Topiramate reduces excitotoxic and ischemic injury in the rat retina

Brain Research 967 (2003) 257–266 www.elsevier.com / locate / brainres

Research report

Topiramate reduces excitotoxic and ischemic injury in the rat retina Shinji Yoneda, Etsuko Tanaka, Wakana Goto, Takashi Ota, Hideaki Hara* Glaucoma Group, Research and Development Division, Santen Pharmaceutical Co. Ltd., 8916 -16, Takayama, Ikoma 630 -0101, Japan Accepted 7 January 2003

Abstract The effects of topiramate, a drug used clinically as an anti-epileptic, were investigated in excitotoxin-induced neurotoxicity models involving two different retinal primary cultures and in a rat model of retinal ischemic injury. For the in vitro studies, we used retinal-neuron cultures from rat embryos and purified retinal ganglion cells (RGCs) from newborn rats. In the retinal-neuron cultures, neurotoxicity was induced by a 10-min exposure to 1 mM glutamate or (6)-a-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA). In RGCs, neurotoxicity was induced by incubation for 3 days in a culture medium containing 25 mM glutamate. For the in vivo study, retinal ischemia was induced by elevating intraocular pressure to 130 mmHg for 45 min, and topiramate was administered intraperitoneally before and after the ischemia. Retinal damage was evaluated by measuring the number of cells in the ganglion cell layer (GCL) and the thickness of the inner plexiform layer (IPL), and by examining the a- and b-waves of the electroretinogram (ERG). Topiramate ($1 mM) markedly reduced the neuronal cell death induced by each of the excitotoxins in rat retinal-neuron cultures and in RGCs. Ischemia caused a decrease in GCL cells and in IPL thickness, and a diminution of the ERG waves. Histopathologic and functional analyses indicated that systemic treatment with topiramate prevented ischemia-induced damage in a dose-dependent manner. In conclusion, topiramate was protective against excitotoxic and ischemic retinal-neuron damage in vitro and in vivo, respectively. Therefore, it may be useful for treatment of the retina-related diseases such as central retinal artery occlusion, diabetic retinopathy, and glaucoma.  2003 Elsevier Science B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Degenerative disease: other Keywords: Topiramate; Retinal cell culture; Retinal ischemia; Retinal ganglion cell; Rat

1. Introduction Ischemia-induced retinal injury which plays important roles in retinal or optic nerve diseases such as central retinal artery occlusion, the retinopathy of prematurity, diabetic retinopathy, and possibly glaucoma leads to a characteristic degeneration of retinal ganglion cells and inner retinal neurons. The mechanisms underlying such degeneration are not fully understood, and there are no current efficacious therapies. Glutamate has been established as the major excitatory neurotransmitter in the retina [2,7,8], and glutamate excitotoxicity has been implicated

*Corresponding author. Tel.: 181-743-79-4529; fax: 181-743-794518. E-mail address: [email protected] (H. Hara).

as a pathogenic mechanism leading to neuronal cell death in cases of ischemic retinal injury [26,27,32,52]. Topiramate [2,3:4,5-bis-O-(1-methylethylidene) b-Dfructopyranose sulfamate] is a structurally novel anti-epileptic drug that is effective against various types of epilepsy [28,29,39,40]. In addition, topiramate is undergoing clinical evaluation for use in the treatment of neuropathic pain, bipolar disorder, and migraine. Several pharmacological actions of topiramate have been identified and are thought to contribute to its anti-epileptic and other actions. These include: (a) a blockade of the (6)-a-amino3-hydroxy-5-methyl-4-isoxazole-propionic acid (AMPA) / kainite receptor [13], (b) an attenuation of the voltagegated sodium current [31], and (c) a positive modulation of g-aminobutyric acid (GABA) receptors [48]. In addition to these pharmacological actions, topiramate has a negative modulatory effect on L-type high-voltage-activated cal-

0006-8993 / 03 / $ – see front matter  2003 Elsevier Science B.V. All rights reserved. doi:10.1016 / S0006-8993(03)02270-4

258

S. Yoneda et al. / Brain Research 967 (2003) 257–266

cium channels [53]. Under neuronal ischemic conditions, glutamate efflux from presynaptic neurons is mainly achieved via sodium-dependent, passive transport [38]. In addition, GABA is considered a primary inhibitory neurotransmitter in the mammalian brain and is thought to counterbalance the physiological and toxic effects of glutamate [15,36]. Activation of ionotropic glutamate receptors depolarizes neurons and subsequently activates voltage-dependent calcium channels, which potentiates the excessive influx of calcium ions [9,44]. In view of these pharmacological properties, topiramate would seem to be potentially useful as a neuroprotectant in conditions involving retinal ischemia. In animal models of brain ischemia, topiramate shows evidence of neuroprotective effects. It has been reported to reduce ischemia-induced behavioral disturbances and infarct damage when given 2 h after middle cerebral artery embolization in the rat [49], and to reduce hippocampal neuronal damage in a gerbil model of transient global ischemia [23]. Moreover, it has been reported to promote both neurite outgrowth in vitro and recovery of facial nerve function after crush injury in vivo [42]. In view of the fact that in widespread use in many countries as an anti-epileptic drug, topiramate has been found to have a high tolerability, and that it has been found to exert neuroprotective effects in some neuronal-damage models, it has the potential to be a useful neuroprotectant against optic neuropathies, especially those caused by retinal ischemia. The purpose of the present study was to examine the possible neuroprotective effects of topiramate against excitotoxic cell death among retinal neurons in vitro and against ischemic injury in retinal cells in vivo [by histological examination and by electroretinographic (ERG) assessment of visual function].

2. Material and methods

2.1. Rat retinal-cell culture Primary cultures obtained from the retina of fetal Wistar rats (18–19 days’ gestation) were purchased from the Shizuoka Laboratory Animal Center (Hamamatsu, Japan) and were used for the experiments, as described previously [19,21]. The retinal tissues were dissociated, and singlecell suspensions (1.0310 6 cells / ml) were plated onto plastic coverslips previously placed in 60-mm Coring culture dishes (80 ml of cell suspension). Cultures were incubated in Eagle’s minimum essential medium (EMEM; GIBCO, Grand Island, NY, USA) containing 2 mM glutamine, 11 mM glucose, and 24 mM HEPES. The medium was supplemented with 10% heat-inactivated fetal bovine serum (1–8 days after plating) and 10% heatinactivated horse serum (9–10 days after plating). After a 6-day culture, the growth of non-neuronal cells was

terminated by the addition of 10 25 M cytosine arabinoside. We used only those cultures maintained for 11 days. Cultures were exposed for 10 min to glutamate (1 mM) or AMPA (1 mM) followed by a 1-h incubation in glutamate- or AMPA-free medium, as described previously [19]. Each of three drugs [topiramate (provided by the R.W. Johnson Pharmaceutical Research Institute, Spring House, PA, USA) or (5R,10S)-(1)-5-methyl-10,11dihydro-5H-dibenzo[a,d]cyclo-hepten-5,10-imine hydrogen maleate (MK-801; RBI, Natick, MA, USA) or 6,7dinitroquinoxaline-2,3-dione (DNQX; RBI)] was applied to the cultures for 10 min together with the glutamate or AMPA. The neurotoxic effects of glutamate and AMPA were quantitatively assessed by the Trypan Blue exclusion method after the 1-h incubation in excitotoxin- and drugfree medium, as described previously [10,46]. To this end, cell cultures were stained with 1.5% Trypan Blue solution at room temperature for 10 min, then examined by Hoffman modulation microscopy using Hoffman Modulation Contrast  (Olympus, Tokyo, Japan) (magnification, 3400). More than 200 cells chosen at random were counted to determine the viability of the cell culture, which was assessed by counting the number of unstained cells (viable cells; counted by a person blind to the experimental design) in a masked fashion and expressing it as a percentage of the total number of cells (viable cells plus nonviable cells).

2.2. Retinal ganglion cell culture Retinal ganglion cell (RGC) culture reagents were obtained from GIBCO. The retrograde fluorescent tracer 1,19-dioctadecyl-3,3,39,39-tetramethyl-indocarbocyanine (DiI) and the fluorescent viability agent calcein acetoxymethyl ester (Calcein-AM) were obtained from Molecular Probes (Eugene, OR, USA). A cell-dissection kit was obtained from Worthington Biochemical (Freehold, NJ, USA). Recombinant human brain-derived neurotrophic factor (BDNF) and rat ciliary neurotrophic factor (CNTF) were obtained from Chemicon (Temecula, CA, USA). Unless noted, all other reagents were obtained from Sigma (St Louis, MO, USA). RGCs were purified from 5- to 7-day-old Wistar rats, essentially as previously described [33], with a minor modification. In brief, to determine purity and yield in this procedure, RGCs were labeled in a retrograde manner by injecting 1 mg / ml of the fluorescent tracer DiI, dissolved in N,N-dimethyl formamide, into the superior colliculi of anesthetized 2- to 4-day-old Wistar rats. Pups were decapitated and the eyes enucleated. The tissue was incubated at 37 8C for 30 min in 20 U / ml papain solution, 70 U / ml collagenase, and 2000 U / ml deoxyribonuclease in Earle’s Balanced salt solution containing 0.2 mM ethylenediaminetetraacetic acid (EDTA) and 1 mM L-cysteine. To yield a suspension of single cells, the tissue was then triturated sequentially through a

S. Yoneda et al. / Brain Research 967 (2003) 257–266

narrow-bore Pasteur pipette in a solution containing 1 mg / ml ovomucoid, 2000 U / ml deoxyribonuclease, and 1 mg / ml albumin. After centrifugation at 903g for 5 min, the cells were rewashed in another ovomucoid-albumin solution (10 mg / ml of each agent) through a narrow-bore Pasteur pipette. After centrifugation, the cells were suspended gently with a narrow-bore Pasteur pipette in 0.1% BSA in phosphate-buffered saline (PBS). To isolate RGCs from retinal tissue, we used the twostep panning procedure described previously [18,19,50]. The primary antibodies used were a monoclonal ascites IgG antibody against mouse macrophage SIRP (clone MRC OX 41) and a monoclonal IgG antibody against rat Thy-1.1 (clone OX-7), each purchased from Chemicon. Fifty-milliliter polypropylene tubes (Corning) were incubated with 3 ml PBS containing Thy-1.1 antibody (1:300), and polystyrene 25 cm 2 flasks were incubated with SIRP antibody (1:50), each at 4 8C overnight. The tubes and flasks were washed twice with PBS. Then, to prevent nonspecific binding of cells to the panning tubes and flasks, 3 ml PBS containing 0.1% BSA was applied to the coating area. The retinal suspension was incubated in an SIRP-coated flask at room temperature for 30 min. The non-adherent cells were removed and placed in the Thy-1.1 coated tubes, then incubated as described above. Thirty minutes later, the tubes were gently washed six times with 3 ml PBS. Finally, the adherent cells on SIRP-coated tubes were washed with serum-free culture medium containing Neurobasal Medium with 2 mM glutamine penicillin / streptomycin (100 U / ml–50 mg / ml), B27 supplement (1:50), 50 ng / ml each of BDNF and CNTF, and 10 mM forskolin as described previously [10]. After centrifugation at 903g for 5 min, the cells were seeded onto 12-mm glass coverslips that had previously been coated with 50 mg / ml poly-L-lysine and 10 mg / ml laminin. Purified RGCs were plated at a low density of approximately 1000 cells / cm 2 . This plating density provided cultures in which most RGCs grew in physical isolation from other RGCs. The purified RGCs were cultured in 400 ml serum-free medium. Each drug was diluted with the serum-free medium described above. Cultures were maintained at 37 8C in a humidified atmosphere containing 5% CO 2 and 95% air. In a previous study using purified RGCs, it was demonstrated that cell viability was markedly reduced by exposure to glutamate (25 mM) for 3 days [33]. In the present study, we therefore used a 3-day exposure to glutamate, with or without topiramate or DNQX. Cell viability was determined using 1 mM Calcein-AM as described previously [24]. In all cases, the rate and extent of process-outgrowth correlated closely with the degree of survival. In the present study, a surviving RGC was defined as a cell with a Calcein-AM-stained cell body and a process extending at least two cell-diameters from the cell body. The percentage of surviving RGCs was

259

determined on two or three coverslips for each condition in each experiment. The average relative percentage regeneration in at least six experiments conducted under each condition is expressed in the text and figures as the mean6S.E.

2.3. Experimental animals Male Sprague–Dawley rats (200–300 g), purchased from the Shizuoka Laboratory Animal Center, were kept under controlled light / dark conditions with food and water available ad libitum. Anesthesia was induced by inhalation of 3% halothane, and maintained using 1% halothane in 70% N 2 O and 30% O 2 . Topiramate and MK-801 were dissolved in saline (0.9% NaCl). Topiramate (25, 50, 100 or 200 mg / kg) was injected intraperitoneally at a volume of 1 ml / kg at two time points: 2 h before and 5 min after the period of ischemia (see below). MK-801 (10 mg / kg) was injected intraperitoneally 30 min before the onset of ischemia. In another group, vehicle (saline) was injected in the same way. Retinal ischemia was induced in halothane-anesthetized rats by elevating the intraocular pressure to 130 mmHg. This was achieved by cannulating the anterior chamber of the right eye with a tube connected to an elevated reservoir containing saline, in accordance with the procedure described previously [1,51]. Retinal ischemia was confirmed under the microscope by the whitening of the iris and the loss of the red reflex. Forty-five minutes later, the cannulating needle was removed. With the animals briefly re-anesthetized with halothane, reperfusion of the retinal vasculature was confirmed by examination of the fundus. With the aid of an animal blanket controller and a heating pad (ATB-1100; Nihon Kohden, Tokyo, Japan), body temperature was maintained at approximately 37 8C throughout the experiment and until the animal had recovered from the anesthesia. One drop of atropine 1% ophthalmic solution (Nitten Atropine Sulfate Ophthalmic Solution; Nitten, Nagoya, Japan) and one drop of gentamicin ophthalmic solution (Gentacin  Ophthalmic Solution; Schering-Plough, Osaka, Japan) was applied topically to the right eye before and after cannulation of the anterior chamber, respectively. Throughout these studies, all efforts were made to minimize animal suffering and to reduce the number of animals used. All animal experiments were carried out in accordance with the ARVO Statement on the Use of Animals in Ophthalmic and Vision Research.

2.4. Histological analysis Under anesthesia induced using an intramuscular injection of ketamine (25 mg / kg) and xylazine (10 mg / kg), each eye was enucleated 7 days after the period of ischemia and kept for at least 24 h at 4 8C immersed in a

260

S. Yoneda et al. / Brain Research 967 (2003) 257–266

fixative solution containing 2.5% glutaraldehyde and 2% paraformaldehyde. Eight paraffin-embedded sections (thickness, 3 mm) cut through the optic disc of each eye were prepared in a standard manner and stained with hematoxylin and eosin. Retinal damage was evaluated as described previously [51], three sections for each eye being used for the morphometric analysis. Light-microscope images were photographed, and (a) the cell density in the ganglion cell layer (GCL) at a distance between 1 and 1.5 mm from the optic disc and (b) the thickness of the inner plexiform layer (IPL) were measured on the photographs in a masked fashion. Data from three sections (selected randomly from eight sections) were averaged for each eye and used to evaluate the cell density in the GCL and the thickness of the IPL.

2.5. Measurement of electroretinograms ( ERG) After 3 days of reperfusion, the flashing ERG was recorded, and the amplitude of the a- and b-waves analyzed. The procedures used have been described elsewhere [37]. In brief, in rats anesthetized with halothane the pupils were dilated with phenylephrine hydrochloride and tropicamide (Mydrin-P Ophthalmic Solution; Santen). The scotopic ERG was recorded from animals dark-adapted for at least 60 min by placing a contact-lens type of white light-emitting diode (LED)-recording electrode on the cornea, with a reference electrode wire and a ground electrode wire being connected to the nose and tail, respectively. The cornea was intermittently irrigated with balanced salt solution to maintain adequate electrical contact and to prevent exposure–keratopathy. The LED electrode was irradiated by a photostimulator system (WLS-20; Mayo, Nagoya, Japan). The responses were amplified with a time constant of 0.3 s, and with low- and high-frequency cut filters of 0.5 and 1.5 kHz (SYNAX ER1100; Nihon GE Maruketto Medical System, Tokyo, Japan).

2.6. Statistical analysis Results are expressed as mean6S.E. Data were analyzed using a one-way analysis of variance (ANOVA) followed by Dunnett’s test or a Student’s t-test. Results were considered to show a significant difference at P,0.05.

3. Results

3.1. Glutamate- or AMPA-induced neurotoxicity in rat primary cell cultures The cell viability of retinal cultures (evaluated using Trypan Blue staining) was reduced to approximately 70% of control by a brief exposure of the cells to 1 mM

Fig. 1. Effects of topiramate and MK-801 on glutamate-induced neurotoxicity in rat retinal cell cultures. After glutamate (1 mM) had been present for 10 min, cultures were incubated for 60 min in glutamate-free medium prior to the assessment of cell survival. Topiramate (0.01–100 mM) or MK-801 (10 mM) was applied to cultures for 10 min together with the glutamate. Each value represents mean6S.E., n57 or 8. *P, 0.05 versus control; § P,0.05 versus glutamate alone (Student’s t-test); [ P,0.05 versus glutamate alone (Dunnett’s test).

glutamate followed by a 1-h incubation in standard culture medium (Fig. 1). The protective effects of topiramate and MK-801 were evaluated by adding each drug together with the glutamate. Addition of topiramate (0.01–100 mM) reduced the neurotoxicity in a concentration-dependent manner. The effect of topiramate was statistically significant at concentrations of 1 mM or more, while that of MK-801 was significant at 10 mM (the only concentration tested). Similarly, the cell viability of retinal cultures was reduced to approximately 63% of control by exposure to 1 mM AMPA. Topiramate at concentrations of 0.1 mM or more and DNQX at its only concentration (10 mM) significantly reduced the AMPA-induced neurotoxicity (Fig. 2).

3.2. Glutamate-induced neurotoxicity in rat purified retinal ganglion-cell cultures Our quantitative assessment of the protective effect of topiramate against glutamate-induced neurotoxicity in RGC cultures is shown in Table 1. Cell viability was reduced to approximately 75% of control by a 3-day exposure of the cells to 25 mM glutamate. The addition of topiramate (0.01–10 mM) reduced this neurotoxicity. The effect of topiramate was statistically significant at 1 and 10 mM, while that of DNQX was significant at its only concentration, 10 mM.

S. Yoneda et al. / Brain Research 967 (2003) 257–266

Fig. 2. Effects of topiramate and DNQX on AMPA-induced neurotoxicity in rat retinal cell cultures. After AMPA (1 mM) had been present for 10 min, cultures were incubated for 60 min in AMPA-free medium prior to the assessment of cell survival. Topiramate (0.01–100 mM) or DNQX (10 mM) was applied to cultures for 10 min together with the AMPA. Each value represents mean6S.E., n59 or 10. *P,0.05 versus control; § P,0.05 versus glutamate alone (Student’s t-test); [ P,0.05 versus glutamate alone (Dunnett’s test).

3.3. Neuroprotective effects against ischemia-induced retinal damage in the rat Typical photomicrographs of retinas removed at 7 days after a period of ischemia are shown in Fig. 3. In vehicletreated animals, degenerative changes were selectively observed in the inner layers of the ischemic retina, a characteristic of retinal ischemic atrophy (Fig. 3B). The actual changes observed were a marked reduction in cell density in the GCL and a marked reduction in the

Table 1 Effects of topiramate and 6,7-dinitroquinoxaline-2,3-dione (DNQX) on low-dose glutamate-induced neurotoxicity in purified rat retinal ganglioncell culture Treatment

Viability (%)

No treatment Glutamate alone Glutamate10.01 mM topiramate Glutamate10.1 mM topiramate Glutamate11 mM topiramate Glutamate110 mM topiramate Glutamate110 mM DNQX

81.961.4 60.461.2* 63.460.6 64.762.5 71.162.1 [ 69.461.6 [ 71.462.0 §

Purified RGCs were cultured in serum-free medium containing 50 ng / ml each of BDNF and CNTF, and 10 mM forskolin. After 3 days’ exposure to 25 mM glutamate either alone or with a test drug, the surviving RGCs were counted. Each value represents mean6S.E., n56. *P,0.05 versus No treatment, § P,0.05 versus Glutamate alone (Student’s t-test). [ P,0.05 versus Glutamate alone (Dunnett’s test).

261

thickness of the IPL. At 100 mg / kg, topiramate markedly reduced the retinal damage seen after ischemia (Fig. 3C). Morphometric results for the aforementioned changes are shown in Fig. 4. In normal eyes, about 100 cells were counted in each retina to evaluate the cell density in the GCL. As indicated by the morphologic analysis, in the ischemic retina of vehicle-treated animals, the cell density in the GCL was lower than normal and the IPL was thinner than normal at 7 days after the ischemia. Injection of MK-801, a non-competitive N-methyl-D-aspartate (NMDA)-receptor antagonist, prior to the ischemia had a significant neuroprotective effect against both forms of retinal damage. Treatment with topiramate at doses of 25 to 200 mg / kg, i.p. reduced the retinal damage in a dosedependent manner, the effects being significant at 100 mg / kg or more (cell numbers in GCL), or at 50 mg / kg or more (IPL thickness). Three days after the period of ischemia, the electrical function of the retina was assessed by recording the ERG. Alterations in the a- and b-waves of the electric response were analyzed as shown in Fig. 5. The amplitudes of the aand b-waves of the ERG were markedly reduced (Fig. 5B). Injection of topiramate at 100 mg / kg diminished the ischemia-induced decrease in the amplitude of the b-wave. (Fig. 5C). A summary of the ERG results is shown in Fig. 6. The amplitudes of the a- and b-waves were approximately 50 and 25%, respectively, of the normal values 3 days after the start of reperfusion. In the groups treated with 50 or 100 mg / kg topiramate, although at 50 mg / kg topiramate only tended to diminish the decrease in a- and b-wave amplitude, the improvement was statistically significant (a-wave; P50.042, b-wave; P50.013) in the group treated with 100 mg / kg (Fig. 6).

4. Discussion This study showed that topiramate protected against excitotoxin-induced retinal-cell death in vitro and against high-intraocular pressure-induced ischemic damage in the retina in vivo. Topiramate has several pharmacological actions via which it may exert neuroprotective effects against ischemic neuronal damage. Although the exact mechanism(s) is unclear, multiple properties may contribute to these neuroprotective effects. Firstly, topiramate inhibits the AMPA / kainite subtype of glutamate receptor without significantly affecting the NMDA receptor [13], and AMPA / kainate-receptor antagonists have been shown to prevent ischemia-induced retinal damage by some investigators [25,30]. Secondly, topiramate inhibits the voltage-activated sodium channel [31]. Blockade of this channel may reduce the depolarization triggered by ischemia, which will lead to an inhibition of synaptic glutamate release. Interestingly, in spontaneously epileptic rats, topiramate reduced the abnormally high extracellular

262

S. Yoneda et al. / Brain Research 967 (2003) 257–266

Fig. 3. Light micrographs of transverse sections of rat retinas 7 days after a 45-min period of ischemia. (A) Normal retina; (B) Ischemic retina from vehicle-treated animal; (C) Ischemic retina from topiramate (100 mg / kg, i.p.)-treated animal. Topiramate was injected at two time points: 2 h before and 5 min after the ischemia. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer. Scale bar550 mm.

levels of glutamate and aspartate [17], and riluzole, which exerts blocking action towards sodium channels, acts as a neuroprotective drug against high-intraocular pressure-induced retinal ischemia [12]. Thirdly, topiramate seems to exert a pharmacological effect by potentiating GABA responses [48]. This finding is supported by studies [22,35] showing that it increases the brain GABA content in healthy humans and in epilepsy patients. A potentiation of a GABAergic influence may tend to counterbalance the toxic effects of glutamate seen following ischemia. Further studies will be needed to identify the exact mechanism(s) underlying the neuroprotective effects of topiramate in the retina. Nevertheless, our study is the first to show that neuroprotection can be achieved with topiramate in a retinal-injury model. In the present study, we used two different primary cultures, a rat retinal-neuron culture containing several types of retinal neurons and a purified RGC culture. Retinal-neuron cultures are commonly used in the evaluation of the neuroprotective effects of a variety of compounds [18,19,47,50]. The majority of cells in retinalneuron cultures have been reported to be amacrine cells, and the population of RGCs or photoreceptors was small, which was confirmed using an immunohistochemical method [19]. In this culture, topiramate showed clear evidence of a neuroprotective effect against AMPA-induced neurotoxicity, suggesting that it may exert a negative modulatory influence over AMPA / kainate receptors in amacrine cells. In our previous preliminary study, MK801, but not DNQX, blocked glutamate-induced neuronal death in rat retinal-neuron culture, while in the present study topiramate reduced glutamate-induced toxicity. These results suggest that mechanisms other than blockade

of AMPA / kainate receptors may be associated with the neuroprotective effect of topiramate against glutamateinduced neurotoxicity. RGC death is a common feature of some ophthalmic disorders, especially glaucoma, and is associated with visual-field loss [20]. We therefore evaluated the effect of topiramate on glutamate-induced neurotoxicity using a purified RGC culture. In this culture, a low dose of glutamate can activate AMPA / kainate receptors in RGCs, and this increases intracellular calcium and decreases cell survival [33]. Topiramate prevented glutamateinduced RGC death at concentrations of 1 mM or more. Following oral administration of 100 mg to healthy volunteers, the peak plasma concentration is approximately 5 mM, and the mean half-life of topiramate in humans is between 19 and 23 h regardless of dose [5]. Taken together with our results using rat retinal-neuron culture and purified RGC culture, this suggests that topiramate may be clinically effective in preventing disease states associated with ischemia and / or excitotoxic neuronal death in the retina. In our in vivo experiment, we examined the neuroprotective effects of topiramate in the rat retina using a model of high-intraocular pressure-induced retinal ischemia and reperfusion. This is both the simplest and the most frequently used model for the investigation of the mechanisms underlying retinal ischemia and for the testing of possible therapies. In this model, histological evaluation showed that the ischemia-induced damage occurs mainly in the inner retinal layer, that is the GCL and the IPL. After such an ischemic insult, in addition to the histological changes in the inner retinal layers, there were reductions in amplitude in both the a- and b-waves of the ERG. The a-wave reflects the activity of photoreceptor cells,

S. Yoneda et al. / Brain Research 967 (2003) 257–266

Fig. 4. Effects of topiramate and MK-801 on ischemia-induced retinal injury in rats. Topiramate (25–200 mg / kg), MK-801 (10 mg / kg), or vehicle (saline) was injected intraperitoneally. Seven days later, the eyes were enucleated. Cell density in the ganglion cell layer (GCL) and the thickness of the inner plexiform layer (IPL) were measured. Each value represents mean6S.E., n512. *P,0.05 versus normal; § P,0.05 versus vehicle (Student’s t-test); [ P,0.05 versus vehicle (Dunnett’s test).

¨ while the b-wave directly reflects the activity of Muller cells and indirectly reflects the activity of neurons in the inner nuclear layer. Therefore, retinal ischemia also induced functional damage in the outer retinal layers. For this reason, we evaluated the neuroprotective effect of topiramate by assessing both histological and ERG

263

changes in our in vivo experiments. Intraperitoneal injection of topiramate was found to exert significant neuroprotective effects at doses of 50 mg / kg or more (histological evaluation) and at 100 mg / kg (functional evaluation by measurement of the ERG). These results indicate that topiramate acts as a neuroprotectant in both the inner and outer retinal layers after an ischemic insult. The dose-range over which topiramate was protective in this study is similar to those reported for its anti-epileptic effects on amygdaloid kindling in rats [3] and for its neuroprotective effects on focal ischemia in rats [49] and on global ischemia in gerbils [23]. It is well known that NMDA antagonists are neuroprotective after an ischemic or excitotoxic insult in retinal cells in vitro [34,43]and in vivo [6,41]. Pharmacologically, the development of clinically useful NMDA antagonists has been difficult, because many of these antagonists have significant undesirable side effects on the central nervous system (such as neuronal vacuolization and ataxia) [4]. Unlike these drugs, topiramate seems to produce its effects via actions that do not involve the NMDA receptor, as described above. Moreover, topiramate has been used clinically in more than 60 countries for several years, and has been shown to have a good pharmacokinetic profile and tolerability. In this study, although no specific test was carried out, no behavioral disturbance was noted after the injection of topiramate. On the other hand, with MK-801 severe adverse effects such as ataxia were observed in all rats. An elevation in intraocular pressure is thought to be a major risk factor for glaucomatous optic neuropathy, on the basis of mechanical or vascular theory [45], and current therapies for glaucoma are aimed at lowering the intraocular pressure. The medical treatment of glaucoma has consisted of the administration of a variety of ocular hypotensive agents such as prostaglandins, b-blockers, adrenergic agents, miotics, and carbonic anhydrase inhibitors [16]. In addition to the pharmacological actions of topiramate that seem to contribute to the neuroprotective effects of this drug (that is, modulatory actions towards the AMPA / kainate receptor, GABA receptor, sodium channels and calcium channels), topiramate has been shown to act as an inhibitor of some carbonic anhydrase isozymes [11]. Oral and topical carbonic anhydrase inhibitors have been used extensively in the treatment of all types of glaucoma [14]. These agents reduce intraocular pressure by suppressing the rate of aqueous humor formation. Therefore, topiramate has the potential to induce such an ocular hypotensive effect in clinical use. Topiramate may prove to be an ideal anti-glaucoma drug if it really does have these dual effects, a direct neuroprotective effect in the retina and an ocular hypotensive effect. In conclusion, topiramate inhibited neuronal cell death in two different retinal primary cultures, and also prevented the ischemia-induced neuronal cell damage and functional damage in the rat eye in vivo. Thus, topiramate

264

S. Yoneda et al. / Brain Research 967 (2003) 257–266

Fig. 5. Representative ERG waveforms obtained 3 days after a 45-min period of ischemia. (A) Trace from normal eye; (B) trace from ischemic retina of vehicle-treated animal; (C) trace from ischemic retina of topiramate (100 mg / kg, i.p.)-treated animal. Topiramate was injected at two time points: 2 h before and 5 min after the ischemia.

shows promise for the treatment of a variety of ischemic or traumatic retinopathies, and possibly glaucoma.

Acknowledgements The authors thank the R.W. Johnson Pharmaceutical Research Institute for providing topiramate and for helpful advice, and Dr Masamitsu Shimazawa, Izumi Furutani, Rie Yamamoto and Rumi Ishida for their technical assistance and advice.

References

Fig. 6. Effects of topiramate on ischemia-induced diminution in a- and b-waves of ERG. Topiramate (50 or 100 mg / kg) or vehicle (saline) was injected intraperitoneally at two time points: 2 h before and 5 min after the ischemia. Three days later, the amplitudes of the a- and b-waves were measured. Each value represents mean6S.E., n511 or 12. *P,0.05 versus normal; [ P,0.05 versus vehicle (Dunnett’s test).

[1] K. Adachi, Y. Fujita, C. Morizane, A. Akaike, M. Ueda, M. Satoh, H. Masai, S. Kashii, Y. Honda, Inhibition of NMDA receptors and nitric oxide synthase reduces ischemic injury of the retina, Eur. J. Pharmacol. 350 (1998) 53–57. [2] E. Aizenman, M.P. Frosch, S.A. Lipton, Responses mediated by excitatory amino acid receptors in solitary retinal ganglion cells from rat, J. Physiol. (Lond.) 396 (1988) 75–91. [3] K. Amano, K. Hamada, K. Yagi, M. Seino, Anti-epileptic effects of topiramate on amygdaloid kindling in rats, Epilepsy Res. 31 (1998) 123–128. [4] R.N. Auer, Assessing structural changes in the brain to evaluate neurotoxicological effects of NMDA receptor antagonists, Psychopharmacol. Bull. 30 (1994) 585–591. [5] M. Bialer, Comparative pharmacokinetics of the newer anti-epileptic drugs, Clin. Pharmacokinet. 24 (1993) 441–452. [6] F. Block, G. Pergande, M. Schwarz, Flupirtine protects against ischaemic retinal dysfunction in rats, Neuroreport 5 (1994) 2630– 2632. [7] S.A. Bloomfield, J.E. Dowling, Roles of aspartate and glutamate in synaptic transmission in rabbit retina, I: outer plexiform layer, J. Neurophysiol. 53 (1985) 699–713. [8] S.A. Bloomfield, J.E. Dowling, Roles of aspartate and glutamate in synaptic transmission in rabbit retina, II: inner plexiform layer, J. Neurophysiol. 53 (1985) 714–725. [9] S.G. Carriedo, H.Z. Yin, S.L. Sensi, J.H. Weiss, Rapid Ca 21 entry through Ca 21 -permeable AMPA / kainate channels triggers marked intracellular Ca 21 rises and consequent oxygen radical production, J. Neurosci. 18 (1998) 7727–7738.

S. Yoneda et al. / Brain Research 967 (2003) 257–266 [10] D.W. Choi, G.M. Maulucci, A.R. Kreigstein, Glutamate neurotoxicity in cortical cell culture, J. Neurosci. 7 (1987) 357–368. [11] S.J. Dodgson, R.P. Shank, B.E. Maryanoff, Topiramate as an inhibitor of carbonic anhydrase isoenzymes, Epilepsia 41 (2000) S35–39. [12] M. Ettaiche, K. Fillacier, C. Widmann, C. Heurteaux, M. Lazdunski, Riluzole improves functional recovery after ischemia in the rat retina, Invest. Ophthalmol. Vis. Sci. 40 (1999) 729–736. [13] J.W. Gibbs 3rd, S. Sombati, R.J. DeLorenzo, D.A. Coulter, Cellular actions of topiramate: blockade of kainate-evoked inward currents in cultured hippocampal neurons, Epilepsia 41 (2000) S10–S16. [14] J.A. Hiett, C.A. Dockter, Topical carbonic anhydrase inhibitors: a new perspective in glaucoma therapy, Optom. Clin. 2 (1992) 97– 112. [15] T. Hirayama, H. Ono, H. Fukuda, Effects of excitatory and inhibitory amino acid agonists and antagonists of ventral horn cells in slices of spinal cord isolated from adult rats, Neuropharmacology 29 (1990) 1117–1122. [16] P.F. Hoyng, L.M. van Beek, Pharmacological therapy for glaucoma: a review, Drugs 59 (2000) 411–434. [17] T. Kanda, M. Kurokawa, S. Tamura, J. Nakamura, A. Ishii, Y. Kuwana, T. Serikawa, J. Yamada, K. Ishihara, M. Sasa, Topiramate reduces abnormally high extracellular levels of glutamate and aspartate in the hippocampus of spontaneously epileptic rats (SER), Life Sci. 59 (1996) 1607–1616. [18] S. Kashii, M. Mandai, M. Kikuchi, Y. Honda, Y. Tamura, K. Kaneda, A. Akaike, Dual actions of nitric oxide in N-methyl-D-aspartate receptor-mediated neurotoxicity in cultured retinal neurons, Brain Res. 711 (1996) 93–101. [19] S. Kashii, M. Takahashi, M. Mandai, H. Shimizu, Y. Honda, M. Sasa, H. Ujihara, Y. Tamura, T. Yokota, A. Akaike, Protective action of dopamine against glutamate neurotoxicity in the retina, Invest. Ophthalmol. Vis. Sci. 35 (1994) 685–695. [20] L.A. Kerrigan-Baumrind, H.A. Quigley, M.E. Pease, D.F. Kerrigan, R.S. Mitchell, Number of ganglion cells in glaucoma eyes compared with threshold visual field tests in the same persons, Invest. Ophthalmol. Vis. Sci. 41 (2000) 741–748. [21] M. Kikuchi, S. Kashii, Y. Honda, H. Ujihara, M. Sasa, Y. Tamura, A. Akaike, Protective action of zinc against glutamate neurotoxicity in cultured retinal neurons, Invest. Ophthalmol. Vis. Sci. 36 (1995) 2048–2053. [22] R. Kuzniecky, H. Hetherington, S. Ho, J. Pan, R. Martin, F. Gilliam, J. Hugg, E. Faught, Topiramate increases cerebral GABA in healthy humans, Neurology 51 (1998) 627–629. [23] S.R. Lee, S.P. Kim, J.E. Kim, Protective effect of topiramate against hippocampal neuronal damage after global ischemia in the gerbils, Neurosci. Lett. 281 (2000) 183–186. [24] L.A. Levin, J.A. Clark, L.K. Johns, Effect of lipid peroxidation inhibition on retinal ganglion cell death, Invest. Ophthalmol. Vis. Sci. 37 (1996) 2744–2749. [25] G. Lombardi, F. Moroni, F. Moroni, Glutamate receptor antagonists protect against ischemia-induced retinal damage, Eur. J. Pharmacol. 271 (1994) 489–495. [26] P. Louzada Jr, J.J. Dias, W.F. Santos, J.J. Lachat, H.F. Bradford, J. Coutinho-Netto, Glutamate release in experimental ischaemia of the retina: an approach using microdialysis, J. Neurochem. 59 (1992) 358–363. [27] D.R. Lucas, J.P. Newhouse, The toxic effect of sodium L-glutamate on the inner layers of the retina, Am. Med. Arch. Ophthalmol. 58 (1957) 193–201. [28] J.E. Markind, Topiramate: a new anti-epileptic drug, Am. J. Health Syst. Pharm. 55 (1998) 554–562. [29] B.E. Maryanoff, M.J. Costanzo, S.O. Nortey, M.N. Greco, R.P. Shank, J.J. Schupsky, M.P. Ortegon, J.L. Vaught, Structure-activity studies on anticonvulsant sugar sulfamates related to topiramate. Enhanced potency with cyclic sulfate derivatives, J. Med. Chem. 41 (1998) 1315–1343.

265

[30] P. Matini, F. Moroni, G. Lombardi, M.S. Faussone-Pellegrini, F. Moroni, Ultrastructural and biochemical studies on the neuroprotective effects of excitatory amino acid antagonists in the ischemic rat retina, Exp. Neurol. 146 (1997) 419–434. [31] M.J. McLean, A.A. Bukhari, A.W. Wamil, Effects of topiramate on sodium-dependent action-potential firing by mouse spinal cord neurons in cell culture, Epilepsia 41 (2000) S21–S24. [32] M.J. Neal, J.R. Cunningham, P.H. Hutson, J. Hogg, Effects of ischaemia on neurotransmitter release from the isolated retina, J. Neurochem. 62 (1994) 1025–1033. [33] Y. Otori, J.Y. Wei, C.J. Barnstable, Neurotoxic effects of low doses of glutamate on purified rat retinal ganglion cells, Invest. Ophthalmol. Vis. Sci. 39 (1998) 972–981. [34] I.H. Pang, E.M. Wexler, S. Nawy, L. DeSantis, M.A. Kapin, Protection by eliprodil against excitotoxicity in cultured rat retinal ganglion cells, Invest. Ophthalmol. Vis. Sci. 40 (1999) 1170–1176. [35] O.A. Petroff, F. Hyder, D.L. Rothman, R.H. Mattson, Topiramate rapidly raises brain GABA in epilepsy patients, Epilepsia 42 (2001) 543–548. [36] E. Roberts, Gamma-aminobutyric acid and nervous system function—a perspective, Biochem. Pharmacol. 23 (1974) 2637–2649. [37] H. Sakai, Y. Tani, E. Shirasawa, Y. Shirao, K. Kawasaki, Development of electroretinographic alterations in streptozotocin-induced diabetes in rats, Ophthalmic Res. 27 (1995) 57–63. [38] J. Sanchez-Prieto, T.S. Sihra, D.G. Nicholls, Characterization of the exocytotic release of glutamate from guinea-pig cerebral cortical synaptosomes, J. Neurochem. 49 (1987) 58–64. [39] R.P. Shank, J.F. Gardocki, A.J. Streeter, B.E. Maryanoff, An overview of the preclinical aspects of topiramate: pharmacology, pharmacokinetics, and mechanism of action, Epilepsia 41 (2000) S3–9. [40] R.P. Shank, J.F. Gardocki, J.L. Vaught, C.B. Davis, J.J. Schupsky, R.B. Raffa, S.J. Dodgson, S.O. Nortey, B.E. Maryanoff, Topiramate: preclinical evaluation of structurally novel anticonvulsant, Epilepsia 35 (1994) 450–460. [41] R. Siliprandi, R. Canella, G. Carmignoto, N. Schiavo, A. Zanellato, R. Zanoni, G. Vantini, N-methyl-D-aspartate-induced neurotoxicity in the adult rat retina, Vis. Neurosci. 8 (1992) 567–573. [42] V.L. Smith-Swintosky, B. Zhao, R.P. Shank, C.R. Plata-Salaman, Topiramate promotes neurite outgrowth and recovery of function after nerve injury, Neuroreport 12 (2001) 1031–1034. [43] N.J. Sucher, E. Aizenman, S.A. Lipton, N-methyl-D-aspartate antagonists prevent kainate neurotoxicity in rat retinal ganglion cells in vitro, J. Neurosci. 11 (1991) 966–971. [44] N.J. Sucher, S.Z. Lei, S.A. Lipton, Calcium channel antagonists attenuate NMDA receptor-mediated neurotoxicity of retinal ganglion cells in culture, Brain Res. 551 (1991) 297–302. [45] M.F. Sugrue, New approaches to antiglaucoma therapy, J. Med. Chem. 40 (1997) 2793–2809. [46] Y. Tamura, Y. Sato, T. Yokota, A. Akaike, M. Sasa, S. Takaori, Ifenprodil prevents glutamate cytotoxicity via polyamine modulatory sites of N-methyl-D-aspartate receptors in cultured cortical neurons, J. Pharmacol. Exp. Ther. 265 (1993) 1017–1025. [47] N. Toriu, A. Akaike, H. Yasuyoshi, S. Zhang, S. Kashii, Y. Honda, M. Shimazawa, H. Hara, Lomerizine, a Ca 21 channel blocker, reduces glutamate-induced neurotoxicity and ischemia / reperfusion damage in rat retina, Exp. Eye Res. 70 (2000) 475–484. [48] H.S. White, S.D. Brown, J.H. Woodhead, G.A. Skeen, H.H. Wolf, Topiramate modulates GABA-evoked currents in murine cortical neurons by a nonbenzodiazepine mechanism, Epilepsia 41 (2000) S17–S20. [49] Y. Yang, A. Shuaib, Q. Li, M.M. Siddiqui, Neuroprotection by delayed administration of topiramate in a rat model of middle cerebral artery embolization, Brain Res. 804 (1998) 169–176. [50] H. Yasuyoshi, S. Kashii, S. Zhang, A. Nishida, T. Yamauchi, Y. Asano, S. Sato, A. Akaike, Protective effect of bradykinin against glutamate neurotoxicity in cultured rat retinal neurons, Invest. Ophthalmol. Vis. Sci. 41 (2000) 2273–2278.

266

S. Yoneda et al. / Brain Research 967 (2003) 257–266

[51] S. Yoneda, H. Tanihara, N. Kido, Y. Honda, W. Goto, H. Hara, N. Miyawaki, Interleukin-1b mediates ischemic injury in the rat retina, Exp. Eye Res. 73 (2001) 661–667. [52] G.D. Zeevalk, A.G. Hyndman, W.J. Nicklas, Excitatory amino acidinduced toxicity in chick retina: amino acid release, histology, and

effects of chloride channel blockers, J. Neurochem. 53 (1989) 1610–1619. [53] X. Zhang, A.A. Velumian, O.T. Jones, P.L. Carlen, Modulation of high-voltage-activated calcium channels in dentate granule cells by topiramate, Epilepsia 41 (2000) S52–S60.