June 2009
Volume 50, Issue 6
Free
Physiology and Pharmacology  |   June 2009
Experimental Glutamatergic Excitotoxicity in Rabbit Retinal Ganglion Cells: Block by Memantine
Author Affiliations
  • William A. Hare
    From Allergan, Inc., Irvine, California.
  • Larry Wheeler
    From Allergan, Inc., Irvine, California.
Investigative Ophthalmology & Visual Science June 2009, Vol.50, 2940-2948. doi:https://doi.org/10.1167/iovs.08-2103
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      William A. Hare, Larry Wheeler; Experimental Glutamatergic Excitotoxicity in Rabbit Retinal Ganglion Cells: Block by Memantine. Invest. Ophthalmol. Vis. Sci. 2009;50(6):2940-2948. https://doi.org/10.1167/iovs.08-2103.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. Excessive activity of NMDA-type (N-methyl-d-aspartate) glutamatergic channels has been implicated as a mechanism for neuronal injury in neurologic disorders, including glaucoma, and retinal disease. This study was designed to characterize the retinal response to experimental manipulations that mimic features of glutamatergic excitotoxic insult and also to determine whether memantine, an NMDA-type glutamatergic channel blocker, is effective in reversing experimental excitotoxicity.

methods. Recordings of the electroretinogram (ERG) and spiking activity of single retinal ganglion cells (RGCs) were made from rabbit retinas. Excitotoxic insult was induced by either (1) application of NMDA, a selective NMDA receptor agonist; (2) application of TBOA (dl-threo-β-benzyloxyaspartic acid), a selective inhibitor of glutamate transporters, or (3) perfusion with magnesium-free medium. For each condition, memantine was coapplied to determine its efficacy for reversal of experimental excitotoxicity. Memantine was also applied in isolation to characterize any effect on retinal responses to light stimuli.

results. All three experimental manipulations were associated with an increase in the tonic level of RGC spiking activity, a reduction in RGC spike amplitude, and, in some cells, block of spike generation. Experimental excitotoxicity had little or no effect on ERG responses. Coapplication of memantine was associated with recovery of RGC tonic spiking activity and spike amplitude toward control levels. Application of memantine in isolation was associated with a dose-dependent effect on the timing of ERG and RGC-OFF responses.

conclusions. Memantine was effective in reversing acute experimental excitotoxicity at concentrations that have little effect on retinal light signaling.

Glutamate is the principal excitatory neurotransmitter in the central nervous system (CNS). However, excessive activation of glutamate-gated membrane channels may lead to irreversible injury of CNS neurons. Glutamatergic excitotoxicity, originally described in the retina, 1 results primarily from intracellular accumulation of sodium and calcium ions. Because of the relatively high permeability of N-methyl-d-aspartate (NMDA)-type glutamate-gated channels to calcium ions, neurons are particularly sensitive to injury associated with excessive activity of this channel subtype. 2 Retinal ganglion cells (RGCs) are known to express NMDA-type channels, 3 4 5 6 and glutamatergic excitotoxicity, mediated by NMDA channels, has been demonstrated to contribute significantly to RGC injury in both in vitro 7 8 9 10 11 and animal 12 13 14 15 16 17 models. However, in a recent study, RGCs were found to be relatively insensitive to excitotoxic injury from exposure to either glutamate or NMDA. 18 Treatment with memantine (1-amino-3,5-dimethyladamantane), a noncompetitive NMDA channel blocker, has been shown to reduce RGC injury in a wide range of in vitro 19 20 and animal 21 22 23 24 25 26 models including experimental glaucoma in mouse, 27 28 rat, 25 29 and monkey. 30 31  
NMDA channels play an important role in neuronal signaling within the CNS, and the properties of NMDA channel function have been well characterized. Binding of glutamate to its receptor is associated with a switch of the channel to an “open” configuration. However, at highly negative membrane potential, the channel is blocked by binding of a magnesium ion within the conductance pore. Membrane depolarization is associated with a reduction of this block and increased conductance of sodium and calcium ions. The excitatory action of glutamate at NMDA channels is therefore greatest when the neuronal membrane is depolarized either by the action of other excitatory membrane channels or under pathologic conditions in which normal resting membrane potential is lost. 32 33 34 Extracellular glutamate has been shown to be elevated after acute experimental retinal ischemia, 35 36 37 38 39 in both human 40 and experimental 41 diabetic retina and in human retinal detachment. 42 Evidence of dysregulation of extracellular glutamate buffering has also been observed in animal models of glaucoma. 43 44 45 46 47 Increases in local extracellular levels of glutamate may drive increased activity of both NMDA and non-NMDA glutamate channels. Under pathologic conditions resulting in neuronal membrane depolarization, NMDA channel activity may reach excitotoxic levels, due to loss of loss of voltage-dependent magnesium block, even in the presence of normal extracellular glutamate concentrations. 48  
Memantine acts to block NMDA channel conductance by binding to a site within the channel pore. In order for memantine to access its binding site, glutamate (or NMDA) must first bind to its receptor to open the channel and magnesium must leave the pore. “Open-channel block” by memantine is therefore dependent on both the neuronal membrane potential and the concentration of extracellular glutamate. 49 50 51 52 Thus, the efficacy for memantine block of NMDA channel activity is greatest under conditions of pathologically elevated glutamate levels and/or neuronal membrane depolarization but relatively less under normal conditions of glutamate signaling at the RGC membrane. 
For the experiments described in this study, simultaneous recordings of the electroretinogram (ERG) and single-unit RGC activity in rabbit retina were used to characterize the retinal response to experimental manipulations that mimic features of glutamatergic excitotoxic insult associated with retinal disease. Application of NMDA was used to simulate the action of increased extracellular glutamate levels at NMDA channels. Inhibition of glutamate transporters was used to disrupt normal mechanisms for regulation of extracellular glutamate. Perfusion with magnesium-free medium was used to simulate the loss of voltage-dependent magnesium block of NMDA channels that may be associated with pathologically depolarized neuronal membrane potential. In each of these experimental conditions, memantine application was tested for its ability to block excitotoxic activity in RGCs. Memantine application in isolation was also used to characterize any effect of memantine treatment on retinal responses to light stimulation under normal conditions. 
Methods
Ex Vivo Retina Preparation
All procedures were conducted in accordance with guidelines outlined in the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were also approved by an internal institutional review committee. 
Adult Dutch belted rabbits (Charles River; Wilmington, MA) were anesthetized with intramuscular injection of ketamine (50 mg/kg) in combination with acepromazine (21 mg/kg). Deep anesthesia was then obtained using inhalation of 4.0% isoflurane. The eyes were removed in bright room illumination and placed in room temperature phosphate-buffered saline for approximately 3 minutes. Under dim red illumination, the anterior segment including the lens was removed. A series of radial cuts, extending approximately 5 to 10 mm from the peripheral edge of the eye cup toward the posterior pole, were used to flatten the eye cup. After careful removal of the vitreous, a circular sample of retina-choroid-sclera, 11 mm in diameter and extending inferiorly from the inferior edge of the optic nerve head, was cut from the tissue as shown in Figure 1A . This tissue sample was then mounted, retina side up, in a custom perfusion chamber, as illustrated in Figure 1B , and perfused continuously at a rate of 2.5 mL/min with Ames medium (Sigma-Aldrich; St. Louis, MO), which was saturated with a combination of 95% O2 and 5% CO2. The tissue was always oriented within the chamber in the same position so that microelectrode recordings were made in the same area immediately inferior to the visual streak. The preparation was maintained at a temperature of 35°C in complete darkness (with the exception of periodic light stimulation) inside a light-tight Faraday cage. Test solutions containing NMDA (Sigma-Aldrich), AP5 (d-(–)-2-amino-5-phosphonopentanoic acid; Tocris Bioscience, Ellisville, MO), TBOA (dl-threo-β-benzyloxyaspartic acid; Tocris Bioscience), or memantine (3,5-dimethyladamantane-1-amine hydrochloride; Allergan Inc., Irvine, CA) were prepared by adding the test agents directly to the perfusion medium. Magnesium-free medium was obtained by preparation of a modified Ames medium from which magnesium chloride was omitted. The preparation was dark adapted for approximately 1.5 hours before recording. The results described in this study were obtained from more than 45 retina preparations. 
Light Stimulation and Electrophysiological Recording
Dim diffuse light stimuli of either 10-ms or 1-second duration were generated with an LED (λp = 504 nm; Nichia Corp., Detroit, MI). The ERG was recorded as the voltage between an electrode located on the scleral side of the tissue and the bath electrode. ERG signals were band-pass filtered from 1 Hz to 1 kHz and amplified (model CP511 Grass amplifier; Astro Med, West Warwick, RI). Spiking activity of RGCs was recorded with tungsten microelectrodes having an impedance of approximately 4 MΩ and with the bath electrode used as the reference. RGC signals were band-pass filtered from 0.5 to 5 kHz and amplified with an amplifier with custom modifications (model IX1; Dagan Corp., Minneapolis, MN). ERG and RGC signals were recorded simultaneously at a sampling rate of 50 kHz using two channels of a signal processor (CED Model 1400; Cambridge Electronic Design, Cambridge, UK). The microelectrodes were positioned by using infrared illumination and a video monitor located outside the Faraday cage. Isolation of single-unit activity was verified by inspection of action potential amplitude and waveform. For averaged responses, the RGC signal was used to trigger a threshold discriminator (model 121; World Precision Instruments, Sarasota, FL) that converted each spike to a positive square pulse of 1-ms duration and 5-V amplitude. RGCs were characterized as either ON, OFF, or ON-OFF according to the responses to dim diffuse stimuli of 1-second duration. ON cells responded to this stimulus with a burst of spikes at stimulus onset, OFF cells responded with a burst of spikes at stimulus offset, and ON-OFF cells responded with a burst of spikes at both stimulus onset and stimulus offset. An example of an OFF cell response is shown in 2 3 4 5 Figure 6
Results
Application of NMDA
Application of NMDA at 50 μM (n = 14), 100 μM (n = 23) or 200 μM (n = 12) concentration was not associated with any effect on measures of time-to-peak for ERG b-wave responses to dim flash stimuli (2.0 × 104 photons/flash-cm2). ERG b-wave amplitude, measured from baseline to peak positive voltage, was also not significantly affected by either 50 or 100 μM NMDA but was increased, on average, by 13.7% (P = 0.03; paired t-test) after application of 200 μM NMDA. 
NMDA, at concentrations ranging from 50 to 200 μM, was tested on a total of 23 ON cells, 27 OFF cells, and 4 ON-OFF cells, as shown in Table 1which summarizes all RGC cell types exposed to experimental excitotoxic treatments for this study. In an initial series of experiments, RGC sensitivity to NMDA was tested at concentrations of 50, 100, and 200 μM. Some cells showed little or no effect at either 50 or 100 μM concentration. However, application of 200 μM NMDA was followed by a rapid increase in tonic spiking activity in all cells tested. These initial recordings showed that RGCs exhibit a rather wide range of sensitivity to NMDA though, in general, ON cells showed greater sensitivity to lower concentrations than OFF cells displayed. For later experiments, NMDA was always tested at a concentration of 200 μM. At this concentration, the NMDA-induced increase in tonic spiking activity in some cells was maintained for more than 10 minutes with only a moderate reduction in spike amplitude, whereas other cells showed a rapid decline in spike amplitude resulting in complete loss of spiking activity including light-elicited responses. In all cells tested, these effects were rapidly and completely reversible after switch back to control perfusion medium (NMDA washout). 
Results for application of 200 μM NMDA to an OFF RGC are shown in Figure 2 . ERG and RGC responses to a dim flash, delivered at 400 ms under control conditions, are represented as the top and bottom traces, respectively (Fig. 2A) . At 2 minutes after the switch to perfusion medium containing 200 μM NMDA, the tonic level of RGC spiking activity had increased, and spike amplitude had declined by approximately 25% (Fig. 2B) . At 6 minutes after application of NMDA, RGC spiking activity was almost completely blocked (Fig. 2C) . RGC spike generation recovered partially by 2 minutes after the switch from NMDA to the combination of 200 μM NMDA and 10 μM memantine (Fig. 2D) , whereas both the amplitude and tonic level of spiking activity recovered by 4 minutes after the switch (Fig. 2E) . In all 33 cells tested (see Table 1 ), memantine at a concentration of either 10 or 30 μM reversed the excitotoxic effects of NMDA. 
Results for application of 200 μM NMDA to a different OFF-RGC are shown in Figure 3 . This cell exhibited a low level of tonic spike activity under control conditions (Fig. 3A) . The NMDA-induced increase in tonic spike activity was somewhat less than that observed for the cell of Figure 2 , and spike amplitude was reduced by only approximately 30% at 10 minutes after the switch to NMDA (Fig. 3B) . At 12 minutes after the switch to medium containing NMDA in combination with 30 μM AP5, a selective NMDA receptor antagonist, RGC tonic spike activity and spike amplitude recovered to near control levels (Fig. 3C) . AP5 at a concentration of 30 μM was effective in reversing NMDA-induced excitotoxicity in all five cells tested (see Table 1 ). 
Perfusion with Zero-Mg2+ Medium
Removal of Mg2+ from the perfusion medium was associated with excitotoxic activity in RGCs that was similar to that observed after application of NMDA. As was seen with NMDA, there was considerable variation among the six RGCs tested (Table 1)with respect to their sensitivity to zero Mg2+. However, all six cells responded with an increase in tonic spiking activity and reduction of spike amplitude. Figures 4A 4B 4Cshow that the switch to zero-Mg2+ medium was associated with an increase in the tonic level of spiking activity for this ON RGC. This was accompanied by an approximate 50% decrease in spike amplitude at 19 minutes after Mg2+ washout. This excitotoxic response developed more slowly than the typical NMDA-induced response (Figs. 2A 2B 2C) . The switch from zero-Mg2+ medium to the combination of zero Mg2+ and 10 μM memantine was associated with recovery of tonic spike activity and spike amplitude toward control levels (Fig. 4D) . Zero Mg2+-induced excitotoxicity was reversed by coapplication of 10 μM memantine in all six cells tested. Perfusion with zero Mg2+ was associated with no significant effect on the ERG response to dim stimuli. 
Blockade of Glutamate Transporters
TBOA, a potent and selective inhibitor of excitatory amino acid transporters, 53 54 55 was used to block intrinsic mechanisms for retinal glutamate buffering and its application was associated with excitotoxic responses in all 17 RGCs tested (Table 1) . As for NMDA or zero Mg2+, there was a wide range in the severity of TBOA-induced excitotoxicity. Figure 5summarizes results from application of 50 μM TBOA to an OFF RGC. At 3 minutes after the switch to TBOA, tonic spiking activity increased dramatically and spike amplitude decreased by approximately 65% (Fig. 5B) . At 4 minutes after the switch, RGC spiking activity was completely blocked (Fig. 5C) . The switch to medium containing TBOA in combination with 10 μM memantine was associated with recovery of tonic spiking activity and spike amplitude toward control levels (Fig. D5) . Memantine washout was followed by a rapid return of TBOA-induced excitotoxic spiking activity (Fig. 5E) . In all 13 cells tested, 10 μM memantine was effective in reversing TBOA-induced excitotoxicity. TBOA was not associated with any significant effect on the ERG response. 
Effects of Memantine on Retinal Light Signaling
Memantine was tested in isolation on a total of 25 OFF and 12 ON RGCs from 34 retinal preparations at concentrations ranging from 1 to 30 μM for effects on normal retinal light signaling. In these experiments, individual RGC spikes were recorded as unitary events and response averages were used for analysis. Responses to dim stimuli of 10-ms and 1-second duration were recorded both before (CONTROL) and at approximately 10 minutes after continuous application of memantine. Results for application of 30 μM memantine, obtained from an OFF RGC, are illustrated in Figure 6 . For this cell, application of memantine was associated with a delay of approximately 10 ms for both the peak of the ERG b-wave and onset (latency) of the RGC OFF response to the flash stimulus (Fig. 6 , top). The RGC response seen at offset of the dim 1-second stimulus (Fig. 6 , bottom) was delayed by approximately 6 ms in the presence of memantine. 
Memantine was tested in isolation at 1-, 3-, 10-, and 30-μM concentration, with results summarized in Figure 7 , where measures obtained in the presence of memantine are normalized with respect to measures obtained immediately before memantine application (100%). Memantine was not associated with any significant effect on either the amplitude of the ERG response or timing of ON RGC responses. It was associated, in a dose-dependent manner, with delays in the timing for both the ERG and OFF RGC responses. At 10 μM, a concentration shown to effectively reverse experimental excitotoxicity (Figs. 2 4 5) , the b-wave peak and OFF response were delayed on average by approximately 3 and 9 ms, respectively. Increasing the concentration of memantine to 30 μM resulted, on average, in ERG and OFF RGC response delays of approximately 7 and 22 ms, respectively. At the 1- and 3-μM concentrations, no significant effect was observed on any measured response. ERG and OFF RGC response timing recovered toward control levels after memantine washout in all cells tested (data not shown). 
Discussion
Experimental Excitotoxicity
ERG Responses.
The dim-flash stimuli used in this study elicited dark-adapted ERG responses consisting of an early positive peak (b-wave) followed by a negative phase. The b-wave is generated primarily by direct contributions from activity in ON and OFF retinal bipolar cells 56 and, since bipolar cells do not express NMDA receptors, NMDA would not be expected to have large effects on this response. Application of NMDA had no significant effect on b-wave timing. NMDA at a concentration of 200 μM increased b-wave amplitude, on average, by approximately 14%, whereas lower concentrations were without effect. At 200 μM, NMDA was associated with increased tonic spiking activity and reduced spike amplitude in all RGCs tested and resulted in complete block of spike generation in many RGCs (Fig. 2) . Blockade of spiking activity in third-order neurons with TTX increased the amplitude of ERG b-wave responses to dim flash stimuli in dark-adapted rabbits. 57 The small NMDA-induced increase in b-wave amplitude in the present study may therefore have resulted from inhibition of spiking activity in third-order retinal neurons. Intravitreous application of NMDA in primate eyes has been shown to have little or no effect on the b-wave of ERG responses in primates. 58  
RGC Responses.
The application of exogenous NMDA was associated with increased tonic spiking activity in all RGCs tested (Figs. 2 3) . There was a considerable range in the observed sensitivity to NMDA, although all cells tested showed excitotoxic responses to NMDA at 200 μM concentration. Massey and Miller 4 also found that NMDA increased spiking activity in all RGC types they tested in rabbit retina and also showed that the EC50 for this effect in a large field unit was equal to approximately 200 μM. In the present study, there was also a wide variation in the degree of spike amplitude reduction associated with NMDA-induced increases in spike activity. Dark-adapted RGCs were characterized as ON-, OFF-, or ON-OFF-type, according to their responses to dim diffuse stimuli of 10-ms and 1-second duration (Fig. 6) . The cells were not further classified with respect to their level of resting spiking activity, temporal response properties (i.e., transient versus sustained), or other receptive field properties. The observed range in sensitivity to NMDA likely reflects a diversity of functionally distinct RGC subtypes within the study sample. 
The term excitotoxicity describes the NMDA-induced increase in tonic spiking activity and associated reduction in spike amplitude. In a functional sense, this NMDA-induced effect is clearly toxic. However, it is not clear whether this acute disruption of normal RGC function was associated with irreversible neuronal injury. These NMDA-induced effects were rapidly reversible (within minutes) after NMDA washout (data not shown). Application of NMDA for longer durations may have led to irreversible loss of function. It is also possible that, even with these relatively short exposures to NMDA, RGCs had sustained irreversible injury. Function may recover and persist for some time after activation of intracellular pathways, leading ultimately to cell death. The precise mechanism of the NMDA-induced excitotoxicity observed in this study is also unclear. RGCs are known to express NMDA receptors, 3 4 5 6 and NMDA may act directly at the RGC plasma membrane to activate a depolarizing (inward) current with consequent increase in the cytoplasmic concentration of calcium and sodium ions. Constantly high levels of spiking activity could also lead to the accumulation of intracellular sodium ions and contribute to loss of the sodium gradient that is necessary for generation of action potentials (spikes) as well as the function of sodium-coupled membrane transporters. It is also possible that NMDA may act pre-synaptically to enhance excitatory inputs to the RGC. However, the failure to observe large effects of NMDA on the ERG b-wave suggests that RGC excitotoxicity is not mediated primarily by indirect actions on either photoreceptors or bipolar cells. 
Perfusion with magnesium free medium was associated with RGC excitotoxicity that was similar to that observed with NMDA (Fig. 4) . This observation is consistent with the notion that magnesium ions act to regulate the conductance of the NMDA channel by blocking the channel pore in a voltage-dependent manner. 32 33 At normal RGC resting membrane potential, the effect of extracellular glutamate to increase NMDA channel conductance is partially blocked by magnesium. Membrane depolarization associated with a pathologic effect (i.e., metabolic insult) would also reduce magnesium block of the NMDA channel. In this sense, perfusion with magnesium leads to an increase in RGC NMDA channel activity in a manner similar to that which occurs under pathologic conditions associated with loss of the normal RGC resting membrane potential. 
Application of TBOA, a selective blocker of excitatory amino acid transporters, resulted in RGC excitotoxicity (Figs. 5A 5B 5C)that was similar to that associated with application of NMDA or perfusion with magnesium-free medium. Glutamate transporters play an important role in regulating levels of extracellular retinal glutamate and normally function to maintain the extracellular concentration of glutamate at a low level. In mixed retinal cell culture, pharmacologic blockade of glutamate transporter function has been shown to result primarily in injury to RGCs. 59 In the intact retina, loss of transporter function leads to increased levels of extracellular glutamate that act at all retinal glutamate receptor subtypes. 60 61 Thus, TBOA-induced RGC excitotoxicity probably reflects contributions from both NMDA and non-NMDA type excitatory (depolarizing) glutamatergic membrane conductances and, in this regard, is analogous to retinal disease associated with dysregulation of extracellular glutamate buffering. 
Block of Experimental RGC Excitotoxicity by Memantine
Application of memantine, a use-dependent blocker of the NMDA-type glutamatergic channel, was associated with reversal of RGC excitotoxicity induced by application of NMDA (Fig. 2)or perfusion with magnesium-free medium (Fig. 4) . In all cells tested, recovery of spike amplitude and the tonic level of spiking activity was observed with perfusion of memantine at a concentration of 10 or 30 μM. This recovery was observed even in cells where spike generation was completely blocked as a result of experimental excitotoxicity. Memantine was also able to reverse excitotoxicity resulting from blockade of glutamate transporters with TBOA (Fig. 5) . Any TBOA-induced increase in extracellular glutamate may act at both NMDA and non-NMDA-type glutamatergic receptors to drive excitotoxic responses in RGCs. The fact that memantine was effective to reverse TBOA-induced RGC excitotoxicity suggests that, in the cells examined in this study, NMDA receptors make a major contribution to excitotoxicity associated with dysregulation of extracellular glutamate. The magnitude of this contribution depends on the relative density and distribution of NMDA receptors in the RGC plasma membrane. 
Safety of Memantine
Memantine has been shown to be safe and effective (and has been approved by the FDA) for the treatment of Alzheimer’s dementia. This safety profile is thought to reflect properties for the interaction of memantine with its binding site within the NMDA channel pore. 49 Because the NMDA channel must be open for memantine to access its binding site, memantine would be relatively more effective in blocking channel conductance under conditions of high extracellular glutamate concentration (use-dependent block). Also, since magnesium must leave the channel pore for memantine to bind, memantine would be more effective in blocking the channel under conditions of relative neuronal membrane depolarization. 
The results of this study show that 10 μM memantine was able to reverse severe acute experimental excitotoxicity in RGCs. Applied in isolation, this concentration of memantine was associated, in a dose-dependent manner (Fig. 7) , with delays in the ERG b-wave and RGC OFF responses but no effect on either the b-wave amplitude or timing of RGC ON responses. Response delays were not observed at memantine concentrations below 3 μM, and the delays at the 30-μM concentration were greater on average than those at the 10-μM concentration. Delays in RGC response timing may be expected to have consequences for temporal properties of visual signal processing such as motion perception. However, oral dosing with memantine, a compound known to cross the blood–brain barrier, has been used clinically with no reports of visual disturbance. Treatment efficacy has been demonstrated in human subjects with memantine serum concentrations of less than 1 μM, 62 whereas efficacy in a primate model of experimental glaucoma was observed with serum concentrations of less than 2 μM. 30 In the primate study, this level of memantine was found to have no effect on measures of amplitude or timing for any component of the ERG or visually evoked cortical potential (VECP). Clinical memantine treatment is thus not likely to produce plasma concentrations high enough to interfere with normal retinal signal processing. 
Summary
Experimental manipulations designed to enhance activity of NMDA-type glutamatergic receptors are associated with increased levels of tonic spiking activity in RGCs and interference with the mechanism for generation of action potentials. Under these conditions, application of memantine, a use-dependent blocker of the NMDA channel, results in the return of RGC tonic spiking activity and spike amplitude toward normal levels. Concentrations of memantine that are effective in the reversal of NMDA-type glutamatergic excitotoxicity in RGCs have little effect on the normal processing of visual signals in the retina. These results support a conclusion that memantine may provide a safe and effective treatment for retinal disorders associated with excessive activity of NMDA-type glutamatergic channels. 
 
Figure 1.
 
Illustration of the retinal preparation used for electrophysiological recordings. An 11-mm diameter sample of retina-choroid-sclera was cut from the posterior fundus (A). The superior border of this sample included the inferior rim of the optic nerve head (ONH). All recordings were obtained from a region located just inferior to the visual streak (VS). The retina-choroid-sclera sample was mounted retina side up in the perfusion chamber (B). Silver chloride electrodes, located beneath the preparation and in the bath, provided for measures of transretinal voltage (ERG). A tungsten microelectrode, using the bath electrode as the reference, provided measures of spiking activity (action potentials) from single RGCs. Diffuse light stimuli were generated with a blue-green LED.
Figure 1.
 
Illustration of the retinal preparation used for electrophysiological recordings. An 11-mm diameter sample of retina-choroid-sclera was cut from the posterior fundus (A). The superior border of this sample included the inferior rim of the optic nerve head (ONH). All recordings were obtained from a region located just inferior to the visual streak (VS). The retina-choroid-sclera sample was mounted retina side up in the perfusion chamber (B). Silver chloride electrodes, located beneath the preparation and in the bath, provided for measures of transretinal voltage (ERG). A tungsten microelectrode, using the bath electrode as the reference, provided measures of spiking activity (action potentials) from single RGCs. Diffuse light stimuli were generated with a blue-green LED.
Table 1.
 
Summary of All RGCs Tested for Experimental Excitotoxicity
Table 1.
 
Summary of All RGCs Tested for Experimental Excitotoxicity
Treatment RGC Type
ON OFF ON-OFF
NMDA (50–200 μM) 23 27 4
 +Memantine (10, 30 μM) 16 17 4
 +AP5 (30 μM) 1 4
Zero Mg2+ 4 2
 +Memantine (10 μM) 4 2
TBOA (10–75 μM) 7 8 2
 +Memantine (10 μM) 5 8
Figure 2.
 
NMDA-induced RGC excitotoxicity is reversed by memantine. (A) Recordings of the ERG (top trace) and activity of a single RGC (bottom trace) obtained under control conditions. (B) Increased tonic RGC spiking activity and reduction of spike amplitude seen at 2 minutes after application of 200 μM NMDA. (C) Complete block of RGC spike generation at 6 minutes after application of NMDA. (D, E) Recovery of RGC spike amplitude and tonic spiking activity to near control levels after switch to perfusion medium containing NMDA in combination with 10 μM memantine. All traces were recorded from the same OFF-RGC in the order (A)→(D). In all panels, a dim stimulus of 10-ms duration was delivered at 0.4 second (vertical solid line). Stimulus intensity = 1.2 × 105 photons/flash-cm2. Scale for ERG is in microvolts; the scale for RGCs is in millivolts.
Figure 2.
 
NMDA-induced RGC excitotoxicity is reversed by memantine. (A) Recordings of the ERG (top trace) and activity of a single RGC (bottom trace) obtained under control conditions. (B) Increased tonic RGC spiking activity and reduction of spike amplitude seen at 2 minutes after application of 200 μM NMDA. (C) Complete block of RGC spike generation at 6 minutes after application of NMDA. (D, E) Recovery of RGC spike amplitude and tonic spiking activity to near control levels after switch to perfusion medium containing NMDA in combination with 10 μM memantine. All traces were recorded from the same OFF-RGC in the order (A)→(D). In all panels, a dim stimulus of 10-ms duration was delivered at 0.4 second (vertical solid line). Stimulus intensity = 1.2 × 105 photons/flash-cm2. Scale for ERG is in microvolts; the scale for RGCs is in millivolts.
Figure 3.
 
Reversal of NMDA-induced RGC excitotoxicity by AP-5. (A) ERG and RGC responses recorded under control conditions. (B) Application of NMDA is associated with increased tonic RGC spiking activity and decreased spike amplitude. (C) Application of AP5 in the presence of NMDA resulted in reversal of NMDA-induced RGC excitotoxicity. (D) ERG and RGC responses obtained after washout of NMDA+AP5. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity was 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scales are microvolts and millivolts, respectively.
Figure 3.
 
Reversal of NMDA-induced RGC excitotoxicity by AP-5. (A) ERG and RGC responses recorded under control conditions. (B) Application of NMDA is associated with increased tonic RGC spiking activity and decreased spike amplitude. (C) Application of AP5 in the presence of NMDA resulted in reversal of NMDA-induced RGC excitotoxicity. (D) ERG and RGC responses obtained after washout of NMDA+AP5. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity was 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scales are microvolts and millivolts, respectively.
Figure 4.
 
(AC) Perfusion with zero magnesium medium was associated with RGC excitotoxicity that is similar to that induced by NMDA. (D) Application of memantine (10 μM) in the presence of zero Mg2+ resulted in recovery of RGC spiking activity and spike amplitude toward control levels. (E) ERG and RGC responses obtained after washout of memantine and zero Mg2+. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity was 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scales are microvolts and millivolts, respectively.
Figure 4.
 
(AC) Perfusion with zero magnesium medium was associated with RGC excitotoxicity that is similar to that induced by NMDA. (D) Application of memantine (10 μM) in the presence of zero Mg2+ resulted in recovery of RGC spiking activity and spike amplitude toward control levels. (E) ERG and RGC responses obtained after washout of memantine and zero Mg2+. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity was 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scales are microvolts and millivolts, respectively.
Figure 5.
 
(AC) Blockade of excitatory amino acid transporters with TBOA was associated RGC excitotoxicity that was similar to that observed after application of either NMDA or zero Mg2+. Tonic spiking activity was increased and, in this cell, spike generation was blocked. (D) Application of memantine (10 μM) in the presence of TBOA resulted in the recovery of RGC spiking activity and amplitude toward control levels. (E) Memantine washout, in the presence of continued perfusion with TBOA, resulted in the return of RGC excitotoxicity. (F) TBOA washout was followed by the recovery of RGC spiking activity and amplitude toward control levels. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity is 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scale are microvolts and millivolts, respectively.
Figure 5.
 
(AC) Blockade of excitatory amino acid transporters with TBOA was associated RGC excitotoxicity that was similar to that observed after application of either NMDA or zero Mg2+. Tonic spiking activity was increased and, in this cell, spike generation was blocked. (D) Application of memantine (10 μM) in the presence of TBOA resulted in the recovery of RGC spiking activity and amplitude toward control levels. (E) Memantine washout, in the presence of continued perfusion with TBOA, resulted in the return of RGC excitotoxicity. (F) TBOA washout was followed by the recovery of RGC spiking activity and amplitude toward control levels. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity is 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scale are microvolts and millivolts, respectively.
Figure 6.
 
Effect of memantine on retinal light responses. Left: averaged ERG (top traces) and RGC (bottom traces) responses to dim stimuli of either 10-ms (top; indicated as vertical hatch mark at 0.4 second) or 1-second (bottom; indicated as solid horizontal bar extending from 0.4 to 1.4 seconds) duration obtained under control conditions. The 10-ms flash delivered 2.0 × 104 photons/flash · cm−2 while the 1-second stimulus delivered 2.8 × 105 photons/s · cm−2. Averages of 11 responses were used for all traces. Individual spikes in the RGC response are represented as unitary voltage pulses of 1-ms duration. Right: averaged responses from the same cell obtained at approximately 10 minutes after continuous application of 30 μM memantine. Memantine was associated with slight delays in the timing of the ERG b-wave and RGC OFF responses. ERG scale is in microvolts; RGC scale is in arbitrary units.
Figure 6.
 
Effect of memantine on retinal light responses. Left: averaged ERG (top traces) and RGC (bottom traces) responses to dim stimuli of either 10-ms (top; indicated as vertical hatch mark at 0.4 second) or 1-second (bottom; indicated as solid horizontal bar extending from 0.4 to 1.4 seconds) duration obtained under control conditions. The 10-ms flash delivered 2.0 × 104 photons/flash · cm−2 while the 1-second stimulus delivered 2.8 × 105 photons/s · cm−2. Averages of 11 responses were used for all traces. Individual spikes in the RGC response are represented as unitary voltage pulses of 1-ms duration. Right: averaged responses from the same cell obtained at approximately 10 minutes after continuous application of 30 μM memantine. Memantine was associated with slight delays in the timing of the ERG b-wave and RGC OFF responses. ERG scale is in microvolts; RGC scale is in arbitrary units.
Figure 7.
 
Summary results of effects of memantine on retinal light responses. Results obtained after the switch to a 1-, 3-, 10-, or 30-μM concentration are shown for measures of ERG b-wave amplitude (b-Amp), b-wave time-to-peak (b-Tpeak), RGC ON response latency (ON-Lat), and RGC OFF response latency (OFF-Lat). RGC response timing was measured as the interval between the onset of the 10-ms flash and the onset of the response as shown in the Figure 6 , top. In each case, measures are normalized with respect to those obtained immediately preceding the switch to memantine. Statistical analysis was made using the paired t-test (*P < 0.05, **P < 0.01, ***P < 0.001). The number of cells tested at each concentration is indicated at the base of each bar plot. Mean value ± SEM for control measures: b-Amp μV (1 μM = 94.6 ± 10.2; 3 μM = 87.6 ± 9.5; 10 μM = 83.1 ± 9.0; 30 μM = 71.7 ± 14.9), b -Tpeak ms (1 μM = 95.1 ± 1.4; 3 μM = 95.9 ± 1.3; 10 μM = 103.7 ± 2.1; 30 μM = 101.2 ± 3.0), ON-Lat ms (3 μM = 47.9 ± 4.2; 10 μM = 44.0 ± 4.7) and OFF-Lat ms (1 μM = 115.5 ± 12.4; 3 μM = 115.5 ± 12.4; 10 μM = 136.8 ± 8.8; 30 μM = 125.5 ± 9.1).
Figure 7.
 
Summary results of effects of memantine on retinal light responses. Results obtained after the switch to a 1-, 3-, 10-, or 30-μM concentration are shown for measures of ERG b-wave amplitude (b-Amp), b-wave time-to-peak (b-Tpeak), RGC ON response latency (ON-Lat), and RGC OFF response latency (OFF-Lat). RGC response timing was measured as the interval between the onset of the 10-ms flash and the onset of the response as shown in the Figure 6 , top. In each case, measures are normalized with respect to those obtained immediately preceding the switch to memantine. Statistical analysis was made using the paired t-test (*P < 0.05, **P < 0.01, ***P < 0.001). The number of cells tested at each concentration is indicated at the base of each bar plot. Mean value ± SEM for control measures: b-Amp μV (1 μM = 94.6 ± 10.2; 3 μM = 87.6 ± 9.5; 10 μM = 83.1 ± 9.0; 30 μM = 71.7 ± 14.9), b -Tpeak ms (1 μM = 95.1 ± 1.4; 3 μM = 95.9 ± 1.3; 10 μM = 103.7 ± 2.1; 30 μM = 101.2 ± 3.0), ON-Lat ms (3 μM = 47.9 ± 4.2; 10 μM = 44.0 ± 4.7) and OFF-Lat ms (1 μM = 115.5 ± 12.4; 3 μM = 115.5 ± 12.4; 10 μM = 136.8 ± 8.8; 30 μM = 125.5 ± 9.1).
The authors thank Peter Agey for assistance with animal anesthesia and enucleation of the eyes used in this study. 
LucasDR, NewhouseJP. The toxic effect of sodium l-glutamate on the inner layers of the retina. Arch Ophthalmol. 1957;58:193–201. [CrossRef]
ChoiDW. Glutamate neurotoxicity and diseases of the nervous system. Neuron. 1988;1:623–234. [CrossRef] [PubMed]
AizenmanE, FroschMP, LiptonSA. Responses mediated by excitatory amino acid receptors in solitary retinal ganglion cells from rat. J Physiol. 1988;396:75–91. [CrossRef] [PubMed]
MasseySC, MillerRF. N-methyl-D-aspartate receptors of ganglion cells in rabbit retina. J Neurophysiol. 1990;63(1)16–30. [PubMed]
CohenED, ZhouZJ, FainGL. Ligand-gated currents of alpha and beta ganglion cells in the cat retinal slice. J Neurophysiol. 1994;72(3)1260–1269. [PubMed]
GrunertU, HaverkampS, FletcherEL, WässleH. Synaptic distribution of ionotropic glutamate receptors in the inner plexiform layer of the primate retina. J Comp Neurol. 2002;447(2)138–151. [CrossRef] [PubMed]
LevyDI, LiptonSA. Comparison of delayed administration of competitive and uncompetitive antagonists in preventing NMDA receptor-mediated neuronal death. Neurology. 1990;40(4)852–855. [CrossRef] [PubMed]
SucherNJ, LeiSZ, LiptonSA. Calcium channel antagonists attenuate NMDA receptor-mediated neurotoxicity of retinal ganglion cells in culture. Brain Res. 1991;551(1–2)297–302. [CrossRef] [PubMed]
KitanoS, MorganJ, CaprioliJ. Hypoxic and excitotoxic damage to cultured rat retinal ganglion cells. Exp Eye Res. 1996;63(1)105–112. [CrossRef] [PubMed]
PangIH, WexlerEM, NawyS, DeSantisL, KapinMA. Protection by eliprodil against excitotoxicity in cultured rat retinal ganglion cells. Invest Ophthalmol Vis Sci. 1999;40(6)1170–1176. [PubMed]
LuoX, HeidingerV, PicaudS, et al. Selective excitotoxic degeneration of adult pig retinal ganglion cells in vitro. Invest Ophthalmol Vis Sci. 2001;42(5)1096–1106. [PubMed]
TungNN, MorganIG, EhrlichD. A quantitative analysis of the effects of excitatory neurotoxins on retinal ganglion cells in the chick. Vis Neurosci. 1990;4(3)217–223. [CrossRef] [PubMed]
SiliprandiR, CanellaR, CarmignotoG, et al. N-methyl-d-aspartate-induced neurotoxicity in the adult rat retina. Vis Neurosci. 1992;8(6)567–573. [CrossRef] [PubMed]
SabelBA, SautterJ, StoehrT, SiliprandiR. A behavioral model of excitotoxicity: retinal degeneration, loss of vision, and subsequent recovery after intraocular NMDA administration in adult rats. Exp Brain Res. 1995;106(1)93–105. [PubMed]
LiY, SchlampCL, NickellsRW. Experimental induction of retinal ganglion cell death in adult mice. Invest Ophthalmol Vis Sci. 1999;40(5)1004–1008. [PubMed]
KidoN, TaniharaH, HonjoM, et al. Neuroprotective effects of brain-derived neurotrophic factor in eyes with NMDA-induced neuronal death. Brain Res. 2000;884(1–2)59–67. [CrossRef] [PubMed]
ChidlowG, OsborneNN. Rat retinal ganglion cell loss caused by kainate, NMDA and ischemia correlates with a reduction in mRNA and protein of Thy-1 and neurofilament light. Brain Res. 2003;963(1–2)298–306. [CrossRef] [PubMed]
UllianEM, BarkisWB, ChenS, DiamondJS, BarresBA. Invulnerability of retinal ganglion cells to NMDA excitotoxicity. Mol Cell Neurosci. 2004;26:544–557. [CrossRef] [PubMed]
PellegriniJW, LiptonSA. Delayed administration of memantine prevents N-methyl-D-aspartate receptor-mediated neurotoxicity. Ann Neurol. 1993;33(4)403–407. [CrossRef] [PubMed]
LiptonSA. Memantine prevents HIV coat protein-induced neuronal injury in vitro. Neurology. 1992;42(7)1403–1405. [CrossRef] [PubMed]
LagrèzeWA, KnörleR, BachM, FeuersteinTJ. Memantine is neuroprotective in a rat model of pressure-induced retinal ischemia. Invest Ophthalmol Vis Sci. 1998;39(6)1063–1066. [PubMed]
OsborneNN. Memantine reduces alternations to the mammalian retina, in situ, induced by ischemia. Vis Neurosci. 1999;16(1)45–52. [PubMed]
LagrèzeWA, OttoT, FeuersteinTJ. Neuroprotection in ischemia of the retina in an animal model. Ophthalmologe. 1999;96(6)370–374. [CrossRef] [PubMed]
KimTW, KimDM, ParkKH, KimH. Neuroprotective effect of memantine in a rabbit model of optic nerve ischemia. Korean J Ophthalmol. 2002;16(1)1–7. [CrossRef] [PubMed]
WoldeMussieE, YolesE, SchwartzM, RuizG, WheelerLA. Neuroprotective effect of memantine in different retinal injury models in rats. J Glaucoma. 2002;11(6)474–480. [CrossRef] [PubMed]
KusariJ, ZhouS, PadilloE, ClarkeKG, GilDW. Effect of memantine on neuroretinal function and retinal vascular changes of streptozotocin-induced diabetic rats. Invest Ophthalmol Vis Sci. 2007;48(11)5152–5159. [CrossRef] [PubMed]
SchuettaufF, QuintoK, NaskarR, ZurakowskiD. Effects of anti-glaucoma medications on ganglion cell survival: the DBA/2J mouse model. Vision Res. 2002;42(20)2333–2237. [CrossRef] [PubMed]
ZhongL, BradleyJ, SchubertW, et al. Erythropoietin promotes survival of retinal ganglion cells in DBA/2J glaucoma mice. Invest Ophthalmol Vis Sci. 2007;48(3)1212–1218. [CrossRef] [PubMed]
GuZ, YamamotoT, KawaseC, et al. Neuroprotective effect of N-methyl-D-aspartate receptor antagonists in an experimental glaucoma model in the rat (in Japanese). Nippon Ganka Gakkai Zasshi. 2000;104(1)11–16. [PubMed]
HareWA, WoldeMussieE, LaiR, et al. Efficacy and safety of memantine treatment for reduction of changes associated with experimental glaucoma in monkey, I: Functional measures. Invest Ophthalmol Vis Sci. 2004;45(8)2625–2639. [CrossRef] [PubMed]
HareWA, MoldeMussieE, WeinrebRN, et al. Efficacy and safety of memantine treatment for reduction of changes associated with experimental glaucoma in monkey, II: Structural measures. Invest Ophthalmol Vis Sci. 2004;45(8)2640–2651. [CrossRef] [PubMed]
NowakL, BregestovskiP, AscherP, HerbertA, ProchiantzA. Magnesium gates glutamate-activated channels in mouse central neurons. Nature. 1984;307:462–465. [CrossRef] [PubMed]
MayerML, WestbrookGL, GuthriePB. Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurons. Nature. 1984;309:261–263. [CrossRef] [PubMed]
ZhangL, RzigalinskiBA, EllisEF, SatinLS. Reduction of voltage-dependent Mg2+ blockade of NMDA current in mechanically injured neurons. Science. 1996;274:1921–1923. [CrossRef] [PubMed]
DelbarreG, DelbarreB, CalinonF, FergerA. Accumulation of amino acids and hydroxyl free radicals in brain and retina of gerbil after transient ischemia. J Ocul Pharmacol. 1991;7(2)147–155. [CrossRef] [PubMed]
Louzada-JùniorP, DiasJJ, SantosWF, LachatJJ, BradfordHF, Coutinho-NettoJ. Glutamate release in experimental ischemia of the retina: an approach using microdialysis. J Neurochem. 1992;59(1)358–363. [CrossRef] [PubMed]
NealMJ, CunninghamJR, HutsonPH, HoggJ. Effects of ischemia on neurotransmitter release from the isolated retina. J Neurochem. 1994;62(3)1025–1033. [PubMed]
DonelloJE, PadilloEU, WebsterML, WheelerLA, GilDW. Alpha(2)-adrenoceptor agonists inhibit vitreal glutamate and aspartate accumulation and preserve retinal function after transient ischemia. J Pharmacol Exp Ther. 2001;296(1)216–223. [PubMed]
NucciC, TartaglioneR, RombolàL, MorroneLA, FazziE, BagettaG. Neurochemical evidence to implicate elevated glutamate in the mechanisms of high intraocular pressure (IOP)-induced retinal ganglion cell death in rat. Neurotoxicology. 2005;26(5)935–941. [CrossRef] [PubMed]
LuMJ, PulidoJS, McCannelCA, et al. Detection of elevated signaling amino acids in human diabetic vitreous by rapid capillary electrophoresis. Exp Diabetes Res. 2007;2007:39765. [PubMed]
KowluruRA, EngermanRL, CaseGL, KernTS. Retinal glutamate in diabetes and effect of antioxidants. Neurochem Int. 2001;38(5)385–390. [CrossRef] [PubMed]
BertramKM, BulaDV, PulidoJS, et al. Amino-acid levels in subretinal and vitreous fluid of patients with retinal detachment. Eye. 2008;22(4)582–589. [CrossRef] [PubMed]
MartinKR, Levkovitch-VerbinH, ValentaD, BaumrindL, PeaseME, QuigleyHA. Retinal glutamate transporter changes in experimental glaucoma and after optic nerve transaction in the rat. Invest Ophthalmol Vis Sci. 2002;43(7)2236–2243. [PubMed]
WoldeMussieE, WijonoM, RuizG. MÜller cell response to laser-induced increase in intraocular pressure in rats. Glia. 2004;47(2)109–119. [CrossRef] [PubMed]
MorenoMC, SandeP, MarcosHA, de ZavaliaN, Keller-SarmientoMI, RosensteinRE. Effect of glaucoma on the retinal glutamate/glutamine cycle activity. FASEB J. 2005;19(9)1161–1162. [PubMed]
Carter-DawsonL, ShenFF, HarwerthRS, CrawfordML, SmithEL, 3rd, WhitetreeA. Glutathione content is altered in Müller cells of monkey eyes with experimental glaucoma. Neurosci Lett. 2004;364(1)7–10. [CrossRef] [PubMed]
SullivanRK, WoldeMussieE, MacnabL, RuizG, PowDV. Evoked expression of the glutamate transporter GLT-1c in retinal ganglion cells in human glaucoma and in a rat model. Invest Ophthalmol Vis Sci. 2006;47(9)3853–3859. [CrossRef] [PubMed]
ZeevalkGD, NicklasWJ. Evidence that the loss of the voltage-dependent Mg2+ block at the N-methyl-D-aspartate receptor underlies receptor activation during inhibition of neuronal metabolism. J Neurochem. 1992;59(4)1211–1120. [CrossRef] [PubMed]
ChenHS, PellegriniJW, AggarwalSk, et al. Open-channel block of N-methyl-D-aspartate (NMDA) responses by memantine: therapeutic advantage against NMDA receptor-mediated neurotoxicity. J Neurosci. 1992;12(11)4427–4436. [PubMed]
ParsonsCG, QuackG, BresinkI, et al. Comparison of the potency, kinetics and voltage-dependency of a series of uncompetitive NMDA receptor antagonists in vitro with anticonvulsive and motor impairment activity in vivo. Neuropharmacology. 1995;34(10)1239–1258. [CrossRef] [PubMed]
FrankiewiczT, PotierB, BashirZI, CollingridgeGL, ParsonsCG. Effects of memantine and MK-801 on NMDA-induced currents in cultured neurons and on synaptic transmission and LTP in area CA1 of rat hippocampal slices. Br J Pharmacol. 1996;117(4)689–697. [CrossRef] [PubMed]
ChenHS, LiptonSA. Mechanism of memantine block of NMDA-activated channels in rat retinal ganglion cells: uncompetitive antagonism. J Physiol. 1997;499:27–46. [CrossRef] [PubMed]
ShimamotoK, LebrunB, Yasuda-KamataniY, et al. dl-Threo-beta-benzyloxyaspartate, a potent blocker of excitatory amino acid transporters. Mol Pharmacol. 1998;53(2)195–201. [PubMed]
IzumiY, ShimamotoK, BenzAM, HammermanSB, OlneyJW, ZorumskiCF. Glutamate transporters and retinal excitotoxicity. Glia. 2002;39(1)58–68. [CrossRef] [PubMed]
NelsonRM, LambertDG, Richard GreenA, HainsworthAH. Pharmacology of ischemia-induced glutamate efflux from rat cerebral cortex in vitro. Brain Res. 2003;964(1)1–8. [CrossRef] [PubMed]
BushRA, SievingPA. A proximal retinal component in the primate photopic ERG a-wave. Invest Ophthalmol Vis Sci. 1994;35(2)635–645. [PubMed]
DongC-J, HareWA. Contribution to the kinetics and amplitude of the electroretinogram b-wave by third order retinal neurons in the rabbit retina. Vision Res. 2000;40:579–589. [CrossRef] [PubMed]
RangaswamyNV, FrishmanLJ, DorotheoEU, et al. Photopic ERGs in patients with optic neuropathies: Comparison with primate ERGs after pharmacologic blockade of inner retina. Invest Ophthalmol Vis Sci. 2004;45(10)3827–3837. [CrossRef] [PubMed]
LuoX, LambrouGN, SabelJA, HicksD. Hypoglycemia induces general neuronal death, whereas hypoxia and glutamate transport blockade lead to selective retinal ganglion cell in vitro. Invest Ophthalmol Vis Sci. 2001;42(11)2695–2705. [PubMed]
SembaJ, WakutaMS. Regional differences in the effects of glutamate uptake inhibitor l-trans-pyrrolidine-2,4-dicarboxylic acid on extracellular amino acids and dopamine in rat brain: an in vivo microdialysis study. Gen Pharmacol. 1998;31(3)399–404. [CrossRef] [PubMed]
JabaudonD, ShimamotoK, Yasuda-KamataniY, ScanzianiM, GähwilerBH, GerberU. Inhibition of uptake unmasks rapid extracellular turnover of glutamate of nonvesicular origin. Proc Natl Acad Sci USA. 1999;96(15)8733–8738. [CrossRef] [PubMed]
KornhuberJ, QuackG. Cerebrospinal fluid and serum concentrations of the N-methyl-d-aspartate (NMDA) receptor antagonist memantine in man. Neurosci Lett. 1995;195(2)137–139. [CrossRef] [PubMed]
Figure 1.
 
Illustration of the retinal preparation used for electrophysiological recordings. An 11-mm diameter sample of retina-choroid-sclera was cut from the posterior fundus (A). The superior border of this sample included the inferior rim of the optic nerve head (ONH). All recordings were obtained from a region located just inferior to the visual streak (VS). The retina-choroid-sclera sample was mounted retina side up in the perfusion chamber (B). Silver chloride electrodes, located beneath the preparation and in the bath, provided for measures of transretinal voltage (ERG). A tungsten microelectrode, using the bath electrode as the reference, provided measures of spiking activity (action potentials) from single RGCs. Diffuse light stimuli were generated with a blue-green LED.
Figure 1.
 
Illustration of the retinal preparation used for electrophysiological recordings. An 11-mm diameter sample of retina-choroid-sclera was cut from the posterior fundus (A). The superior border of this sample included the inferior rim of the optic nerve head (ONH). All recordings were obtained from a region located just inferior to the visual streak (VS). The retina-choroid-sclera sample was mounted retina side up in the perfusion chamber (B). Silver chloride electrodes, located beneath the preparation and in the bath, provided for measures of transretinal voltage (ERG). A tungsten microelectrode, using the bath electrode as the reference, provided measures of spiking activity (action potentials) from single RGCs. Diffuse light stimuli were generated with a blue-green LED.
Figure 2.
 
NMDA-induced RGC excitotoxicity is reversed by memantine. (A) Recordings of the ERG (top trace) and activity of a single RGC (bottom trace) obtained under control conditions. (B) Increased tonic RGC spiking activity and reduction of spike amplitude seen at 2 minutes after application of 200 μM NMDA. (C) Complete block of RGC spike generation at 6 minutes after application of NMDA. (D, E) Recovery of RGC spike amplitude and tonic spiking activity to near control levels after switch to perfusion medium containing NMDA in combination with 10 μM memantine. All traces were recorded from the same OFF-RGC in the order (A)→(D). In all panels, a dim stimulus of 10-ms duration was delivered at 0.4 second (vertical solid line). Stimulus intensity = 1.2 × 105 photons/flash-cm2. Scale for ERG is in microvolts; the scale for RGCs is in millivolts.
Figure 2.
 
NMDA-induced RGC excitotoxicity is reversed by memantine. (A) Recordings of the ERG (top trace) and activity of a single RGC (bottom trace) obtained under control conditions. (B) Increased tonic RGC spiking activity and reduction of spike amplitude seen at 2 minutes after application of 200 μM NMDA. (C) Complete block of RGC spike generation at 6 minutes after application of NMDA. (D, E) Recovery of RGC spike amplitude and tonic spiking activity to near control levels after switch to perfusion medium containing NMDA in combination with 10 μM memantine. All traces were recorded from the same OFF-RGC in the order (A)→(D). In all panels, a dim stimulus of 10-ms duration was delivered at 0.4 second (vertical solid line). Stimulus intensity = 1.2 × 105 photons/flash-cm2. Scale for ERG is in microvolts; the scale for RGCs is in millivolts.
Figure 3.
 
Reversal of NMDA-induced RGC excitotoxicity by AP-5. (A) ERG and RGC responses recorded under control conditions. (B) Application of NMDA is associated with increased tonic RGC spiking activity and decreased spike amplitude. (C) Application of AP5 in the presence of NMDA resulted in reversal of NMDA-induced RGC excitotoxicity. (D) ERG and RGC responses obtained after washout of NMDA+AP5. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity was 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scales are microvolts and millivolts, respectively.
Figure 3.
 
Reversal of NMDA-induced RGC excitotoxicity by AP-5. (A) ERG and RGC responses recorded under control conditions. (B) Application of NMDA is associated with increased tonic RGC spiking activity and decreased spike amplitude. (C) Application of AP5 in the presence of NMDA resulted in reversal of NMDA-induced RGC excitotoxicity. (D) ERG and RGC responses obtained after washout of NMDA+AP5. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity was 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scales are microvolts and millivolts, respectively.
Figure 4.
 
(AC) Perfusion with zero magnesium medium was associated with RGC excitotoxicity that is similar to that induced by NMDA. (D) Application of memantine (10 μM) in the presence of zero Mg2+ resulted in recovery of RGC spiking activity and spike amplitude toward control levels. (E) ERG and RGC responses obtained after washout of memantine and zero Mg2+. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity was 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scales are microvolts and millivolts, respectively.
Figure 4.
 
(AC) Perfusion with zero magnesium medium was associated with RGC excitotoxicity that is similar to that induced by NMDA. (D) Application of memantine (10 μM) in the presence of zero Mg2+ resulted in recovery of RGC spiking activity and spike amplitude toward control levels. (E) ERG and RGC responses obtained after washout of memantine and zero Mg2+. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity was 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scales are microvolts and millivolts, respectively.
Figure 5.
 
(AC) Blockade of excitatory amino acid transporters with TBOA was associated RGC excitotoxicity that was similar to that observed after application of either NMDA or zero Mg2+. Tonic spiking activity was increased and, in this cell, spike generation was blocked. (D) Application of memantine (10 μM) in the presence of TBOA resulted in the recovery of RGC spiking activity and amplitude toward control levels. (E) Memantine washout, in the presence of continued perfusion with TBOA, resulted in the return of RGC excitotoxicity. (F) TBOA washout was followed by the recovery of RGC spiking activity and amplitude toward control levels. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity is 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scale are microvolts and millivolts, respectively.
Figure 5.
 
(AC) Blockade of excitatory amino acid transporters with TBOA was associated RGC excitotoxicity that was similar to that observed after application of either NMDA or zero Mg2+. Tonic spiking activity was increased and, in this cell, spike generation was blocked. (D) Application of memantine (10 μM) in the presence of TBOA resulted in the recovery of RGC spiking activity and amplitude toward control levels. (E) Memantine washout, in the presence of continued perfusion with TBOA, resulted in the return of RGC excitotoxicity. (F) TBOA washout was followed by the recovery of RGC spiking activity and amplitude toward control levels. Vertical solid line: light stimulus of 10-ms duration delivered at 0.4 second. Stimulus intensity is 1.2 × 105 photons/flash-cm2. ERG and RGC vertical scale are microvolts and millivolts, respectively.
Figure 6.
 
Effect of memantine on retinal light responses. Left: averaged ERG (top traces) and RGC (bottom traces) responses to dim stimuli of either 10-ms (top; indicated as vertical hatch mark at 0.4 second) or 1-second (bottom; indicated as solid horizontal bar extending from 0.4 to 1.4 seconds) duration obtained under control conditions. The 10-ms flash delivered 2.0 × 104 photons/flash · cm−2 while the 1-second stimulus delivered 2.8 × 105 photons/s · cm−2. Averages of 11 responses were used for all traces. Individual spikes in the RGC response are represented as unitary voltage pulses of 1-ms duration. Right: averaged responses from the same cell obtained at approximately 10 minutes after continuous application of 30 μM memantine. Memantine was associated with slight delays in the timing of the ERG b-wave and RGC OFF responses. ERG scale is in microvolts; RGC scale is in arbitrary units.
Figure 6.
 
Effect of memantine on retinal light responses. Left: averaged ERG (top traces) and RGC (bottom traces) responses to dim stimuli of either 10-ms (top; indicated as vertical hatch mark at 0.4 second) or 1-second (bottom; indicated as solid horizontal bar extending from 0.4 to 1.4 seconds) duration obtained under control conditions. The 10-ms flash delivered 2.0 × 104 photons/flash · cm−2 while the 1-second stimulus delivered 2.8 × 105 photons/s · cm−2. Averages of 11 responses were used for all traces. Individual spikes in the RGC response are represented as unitary voltage pulses of 1-ms duration. Right: averaged responses from the same cell obtained at approximately 10 minutes after continuous application of 30 μM memantine. Memantine was associated with slight delays in the timing of the ERG b-wave and RGC OFF responses. ERG scale is in microvolts; RGC scale is in arbitrary units.
Figure 7.
 
Summary results of effects of memantine on retinal light responses. Results obtained after the switch to a 1-, 3-, 10-, or 30-μM concentration are shown for measures of ERG b-wave amplitude (b-Amp), b-wave time-to-peak (b-Tpeak), RGC ON response latency (ON-Lat), and RGC OFF response latency (OFF-Lat). RGC response timing was measured as the interval between the onset of the 10-ms flash and the onset of the response as shown in the Figure 6 , top. In each case, measures are normalized with respect to those obtained immediately preceding the switch to memantine. Statistical analysis was made using the paired t-test (*P < 0.05, **P < 0.01, ***P < 0.001). The number of cells tested at each concentration is indicated at the base of each bar plot. Mean value ± SEM for control measures: b-Amp μV (1 μM = 94.6 ± 10.2; 3 μM = 87.6 ± 9.5; 10 μM = 83.1 ± 9.0; 30 μM = 71.7 ± 14.9), b -Tpeak ms (1 μM = 95.1 ± 1.4; 3 μM = 95.9 ± 1.3; 10 μM = 103.7 ± 2.1; 30 μM = 101.2 ± 3.0), ON-Lat ms (3 μM = 47.9 ± 4.2; 10 μM = 44.0 ± 4.7) and OFF-Lat ms (1 μM = 115.5 ± 12.4; 3 μM = 115.5 ± 12.4; 10 μM = 136.8 ± 8.8; 30 μM = 125.5 ± 9.1).
Figure 7.
 
Summary results of effects of memantine on retinal light responses. Results obtained after the switch to a 1-, 3-, 10-, or 30-μM concentration are shown for measures of ERG b-wave amplitude (b-Amp), b-wave time-to-peak (b-Tpeak), RGC ON response latency (ON-Lat), and RGC OFF response latency (OFF-Lat). RGC response timing was measured as the interval between the onset of the 10-ms flash and the onset of the response as shown in the Figure 6 , top. In each case, measures are normalized with respect to those obtained immediately preceding the switch to memantine. Statistical analysis was made using the paired t-test (*P < 0.05, **P < 0.01, ***P < 0.001). The number of cells tested at each concentration is indicated at the base of each bar plot. Mean value ± SEM for control measures: b-Amp μV (1 μM = 94.6 ± 10.2; 3 μM = 87.6 ± 9.5; 10 μM = 83.1 ± 9.0; 30 μM = 71.7 ± 14.9), b -Tpeak ms (1 μM = 95.1 ± 1.4; 3 μM = 95.9 ± 1.3; 10 μM = 103.7 ± 2.1; 30 μM = 101.2 ± 3.0), ON-Lat ms (3 μM = 47.9 ± 4.2; 10 μM = 44.0 ± 4.7) and OFF-Lat ms (1 μM = 115.5 ± 12.4; 3 μM = 115.5 ± 12.4; 10 μM = 136.8 ± 8.8; 30 μM = 125.5 ± 9.1).
Table 1.
 
Summary of All RGCs Tested for Experimental Excitotoxicity
Table 1.
 
Summary of All RGCs Tested for Experimental Excitotoxicity
Treatment RGC Type
ON OFF ON-OFF
NMDA (50–200 μM) 23 27 4
 +Memantine (10, 30 μM) 16 17 4
 +AP5 (30 μM) 1 4
Zero Mg2+ 4 2
 +Memantine (10 μM) 4 2
TBOA (10–75 μM) 7 8 2
 +Memantine (10 μM) 5 8
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×