A study of retinal oxidation : the effects of melatonin and the roles of nitric oxide and hydroxyl radicals

Abstract

Purpose The retina has a high concentration of poly-unsaturated fatty acid predisposed to uncontrolled oxidation, causing cellular damage and hence tissue dysfunction. The retina contains at least two potential antioxidants, vitamin E, a potent lipid-soluble antioxidant which the body cannot produce naturally, and melatonin (MaT), a lipid-and aqueous-soluble antioxidant locally produced in the retina. However, the roles of vitamin E and MaT as retinal antioxidants are unclear. The work reported here was to investigate free radical-induced damage in the retinal tissues and related protective phenomenon. Nitric oxide (NO*) was studied as a possible causative factor of retinal oxidation. The damaging effects of NO* on the retinal tissues were compared with those of hydroxyl radicals (*OH). The protective role of MaT (and its structurally related substances) was compared with that of vitamin E in vitro. Lastly, the anti-oxidative capability of MaT against *OH- and NO*-induced retinal damage was evaluated in vivo. Methods Albino Sprague-Dawley and Wistar rats were used in all experiments. Argon laser irradiation was applied on 2 occasions, 7 days apart, to the episceleral and limbal drainage veins (and trabecular meshwork) of the right eyes. The intra-ocular pressure (IOP) was measured on Day 21, Day 28 and Day 35 after the second laser treatment. The retinal ganglion cell (RGC) counts and the total nitrite/nitrate levels (nitric oxide synthase activity) were determined after 35 days. In the second experiment, the extent of *OH and NO*-induced cell damage was determined after 60 min incubation of retinal homogenates with different concentrations (0, 2, 20 or 200 pM) of iron (II) sulphate (*OH generator) or sodium nitroprusside (SNP; NO* donor) at 37 C. Retinal susceptibility to free radical damage was also compared with those of two other tissues - kidney and liver, the levels of malondialdehyde (MDA) produced per unit weight protein providing an index of lipid peroxidation (LPO). In the third and fourth set of experiments, the protective effects of MaT on the *OH and NO*-induced retinal LPO were compared with that of vitamin E (Trolox). The protective efficiencies of indoleamines (i.e. MaT, N-acetyl-serotonin and pinoline), vitamin E, and of their combinations at different concentrations were also evaluated. In the final set of experiments, MaT was injected intra-peritoneally into the animals prior to the application of either whole body ionising radiation (inducing *OH) or intra-vitreal injection of SNP (inducing NO*). Retinal damage was assessed morphologically by the measurement of inner retinal layer (IRL) thickness, histologically by the expression of hyperchromatic nucleus and bio-chemically by apoptotic assay. Results The laser treatment resulted in significantly higher IOP in the ensuing 3-5 weeks (p<0.001). On Day 35, the mean IOP of the laser treated eyes was 23.6+-0.61 mmHg compared with 10.9+-0.46 mmHg in the control eyes. The IOP-elevated retinas expressed a statistically significant 2.4x higher retinal nitrite/nitrate level (p=0.011) and 28.2% less RGC density (p=0.01). Following the incubation with *OH or NO*, MDA levels showed that the retina resulted in several times higher amount of free radical-induced LPO than the kidney (p<0.001) and liver (p<0.001). The co-incubation of MaT (and 2 other indoleamines), vitamin E, and their mixtures significantly suppressed retinal LPO inflicted by both *OH and NO* (p=0.0001) in a dose-dependent manner (p=0.0001). The protective efficiency of MaT was approximately 15% of that of vitamin E in both systems, however mixtures of indoleamines and vitamin E demonstrated synergistic protection against NO*-induced retinal LPO. Administration of MaT prior to acute elevation of *OH protected the retina from developing IRL oedema (p<0.001) in the first 12 hr in vivo. Similarly, MaT prevented an increase in IRL thickness over a 24 hr period (p=0.039) compared with the control animals, but not over 96 hr (p=0.432) after the introduction of NO* in vivo. MaT also suppressed the development of IRL hyperchromatic nuclear expression and ganglion cell apoptosis for the first 24 hr only. Conclusion An approximately two-fold elevation in IOP increases retinal NO* production and decreases the viable RGC density, supporting an alternative glaucoma model induced by laser. Elevated NO* production is potentially threatening to the retina, which is more susceptible to free radical-induced LPO than kidney and liver. A role for free radical oxidation in the pathogenesis of glaucoma is further supported by the dose-dependent increase in retinal LPO damage which occurred after incubation of NO* in vitro. MaT, and other structurally related substances, protected the retina from *OH-and NO*-induced cellular LPO to an extent less than the potent lipid-soluble antioxidant vitamin E. LPO damage can be minimised effectively by mixtures of vitamin E and indoleamines. Intra-peritoneal administration of MaT protected the retina for the first 12-24 hr in vivo, however, the emergence of delayed retinal change suggests the mechanism of either a secondary generation of long-life free radical species, an exhaustion of MaT, or both. Further studies are indicated to investigate the role of oxidation in retinal damage and the applications of MaT (and other indoleamines) in retinal oxidative disorders, with particular attention to its pharmacological uptake and anti-oxidative mechanisms.