Mechanisms of glutamate metabolic signaling in retinal glial (Müller) cells.

Retinal Müller (glial) cells metabolize glucose to lactate, which is preferentially taken up by photoreceptor neurons as fuel for their oxidative metabolism. We explored whether lactate supply to neurons is a glial function controlled by neuronal signals. For this, we used subcellular fluorescence imaging and either amperometric or optical biosensors to monitor metabolic responses simultaneously from mitochondrial and cytosolic compartments of individual Müller cells from salamander retina. Our results demonstrate that lactate production and release is controlled by the combined action of glutamate and NH(4)(+), both at micromolar concentrations. Transport of glutamate by a high-affinity carrier can produce in Müller cells a rapid rise of glutamate concentration. In our isolated Müller cells, glutamine synthetase (GS) converted transported glutamate to glutamine that was released. This reaction, predominant when enough NH(4)(+) is available, was limited at micromolar concentrations of NH(4)(+), and more glutamate entered then as substrate into the mitochondrial tricarboxylic acid cycle (TCA). Increased production of glutamine by GS leads to increased utilization of ATP, some of which is generated glycolytically. Methionine sulfoximine, a specific inhibitor of GS, suppressed the stimulatory effect of glutamate and NH(4)(+) on glycolysis and induced massive entry of glutamate into the TCA cycle. The rate of glutamine production also determined the amount of pyruvate transaminated by glutamate to alanine. Lactate, alanine, and glutamine can be taken up and metabolized by photoreceptor neurons. We conclude that a major function of Müller glial cells is to nourish retinal neurons and to metabolize the neurotoxic ammonia and glutamate.

Kinetics of oxygen consumption after a single flash of light in photoreceptors of the drone (Apis mellifera).

The time course of the rate of oxygen consumption (QO2) after a single flash of light has been measured in 300-micrometers slices of drone retina at 22 degrees C. To measure delta QO2(t), the change in QO2 from its level in darkness, the transients of the partial pressure of O2 (PO2) were recorded with O2 microelectrodes simultaneously in two sites in the slice and delta QO2 was calculated by a computer using Fourier transforms. After a 40-ms flash of intense light, delta QO2, reached a peak of 40 microliters O2/g.min and then declined exponentially to the baseline with a time constant tau 1 = 4.96 +/- 0.49 s (SD, n = 10). The rising phase was characterized by a time constant tau 2 = 1.90 +/- 0.35 s (SD, n = 10). The peak amplitude of delta QO2 increased linearly with the log of the light intensity. Replacement of Na+ by choline, known to decrease greatly the light-induced transmembrane current, caused a 63% decrease of delta QO2. With these changes, however, the kinetics of delta QO2 (t) were unchanged. This suggest that the recovery phase is rate-limited by a single reaction with apparent first-order kinetics. Evidence is provided that suggests that this reaction may be the working of the sodium pump. Exposure of the retina to high concentrations of ouabain or strophanthidin (inhibitors of the sodium pump) reduced the peak amplitude of delta QO2 by approximately 80% and increased tau 1. The increase of tau 1 was an exponential function of the time of exposure to the cardioactive steroids. Hence, it seems likely that the greatest part of delta QO2 is used for the working of the pump, whose activity is the mechanism underlying the rate constant of the descending limb of delta QO2 (t).

Light-induced oxygen consumption in Limulus ventral photoreceptors does not result from a rise in the intracellular sodium concentration.

Illumination of Limulus ventral photoreceptors leads to an increase in the intracellular concentration of sodium, [Na+]i, and to an increase in the consumption of O2 (delta QO2). After a 1-s light flash, it takes approximately 480 s for [Na+]i to return to within 10% of its preillumination level, whereas delta QO2 takes approximately 90 s. Thus, the delta QO2 is complete long before [Na+]i has returned to its resting level. Pressure injection of Na+ into the cell in order to elevate [Na+]i to the same levels as attained by illumination causes a rise in [Na+]i that returns to baseline with the same time course as the light-induced rise in [Na+]i. However, the injection of Na+ does not lead to an increase of the consumption of O2. We conclude that activation of the Na pump by a rise in [Na+]i is not a factor involved in the light-induced activation of O2 consumption in these cells.

The light-induced increase of carbohydrate metabolism in glial cells of the honeybee retina is not mediated by K+ movement nor by cAMP.

The retina of the honeybee drone is a nervous tissue in which glial cells and photoreceptor neurons constitute two distinct metabolic compartments. The phosphorylation of glucose and its subsequent incorporation into glycogen occur essentially in glia, whereas O2 consumption occurs in the photoreceptors. After [3H] glucose loading of superfused retinal slices, light stimulation induced a significant rise in [3H] glycogen turnover in the glia. This occurs without a concomitant covalent modification of glycogen enzymes. Probably only an increase or a decrease of the availability of [3H] glycosyls that are incorporated into glycogen is necessary. As only photoreceptors are directly excitable by light, we searched for a signal that stimulates glycogen metabolism in the glia. Although K+ in extracellular space and glia increases after repetitive light stimulation, increasing bath K+ in the dark did not mimic the metabolic effects of light, despite an equivalent increase of K+ in the extracellular space and glia. We subsequently explored the role of cAMP, a universal intracellular second messenger. Exposure of retinal slices to the adenylate-cyclase activator forskolin induced an expected increase in the rate of formation of cAMP, but only partially mimicked the metabolic effects of light. Furthermore, light stimulation failed to induce a rise in the rate of formation of cAMP. We conclude that in this nervous system, without synapses, neither K+ nor cAMP mediates the effect of light stimulation on intraglial glucose metabolism.

Diffusion and consumption of oxygen in the superfused retina of the drone (Apis mellifera) in darkness.

Double-barreled O2 microelectrodes were used to study O2 diffusion and consumption in the superfused drone (Apis mellifera) retina in darkness at 22 degrees C. Po2 was measured at different sites in the bath and retinas. It was found that diffusion was essentially in one dimension and that the rate of O2 consumption (Q) was practically constant (on the macroscale) down to Po2 s less than 20 mm Hg, a situation that greatly simplified the analysis. The value obtained for Q was 18 +/- 0.7 (SEM) microliter O2/cm3 tissue . min (n = 10), and Krogh's permeation coefficient (alpha D) was 3.24 +/- 0.18 (SEM) X 10(-5) ml O1/min . atm . cm (n = 10). Calculations indicate that only a small fraction of this Q in darkness is necessary for the energy requirements of the sodium pump. the diffusion coefficient (D) in the retina was measured by abruptly cutting off diffusion from the bath and analyzing the time-course of the fall in Po2 at the surface of the tissue. The mean value of D was 1.03 +/- 0.08 (SEM) X 10(-5) cm2/s (n = 10). From alpha D and D, the solubility coefficient alpha was calculated to be 54 +/- 4.0 (SEM) microliter O2 STP/cm3 . atm (n = 10), approximately 1.8 times that for water.

The response to monochromatic light flashes of the oxygen consumption of honeybee drone photoreceptors.

Local measurements of the fall in oxygen pressure on stimulation of slices of the retina of the honeybee drone by flashes of light were made with oxygen microelectrodes and used to calculate the kinetics of the extra oxygen consumption (delta QO2) induced by each flash. The action spectrum for delta QO2 was obtained from response-intensity curves in response to brief (40 ms) monochromatic light flashes. The action spectrum of receptor potentials was obtained with the same experimental conditions. The two action spectra match closely: they deviate slightly from the photosensitivity spectrum of the drone rhodopsin (R). The deviation is thought to be due to wavelength-dependent light scattering and absorption in the preparation. In these experiments, the visual pigment was first illuminated with orange light, which is known to convert the bistable drone photopigment predominantly to the R state from the metarhodopsin (M) state. When long (300-900 ms) light flashes were used to elicit delta QO2, the responses to different wavelengths could not be matched in time course (as for the short flashes). Flashes producing large R-to-M conversions produced a prolonged delta QO2. The prolongation did not occur after double flashes, which produced both large R-to-M and M-to-R conversions. Similar changes in the length of afterpotentials in the photoreceptor cells and in a long-lasting decrease in photoreceptor intracellular K+ activity were found after long single or double flashes. The results are interpreted to show that the initial event for stimulation by light of metabolism in the drone retina is the same as that for stimulation of electrical responses (i.e., absorption of photons by R). Absorption of photons by M can produce an inhibitory effect on this stimulation.

The increase of oxygen consumption after a flash of light is tightly coupled to sodium pumping in the lateral ocellus of barnacle.

In the lateral ocellus of the barnacle, we have tested the hypothesis that the transient increase of oxygen consumption (delta QO2) induced by light results from an increase in the rate of Na+ pumping. With a Na(+)-sensitive microelectrode, we measured the intracellular concentration of Na+ (Nai) in the photoreceptor cells. Nai was 17.6 +/- 1.2 mM (SE; n = 18) in darkness and it increased transiently by 10-20 mM after an 80-ms flash of intense light. The increase of Nai recovered in about the same time as the delta QO2, and the Na+/O2 ratio was 19.2 +/- 3.8 (SE; n = 6). Removing Na+ from the bath caused the delta QO2 to decrease by 79 +/- 3% (SE; n = 5). Exposure to 25 microM ouabain inhibited Na+ pumping and abolished the delta QO2. Removal of K+ from the bathing solution inhibited Na+ pumping in darkness, but mostly shortened the duration of the delta QO2; with a K(+)-sensitive microelectrode, we measured pericellular [K+] and found that it increased after the flash for about the same time as the delta QO2. Increasing Na+ pumping in darkness by reintroducing K+ in the bath or by injecting Na+ into one of the photoreceptor cells induced a delta QO2. Finally, intracellular injection of adenosine diphosphate and inorganic phosphate (ADP + Pi), the metabolic products of ATP splitting by the Na+ pump, also induced a delta QO2 in darkness. We conclude that all the results obtained are consistent with the formulated hypothesis.

Kinetics of oxygen consumption and light-induced changes of nucleotides in solitary rod photoreceptors.

We made simultaneous measurements of light-induced changes in the rate of oxygen consumption (QO2) and transmembrane current of single salamander rod photoreceptors. Since the change of PO2 was suppressed by 2 mM Amytal, an inhibitor of mitochondrial respiration, we conclude that it is mitochondrial in origin. To identify the cause of the change of QO2, we measured, in batches of rods, the concentrations of ATP and phosphocreatine (PCr). After 3 min of illumination, when the QO2 had decreased approximately 25%, ATP levels did not change significantly; in contrast, the amount of PCr had decreased approximately 40%. We conclude that either the light-induced decrease of QO2 is not caused by an increase in [ATP] or [PCr], or that the light-induced change of [PCr] is highly heterogeneous in the rod cell.

Trafficking of molecules and metabolic signals in the retina.

Photoreceptors need the support of pigment epithelial (PE) and Müller glial cells in order to maintain visual sensitivity and neurotransmitter resynthesis. In rod outer segments (ROS), all-trans-retinal is transformed to all-trans-retinol by retinol dehydrogenase using NADPH. NADPH is restored in ROS by the pentose phosphate pathway utilizing high amounts of glucose supplied by choriocapillaries. The retinal formed is transported to PE cells where regeneration of 11-cis-retinal occurs. Müller cells take up and metabolize glucose predominantly to lactate which is massively released into the extracellular space (ES). Lactate is taken up by photoreceptors, where it is transformed to pyruvate which, in turn, enters the Krebs cycle in mitochondria of the inner segment. Stimulation of neurotransmitter release by darkness induces 130% rise in the amount of glutamate released into ES. Glutamate is transported into Müller cells where it is predominantly transformed to glutamine. Stimulation of photoreceptors induces an eightfold increase in glutamine formation. It appears, therefore, that there is a signaling function in the transfer of amino acids from Müller cells to photoreceptors. Work on the model-system of the honeybee retina demonstrated that photoreceptors release NH4+ and glutamate in a stimulus-dependent manner which, in turn, contribute to the biosynthesis of alanine in glia. Alanine released into the extracellular space is taken up and used by photoreceptors. Glial cells take glutamate by high-affinity transporters. This uptake induces a transient change in glial cell metabolism. The transformation of glutamate to glutamine is possibly also controlled by the uptake of NH4+ which directly affects cellular metabolism.

Glucose metabolism in freshly isolated Müller glial cells from a mammalian retina.

Glucose metabolism was studied in isolated retinal Müller glial cells from the juvenile guinea pig. Cells, once enzymatically isolated and purified, were identified by morphological criteria, positive vimentin immunoreactivity, and histochemical staining for glycogen. Purified suspensions of Müller cells were obtained in quantities sufficient for biochemical analysis (approximately 2 x 10(5)/pair of retinas) and light microscopic autoradiography. In bicarbonate-buffered Ringer's medium containing 3H-2-deoxyglucose and no glucose, greater than or equal to 80% of the glucose analogue taken up intracellularly by Müller cells was phosphorylated to 3H-2-deoxyglucose-6-phosphate. In autoradiographs, this non-metabolized product provided visual evidence of glucose phosphorylation: the distribution of cell grains mirrored the morphology of individual Müller cells in situ. Exposure to the glycolytic inhibitor iodoacetate (500 microM) caused an 85% decrease in adenosine triphosphate (ATP) content; concomitantly, 3H-2-deoxyglucose-6-phosphate decreased by 90% and paralleled a dramatic decrease of cell labelling in autoradiographs, while levels of 3H-2-deoxyglucose did not change. In the continual absence of glucose, glycogen content decreased with time and this decrease was slowed by 36% in the presence of iodoacetate. This indicated that, in control conditions, glycosyl units from glycogen sustain cellular metabolism, and hence 3H-2-deoxyglucose phosphorylation. 3H-2-deoxyglucose-6-phosphate concentration was 43-fold less than that of ATP in the control conditions so that depletion of ATP during iodoacetic acid (IAA)-blocked glycolysis was not due to hexokinase activity. These results demonstrate that this preparation is adequate for quantitative studies of glucose metabolism at the cellular and molecular level in an important metabolic compartment of the mammalian retina.

Effect of nitroprusside on arteriolar constriction after retinal branch vein occlusion.

PURPOSE: The development of extended areas of nonperfused capillaries after branch vein occlusion (BVO) is correlated to the secondary constriction of the arteriole crossing the occluded area. The decrease in nitric oxide (NO) in tissue that occurs early after BVO accounts for the secondary arteriolar constriction. The present study shows that the administration of an NO donor can reverse the secondary arteriolar vasoconstriction observed after BVO. METHODS: Simultaneous preretinal NO profiles and arteriolar diameter measurements were performed in miniature pigs after experimental BVO. The effect of preretinal microinjections of the NO donor sodium nitroprusside (SNP) on the arteriolar diameter was studied. RESULTS: Significant arteriolar vasoconstriction (mean arteriolar diameter, 92.1% +/- 3.3% of control; n = 7; P = 7.4 x 10(-5)) and a simultaneous decrease in the preretinal NO concentration ([NO]) (preretinal [NO], 20% +/- 15.6% of control; n = 5; P = 0.0003) were observed 4 hours after BVO. Microinjection of the NO donor SNP (1 mM applied by puffer) near the constricted retinal arteriole caused a segmental, reversible arteriolar dilation that reached its maximum 20 minutes after the injection (mean arteriolar diameter; 110.8% +/- 7.5% of control; n = 6; P = 0.02) and was completely reversed 60 minutes later (n = 6). CONCLUSIONS: Local administration of NO donors may contribute to the restoration of the retinal arteriolar blood flow after BVO and thus may improve the supply of oxygen and nutrients to the injured tissue.

Decreased nitric oxide production accounts for secondary arteriolar constriction after retinal branch vein occlusion.

PURPOSE: After retinal branch vein occlusion (BVO), the arteriole crossing the occluded territories is often constricted. This constriction persists up to several weeks and is correlated with the development of extended territories of nonperfused capillaries. These are results of an investigation supporting the hypothesis that decrease in the production of nitric oxide (NO) accounts for the observed arteriolar constriction. METHODS: Preretinal [NO] was measured using an NO microprobe in the anesthetized miniature pigs, before and during the first 4 hours after experimental branch vein occlusion. Modifications of arteriolar diameter were correlated to preretinal [NO] changes. The retinal arteriolar sensitivity to constitutive NO was checked by applying preretinal puff injections of nitro-L-arginine (L-NA) after both systemic hypoxia and branch vein occlusion. RESULTS: Two hours after branch vein occlusion there was a 73.7 +/- 4% decrease in preretinal [NO] and a simultaneous 25.4 +/- 3.4% decrease in the diameter of the arteriole in the affected territory. Both persisted for at least 4 hours after branch vein occlusion. Applying a puff of L-NA to an arteriole previously dilated by systemic hypoxia induced a vasoconstriction. However, no arteriolar constriction was observed when a puff was applied to an arteriole after branch vein occlusion. CONCLUSIONS: These results show that experimental branch vein occlusion induces in the affected retina an impairment in the release of constitutive NO and an arteriolar constriction, which, in turn, contributes to the development of hypoxia in tissue and neuronal swelling and death in the inner retina.

Microinjection of L-lactate in the preretinal vitreous induces segmental vasodilation in the inner retina of miniature pigs.

PURPOSE: The authors investigated the hypothesis that the retinal vasomotor effect of acute hypoxia is mediated by lactate. METHODS: Retinal vasomotor arteriolar response was measured in the intact eyes of miniature pigs after systemic administration and after local preretinal juxta-arteriolar microinjection of lactate. RESULTS: Injection of L-lactate (physiologically produced lactate) into the systemic circulation decreased the arterial blood pH but did not dilate the retinal arterioles. By contrast, microinjections of L-lactate (0.5 mol/l, pH 2) into the juxta-arteriolar vitreous induced a reversible segmental vasodilation of 32 +/- 4% (standard deviation). This vasodilation did not depend on periarteriolar pH lowering because microinjections of a 0.5 mol/l L-lactate at neutral pH also dilated segmentally the retinal arterioles (37 +/- 5.5%). The effect of lactate was stereospecific because microinjections of the isomer D-lactate (0.5 mol/l, pH 2) did not affect the arteriolar caliber (P = 0.63). Perfusion of the eye with the cyclo-oxygenase inhibitor indomethacin, through cannulization of the sublingual artery, caused a generalized reversible arteriolar vasoconstriction of 51 +/- 9.8% but did not inhibit the segmental vasodilator effect of locally microinjected L-lactate. CONCLUSIONS: It is known that acute hypoxia in the isolated retina causes an increase in lactate production. In the intact eye, there is a retinal vasodilation, which is not inhibited by indomethacin. Hence, it was concluded that retinal, but not blood, lactate is a possible mediator of the acute hypoxia-induced vasodilation.

Effect of laser photocoagulation on oxygenation of the retina in miniature pigs.

A large area of the posterior pole of the retina of the miniature pig was photocoagulated, and 2-3 wk later the PO2 in the preretinal vitreous was mapped with O2-sensitive microelectrodes. In the control retina, the PO2 was highly heterogeneous being much higher close to an artery than opposite an intervascular zone. After photocoagulation, PO2 opposite an intervascular zone was found to be significantly increased. A quantitative histologic analysis showed that in the photocoagulated areas more than 28% of the outer retina was destroyed. The authors conclude from these results that photocoagulation, by partially destroying the pigment epithelium-photoreceptor complex, causes an increase of the oxygenation of the retina.

Nitric oxide controls arteriolar tone in the retina of the miniature pig.

PURPOSE: Experimental evidence indicates that the retinal microcirculation is mainly controlled by factors released from the tissue surrounding the arterioles. This study explores whether nitric oxide (NO), a possible factor, is released in the retina and controls the arteriolar tone. METHODS: Using a NO microprobe, the authors measured [NO] in the preretinal vitreous of miniature pigs as a function of distance from the retinal surface. Additionally, the NO-synthase inhibitor nitro-L-arginine was pressure injected. Finally, the retinal pool size of arginine and its biosynthesis from 14C(U)-glucose were biochemically assessed on retinal tissue and acutely isolated Müller cells. RESULTS: At the retinal surface, [NO] measured 6 to 9 microM, and, in the vitreous, it fell to zero approximately 180 microns away from the retina. Therefore, NO is degraded faster in the vitreous (65 to 80 microM.minute-1) than in aqueous solution. Light flicker stimulation of the dark-adapted retina induced a reversible increase of [NO] (approximately 1.6 microM). Preretinal juxta-arteriolar microinjections of nitro-L-arginine (0.6 mM) induced a segmental and reversible arteriolar vasoconstriction of 45%; in contrast, intravenous infusion of nitro-L-arginine had no measurable effect on arteriolar diameter. The retinal pool size of arginine was small (< or = 200 microM), but there was an important rate of arginine biosynthesis in Müller cells. CONCLUSIONS: These results strongly suggest that cells in the retina, other than endothelial cells, produce and release NO, which in turn controls the basal dilating arteriolar tone in the inner retina.

Regulation of local oxygen tension and blood flow in the inner retina during hyperoxia.

A study has been undertaken to determine whether local changes in PO2 could be a factor in the regulation of retinal blood flow during 100% O2 breathing. For this purpose we have measured simultaneously in eyes of anesthetized and artificially ventilated miniature pigs the change in local preretinal PO2 and retinal blood flow using O2-sensitive microelectrodes and laser Doppler velocimetry. Although preretinal intervascular PO2 changed little, periarteriolar PO2 increased markedly during 100% O2 breathing. When measured less than 50 microns from the arteriolar wall, the time course of this increase preceded that of the decrease in retinal blood flow. The results indicate that O2 diffusing through the wall of the large retinal arterioles represents the most important component of periarteriolar PO2. This diffusion coupled with the decrease in retinal blood flow could play a major role in the regulation of inner retinal PO2. The data also suggest that diffusional shunting of O2 between retinal arterioles and veins could explain the unexpected increase in venous O2 saturation during 100% O2 breathing that has been previously reported by others. The finding that retinal periarteriolar PO2 is always higher than the PO2 in the outer retina does not support the previously formulated hypothesis that O2 from the choroid is responsible for the constriction of the arterioles during hyperoxia.

Metabolic signaling between photoreceptors and glial cells in the retina of the drone (Apis mellifera).

Experimental evidence showing metabolic interaction and signaling between photoreceptors-neurons and glial cells of the honeybee drone retina is presented. In this tissue [3H]2-deoxyglucose ([3H]2DG) in the dark and during repetitive light stimulation is phosphorylated to [3H]2-deoxyglucose-6P ([3H]2DG-6P) almost exclusively in the glial cells. Hence, stimulus-induced changes in the rate of formation of [3H]2DG-6P occurs predominantly in the glial cells. Repetitive stimulation of the photoreceptors with light flashes induced about a 47% rise in the rate of formation of [3H]2DG-6P in the glial cells and this effect is probably due to the activation of hexokinase. The potent inhibitor of glycolysis iodoacetic acid (IAA), inhibited this phosphorylation by about 75%. Probably this was largely due to an about 70% decrease of adenosine triphosphate (ATP). Exposure of the retina to IAA suppressed the transient rise in oxygen consumption (delta QO2) in the photoreceptors and subsequently the light-induced receptor potential. This indicates that the supply of a glycolytic substrate by glial cells to the photoreceptors is greatly reduced by IAA. Anoxia, by rapidly suppressing QO2, abolished the receptor potential of the photoreceptors and caused a rapid drop of about 50% in the ATP content of the retina. At the same time the formation of [3H]2DG-6P was inhibited by about 30%. This indicates that respiring photoreceptors send a metabolic signal to glial cells which is suppressed by anoxia.

Clearance of extracellular potassium: evidence for spatial buffering by glial cells in the retina of the drone.

Work with ion-selective microelectrodes on the retina of the honeybee drone has shown that potassium is released from photoreceptors during activity and enters glial cells. Measurements of the extracellular voltage gradients indicate that, in this preparation, currents flowing through the glial cells in the 'spatial buffer' pattern account for a large fraction of the glial K+ entry in the active region of the tissue.

Glial (Müller) cells take up and phosphorylate [3H]2-deoxy-D-glucose in mammalian retina.

[3H]2-Deoxy-D-glucose-6-PO4 ([3H]2DG-6P) was visualised at the level of single cells in freeze-dried guinea pig retinal sections by in vitro light microscopic autoradiography after incubation with [3H]2-deoxy-D-glucose ([3H]2DG). In the dark, autoradiographs revealed heterogeneous labeling within individual retinal layers. Labeling, representing [3H]2DG-6P, was preferentially located over Müller (glial) cells. Labelling over identified neurones in the inner nuclear layer, in contrast, was scarce and over ganglion cells was exceptional. Our observations indicate that Müller cells in the mammalian retina phosphorylate [3H]2DG to [3H]2DG-6P, the first step in glycolysis.

Light induced sodium dependent accumulation of calcium and potassium in the extracellular space of bee retina.

Intense illumination of long duration induced a large transient increase in extracellular calcium (delta[Ca2+]o) and potassium (delta[K+]o) during and after light in bee retina when measured with ion-selective microelectrodes. Whenever a large delta[Ca2+]o appeared, it was accompanied by a transient afterdepolarization (TA). Both the increase in [Ca2+]o, [K+]o and the TA were reduced or abolished when sodium was replaced by arginine, choline or lithium (Li+) ions. At 0-Na conditions a Na independent decrease in [Ca2+]o was observed during illumination only. A pronounced transient depolarization of the photoreceptor in the dark due to transient anoxia did not result in a significant change in [Ca2+]o. In some retinae the elevated level of [K+]o after light was absent, however a small Na-dependent TA was still observed. The above findings suggest that intense long illumination induces a large Ca2+ influx into the photoreceptors which is followed by Na-dependent Ca2+ efflux due to Na-Ca exchange. The light-induced afterdepolarization arises mainly from K+ accumulation in the extracellular space but partially from the electrogenicity of Na-Ca exchange.