Hydrogen Peroxide Affects the Electroretinogram of Isolated Perfused Rat Retina

Abstract

Purpose

To evaluate the effects of H₂O₂ as an oxidant on the electroretinogram (ERG) in isolated rat retina.

Methods

Retinas were isolated from rat eyes and perfused with a nutrient solution. ERGs were recorded every 3 min. Once the signal was at a steady state, H₂O₂ was added to the perfusion solution.

Results

H₂O₂ caused instantaneous and transient changes in amplitudes and implicit times of the ERG, followed by changes in retinal survival curves. H₂O₂ 0.2 mM produced a rapid increase in b-wave amplitude, followed by a return to the initial value and a survival curve above the control (without H₂O₂). A slight increase in a-wave was observed, followed by a decrease and a recovery above the control. The slow PIII decreased and then recovered to the initial value. H₂O₂ 0.6 mM induced a small increase in b-wave amplitude, followed by a rapid decrease without recovery. The a-wave and slow PIII decreased rapidly without recovery. The implicit times of the a-wave and b-wave increased moderately with a low dose of H₂O₂, whereas they significantly increased with a high dose. Whatever the dose, the slow PIII implicit time increased significantly, followed by a return to the initial value. Barium increased the a-wave and b-wave, and then H₂O₂ reduced the two waves with a stronger effect on the a-wave. Aspartate and barium isolated the fast PIII, which decreased after H₂O₂ application.

Conclusions

H₂O₂ affects retinal function as shown by ERGs in isolated rat retina. The response differs with the dose of H₂O₂, suggesting that mechanisms underlying the action at low doses might be different from those at high doses. Our results also suggest an effect of H₂O₂ on ionic currents and/or neurotransmitter releases involved in the generation of the ERG and indicate a more pronounced effect on photoreceptors than on postsynaptic cells.

Keywords:

• Hydrogen peroxide
• H₂O₂
• electroretinogram
• isolated retina
• rat

Introduction

Oxidative stress is involved in many ocular pathologies such as age-related macular degeneration (ARMD), various forms of retinopathies, glaucoma, and cataracts.¹ Oxidative stress corresponds to an imbalance between reactive oxygen species (ROS) production and anti-oxidative defense mechanisms, causing damage to deoxyribonucleic acids, proteins, and lipids. Hydrogen peroxide (H₂O₂) is a ROS that is naturally produced in the eye and whose level is increased under certain circumstances. For instance, H₂O₂ is produced during photo-oxidation of lipofuscin, a fluorescent brown pigment that accumulates in the retina during aging, and lipofuscin accumulation has been associated with ARMD.² H₂O₂ has also been shown to increase in retinas exposed to blue light, an identified risk factor for ARMD,³ and in diabetic rat retinas.⁴ In addition, H₂O₂ has been reported to be 15-fold higher in the vitreous of rabbit eyes after prolonged light exposure,⁵ and 2-3 higher in the vitreous of human cataract patients.⁶ Although there is increasing evidence that H₂O₂ levels are increased in many ocular diseases, the consequences of the action of H₂O₂ on normal visual function remain unexplored to date.

Hydrogen peroxide is often considered harmful to the cells, but this oxidant is also involved in the regulation of various physiological processes ranging from normal cellular activity to cell death. Numerous studies conducted in the central nervous system have highlighted the involvement of H₂O₂ as a potent and rapid modulator of neuronal activity,^7–10 ion channel function^11–15 and synaptic neurotransmission.^16,17 In the retina, H₂O₂ has been widely used in vitro as an inducer of oxidative stress in retinal cell cultures, however its ex vivo effect in isolated retinas has been poorly investigated.^18,19 To our knowledge, only one in vivo study has evaluated the effect of H₂O₂ on the retina in mice.²⁰ In these studies, retinal structure and cellular integrity were analyzed, but retinal function was never assessed. Thus, the consequences of the action of H₂O₂ on retinal function remain unknown.

The electroretinogram (ERG) is an excellent tool for studying retinal function. It represents the electrical signal of the retina in response to a light stimulus. This signal is composed of a series of waves originating from different retinal cells and generated by ion currents. The aim of the present work was to determine the effects of H₂O₂ on the ERG, using a model of isolated perfused rat retina ERG recording. This is the first study on the effects of H₂O₂ on retinal function ex vivo.

Materials and methods

Animals

Male Wistar rats weighing 160–180 g were obtained from Depre (Saint-Doulchard, France). Rats were maintained in an animal house with a 12/12 h light/dark cycle. Food and water were supplied ad libitum. All the experiments were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmologic and Visual Research.

Isolated retina model

Retina was isolated according to the procedure described elsewhere.^21,22 Briefly, the rat was dark-adapted overnight and then sacrificed. Eyes were enucleated, dissected at the equator, and the eyecup was placed in a solution with the following composition: 121.5 mM NaCl, 3.1 mM KCl, 0.4 mM CaCl₂, 0.6 mM MgSO₄, 24.4 mM NaHCO₃, 0.5 mM NaH₂PO₄, 0.5 mM sodium glutamate, 28 mM glucose, 3% bovine serum. The retina was gently detached from the retinal pigment epithelium and released by cutting at the optic nerve level. The isolated retina was then placed in a chamber continuously perfused with the same solution described above. The perfusion liquid, maintained at 37 °C, was permanently bubbled by a mixture of 95% O₂ and 5% CO₂. All the preparation procedures were carried out under dim red light.

Electroretinogram (ERG) recording

ERGs were recorded with two Ag/AgCl electrodes connected on either side of the retina, amplified, processed, and stored. The recording bandwidth was set to 0.1-50 Hz. ERGs were elicited every 3 min using a white-light stimulation (0.025 lx.s scotopic, 1 ms). The light stimulus used to generate ERGs was delivered by an optic fiber on the rod outer segment side of the retina. Once the ERG was at a steady state (maximum or 100% of the response), H₂O₂ was added to the perfusion solution at various concentrations (0.2, 0.3, and 0.6 mM) for 3 h. The amplitudes and implicit times of the b-wave a-wave and slow PIII components versus times were plotted to obtain “the survival curves”.

Data analysis

The amplitude of the a-wave was measured from the baseline to the trough of the a-wave, the amplitude of the b-wave was measured from the trough of the a-wave to the peak of the b-wave, and the amplitude of the slow PIII component was measured from the baseline to the trough of the slow PIII. The implicit times (times to peak) of the a-wave, b-wave, and slow PIII components were measured from the flash onset to the peaks of the waves, respectively. Statistical analysis was performed using a two-way analysis of variance (ANOVA) followed by Tukey’s post hoc test for multiple comparisons. Differences with p-values <0.05 were considered as statistically significant.

Results

Our model of isolated retina affords an ERG in response to light stimulus representing the electrical signal of the neural retina. In our experimental conditions, the signal (Figure 1A) is composed of (i) an early negative a-wave reflecting the hyperpolarization of the photoreceptors, (ii) a positive b-wave corresponding primarily to the depolarization of the bipolar cells, and (iii) a negative slow PIII component related to the activities of Müller cells and reflecting K⁺ currents through Kir4.1 channels.²³

Figure 1. Electroretinographic recordings on isolated retina. Isolated rat retina was continuously perfused with a nutrient solution and ERGs were recorded every 3 min. A Example of a single ERG recorded on isolated rat retina in response to white-light stimulation (0.025 lx.s scotopic, 1 ms). The a-wave amplitude was measured from the baseline to the trough of the a-wave, the b-wave amplitude was measured from the trough of the a-wave to the peak of the b-wave, and the amplitude of the slow PIII component was measured from the basal line to the trough of the slow PIII wave. B Example of a b-wave survival curve showing the evolution of the b-wave amplitude (µV) versus time (min) from a representative experiment.

The light stimulus used to generate ERGs was repeated every 3 min, and the amplitudes and implicit times of the waves versus time were plotted to obtain “the survival curves”. Figure 1B shows the ERG b-wave amplitude survival curve of one untreated isolated retina, representing the evolution of the b-wave amplitude as a function of time in the control condition. The pattern of the survival curve is described as follows: first, the b-wave amplitude increases during 30-45 min due to the adaptation of the retina to the environmental conditions; then the amplitude reaches a plateau (stationary phase for approx. 30 min), following by a progressive decrease until the end of the retina survival (150–180 min). We set the time at 0 and the b-wave amplitude at 100% at the beginning of the stationary phase.

Under these conditions, we were able to follow the evolution of the ERG waves of the retina exposed to H₂O₂. Our results show that application of H₂O₂ on isolated retina induces biphasic effects in ERG, characterized by immediate changes in the amplitudes and implicit times of the ERG waves followed by changes in retinal survivals. The effects differ according to the low ("physiological") or high ("pathological") dose of H₂O₂ used. Figure 2 shows an example of ERGs recorded before (Figure 2a) and after (Figure 2b–d) application of H₂O₂ at 0.2 mM (low dose) on an isolated rat retina. Shortly after the addition of H₂O₂ (Figure 2b–c), we observed a significant increase in the b-wave amplitude and a significant reduction in the amplitudes of the a-wave and slow PIII components. The implicit time of the a-wave remained unchanged, that of the b-wave showed a moderate increase, whereas that of the slow PIII component was significantly extended. Ninety minutes after the addition of H₂O₂ (Figure 2d), all the ERG parameters returned to their initial values.

Figure 2. Representative effect of H₂O₂ 0.2 mM on the ERG waves. Isolated rat retina was continuously perfused with a nutrient solution, and ERGs were recorded in response to white-light stimulation (0.025 lx.s scotopic, 1 ms). ERGs were recorded (a) before application of H₂O₂, (b) 6 min, (c) 12 min, and (d) 90 min following application of H₂O₂.

The evolution of the ERG wave's relative amplitudes in the presence of H₂O₂ is displayed in Figure 3A. Specifically, the application of H₂O₂ at 0.2 mM on isolated perfused retinas produced an immediate, significant, and transient increase in the amplitude of the b-wave, followed by a decrease and recovery to the initial value. Retinal survival was not affected and even remained above the control thereafter. A similar slight and transient increase in the a-wave amplitude was observed, followed by a decrease and recovery above the control. The amplitude of the slow PIII component decreased and recovered to the initial value. H₂O₂ at 0.3 mM also induced a rapid increase of the b-wave amplitude, but to a lesser extent than at 0.2 mM, followed by a significant decrease with no recovery. The amplitudes of the a-wave and slow PIII decreased significantly with no recovery. H₂O₂ delivered at a higher dose of 0.6 mM led to a small increase of the b-wave amplitude, followed by a rapid and significant decrease without recovery. The amplitudes of the a-wave and slow PIII decreased more rapidly and significantly, again without recovery.

Figure 3. Effects of H₂O₂ on the evolution of the amplitudes and implicit times of the ERG waves. Isolated rat retinas were continuously perfused with a nutrient solution and ERGs were recorded every 3 min in response to white-light stimulation (0.025 lx.s scotopic, 1 ms). ERGs were measured in the absence (CTRL) or in the presence of H₂O₂ at concentrations of 0.2 mM (H 0.2), 0.3 mM (H 0.3) and 0.6 mM (H 0.6). A The survival curves represent (a) the evolution of the b-wave amplitudes (µV) versus time (min), (b) the evolution of the a-wave amplitudes (µV) versus time (min) and (c) the evolution of the slow PIII-wave amplitudes (µV) versus time (min). B The survival curves represent (a) the evolution of the b-wave implicit times (IT, ms) versus time (min), (b) the evolution of the a-wave implicit times (IT, ms) versus time (min) and (c) the evolution of the slow PIII-wave implicit times (IT, ms) versus time (min). Results are expressed in percentage relative to the initial value (100% of the wave amplitude or implicit time 6 min before injection). Each survival curve represents the average of at least four separate experiments. Dotted lines indicate the time of injection (t = 0 min), 6 min after injection of H₂O₂ (t = 6 min) and 72 min after injection of H₂O₂ (t = 72 min).

Figure 3B shows the evolution of the ERG wave's relative implicit times in the presence of H₂O₂. Application of H₂O₂ at 0.2 mM and 0.3 mM produced a moderate increase in the implicit time of the b-wave, which persisted thereafter. Perfusion with 0.6 mM of H₂O₂ produced a higher increase in the b-wave implicit time. Regarding the effects on the a-wave implicit time, no significant changes were observed after the addition of H₂O₂ at 0.2 and 0.3 mM, whereas an increase was observed with the concentration of 0.6 mM. On the other hand, the implicit time of the slow PIII component showed a marked and significant increase, followed by a return to the initial value, for all the concentrations tested.

The mean values of the relative amplitudes and implicit times of the ERG waves, calculated at the time of injection (t = 0 min), after 6 min of perfusion with H₂O₂ (t = 6 min) and after 72 min of perfusion (t = 72 min), are displayed as a histogram in Figure 4.

Figure 4. Mean values of the relative amplitudes and implicit times of the ERG-waves at the time of injection, and after 6 and 72 min of perfusion with H₂O₂. Isolated rat retinas were continuously perfused with a nutrient solution and ERGs were recorded in response to white-light stimulation (0.025 lx.s scotopic, 1 ms). ERGs were measured in the absence (CTRL) or in the presence of H₂O₂ at 0.2 mM (H 0.2), 0.3 mM (H 0.3) and 0.6 mM (H 0.6). Histograms represent the averaged values of (A) the amplitudes of the b-wave, the a-wave and the slow PIII, and (B) the implicit times of the b-wave, the a-wave and the slow PIII. Values were calculated at the time of injection (t = 0 min) and after 6 min (t = 6 min) and 72 min (t = 72 min) of perfusion with H₂O₂. Results are expressed in percentage relative to the initial value (100% of the wave amplitude or implicit time 6 min before injection). The data are presented as means ± SD (n ≥ 4 separate experiments). *p < 0.05, **p < 0.01, ***p < 0.001, ^#p < 0.05 and ^###p < 0.001. ANOVA followed by Tukey’s posthoc test.

The b-wave recorded from the isolated retina can be masked by the following large, slow and negative component, named the slow PIII. When barium, known to block the slow PIII,^24–26 was added to the perfusion solution (Figure 5), only the negative a-wave (hyperpolarization of photoreceptors) and the subsequent positive b-wave (depolarization of bipolar cells) could be recorded. After the addition of barium 1 mM in the solution, the amplitudes and implicit times of the a-wave and b-wave increased (Figure 5A–B). When H₂O₂ 0.3 mM was applied in the solution, the amplitudes of the two waves decreased and the implicit times increased (Figure 5A-B). The ratio of the b-wave amplitude to a-wave amplitude (b/a ratio), which indicates the relation of bipolar cells in response to photoreceptors was calculated. We observed that, whereas a-wave and b-wave amplitudes were reduced, the b/a ratio increased from 4 ± 1 to 6 ± 1 (n = 3). This shows that the amplitude reduction was higher for the a-wave than for the b-wave and may be due to a more pronounced effect of H₂O₂ on the photoreceptors than on the post-receptoral system.

Figure 5. ERG responses in the presence of H₂O₂ after barium chloride application. Isolated rat retinas were continuously perfused with a nutrient solution and ERGs were recorded every 3 min in response to white-light stimulation (0.025 lx.s scotopic, 1 ms). Barium chloride (BaCl₂, 1 mM) was added to the perfusion solution to suppress the slow PIII and to improve the b-wave. Hydrogen peroxide (H₂O₂, 0.3 mM) was then applied in the perfusion solution. A Representative ERGs recorded (a) before application of BaCl₂ and H₂O₂, (b) after 6 min of perfusion with BaCl₂, (c) after 36 min of perfusion with BaCl₂, (d) 15 min following application of H₂O₂ after 51 min of perfusion with BaCl₂, (e) 96 min following application of H₂O₂ after 132 min of perfusion with BaCl₂. B Representative survival curves of (a) the a-wave amplitude, (b) the b-wave amplitude, (c) the a-wave implicit time and (d) the b-wave implicit time.

To assess the effect of H₂O₂ directly on photoreceptors, aspartate 5 mM and barium 1 mM were applied in the perfusion solution (Figure 6). Aspartate, known to inhibit synaptic transmission from photoreceptors to bipolar cells,^27,28 suppresses the b-wave and isolates the PIII-wave (Figure 6b). The PIII-wave is composed of the fast PIII component (photoreceptor current) and the slow PIII component (Figure 6c). Barium, which blocks the slow PIII, was then added to the perfusion medium to isolate the fast PIII, as illustrated in Figure 6d. We observed that the amplitude of the fast PIII component remained unchanged during a 75-minutes period incubation (Figure 6d). When H₂O₂ 0.3 mM was applied in the solution, a decrease in the amplitude was observed after 15 min of incubation, demonstrating the direct action of H₂O₂ on the photoreceptors. Simultaneously, the implicit time of the fast PIII increased probably due to the decrease of the amplitude.

Figure 6. ERG responses in the presence of H₂O₂ after aspartate and barium chloride applications. Isolated rat retinas were continuously perfused with a nutrient solution and ERGs were recorded every 3 min in response to white-light stimulation (0.025 lx.s scotopic, 1 ms). Aspartate (Asp, 5 mM) and barium chloride (BaCl₂, 1 mM), known to suppress the b-wave and the slow PIII respectively, were added to the perfusion solution to isolate the fast PIII component. Hydrogen peroxide (H₂O₂, 0.3 mM) was then applied in the perfusion solution. Representative ERGs recorded (a) before application of Asp, BaCl₂ and H₂O₂, (b) after perfusion with Asp, (c) after perfusion with BaCl₂ and Asp, (d) following application of H₂O₂ after perfusion with Asp and BaCl₂. The smallest dotted line (slow PIII) in part c was calculated by subtracting the solid line (fast PIII) from the largest dotted line (PIII).

Discussion

Hydrogen peroxide (H₂O₂) is a ROS that is naturally produced under various physiological processes and whose level increases under certain pathological conditions.²⁹ H₂O₂ is generated by the photoexcited pigment lipofuscin, which accumulates in the retina during aging, and its accumulation is strongly associated with ARMD.² The level of H₂O₂ has also been reported to be increased in the retinas exposed to blue light, an ARMD risk factor, as well as in the retinas of diabetic rats.^3,4 Similarly, the concentration of H₂O₂ was shown to be higher in the vitreous of rabbit eyes after prolonged light exposure⁵ and in the vitreous of human cataract patients.⁶ There is increasing evidence that H₂O₂ is involved in many ocular diseases; however, the consequences of the action of this peroxide on retinal function remain unexplored.

Numerous studies in the central nervous system have demonstrated the implication of H₂O₂ in the modulation of neuronal activity,^7–10 synaptic neurotransmission^16,17 and ion channel activity.^11–15 The work published in the literature on the effects of H₂O₂ in the retina focuses mainly on retinal degeneration models. Hydrogen peroxide has been widely used as an inducer of oxidative stress in in vitro models of retinal cell culture. A few studies have examined its degenerative effects in ex vivo models of isolated perfused retinas,^18,19 and only one in vivo study has evaluated the effect of intravitreal H₂O₂ injection on retinas in mice.²⁰ In these studies, the authors assessed the cellular integrity and the retinal structure by measuring cell viability, histological assessment and in vivo imaging. Cell apoptosis, oxidative stress and expression of various markers of inflammation were also measured. To the best of our knowledge, there is no study in the literature that has focused on the action of H₂O₂ on retinal function as assessed by electroretinography. It is therefore essential to determine the functional effect of H₂O₂, which should lead to a better understanding of the pathophysiology of the retina under oxidative stress, as well as to the determination of new therapeutic strategies.

In our study, we report that the exposure of isolated rat retina to H₂O₂ affects retinal function, as shown by electroretinography. H₂O₂ causes instantaneous and transient changes in amplitudes and implicit times of some of the ERG waves (phase 1), followed by changes in retinal survival curves (phase 2). The observed effects differ according to the dose of H₂O₂. Our data show for the first time the impact of low (physiological) to high (pathological) concentrations of H₂O₂ on ERG properties and suggest a direct effect of this oxidant on ionic currents and/or neurotransmitter release involved in the generation of the signal.

Our results are consistent with the study of LeDay et al. (2004), who reported that exogenous application of H₂O₂ on isolated bovine retina can alter the release of neurotransmitter glutamate.¹⁸ The authors observed that H₂O₂ at low concentration (100 µM) caused an inhibition of K⁺-induced [³H]D-aspartate release, whereas higher concentration (1 mM) had no inhibitory effect. We know that hyperpolarization of photoreceptors reduces glutamate release and leads to the appearance of the positive b-wave, so we speculate that the decrease in glutamate release generated by photoreceptor hyperpolarization could add to that induced by H₂O₂ at low concentrations and thus explain the greater increase in b-wave amplitude observed with H₂O₂ at 0.2 mM. Similarly, the lack of inhibitory effect of H₂O₂ at high concentrations on neurotransmitter release could explain the smaller increase in b-wave amplitude at 0.6 mM.

Furthermore, we observed that the survival curve of the retina exposed to H₂O₂ at 0.2 mM was unaffected and even remained above the control survival curve. This could be explained by the potential neuroprotective role of H₂O₂ at low concentrations.^8,30 The H₂O₂-induced decrease in neurotransmitter release may play a protective role against excitotoxicity. In contrast, the survival curve of the retina exposed to high concentrations of H₂O₂ was affected and decreased significantly. This is consistent with the fact that high levels of H₂O₂ induce neurotoxicity. Taken together, these results suggest that the mechanisms underlying the action of H₂O₂ at low concentrations might be different from those at high concentrations. We speculate that H₂O₂ may use more than one mechanism or pathway to affect glutamatergic transmission in the retina.

Glutamate is the main neurotransmitter of the retina. It is released from photoreceptors and bipolar cells via depolarization-evoked activation of L-type calcium channels.^31–34 It has been shown in various tissues, including the retina,³⁵ that H₂O₂ can regulate intracellular calcium homeostasis via voltage-gated calcium channels³⁶ and intracellular calcium stores.³⁷ Additionally, H₂O₂ was reported to alter glutamate release in the brain and in the retina.^18,38,39 Thus, our results combined with data from the literature showing the effects of H₂O₂ on intracellular calcium homeostasis and glutamate release, strongly suggest an effect on photoreceptor/bipolar cell synapse (modulation of presynaptic glutamate release and/or postsynaptic glutamate response).

To better understand how H₂O₂ induces changes in ERG responses, we evaluated the effect of this peroxide on the different components of the signal that were isolated with pharmacological agents. As a consequence of barium addition, which blocks the slow PIII component,^24–26 the amplitudes of the a-wave and b-wave improved and stabilized, and the implicit times increased. Application of H₂O₂ 0.3 mM reduced the amplitudes of the two waves with a stronger effect on the a-wave, suggesting a more pronounced effect of H₂O₂ on photoreceptors than on postsynaptic cells. Simultaneously, the implicit times increased. In another set of experiments, the use of aspartate, known to block the synaptic transmission from photoreceptors to bipolar cells,^27,28 isolated the PIII wave composed of the fast PIII (photoreceptor current) and the slow PIII (Müller glial cells response). Following the addition of barium, the fast PIII component was isolated and remained stable for over one hour. A few minutes after the addition of H₂O₂, a decrease in the amplitude of the fast PIII was observed, demonstrating the action of H₂O₂ directly on the photoreceptors.

When aspartate was used to isolate the PIII-wave and H₂O₂ was applied, we observed a reduction in the amplitude of the PIII component and no immediate effect on the implicit time (data not shown). This suggests that the instantaneous and transient increase in the implicit time of the slow PIII observed immediately after application of H₂O₂ alone, may be a direct consequence of the effect of H₂O₂ on photoreceptors. Because aspartate blocks photoreceptor responses to second-order neurons in the retina,^27,28 the delayed implicit time of the slow PIII observed after application of H₂O₂ alone may arise from changes in synaptic transmission. The delayed implicit time of the slow PIII could also originate from changes in the homeostasis in Müller glial cells. Indeed, the slow PIII reflects the potassium conductance carried by Kir4.1 channels in Müller cells as a secondary response to the light-induced activity of photoreceptors.²³ It has been reported that oxidative stress induced by H₂O₂ can inhibit the activity of Kir4.1 channels and that acute administration of H₂O₂ causes osmotic swelling of Müller cells.^40,41 The Müller cell swelling reflects an impairment in the rapid osmotically driven water movement across Müller cell membranes after functional inactivation of Kir4.1 channels.⁴⁰ Functional inactivation or downregulation of Kir4.1 channels have been reported in experimental models of various retinal diseases, such as retinal ischemia reperfusion, ocular inflammation, diabetic retinopathy, retinal detachment and blue light-evoked retinal degeneration.⁴⁰

A few studies have evaluated the effect of H₂O₂ on ex vivo models of isolated retina from different species, such as mice, rats or pigs^19,42,43 with sometimes contradictory results. In these studies, isolated retinas were exposed for 3 h to H₂O₂ at doses ranging from 100 to 1000 µM, and immunohistochemical and western-blot analyses were performed after several hours or several days. Surprisingly, the effect of H₂O₂ on photoreceptors was not described in these studies. For instance, Hurst and collaborators (2017) have evaluated the degenerative effects of oxidative stress on the porcine ex vivo retina treated with 100 to 500 µM H₂O₂ for 3 h.¹⁹ While bipolar and amacrine cells were not significantly affected, a degeneration of retinal ganglion cells was observed at all H₂O₂ concentrations. Additionally, activation of microglia was detected for H₂O₂ 100 µM. Macroglia was not significantly affected by H₂O₂. In another study conducted in 2020,⁴² porcine retinal explants were exposed to H₂O₂ 300 µM for 3 h. The authors observed a loss of bipolar, amacrine and ganglion cells, a late effect on microglia and no effect on macroglial response. In the mouse retina, Ulyanova et al. (2001) observed that H₂O₂ induced a strong heme oxygenase-1 immunoreactivity in Müller cells, but not in other retinal cells, whereas GFAP expression staining in Müller cells was not altered.⁴³ One in vivo study has also examined the effect of intravitreal injection of H₂O₂ (5-10 µg/µl) in mice.²⁰ The authors observed cell degeneration in all layers of the retina with photoreceptors being the most severely damaged. The ganglion cell layer was only poorly affected.

The concentrations of H₂O₂ used in our study are similar to those reported in previous ex vivo studies (100-1000 µM) evaluating the effect of H₂O₂ on isolated retinas,^18,19,42,43 and are in the range of concentrations expected in the retina. For instance, H₂O₂ in the vitreous of rabbit eyes was estimated at 110 µM after prolonged light exposure⁵ and might be even higher in the retina since H₂O₂ is released by illuminated photoreceptors. Likewise, the H₂O₂ level in the vitreous of human patients with cataract was found to be 160 µM.⁵

In summary, we demonstrate that exposure of isolated rat retina to H₂O₂ affects retinal function, as shown by ERG recordings. H₂O₂ caused instantaneous and transient changes in amplitudes and implicit times of some of the ERG waves, followed by changes in retinal survival curves. The effects differ with the dose of H₂O₂, suggesting that the mechanisms underlying the action of H₂O₂ at low doses might be different from those at high doses. Our results also suggest a direct effect of H₂O₂ on ionic currents and/or neurotransmitter releases involved in the generation of the ERG and indicate a more pronounced effect of H₂O₂ on photoreceptors than on postsynaptic cells. It is possible that the effects of H₂O₂ occur directly and exclusively on photoreceptors (i.e. with no direct effect on postsynaptic cells). Decreased photoreceptor responses would affect the subretinal potassium change, decreasing the slow PIII, as well as the synaptic drive to bipolar cells, decreasing the b-wave. However, it could not be proved that there were no effects on other cells. Further studies will be necessary to tackle this point and to examine in more details the mechanisms by which H₂O₂ induces changes in ERG waves (modulation of ion conductance/synaptic transmission in retinal cells). Interestingly, Sies et al. (2017, 2020) have recently pointed out^44,45 the pleiotropic and dose-dependent role of H₂O₂ in the redox regulation of various activities including receptors, ion channels, enzymes and transcription factors at low physiological levels, and in the disruption of redox signaling leading to oxidative damage at high concentrations, corroborating our observations in the retina.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.