INVOLVEMENT OF ADENOSINE IN RETINAL ISCHEMIA - STUDIES ON THE RAT

Purpose. The aim of this study was to determine whether adenosinergic agents can be used to slow down the changes seen in the rat retina aft er ischemia-reperfusion. Methods. Ischemia-reperfusion injury to the r at retina was induced by raising the intraocular pressure above the sy stolic blood pressure for 45 minutes, followed by reperfusion for 3 da ys. This insult caused a reduction of the b-wave of the electroretinog ram (54%+/-5%, n=23) relative to the contralateral control retina, an expression of glial fibrillary acidic protein (GFAP) in the Muller cel ls, and an alteration in the ''staining'' pattern of the calretinin im munoreactivity. The normal two to three bands of calretinin immunoreac tivity in the inner plexiform layer appeared as a single band. Elevati on of the intraocular pressure for 60 minutes, followed by reperfusion of 2 weeks, caused a 40% reduction in the thickness of the inner nucl ear and plexiform layers. No statistically significant changes in the other retinal layers were recorded. Results. When the adenosine deamin ase inhibitor erythro-9-(2-hydroxyl-3-nonyl) adenine (EHNA) was inject ed into the eye just before ischemia, the ischemia-reperfusion changes in the b-wave and calretinin immunoreactivity were largely prevented. Similar results were observed when the adenosine A(1) receptor agonis t, R-N-6-(2-phenylisopropyl)adenosine (R-PIA), was administered intrap eritoneally just before ischemia. Injection of adenosine deaminase int o the eye before ischemia seemed to potentiate the ischemia-reperfusio n effect because the reduction of the b-wave was almost complete (8%+/ -4%, n=6). The ischemia-reperfusion-induced expression of GFAP in the Muller cells was not reduced by any of the adenosinergic agents tested . This suggests that GFAP expression in the Muller cells is not relate d to a reduction in the b-wave. An injection of EHNA into the eye befo re ischemia reduced the thinning of the inner plexiform and nuclear re tinal layers so that no significant difference between them and the co ntrol retinas existed. However, an injection of R-PIA just before isch emia did not reduce the thinning of the retinal layers in a statistica lly significant way, possibly because the R-PIA. protective effect is less than that of EHNA and is difficult to detect when thickness of th e retinal layers is measured. It may be necessary to use higher concen trations of R-PIA to observe a protective effect. Conclusions. The com bined data show that substances resulting in the activation of adenosi ne A(1) receptors protect the retina against changes induced by ischem ia-reperfusion.