Histological Protection by Donepezil Against Neurodegeneration Induced by Ischemia–Reperfusion in the Rat Retina

Home , Donepezil

J Pharmacol Sci 112, 327 – 335 (2010) Journal of Pharmacological Sciences ©2010 The Japanese Pharmacological Society Full Paper Histological Protection by Donepezil Against Neurodegeneration Induced by Ischemia–Reperfusion in the Rat Retina

Kenji Sakamoto1, *, Kayo Ohki 1 , Maki Saito1 , Tsutomu Nakahara1 , and Kunio Ishii1 1 Department of Molecular Pharmacology, Kitasato University School of Pharmaceutical Sciences, 9-1 Shirokane 5-chome, Minato-ku, Tokyo 108-8641, Japan

Received November 2, 2009; Accepted January 19, 2010

Abstract. Although a blockade of acetylcholine esterase has been reported to suppress neuronal cell death induced by exogenous glutamate and β -amyloid, information is still limited regarding the neuroprotective effects of the acetylcholine esterase inhibitor donepezil. We histologically examined the effects of donepezil on neuronal injury induced by ischemia–reperfusion. Intrave- nous and intravitreous treatment with donepezil 15 min prior to ischemia dramatically reduced the retinal damage. The protective effect of donepezil in the ganglion cell layer was not affected by mecamylamine, a nicotinic acetylcholine-receptor antagonist, nor scopolamine, a muscarinic acetylcholine-receptor antagonist. The protective effect of donepezil in the inner plexiform layer was reduced not by mecamylamine, but by scopolamine. Neostigmine, a choline-esterase inhibitor, and pilocarpine, a muscarinic acetylcholine-receptor agonist, have protective effects in the inner plexiform layer and the inner nuclear layer. These results suggest that not only the activation of acetylcholine receptors but also a mechanism unrelated to acetylcholine-esterase inhibition contribute to the protective effect of donepezil on the ganglion cells in the ischemic–reperfused rat retina. Done- pezil may be useful as a therapeutic drug against retinal diseases that cause neuronal cell death such as glaucoma with high intraocular pressure.

Keywords : donepezil, retina, ischemia, reperfusion, acetylcholine esterase

Introduction However, stimulation of some of the glutamate receptors by an excess amount of glutamate under pathologic Cell death of the retinal ganglion cells and the inner conditions such as hypoxia (6) and ischemia–reperfusion retinal neurons is a characteristic of retinal ischemia– (1) is toxic to neuronal cells. The activation of the N - reperfusion injury. It is believed that ischemia-induced methyl-D-aspartate (NMDA) receptor, a subtype of the injury plays an important role in some retinal diseases glutamate receptor (7, 8), followed by excess Ca2+ influx such as glaucoma, including open angle glaucoma, closed via NMDA receptor–operated channels is thought to be angle glaucoma, and normal-tension glaucoma, and involved in the predominant mechanism of neuronal ex- central retinal vessel occlusions. The mechanism of the citotoxicity. In fact, excitotoxicity caused by the eleva- cell death induced by retinal ischemia is not completely tion of glutamate concentration in the retinal extracellular understood and remains a very interesting area to ex- space near the glutamate-receptor channels is thought to plore. Endogenous substances such as glutamate, oxygen be one of the mechanisms of neuronal cell death induced free radicals, nitric oxide (NO), and calcium are among by glaucoma (9) and MK-801, an NMDA receptor– the pathological causes of ischemia–reperfusion injury channel blocker, prevents retinal damage induced by reti- (1 – 3). nal ischemia–reperfusion (10, 11). Glutamate is the principal excitatory neurotransmitter Donepezil is an acetylcholine esterase inhibitor widely in the central nervous system, including the retina (4, 5). used for treatment of moderate Alzheimer’s disease and known to improve the cognitive performance and global *Corresponding author. [email protected] function in the affected patients (12). Recently, donepezil Published online in J-STAGE on March 2, 2010 (in advance) has been reported to have neuroprotective effects against doi: 10.1254/jphs.09302FP glutamate neurotoxicity, oxygen-glucose deprivation

327 328 K Sakamoto et al

(13), and β -amyloid (14). Both Alzheimer’s disease and 25-μ L gas-tight microsyringe (1702LT; Hamilton, Reno, glaucoma are chronic neurodegenerative diseases. Recent NV, USA), which was inserted approximately 1-mm reports suggest that there are some similarities between behind the corneal limbus. A 5-μ l aliquot of drug solu- Alzheimer’s disease and glaucoma, including activation tion, described below, was administered into both eyes. of caspase and abnormal processing of amyloid precursor All drugs were dissolved with saline and administered 15 protein (15). Bayer et al. (16) reported that patients with min before 60 min of retinal ischemia. When antagonists Alzheimer’s disease and Parkinson’s disease had a larger such as mecamylamine and scopolamine were used, we risk of glaucoma. Recently, donepezil was reported to made a mixed solution with donepezil. have a protective effect against retinal ganglion cell death induced by glutamate and axotomy (17). Histological evaluation In the present study, to clarify the potential utility of Histological evaluation was performed as previously the drug for central retinal vessel occlusion and glaucoma, described (18 – 21). Animals were euthanized by over- we assessed effects of donepezil on retinal ischemia– dosage of pentobarbital sodium 7 days after 60 min of reperfusion injury in rats by comparing them with those ischemia and both eyes were enucleated. Enucleated eyes of neostigmine, a choline-esterase inhibitor. In addition, were fixed with Davidson solution, comprised of 37.5% to clarify the mechanism of the action of donepezil, the ethanol, 9.3% paraformaldehyde, 12.5% acetic acid, and effects of inhibitors of acetylcholine receptors on the 3% glutaraldehyde for 1 – 12 h at room temperature. The protective effects of donepezil and the effect of acetyl- fixed eye was bisected through the optic nerve head in choline-receptor agonists on neurodegeneration induced the vertical meridian with a microtome blade (Histo Cut- by retinal ischemia–reperfusion were also tested. ter Super #35 Type; Micro Glass, Tokyo) and embedded in paraffin after removing a lens. Five-micron horizontal Materials and Methods sections through the optic nerve head of the eye were cut along the vertical meridian of the eye so as to contain the Animals entire retina from the ora serrata in the superior hemi- In the present study, experimental procedures con- sphere to the ora serrata in the inferior hemisphere using formed to the Guiding Principles for the Care and Use of a microtome (HM325; Microm International, Walldorf, Laboratory Animals, approved by The Japanese Pharma- Germany) and a microtome blade (Histo Cutter Super cological Society. Male Sprague-Dawley rats weighing #35 Type, Micro Glass). The sections were stained with 230 – 300 g (Charles River Japan, Kanagawa) were hematoxylin and eosin and subject to morphometry. The anesthetized with pentobarbital sodium (50 mg/kg i.p., sections showed oblique regions were excluded to avoid Nembutal® injection; Abbott Laboratories, North Chi- artifacts. The total number of the cells in the retinal cago, IL, USA), and the rectal temperature of animals ganglion cell layer (GCL) was counted for a length of 1 was maintained at 37°C during the experiment using a mm on either side of the optic nerve head beginning ap- heating pad and heating lamp. The femoral veins were proximately 1 mm from the center of the optic nerve cannulated for administration of the drug, if necessary. head in four independent sections under the direct obser- vation of the HE-stained specimens using a light micro- Induction of retinal ischemia scope (Optiphot-2; Nikon, Tokyo) and a counting ma- Induction of retinal ischemia was performed as previ- chine. No attempt was made to distinguish the cell types ously described (18 – 21). Briefly, the anterior chamber in the GCL, and displaced amacrine cells were not ex- of the one eye was cannulated with a 27-gauge needle cluded from the counts. Measurement of the thickness of connected to a bottle filled with saline. Retinal ischemia the inner plexiform layer (IPL), the inner nuclear layer was induced by raising intraocular pressure to 130 mmHg (INL), and the outer nuclear layer (ONL) was also per- by lifting the bottle for 60 min. The opposite eye of each formed to quantify the degree of cell loss induced by reti- animal served as a non-ischemic control. nal ischemia–reperfusion. Digital photographs with approximately 0.25-mm width of the retinal layers in each Drug injection section at a distance of approximately 1 mm from the Donepezil was administered intravenously and intra- center of the optic nerve head were taken using a digital vitreously. Intravenous injection was performed through camera (DP11; Olympus, Tokyo) connected to a light a cannula indwelling in the femoral vein. Intravitreous microscope. The photographs were printed onto A4 pa- injection was performed as previously described (10, pers. Lines indicating the inner and outer borders of INL 21). Briefly, rats were anesthetized by intraperitoneal and ONL and the bottom of the ganglion cells were injection of sodium pentobarbital (50 mg/kg). Injection drawn on the printed photographs. To know the thick- was performed with a 33-gauge needle connected to a nesses of IPL, INL, and ONL, the distance between the Protection by Donepezil in Rat Retina 329 lines on the paper were measured. Averages for these for multiple comparisons. Differences were considered measurements of thickness taken in five adjacent areas to be statistically significant when the P -values were less (approximately 40-μ m intervals) were calculated. These than 0.05. parameters of each eye subjected to ischemia were normalized with those of the corresponding intact opposite Results eyes and are presented as percentages. We did all of the morphometrical analyses in a blind fashion. Intravenous treatment with donepezil in the retinal ischemia–reperfusion model Statistical analyses First, we determined the effects of donepezil on the The data represent the means ± S.E.M. One way analy- retinal injury induced by ischemia–reperfusion. Typical sis of variance followed by Tukey-Kramer test was used photomicrographs of the retina taken 7 days after reperfusion are shown in Fig. 1. In the vehicle-treated group, degenerative changes were observed in GCL, IPL, and INL of the ischemic eye, a characteristic of retinal ischemic atrophy (Fig. 1B), but such changes were not seen in the contralateral non-ischemic retina (Fig. 1A). Mor- phometric results at 7 days after reperfusion of 3 – 9 independent experiments are shown in Fig. 2. As indicated by the morphologic analysis, the IPL and INL in the ischemic–reperfused eye of the vehicle-treated group (n = 6) were thinner than in the contralateral normal eye at 7 days after reperfusion. No significant change was seen in the thickness of ONL after ischemia–reperfusion in all of the groups. Treatment with 3 mg/kg (n = 9) (Fig. 1F), but not 1 mg/kg (n = 3) (Fig. 1D), donepezil 15 min before 60 min of ischemia significantly reduced the amount of retinal damage. Intravenous treatment with donepezil itself did not affect the retinal morphology (Fig. 1: A, C, and E).

Fig. 2. Effect of intravenous treatment with donepezil on the histo- Fig. 1. Effect of intravenous treatment with donepezil on the histological damage induced by retinal ischemia–reperfusion. Retinal logical damage induced by retinal ischemia–reperfusion. Representa- damage was examined 7 days after 60 min of ischemia. The following tive photomicrographs showing histological appearance of the non- 5 parameters of the ischemic eyes were normalized to those of the ischemic control (A, C, and E), and ischemic retinae 7 days after 60 intact eyes (opposite side of the ischemic eye) and are presented as min of ischemia (B, D, and F). Donepezil administered i.v. 15 min percentages: cell density in the GCL (ganglion cell layer) and thick- before 60 min of ischemia. Severe retinal damage is shown in the ve- nesses of the IPL (inner plexiform layer), INL (inner nuclear layer), hicle-treated (B) and 1 mg/kg donepezil–treated (D) retinae. In the 3 OPL (outer plexiform layer), and ONL (outer nuclear layer). The data mg/kg donepezil–treated group, retinal damage is reduced (F). Scale represent the means ± S.E.M. *P < 0.05 vs. the vehicle-treated bar = 50 μ m. Original magnification is ×200. group. 330 K Sakamoto et al

Intravitreous treatment with donepezil and neostigmine in the retinal ischemia–reperfusion model To verify whether acetylcholine esterase inhibition is involved in the protective effect of donepezil described above, we determined the effect of neostigmine, an acetylcholine esterase inhibitor on the retinal injury induced by ischemia–reperfusion. Because neostigmine is a qua- ternary ammonium compound and never passes through the retinal blood barrier, intravitreous injection was performed. Typical photomicrographs of the retina taken 7 days after reperfusion are shown in Fig. 3. In the vehicle- treated group, degenerative changes were observed in GCL, IPL, and INL of the ischemic eye, a characteristic of retinal ischemic atrophy (Fig. 3B), but such changes were not seen in the contralateral non-ischemic retina (Fig. 3A). No significant change was seen in the thickness of ONL after ischemia–reperfusion in all of the groups. Intravitreous treatment with 50 pmol/eye (Fig. 3F), but not 5 pmol/eye (Fig. 3D), donepezil 15 min before 60 min of ischemia markedly reduced the retinal degeneration. However, intravitreous treatment with 500 pmol/eye (Fig. 3J), but not 50 pmol/eye (Fig. 3H), neostigmine reduced the retinal damages only in IPL and INL. Intravitreous treatment with donepezil and neostigmine itself did not affect the retinal morphology (Fig. 3: A, C, and E). Morphometric results at seven days after reperfusion of five independent experiments are shown in Fig. 4.

Fig. 3. Effect of intravitreous treatment with donepezil and neostigmine on the histological damage induced by retinal ischemia–reperfusion. Representative photomicrographs showing histological appear- Fig. 4. Effect of intravitreous treatment with donepezil and neostig- ance of the non-ischemic control (A, C, E, G, and I) and ischemic mine on the histological damage induced by retinal ischemia–reperfu- retinae 7 days after 60 min of ischemia (B, D, F, H, and J). Donepezil sion. Retinal damage was examined 7 days after 60 min of ischemia. and neostigmine were injected intravitreously. Severe retinal damage The following 5 parameters of the ischemic eyes were normalized to is shown in the vehicle-treated (B) retinae. In the 50 pmol/eye done- those of the intact eyes (opposite side of the ischemic eye) and are pezil–treated group (F), retinal damage is reduced. Neostigmine (500 presented as percentages: cell density in the GCL (ganglion cell layer) pmol/eye) had protective effects in the IPL (inner plexiform layer) and thicknesses of the IPL (inner plexiform layer), INL (inner nuclear and the INL (inner nuclear layer) (J). Scale bar = 50 μ m. Original layer), OPL (outer plexiform layer), and ONL (outer nuclear layer). magnification is ×200. The data represent the means ± S.E.M. *P < 0.05 vs. the vehicle group. Protection by Donepezil in Rat Retina 331

Effects of acetylcholine receptor antagonists on the nicotinic acetylcholine–receptor antagonist, and scopol- protective effect of donepezil in the retinal ischemia– amine, a muscarinic acetylcholine–receptor antagonist, reperfusion model on the retinal protection by donepezil against ischemia– To verify whether acetylcholine receptor stimulation reperfusion injury. Typical photomicrographs of the retina is involved in the protective effect of donepezil described taken 7 days after reperfusion are shown in Fig. 5. Intra- above, we determined the effects of mecamylamine, a

Fig. 6. Effect of mecamylamine on the protective effects of donepezil in the ischemic–reperfused rat retina. Retinal damage was examined 7 days after 60 min of ischemia. The following 5 parameters of the ischemic eyes were normalized to those of the intact eyes (opposite side of the ischemic eye) and are presented as percentages: cell density in the GCL (ganglion cell layer) and thicknesses of the IPL (inner plexiform layer), INL (inner nuclear layer), OPL (outer plexiform layer), and ONL (outer nuclear layer). Severe retinal damage is shown in the vehicle-treated retinae. In the 50 pmol/eye donepezil– treated group, retinal damage is reduced. Mecamylamine (50 pmol/ eye) did not affect the protective effects of donepezil. The data represent the means ± S.E.M. *P < 0.05 vs. the vehicle group.

Fig. 7. Effect of scopolamine on the protective effects of donepezil in the ischemic–reperfused rat retina. Retinal damage was examined 7 days after 60 min of ischemia. The following 5 parameters of the ischemic eyes were normalized to those of the intact eyes (opposite Fig. 5. Effects of mecamylamine (50 pmol/eye) and scopolamine side of the ischemic eye) and are presented as percentages: cell density (15 pmol/eye) on the protective effects of donepezil in the ischemic– in the GCL (ganglion cell layer) and thicknesses of the IPL (inner reperfused rat retina. Representative photomicrographs showing his- plexiform layer), INL (inner nuclear layer), OPL (outer plexiform tological appearance of the non-ischemic control (A, C, E, and G) and layer), and ONL (outer nuclear layer). Severe retinal damage is shown ischemic retinae 7 days after 60 min of ischemia (B, D, F, and H). in the vehicle-treated retinae. In the 50 pmol/eye donepezil–treated Donepezil (C, D, E, F, G, and H), mecamylamine (E and F), and sco- group, retinal damage is reduced. Scopolamine (15 pmol/eye) reduced polamine (G and H) were injected intravitreously. Scale bar = 50 μ m. the protective effect in the IPL. The data represent the means ± S.E.M. Original magnification is ×200. * P < 0.05 vs. the vehicle group. 332 K Sakamoto et al vitreous treatment with mecamylamine (50 and 500 micrographs of the retina taken 7 days after reperfusion pmol/eye) did not affect the protective effect of donepezil are shown in Fig. 8. Intravitreous treatment with nicotine (Fig. 5B). Intravitreous treatment with scopolamine (50 (50 pmol/eye) reduces the retinal damages in INL (Fig. pmol/eye) reduced the retinal protection by donepezil 8B). Intravitreous treatment with pilocarpine (50 pmol/ only in IPL (Fig. 5C). Intravitreous treatment with eye) reduces the retinal damages only in IPL and INL mecamylamine and scopolamine itself did not affect the (Fig. 8C). Intravitreous treatment with nicotine and retinal morphology. Morphometric results at seven days pilocarpine itself did not affect the retinal morphology. after reperfusion of five independent experiments are Morphometric results at seven days after reperfusion of shown in Figs. 6 and 7. five independent experiments are shown in Fig. 9.

Intravitreous treatment with acetylcholine-receptor ago- Dicussion nists in the retinal ischemia–reperfusion model To verify whether acetylcholine-receptor stimulation In the present study, we demonstrated that donepezil, protects against retinal ischemia–reperfusion injury, we an acetylcholine-esterase inhibitor, has a neuroprotective determined the effects of nicotine and pilocarpine, a effect in the ischemic–reperfused rat retina for the first muscarinic acetylcholine–receptor agonist, on the retinal time. Donepezil was reported to increase acetylcholine injury induced by ischemia–reperfusion. Typical photo- content in the central nervous system (22). Acetylcholine activates two different types of acetylcholine receptors: nicotinic acetylcholine receptor and muscarinic acetylcholine receptor. Therefore, it is possible that donepezil protects retinal neurons against ischemia–reperfusion injury via activation of nicotinic acetylcholine receptors and/or muscarinic acetylcholine receptors. Actually, neuroprotection by activation of nicotinic acetylcholine receptors has been reported in various cultured neurons such as rat cortical neuron cultures (23), mixed retinal neuron cultures (24), and purified guinea-pig (25) and rat

Fig. 9. Effect of intravitreous treatment with nicotine and pilocarpine on the histological damage induced by retinal ischemia–reperfusion. Retinal damage was examined 7 days after 60 min of ischemia. The following 5 parameters of the ischemic eyes were normalized to those of the intact eyes (opposite side of the ischemic eye) and are presented as percentages: cell density in the GCL (gan- Fig. 8. Effect of intravitreous treatment with nicotine (50 pmol/eye) glion cell layer) and thicknesses of the IPL (inner plexiform layer), and pilocarpine (50 pmol/eye) on the histological damage induced by INL (inner nuclear layer), OPL (outer plexiform layer), and ONL retinal ischemia–reperfusion. Representative photomicrographs (outer nuclear layer). Nicotine and pilocarpine have protective effects showing histological appearance of the non-ischemic control (A, C, in the IPL and INL and in the INL of the ischemic–reperfused rat ret- and E) and ischemic retinae 7 days after 60 min of ischemia (B, D, ina, respectively. The data represent the means ± S.E.M. *P < 0.05 and F). Scale bar = 50 μ m. Original magnification is ×200. vs. the vehicle-treated group. Protection by Donepezil in Rat Retina 333

(26) retinal ganglion cell cultures. However, neither In addition, donepezil has been reported to have vari- mecamylamine nor scopolamine blocked the neuropro- ous beneficial effects. First, donepezil was reported to tective effects of donepezil on the retinal ganglion cells reduce neuronal damage induced by β -amyloid not via in the present study, suggesting that activation of nico- NMDA-receptor blockade (14), suggesting that donepe- tinic acetylcholine receptors or muscarinic acetylcholine zil has a neuronal protective effect unrelated to NMDA- receptors is not related to the neuroprotective effects of receptor blockade. Second, donepezil was reported to donepezil in the ganglion cells of the ischemic–reperfused block NMDA receptors (34). The activation of NMDA rat retina. The similar phenomenon was reported in cul- receptor channel is reported to play an important role for tured retinal ganglion cells injured by glutamate (17) and ischemia–reperfusion injury in the rat retina (11). There- the brain infarction in the rat left middle cerebral artery fore, NMDA-receptor blockade may possibly be involved occlusion model (27). In contrast, neostigmine and pilo- in the mechanism of the neuroprotection by donepezil. carpine showed neuroprotection in INL and IPL, and so However, other groups reported that donepezil failed to nicotine did in INL, suggesting that stimulation of mus- protect against glutamate-induced neurotoxicity in pri- carinic receptors leads to neuroprotection in IPL and INL mary cultured rat cerebellar granule neurons (35) and and that stimulation of nicotinic receptors leads to neuro- that donepezil reduced neuronal damage induced by β - protection only in INL in the ischemic–reperfused rat amyloid not via NMDA-receptor blockade (14). It is still retina. This hypothesis is consistent with the report indi- unclear whether NMDA-receptor blockade is actually cating that most of the acetylcholine neurons are reported involved in the protective effect of donepezil or not. to be distributed in the IPL and the INL (28). In fact, Third, donepezil was reported to increase glutathione stimulation of nicotinic receptors (29, 30) and muscarinic level and to reduce malondialdehyde level in the strepto- receptors (31, 32) was reported to lead to beneficial ef- zotocin-induced dementia model rat (36). In addition, fects on neuronal cell death (29, 30), memory defects donepezil protected against cell injury induced by oxy- (31), and hypersensitivity after nerve injury (32). Al- gen-glucose deprivation in the rat pheochromocytoma though neither mecamylamine nor scopolamine blocks cells (13). These reports suggest that donepezil reduces the protective effect of donepezil in INL, this may be the injury induced by oxidative stress. To sum up the caused by insufficient concentration of the drugs in INL, reports described above, donepezil protects retinal neu- that is, the farthest area in the inner retina from the vitre- rons against ischemia–reperfusion injury via activation ous body. According to Dureau et al. (33), the volume of of various pathways, including elevating acetylcholine the vitreous body in a two-month-old rat is approximately levels. However, further experiments are needed to 13 μ L. Therefore, the maximum calculated intravitreous clarify the underlying mechanisms. concentrations of donepezil, neostigmine, mecamyla- Various chemicals such as MK-801, an NMDA-type mine, scopolamine, nicotine, and pilocarpine are 3, 30, glutamate-receptor antagonist, and L-NAME, an NO 30, 3, 3 and 3 μ M, respectively. These concentrations of synthase inhibitor (11), have been reported to protect the the drugs are enough to activate or block their targets. retina against ischemia–reperfusion injury. However, However, it is possible that low penetration of mecam- these chemicals are not in practical use because of vari- ylamine and scopolamine into INL causes the insufficient ous problems, including concern about their safety. block of the protective effect of donepezil in INL. Be- Donepezil is a commercially available medicine for the cause we are afraid that higher concentration of the drugs treatment of moderate Alzheimer’s disease, and it has induces non-specific effects on the retina, especially on been reported to cause very few serious systemic side the ganglion cells, that is the nearest to the vitreous body effects (12, 37). It is reported that orally administered and the most important drug target in the treatment of donepezil was absorbed rapidly and that the mean blood glaucoma, we did not try higher concentrations of the level of donepezil reached a peak at 30 minutes after oral drugs in the present study. Further experiments are administration in the rat (38). The reported rate of ab- needed to clarify the underlying mechanisms more sorption of orally administered donepezil is over 95% in clearly. the rat (38). The reported bioavailability of orally admin- In the present study, donepezil reduced the number of istered donepezil is 56.1% in the rat (39). According to the apoptotic cells in the ischemic–reperfused rat retina. these previous data, not only intravenous but also oral In the cultured rat retinal ganglion cells, HA 14-1, a bcl- donepezil will be effective. In fact, orally administered 2 inhibitor, was reported to reduce neuroprotection by donepezil (12 mg/kg) was reported to reduce infarct size donepezil against glutamate neurotoxicity (17). There- in the ischemic–reperfused rat brain (27). Therefore, fore, it is possible that donepezil activates an anti-apop- donepezil has advantages over other new chemicals pres- totic cellular signal transduction pathway related to bcl- ently in development as medicines for treatment of retinal 2. injury induced by high ocular pressure. 334 K Sakamoto et al

Unfortunately, treatment with donepezil at the end of nas by anoxia and extracellular high potassium. Exp Eye Res. 60 min of ischemia did not protect against the ischemic– 1988;46:657–662. reperfused retinal injury in the pilot study (data not 7 Choi DW. Ionic dependence of glutamate neurotoxicity in corti- shown). In our high-ocular-pressure model, no-flow cal cell culture. J Neurosci. 1987;7:369–379. 8 Choi DW. Calcium-mediated neurotoxicity: relationship to spe- complete ischemia is induced in the eyeball. Therefore, cific channel types and role in ischemic damage. Trends Neurosci. the drug injected after ischemia may reach the retina 1988;11:465–469. behind the critical time point for appearance of the pro- 9 Kuehn MH, Fingert JH, Kwon YH. Retinal ganglion cell death in tective effect. The typical ocular pressure of the glau- glaucoma: mechanisms and neuroprotective strategies. Ophthal- comatous patients with high ocular pressure is from 25 to mol Clin North Am. 2005;18:383–395. 60 mmHg, suggesting that the patients keep some degree 10 Siliprandi R, Canella R, Carmignoto G, Schiavo N, Zanellato A, of blood flow in the retina. A low–blood-flow retinal Zanoni R, et al. N-methyl-D-aspartate-induced neurotoxicity in the adult rat retina. Vis Neurosci. 1992;8:567–573. ischemia model is needed to investigate the protective 11 Adachi K, Fujita Y, Morizane C, Akaike A, Ueda M, Satoh M, et effect of the drug treated during a glaucomatous attack. al. Inhibition of NMDA receptors and nitric oxide synthase re- As such an experimental model has not been reported duces ischemic injury of the retina. Eur J Pharmacol. 1998;350: previously, now we are trying to establish a low–blood- 53–57. flow retinal ischemia model. After we establish this 12 Rogers SL, Friedhoff LT. Long-term efficacy and safety of done- model system, we will try to investigate the protective pezil in the treatment of Alzheimer’s disease: an interim analysis effect of donepezil treatment during the induction of high of the results of a US multicentre open label extension study. Eur ocular pressure. Neuropsychopharmacol. 1998;8:67–75. 13 Zhou J, Fu Y, Tang XC. Huperzine A and donepezil protect rat In conclusion, donepezil, which is an acetylcholine- pheochromocytoma cells against oxygen-glucose deprivation. esterase inhibitor and widely used for treatment of Neurosci Lett. 2001;306:53–56. moderate Alzheimer’s disease, reduced ischemic–reper- 14 Kimura M, Komatsu H, Ogura H, Sawada K. Comparison of fused injury in the rat retina. Stimulation of acetylcholine donepezil and memantine for protective effect against amyloid- receptors is not likely to involved in the mechanism of beta(1-42) toxicity in rat septal neurons. Neurosci Lett. 2005;391: the protective effect of donepezil in retinal ganglion 17–21. cells, whereas its protective effect in IPL and INL may 15 McKinnon SJ, Lehman DM, Kerrigan-Baumrind LA, Merges be partially caused by cholinergic stimulation. Although CA, Pease ME, Kerrigan DF, et al. Caspase activation and amyloid precursor protein cleavage in rat ocular hypertension. Invest further studies are needed, the present study shows the Ophthalmol Vis Sci. 2002;43:1077–1087. possibility that donepezil is effective for preventing 16 Bayer AU, Keller ON, Ferrari F, Maag KP. Association of glau- glaucomatous injury to retinal neurons. coma with neurodegenerative diseases with apoptotic cell death: Alzheimer’s disease and Parkinson’s disease. Am J Ophthalmol. Acknowledgments 2002;133:135–137. 17 Miki A, Otori Y, Morimoto T, Okada M, Tano Y. Protective effect Donepezil hydrochloride was kindly gifted by Eisai Co., Ltd. (To- of donepezil on retinal ganglion cells in vitro and in vivo. Curr kyo). This work is supported by the Japanese Ministry of Education, Eye Res. 2006;31:69–77. Culture, Sports, Science, and Technology, a Grant-in-Aid for Young 18 Sakamoto K, Kuwagata M, Nakahara T, Ishii K. Late precondi- Scientists (B) #21790249 (KS) and a Kitasato University Research tioning in rat retina: involvement of adenosine and ATP-sensitive Grant for Young Researchers (KS). K + channel. Eur J Pharmacol. 2001;418:89–93. 19 Sakamoto K, Yonoki Y, Kuwagata M, Saito M, Nakahara T, Ishii References K. Histological protection against ischemia-reperfusion injury by early ischemic preconditioning in rat retina. Brain Res. 2004; 1 Louzada-Júnior P, Dias JJ, Santos WF, Lachat JJ, Bradford HF, 1015:154–160. Coutinho-Netto J. Glutamate release in experimental ischaemia 20 Sakamoto K, Yonoki Y, Kubota Y, Kuwagata M, Saito M, of the retina: an approach using microdialysis. J Neurochem. Nakahara T, et al. Inducible nitric oxide synthase inhibitors 1992;59:358–363. abolished histological protection by late ischemic precondition- 2 Choi DW. Calcium: still center-stage in hypoxic-ischemic neu- ing in rat retina. Exp Eye Res. 2006;82:512–518. ronal death. Trends Neurosci. 1995;18:58–60. 21 Sakamoto K, Kawakami T, Shimada M, Yamaguchi A, Kuwagata 3 Dreyer EB. A proposed role for excitotoxicity in glaucoma. J M, Saito M, et al. Histological protection by cilnidipine, a dual Glaucoma. 1998;7:62–67. L/N-type Ca2+ channel blocker, against neurotoxicity induced by 4 Aizenman E, Frosch MP, Lipton SA. Responses mediated by ischemia-reperfusion in rat retina. Exp Eye Res. 2009;88:974– excitatory amino acid receptors in solitary retinal ganglion cells 982. from rat. J Physiol. 1988;396:75–91. 22 Giacobini E, Zhu XD, Williams E, Sherman KA. The effect of 5 Yang XL. Characterization of receptors for glutamate and GABA the selective reversible acetylcholinesterase inhibitor E2020 on in retinal neurons. Prog Neurobiol. 2004;73:127–150. extracellular acetylcholine and biogenic amine levels in rat cor- 6 David P, Lusky M, Teichberg VI. Involvement of excitatory tex. Neuropharmacology. 1996;35:205–211. neurotransmitters in the damage produced in chick embryo reti- 23 Hejmadi MV, Dajas-Bailador F, Barns SM, Jones B, Wonnacott Protection by Donepezil in Rat Retina 335

S. Neuroprotection by nicotine against hypoxia-induced apopto- tion by donepezil against glutamate neurotoxicity in rat cortical sis in cortical cultures involves activation of multiple nicotinic neurons. J Pharmacol Exp Ther. 2003; 306:772–777. acetylcholine receptor subtypes. Mol Cell Neurosci. 2003;24: 32 Takada-Takatori Y, Kume T, Ohgi Y, Izumi Y, Niidome T, Fujii T, 779–786. et al. Mechanism of neuroprotection by donepezil pretreatment 24 Yasuyoshi H, Kashii S, Kikuchi M, Zhang S, Honda Y, Akaike A. in rat cortical neurons chronically treated with donepezil. J Neu- New insight into the functional role of acetylcholine in develop- rosci Res. 2008; 86:3575–3583. ing embryonic rat retinal neurons. Invest Ophthalmol Vis Sci. 33 Dureau P, Bonnel S, Menasche M, Dufier JL. Abitbol M. Quanti- 2002;43:446–451. tative analysis of intravitreal injections in the rat. Curr Eye Res. 25 Wehrwein E, Thompson SA, Coulibaly SF, Linn DM, Linn CL. 2001;22:74–77. Acetylcholine protection of adult pig retinal ganglion cells from 34 Wang XD, Chen XQ, Yang HH, Hu GY. Comparison of the ef- glutamate-induced excitotoxicity. Invest Ophthalmol Vis Sci. fects of cholinesterase inhibitors on [3 H]MK-801 binding in rat 2004;45:1531–1543. cerebral cortex. Neurosci Lett. 1999;272:21–24. 26 Pereira SP, Medina SV, Araujo EG. Cholinergic activity modu- 35 Li W, Pi R, Chan HH, Fu H, Lee NT, Tsang HW, et al. Novel di- lates the survival of retinal ganglion cells in culture: the role of meric acetylcholinesterase inhibitor bis(7)-tacrine, but not done- M1 muscarinic receptors. Int J Dev Neurosci. 2001;19:559–567. pezil, prevents glutamate-induced neuronal apoptosis by blocking 27 Fujiki M, Kobayashi H, Uchida S, Inoue R, Ishii K. Neuroprotec- N-methyl-D-aspartate receptors. J Biol Chem. 2005;280: tive effect of donepezil, a nicotinic acetylcholine-receptor activa- 18179–18188. tor, on cerebral infarction in rats. Brain Res. 2005;1043:236– 36 Saxena G, Singh SP, Agrawal R, Nath C. Effect of donepezil and 241. tacrine on oxidative stress in intracerebral streptozotocin-induced 28 Hutchins JB. Acetylcholine as a neurotransmitter in the vertebrate model of dementia in mice. Eur J Pharmacol. 2008;581:283– retina. Exp Eye Res. 1987;45:1–38. 289. 29 Dawson GR, Iversen SD. The effects of novel cholinesterase in- 37 Dunn NR, Pearce GL, Shakir SA. Adverse effects associated hibitors and selective muscarinic receptor agonists in tests of with the use of donepezil in general practice in England. J Psy- reference and working memory. Behav Brain Res. 1993;57: chopharmacol. 2000;14:406–408. 143–153. 38 Matsui K, Kagei Y, Mizuo H, Mishima M, Tadano K, Yoshimura 30 Clayton BA, Hayashida K, Childers SR, Xiao R, Eisenach JC. T, et al. Absorption, distribution, metabolism and excretion of Oral donepezil reduces hypersensitivity after nerve injury by a 14 C-donepezil hydrochloride after a single oral administration to spinal muscarinic receptor mechanism. Anesthesiology. 2007; rats. Jpn Pharmacol Ther. 1998;26 Suppl:1339–1355. 106:1019–1025. 39 Eisai Co., Ltd. and Pfizer Inc. [Drug interview form of Aricept® 31 Takada Y, Yonezawa A, Kume T, Katsuki H, Kaneko S, Sugimoto (donepezil hydrochloride).] 2000; p. 40. (in Japanese) H, et al. Nicotinic acetylcholine receptor-mediated neuroprotec-