoperoxidation in the pho-
-8 100 ~
0. E as ~ 50 W
o+-------~------~------,-------,_--------4 2 3 o
Time (h)
Figure 4. Comparison between ERG recorded on diabetic rats after 2 months of diabetes evolution and on EGb 761 treated diabetic rats after the same evolution of diabetes .
>3 Q)
"0
• Diabetic rats 2 months (n
200
= 5)
• Diabetic rats (2 months) treated by Ginkgo biloba extract (n = 4)
150
~
0.100 E as
I
D
50 O+------r----~------r-----,------
o
2
3
4
Time (h)
Figure 5. The same curves than on Figure 4 but after computer adjustment.
305 toreceptors can be responsible for ERG alterations. In a parallel manner, the physiological protective mechanisms (essentially superoxide dismutase and a tocopherol) can also be deficient or overwhelmed (Nishimura and Kuriyamma, 1985b). In this experiment, ERG effects of Ginkgo bi/oba extract can be correlated both with oxygenated free radical reactivity and production. Ginkgo bi/oba extract is rich in benzopyrones such as heterosides, known for free radical scavenger properties (Doly et aI., 1985) and in Ginkgolides A, Band C, (Braquet et aI., 1987), which are potent PAF-acether antagonists; PAF is an immunomediator able to amplify eosinophil and leukocyte response and to induce the release of Oi- and OR' radicals. These radicals may be responsible for retinal membrane impairment and dysfunction in diabetic retinopathy. Conclusion In humans, the retina is one of the most aerated organs. The demand for molecular oxygen by the mitochondria of photoreceptors is particularly high. The rod outer segment membrane contains the highest level of polyunsaturated fatty acids of any tissue in the body. In these conditions, the retina appears as a neuro-sensorial tissue particularly sensitive to the reactivity of oxygenated free radicals. The retinal function, which corresponds to photon detection, is essentially based on membrane potential variations. In other words the quality of light perception is markedly linked to the quality of membranes. One of the most deleterious effects of oxygenated radicals is membrane lipoperoxidation which corresponds to a severe disorganization. Several sudies have demonstrated the exaggerated production of free radicals in the retina of diabetic patients. Here, the impairment of retinal function reported during the evolution of diabetes may be closely related to the reactivity of oxygenated radicals. Such experimental data suggest the possible use of free radical scavengers in order to prevent retinal alterations due to diabetes. The preliminary results obtained with Ginkgo bi/oba extract seem to confirm this possibility and demonstrate that free radical scavengers may play a key role in the prevention of diabetic retinopathy. Anderson, R. E., and Sperling, L. (1971) Lipids of ocular tissues. VII Positional distribution of the fatty acids in the phospholipids of bovine retina rod outer segments. Arch. Biochem. Biophys. 144: 673-677. Bazan, H. E. P., Careaga, M. M., and Bazan, N. G. (1985) Decreased utilization of [2- 3 H] glycerol in phospholipid and neutral glyceride biosynthesis in the retina of streptozotocindiabetic rats. Neurochem. Path. 3: 109. Blair, N. P., Tso, M. O. M., and Dodge, J. T. (1984) Pathologie studies of the blood-retinal barrier in the spontaneously diabetic BB rat. Invest. Ophthalmol. 25, 302-331.
306 Bloodworth, J. H. B., and Moditor, D. L. (1985) Ultrastructural aspect of human and canine diabetic retinopathy. Invest. Ophthalmol. 4: 1037-1048. Braquet, P., Doly, M., Bonhomme, B., and Meyniel, G. (1982) Peroxydation lipidique et activite electrique de la retine isolee: effect protecteur de l'extrait de Ginkgo biloba, in: Compte-rendus des Journees Internationales du Groupe Polyphenols. Toulouse (France), 29-30 September. Braquet, P., Touqui, L., Shen, T. Y., and Vargaftig, B. B. (1987) Perspectives in platelet-activating factor research. Pharmacol. Rev. 39: 97-145. Careaga, M. M., and Bazan, H. E. P. (1981) The rat retina is a useful in vivo model to study membrane lipid synthesis. Rates of biosynthesis of neutral glycerides and phospholipids. Neurochem. Res. 6: 1169-1178. Cohen, G. (1984) Oxy-radical production in alloxan-induced diabetes: an example of an in vivo metal-catalyzed Haber-Weiss reaction, in: Free Radicals in Molecular Biology. Armstrong D., Sohal, R. S., Cuther, R. G., Slater, T. F. eds. Raven Press, New York, pp. 307-316. Cunha-Vaz, J., De Abreu, J. R. F., Campos, A. J., and Figo, G. H. (1975) Early breakdown of the blood-retinal barrier in diabetes. Br. J. Ophthalmol. 59: 649-656. Doly, M., Isabelle, D. B., Vincent, P., Gaillard, G., and Meyniel, G. (1980) Mechanism of the formation of X-ray induced phosphenes. I. Electrophysiological investigations. Rad. Res. 82: 93-105. Doly, M., Braquet, P., Bonhomme, B., and Meyniel, G. (1984) Effects of lipid peroxidation on the isolated rat retina. Ophthalmic Res. 16: 292-296. Doly, M., Braquet, P., Droy, M. T., Bonhomme, B., and Vennat, J. C. (1985) Effects des radicaux libres oxygenes sur l'activite electrophysiologique de la retina isolee de rat. J. Fr. Ophthalmol. 8: 273-277. Doly, M., Droy-Lefaix, M. T., Bonhomme, B., and Braquet, P. (1986) Effets de l'extrait de Ginkgo biloba sur l'electrophysiologie de la retine isolee de rat diabetique. Presse Med. 15: 1480-1483. Doly, M., Braquet, P., Droy, M. T., et al. (1988) Alteration of electrophysiological function of isolated retina from alloxan-induced diabetic rats: effect of treatment with Ginkgo biloba extract. Neurochem. Pathol. 8: 15-26. Farnsworth, C. c., and Dratz, E. A. (1976) Oxidative damage of retinal rod outer segment membranes and the role of Vitamin E. Biochim. Biophys. Acta 443: 556-570. Ishibashi, T., Tanaka, K., and Taniguchi, Y. (1980) Disruption of blood-retinal barrier in experimental diabetic rats: an electron microscopic study. Exp. Eye Res. 30: 401-410. Kagan, V. E., Shvedova, A. A., Novikov, K. N., and Koslov, Y. P. (1973) Lipid-induced free radical oxidation of membrane lipids in photoreceptors of frog retina. Biochim. Biophys. Acta 330: 76-79. Murata, R., Nishida, T., Eto, S., and Mukai, N. (1981) Lipid peroxidation in diabetic rat retina. Metab. Pediat. Ophthalmol. 5: 83-87. Nishimura, c., and Kuriyama, K. (1985a) Alteration of lipid peroxide and endogenous antioxidant contents in retina of streptozotocin-induced diabetic rats: effect of Vitamin A administration. Japan J. Pharmacol. 37: 365-372. Nishimura, c., and Kuriyama, K. (1985b) Alterations in the retinal dopaminergic neuronal system in rats with streptozotocin-induced diabetes. J. Neurochem. 45: 448-455. Novikov, K. N., Kagan, V. E., Shvedova, A. A., and Kozlov, Y. P. (1975) Protein-lipid interactions on peroxide oxidation of lipids in the photoreceptor membrane. Biofizika 20: 1039-1042. Oakley, B., and Pinto, L. H. (1981) [Ca2 +1i modulations of membrane sodium conductance in rod outer segments, in: Cuttent Topics in Membrane and Transport. Muller, ed. Academic Press Inc., New York. p. 405. Palmberg, P. F. (1977) Diabetic Retinopathy. Diabetes 26: 703-711. Papachristodoulou, D., and Heath, H. (1977) Ultrastructural alterations during the development of retinopathy in sucrose-fed and streptozotocin-diabetic rats. Exp. Eye Res. 25: 371-384. Shvedova, A. A., Sidorov, A. S., Novikov, K. N. et al. (1979) Lipid peroxidation and electric activity of the retina. Vision Res. 19: 49-55. Tamai, A., and Tanaka, K. (1973) The ERG of the streptozotocin-diabetic albino rat. Folia Ophthalmol. Japan. 24: 847-850. Tanaka, T. (1981) Effects of various agents on the ERG of the streptozotocin-diabetic rat. Folia Ophthalmol. Japan. 72: 1470-1480.
307 Tanigushi, Y., and Nomura, T. (1968) Fine structure of retinal blood vessels in human diabetics. Acta Soc. Ophthalmol. Japan. 72: 1165-1178. Tomita, T., and Ynagida, T. (1981) Origins of the ERG waves. Visions Res. 21: 1703-1707. Yonemura, D., Aoki, T., and Tsuzuki, K. (1962) Electroretinogram in diabetic retinopathy. Arch. Ophthal. 68: 49-54. Yonemura, D. (1977) An electrophysiological study on activities of neuronal and non-neuronal retinal elements in man with reference to its clinical application. Acta Soc. Ophthalmol. Japan. 81: 1632-1665.
©1992 Birkhiuser Verlag Basel/Switze~and
Oxidative stress in diabetic retina Michel Dol}"" Marie-Therese Droy-Lefaixb and Pierre Braquetb aLaboratoire de Biophysique (lnserm U. 71), Facultes de Medecine et de Pharmacie, BP 38, F-63001 Clermont-Ferrand Cedex, France, and bIBB/IPSEN Laboratoires de Recherche, 17, Avenue Descartes, F-92350 Le Plessis-Robinson, France Summary. The authors describe the alterations usually associated with diabetic retinopathy. They concern the classical thickening of the basal membrane of retinal capillaries and the associated modification of retinal vessel permeability. These alterations correspond to the blood-retinal barrier disruption. The authors then discuss the participation of oxygenated free radicals in the pathogenesis of diabetic retinopathy. They report several experimental studies establishing such a participation and finally describe their own results obtained on a model of retinas isolated from alloxan-induced diabetic rats. After one month of evolution, the electroretinograms (ERG) recorded on isolated retinas from diabetic rats had an amplitude about 20% lower than the controls, whereas after two months of diabetes, this decrease was about 60%. Under these conditons, the authors tested the protective properties of Ginkgo bi/oba extract (EGb 761) on their model. They observed that in EGb-treated animals (100 mg/kg/day), the ERG had a significantly (p < 0.001) greater amplitude than untreated animals after two months of diabetes evolution. In conclusion, the authors discuss the possible utilization of a free radical scavenger, such as EGb 761, in the prevention of the retinal impairment in diabetes.
Introduction
Retinopathy is one of the most serious consequences of diabetes in man. Its gravity is directly linked to irreversible impairment of visual function. At the present time, there is only partial comprehension of the mechanism involved in diabetic retinopathy. Several types of histological impairment have been described. Most notably, a thickening of the basal membrane of the capillaries is observed. More recently, several authors have highlighted a disruption in the blood-retinal barrier both in man and in experimental animals. This disruption is accompanied by an increase in capillary permeability that could play a role in the pathogenesis of diabetic retinopathy. Several recent studies have reported a clear increase in the concentration of oxygenated free radicals in the retinas of diabetic animals (Murata et aI., 1981; Cohen, 1984; Nishimura and Kuriyama, 1985a). Moreover, this increase is correlated with an elevation in the rate of membrane lipid peroxidation. Furthermore, the retina which is a tissue very rich in polyunsaturated fatty acids, is particularly sensitive to the reactivity of oxygenated free radicals (Anderson and Sperling, 1971;
300 Doly et aI., 1984). Excessive production of oxygenated radicals may thus explain the specificity of the retinal impairment involved in diabetes. Retinal alterations in diabetes Diabetic retinopathy has been an important subject in histological studies on animal models. For some time now (Bloodworth and Moditor, 1985; Tanigushi and Nomura, 1968; Papachristodoulou and Health, 1977), it has been known that the characteristic histological finding in retinopathy is the thickening of the basal membrane of the retinal capillaries. More recent studies, in particular those of Ishibashi et ai. (1980), demonstrate that these histological modifications accompany an increase in permeability of certain retinal vessels. Indeed, by injecting horseradish peroxidase (HRP) 30 min before sacrificing animals, a high increase in vesicular transport and the existence of junctional transport between the capillary endothelial cells can be established in rats maintained in a diabetic state for 2 to 6 months. This increase in permeability of the endothelial cells apparently precedes the thickening of the basal membrane. In 1975, Cunha-Vaz et ai. introduced the concept of the blood-retinal barrier by analogy with the blood-brain barrier. They noted that one of the early consequences of diabetes is a disruption of the blood-retinal barrier. These results were subsequently confirmed by several workers, such as Ishibashi et ai. (1980) on streptozotocin-diabetic rats and by Blair et ai. (1984) on a spontaneously diabetic variety of rats. The disruption in the blood-retinal barrier appears to be due to an increase in vesicular transport and a junctional insufficiency between the endothelial cells (Ishibashi et aI., 1980). Its mechanism is still unknown, but hyperglycemia and the related accumulation of sorbitol probably playa primary role (Palmberg, 1977). Aldose reductase, an enzyme responsible for the conversion of glucose to sorbitol, is present in the retina and a significant accumulation of sorbitol is observed in diabetic rat retinas. Thus, aldose reductase inhibitors could delay the disruption of the blood-retinal barrier in experimental animals (Palmberg, 1977). There are fewer studies on the electrophysiological consequences of diabetic retinopathy than there are histological studies on the condition. The first ERG alterations related to diabetes were described at the level of the oscillatory potential (Yonemura et aI., 1962). In humans, these wavelets, superimposed on the ascending branch of the b-wave, decrease in amplitude with the evolution of diabetes and finally disappear. Since this oscillatory potential is directly related to the activity of the amacrin cells, functional impairment at this level has been proposed as the consequences of diabetic retinopathy (Yonemura, 1977; Nishimura and Kuriyama, 1985b).
301 Tamai and Tanaka (1973) have given an account of an electroretinographic study in vivo, performed on streptozotocin-diabetic albino rats. They observed that 4 months after the induction of diabetes, there was an average fall in a-wave amplitude of 40% and 37% for b-wave, which they interpreted as being a moderate impairment of the retinal function. More recently, and still concerning streptozotocin-diabetic rats, Tanaka (1981) demonstrated that injection of dextran sulfate ester sodium had the effect of partially preventing the fall in amplitude of a- and b-waves. Free radicals and diabetic retinopathy The pathogenesis of diabetic retinopathy is still quite obscure but seems to be closely linked with the formation of oxygenated free radicals. The studies of Murata et ai. (1981) performed on alloxan-induced diabetic rats, and by Nishimura and Kuriyama (1985a) on streptozotocin-induced diabetic rats, clearly demonstrate that diabetic retinopathy is accompanied by a high increase in the concentration of oxygenated free radicals in the retina. These radicals react preferentially with unsaturated lipids leading to the formation of lipoperoxide radicals (LOi) inducing an oxidative process which results in membrane lysis (Doly et aI., 1984, 1985). Thus Murata et ai. (1981) reported a 40% increase in the lipid peroxidation rate in the retina of diabetic rats in comparison with control rats. This attack on the membrane has particularly harmful effects on retinal tissue and might be related to the important decrease of lipid replacement observed on streptozotocin-induced diabetic rat retina (Careaga and Bazan, 1981; Bazan et aI., 1985). Indeed, two membranes are implicated in the phototransduction mechanism which generates the electroretinogram (ERG): the plasma membrane and the saccular membrane of the photoreceptor. These two membranes are particularly rich in polyunsaturated fatty acids (Kagan et aI., 1973; Oakley and Pinto, 1981) and so very sensitive to lipoperoxidation. The first one to be impaired and denatured by oxidation see~s to be the saccular membrane. In fact, after lipoperoxidation, rhodopsin is more easily extracted (Novikov et aI., 1975) which may correspond to a weakening of protein-lipid interaction; moreover, regeneration capacity of rhodopsin decreases after peroxidation (Franswooth and Dratz, 1976). Finally, intracellular records performed in the rod outer segment indicate that lipoperoxidative substances interact essentially on the saccular membrane (Shvedova et aI., 1979). Free radical scavengers and diabetic retinopathy In order to demonstrate the participation of free radicals in diabetic retinopathy, experiments have been performed on isolated retina of
302
Ag STIMULATION - - .
Circulation of perfusion solution
Figure 1. Experimental set-up used to record ERG on retina isolated from albino rats.
alloxan-induced diabetic rats (Doly et aI., 1986, 1988). The experimental set-up is shown in Figure 1 and was previously published (Doly et aI., 1980). Experimental diabetes was induced by alloxan injection. The retinal function was assessed by ERG records during the survival of the isolated retina. Alongside the control group, three other groups of rats were examined: the first one comprised animals sacrificed one month after diabetes evolution; the second one was made up of animals sacrificed after two months of diabetes evolution and the third one contained treated animals after two months of diabetes. The treatl)1ent consisted in the daily administration of 100 mg/kg per os Ginkgo bi/oba extract (EGb 761) in aqueous solution. This extract is a potent free radical scavenger and exerts a protective effect on the retina against membrane lipid peroxidation (Braquet et aI., 1982). An example of ERG is shown in Figure 2, so as the evolution of ERG amplitude during the survival of normal retinas in standard conditions. We chose to quantify the ERG by measuring the b-wave amplitude; indeed since the latter is generated at the level of the Muller cells (Tomita and Ynagida, 1981), its development involves all the intraretinal nerve connections, and its amplitude thus reflects the quality of the retinal metabolism. The average survival curve obtained in diabetic animals 1 month after the evolution of diabetes is shown in Figure 3. It can be observed that during the first hour of survival, the ERG amplitudes recorded in I-month diabetic rats are comparable to those of the reference curve, whereas during the remainder of the survival period, the ERGs of the diabetic rats are systematically less in amplitude than the reference ERGs. The amplitude of the ERGs obtained on
303
c.....
o,~----~----~----~----~----~----~ 2 3 4 6 5
Time (hours)
Figure 2. Normal evolution of ERG b-wave amplitude during retina survival in standard conditions. On the upper right an example of electroretinogram (ERG) recorded on isolated retina.
• Normal rats .. Diabetic rats (1 month) o Diabetic rats (2 months)
400
~3oo
..:; Q)
"0 :l
~200 E
as
(!:J
II: W 100
o
------r------rl------TI------TI----------234
~I
o
Times (h)
Figure 3. Evolution of ERG amplitude recorded on retina isolated from diabetic rats (1 month and 2 months) compared to ERG recorded on normal rat retina.
2-month diabetic rat retinas are systematically lower than the reference values over the entire survival period (Fig. 3). It can be observed that the decrease in the ERG amplitude is directly correhlted with the duration of the evolution of diabetes. At the point of maximum amplitude in the survival curve of the control rats, the decrease in the ERG amplitude is around 20% for I-month diabetic rats, whereas it is around 60% for 2-month diabetic rats.
304 The animals that had been treated with Ginkgo bi/oba extract were sacrificed after 2-month evolution of diabetes. In comparison with the non-treated animals, it can be observed that the ERG amplitudes are increased, especially at beginning of the survival period, the second half of the survival curves being superimposable both for treated and nontreated animals (Fig. 4). Figure 5 represents the average survival curves obtained with treated and non-treated rats after computer adjustment; this representation points out graphically the difference between the two curves. On such an alloxanic diabetes model, the increase of capillary permeability responsible for the blood retinal barrier disruption may be correlated with oxygenated free radical release (Murata et aI., 1981). Moreover, the elevated rate of membrane lipoperoxidation in the pho-
-8 100 ~
0. E as ~ 50 W
o+-------~------~------,-------,_--------4 2 3 o
Time (h)
Figure 4. Comparison between ERG recorded on diabetic rats after 2 months of diabetes evolution and on EGb 761 treated diabetic rats after the same evolution of diabetes .
>3 Q)
"0
• Diabetic rats 2 months (n
200
= 5)
• Diabetic rats (2 months) treated by Ginkgo biloba extract (n = 4)
150
~
0.100 E as
I
D
50 O+------r----~------r-----,------
o
2
3
4
Time (h)
Figure 5. The same curves than on Figure 4 but after computer adjustment.
305 toreceptors can be responsible for ERG alterations. In a parallel manner, the physiological protective mechanisms (essentially superoxide dismutase and a tocopherol) can also be deficient or overwhelmed (Nishimura and Kuriyamma, 1985b). In this experiment, ERG effects of Ginkgo bi/oba extract can be correlated both with oxygenated free radical reactivity and production. Ginkgo bi/oba extract is rich in benzopyrones such as heterosides, known for free radical scavenger properties (Doly et aI., 1985) and in Ginkgolides A, Band C, (Braquet et aI., 1987), which are potent PAF-acether antagonists; PAF is an immunomediator able to amplify eosinophil and leukocyte response and to induce the release of Oi- and OR' radicals. These radicals may be responsible for retinal membrane impairment and dysfunction in diabetic retinopathy. Conclusion In humans, the retina is one of the most aerated organs. The demand for molecular oxygen by the mitochondria of photoreceptors is particularly high. The rod outer segment membrane contains the highest level of polyunsaturated fatty acids of any tissue in the body. In these conditions, the retina appears as a neuro-sensorial tissue particularly sensitive to the reactivity of oxygenated free radicals. The retinal function, which corresponds to photon detection, is essentially based on membrane potential variations. In other words the quality of light perception is markedly linked to the quality of membranes. One of the most deleterious effects of oxygenated radicals is membrane lipoperoxidation which corresponds to a severe disorganization. Several sudies have demonstrated the exaggerated production of free radicals in the retina of diabetic patients. Here, the impairment of retinal function reported during the evolution of diabetes may be closely related to the reactivity of oxygenated radicals. Such experimental data suggest the possible use of free radical scavengers in order to prevent retinal alterations due to diabetes. The preliminary results obtained with Ginkgo bi/oba extract seem to confirm this possibility and demonstrate that free radical scavengers may play a key role in the prevention of diabetic retinopathy. Anderson, R. E., and Sperling, L. (1971) Lipids of ocular tissues. VII Positional distribution of the fatty acids in the phospholipids of bovine retina rod outer segments. Arch. Biochem. Biophys. 144: 673-677. Bazan, H. E. P., Careaga, M. M., and Bazan, N. G. (1985) Decreased utilization of [2- 3 H] glycerol in phospholipid and neutral glyceride biosynthesis in the retina of streptozotocindiabetic rats. Neurochem. Path. 3: 109. Blair, N. P., Tso, M. O. M., and Dodge, J. T. (1984) Pathologie studies of the blood-retinal barrier in the spontaneously diabetic BB rat. Invest. Ophthalmol. 25, 302-331.
306 Bloodworth, J. H. B., and Moditor, D. L. (1985) Ultrastructural aspect of human and canine diabetic retinopathy. Invest. Ophthalmol. 4: 1037-1048. Braquet, P., Doly, M., Bonhomme, B., and Meyniel, G. (1982) Peroxydation lipidique et activite electrique de la retine isolee: effect protecteur de l'extrait de Ginkgo biloba, in: Compte-rendus des Journees Internationales du Groupe Polyphenols. Toulouse (France), 29-30 September. Braquet, P., Touqui, L., Shen, T. Y., and Vargaftig, B. B. (1987) Perspectives in platelet-activating factor research. Pharmacol. Rev. 39: 97-145. Careaga, M. M., and Bazan, H. E. P. (1981) The rat retina is a useful in vivo model to study membrane lipid synthesis. Rates of biosynthesis of neutral glycerides and phospholipids. Neurochem. Res. 6: 1169-1178. Cohen, G. (1984) Oxy-radical production in alloxan-induced diabetes: an example of an in vivo metal-catalyzed Haber-Weiss reaction, in: Free Radicals in Molecular Biology. Armstrong D., Sohal, R. S., Cuther, R. G., Slater, T. F. eds. Raven Press, New York, pp. 307-316. Cunha-Vaz, J., De Abreu, J. R. F., Campos, A. J., and Figo, G. H. (1975) Early breakdown of the blood-retinal barrier in diabetes. Br. J. Ophthalmol. 59: 649-656. Doly, M., Isabelle, D. B., Vincent, P., Gaillard, G., and Meyniel, G. (1980) Mechanism of the formation of X-ray induced phosphenes. I. Electrophysiological investigations. Rad. Res. 82: 93-105. Doly, M., Braquet, P., Bonhomme, B., and Meyniel, G. (1984) Effects of lipid peroxidation on the isolated rat retina. Ophthalmic Res. 16: 292-296. Doly, M., Braquet, P., Droy, M. T., Bonhomme, B., and Vennat, J. C. (1985) Effects des radicaux libres oxygenes sur l'activite electrophysiologique de la retina isolee de rat. J. Fr. Ophthalmol. 8: 273-277. Doly, M., Droy-Lefaix, M. T., Bonhomme, B., and Braquet, P. (1986) Effets de l'extrait de Ginkgo biloba sur l'electrophysiologie de la retine isolee de rat diabetique. Presse Med. 15: 1480-1483. Doly, M., Braquet, P., Droy, M. T., et al. (1988) Alteration of electrophysiological function of isolated retina from alloxan-induced diabetic rats: effect of treatment with Ginkgo biloba extract. Neurochem. Pathol. 8: 15-26. Farnsworth, C. c., and Dratz, E. A. (1976) Oxidative damage of retinal rod outer segment membranes and the role of Vitamin E. Biochim. Biophys. Acta 443: 556-570. Ishibashi, T., Tanaka, K., and Taniguchi, Y. (1980) Disruption of blood-retinal barrier in experimental diabetic rats: an electron microscopic study. Exp. Eye Res. 30: 401-410. Kagan, V. E., Shvedova, A. A., Novikov, K. N., and Koslov, Y. P. (1973) Lipid-induced free radical oxidation of membrane lipids in photoreceptors of frog retina. Biochim. Biophys. Acta 330: 76-79. Murata, R., Nishida, T., Eto, S., and Mukai, N. (1981) Lipid peroxidation in diabetic rat retina. Metab. Pediat. Ophthalmol. 5: 83-87. Nishimura, c., and Kuriyama, K. (1985a) Alteration of lipid peroxide and endogenous antioxidant contents in retina of streptozotocin-induced diabetic rats: effect of Vitamin A administration. Japan J. Pharmacol. 37: 365-372. Nishimura, c., and Kuriyama, K. (1985b) Alterations in the retinal dopaminergic neuronal system in rats with streptozotocin-induced diabetes. J. Neurochem. 45: 448-455. Novikov, K. N., Kagan, V. E., Shvedova, A. A., and Kozlov, Y. P. (1975) Protein-lipid interactions on peroxide oxidation of lipids in the photoreceptor membrane. Biofizika 20: 1039-1042. Oakley, B., and Pinto, L. H. (1981) [Ca2 +1i modulations of membrane sodium conductance in rod outer segments, in: Cuttent Topics in Membrane and Transport. Muller, ed. Academic Press Inc., New York. p. 405. Palmberg, P. F. (1977) Diabetic Retinopathy. Diabetes 26: 703-711. Papachristodoulou, D., and Heath, H. (1977) Ultrastructural alterations during the development of retinopathy in sucrose-fed and streptozotocin-diabetic rats. Exp. Eye Res. 25: 371-384. Shvedova, A. A., Sidorov, A. S., Novikov, K. N. et al. (1979) Lipid peroxidation and electric activity of the retina. Vision Res. 19: 49-55. Tamai, A., and Tanaka, K. (1973) The ERG of the streptozotocin-diabetic albino rat. Folia Ophthalmol. Japan. 24: 847-850. Tanaka, T. (1981) Effects of various agents on the ERG of the streptozotocin-diabetic rat. Folia Ophthalmol. Japan. 72: 1470-1480.
307 Tanigushi, Y., and Nomura, T. (1968) Fine structure of retinal blood vessels in human diabetics. Acta Soc. Ophthalmol. Japan. 72: 1165-1178. Tomita, T., and Ynagida, T. (1981) Origins of the ERG waves. Visions Res. 21: 1703-1707. Yonemura, D., Aoki, T., and Tsuzuki, K. (1962) Electroretinogram in diabetic retinopathy. Arch. Ophthal. 68: 49-54. Yonemura, D. (1977) An electrophysiological study on activities of neuronal and non-neuronal retinal elements in man with reference to its clinical application. Acta Soc. Ophthalmol. Japan. 81: 1632-1665.
