Neuropharmacology and analgesia

Activation of the TRPV1 channel attenuates N-methyl-D-asparticacid-induced neuronal injury in the rat retina

Kenji Sakamoto n, Taiyo Kuroki, Yui Okuno, Haruna Sekiya, Akihiro Watanabe,Tomonori Sagawa, Hiroko Ito, Aya Mizuta, Asami Mori, Tsutomu Nakahara, Kunio IshiiDepartment of Molecular Pharmacology, Kitasato University School of Pharmaceutical Sciences, 9-1 Shirokane 5-chome, Minato-ku, Tokyo 108-8641, Japan

a r t i c l e i n f o

Article history:Received 2 November 2013Received in revised form4 March 2014Accepted 13 March 2014Available online 1 April 2014

Keywords:RetinaN-methyl-D-aspartic acidTransient receptor potential vanilloid type1Capsaicin

Chemical compounds studied in this article:Capsaicin (PubChem CID: 1548943)N-methyl-D-aspartic acid (PubChem CID:22880)Calcitonin gene-related peptide (PubChemCID: 44563440)CGRP (8–37) (PubChem CID: 16134896)Capsazepine (PubChem CID: 2733484)Substance P (PubChem CID: 36511)

a b s t r a c t

Capsaicin, a transient receptor potential vanilloid type1 (TRPV1) agonist, has been reported to protectagainst ischemia-reperfusion injury in various organs, including the brain, heart, and kidney, whereasactivation of TRPV1 was also reported to contribute to neurodegeneration, including pressure-inducedretinal ganglion cell death in vitro. We histologically investigated the effects of capsaicin and SA13353,TRPV1 agonists, on retinal injury induced by intravitreal N-methyl-D-aspartic acid (NMDA; 200 nmol/eye) in rats in vivo. Under ketamine/xylazine anesthesia, male Sprague–Dawley rats were subjected tointravitreal NMDA injection. Capsaicin (5.0 nmol/eye) was intravitreally admianeously with NMDAinjection. SA13353 (10 mg/kg) was intraperitoneally administered 15 min before NMDA injection.Morphometric evaluation at 7 days after NMDA injection showed that intravitreal NMDA injectionresulted in ganglion cell loss. Capsaicin and SA13353 almost completely prevented this damage.Treatment with capsazepine (TRPV1 antagonist, 0.5 nmol/eye), CGRP (8–37) (calcitonin gene-relatedpeptide (CGRP) receptor antagonist, 0.5 pmol/eye), or RP67580 (tachykinin NK1 receptor antagonist,0.5 nmol/eye) almost completely negated the protective effect of capsaicin in the NMDA-injected rats.Seven days after intravitreal NMDA injection, the cell number of retinal ganglion cell was significantlysmaller than in the eye that had received capsaicin in B6.Cg-TgN(Thy1-CFP)23Jrs/J transgenic mice thatexpress the enhanced cyan fluorescent protein in retinal ganglion cells in the retina. These resultssuggested that activation of TRPV1 protects retinal neurons from the injury induced by intravitrealNMDA in rats in vivo. Activation of CGRP and tachykinin NK1 receptors is possibly involved in underlyingprotective mechanisms.

& 2014 Elsevier B.V. All rights reserved.

1. Introduction

Retinal ganglion cell death is a characteristic of glaucoma, butthe underlying mechanism is not completely understood. How-ever, it is known that glutamate-receptor stimulation by excessglutamate during hypoxia (David et al., 1988) and ischemia-reperfusion (Louzada-Júnior et al., 1992) is toxic to neurons.Activation of the N-methyl-D-aspartic acid (NMDA) receptor, aglutamate receptor subtype (Choi, 1987, 1988), is followed by alarge Ca2þ influx via NMDA receptor-operated channels. Thisexcess intracellular Ca2þ is involved in predominant neuronalexcitotoxicity mechanisms and is thought to be an underlyingmechanism of glaucoma-induced neuronal cell death (Kuehn et al.,2005).

Capsaicin-sensitive sensory nerves are widely distributedthroughout the cardiovascular system and can be found in the

blood vessels, the heart, and the kidneys (Barja et al., 1983;Wharton et al., 1986). Capsaicin activates transient receptor poten-tial vanilloid type 1 (TRPV1) (Caterina et al., 1997), a non-selectivecation channel, which causes release of neurotransmitters, includ-ing calcitonin gene-related peptide (CGRP) and substance P(Hoover, 1987; Manzini et al., 1989). It was recently shown thatcapsaicin has protective effects against cerebral (Pegorini et al.,2005), myocardial (Wang and Wang, 2005), pulmonary (Wanget al., 2012), hepatic (Harada et al., 2002), and renal (Ueda et al.,2008) ischemia-reperfusion injuries in rats, as well as hypoxia-reoxygenation injury in cultured hippocampus slices (Guo et al.,2008). Anandamide, an endogenous agonist of TRPV1 and canna-binoid receptors, has been reported to protect against ischemia-reperfusion injury in brain, in vivo (Schomacher et al., 2008).

Very strong [3H]-resiniferatoxin labeling has been reported inthe retina (Szallasi et al., 1995), and mRNA and the TRPV1 proteinare expressed in retinal ganglion cells (RGCs) (Sappington et al.,2009). Methylanandamide has also been shown to protect theretina against ischemia-reperfusion injury (Nucci et al., 2007). Incontrast, capsaicin was shown to induce RGC degeneration in

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/ejphar

European Journal of Pharmacology

http://dx.doi.org/10.1016/j.ejphar.2014.03.0350014-2999/& 2014 Elsevier B.V. All rights reserved.

n Corresponding author. Tel./fax: þ81 3 3444 6205.E-mail address: sakamotok@pharm.kitasato-u.ac.jp (K. Sakamoto).

European Journal of Pharmacology 733 (2014) 13–22

preweanling rats (Ritter and Dinh, 1990) and apoptosis via extra-cellular Ca2þ influx in isolated RGCs (Sappington et al., 2009).However, whether capsaicin protects against or aggravates retinalglutamate neurotoxicity in vivo remains uncertain.

SA13353 was originally synthesized by the research laboratoryof Santen Pharmaceutical Co., Ltd. (Nara, Japan), and the com-pound is a potent and orally active inhibitor of TNF-α production(Murai et al., 2008). This effect of the drug was caused by theactivation of TRPV1 in vivo (Murai et al., 2008). The binding affinityof SA13353 to TRPV1 is approximately 15 times higher than that ofcapsaicin (Ueda et al., 2009). SA13353 was reported to prevent theischemic-reperfused kidney injury (Ueda et al., 2009), thelipopolysaccharide-induced acute lung injury (Tsuji et al., 2010a),and the lipopolysaccharide-induced experimental autoimmuneencephalomyelitis (Tsuji et al., 2010b) in the rat or mouse.

In the present study, we first examined whether capsaicin andSA13353 protected against NMDA-induced retinal injury in the rat,in vivo. On the basis of these results, we also evaluated whetherTRPV1, CGRP, and substance P are involved in underlying neuro-protective mechanisms.

2. Materials and methods

2.1. Animals

This study was conducted in accordance with the regulationsset forth in the ARVO Statement for the Use of Animals inOphthalmic and Vision Research. Experimental procedures con-formed to the Regulations for the Care and Use of LaboratoryAnimals and were approved by the Institutional Animal Care andUse Committee of Kitasato University.

Male Sprague–Dawley rats, weighing 230–300 g were pur-chased from Japan SLC, Hamamatsu, Japan. B6.Cg-TgN(Thy1-CFP)23Jrs/J transgenic mice express the enhanced cyan fluorescentprotein (ECFP) in RGCs in the retina (Feng et al., 2000; Dratviman-Storobinsky et al., 2008). Using this transgenic mouse enables usto evaluate the loss of RGCs by various hazardous stimuli includingintravitreal NMDA injection easily without the retrograde labelingof RGCs. Male and female B6.Cg-TgN(Thy1-CFP)23Jrs/J mice, inwhich the Thy1 promoter is linked to a ECFP reporter, werepurchased from Jackson Laboratory (Bar Harbor, ME). The micewere maintained by brother–sister mating in the animal room ofour university. We used the mice aged 8–16 weeks old.

The environment of the animal room was kept at 25 1C with a12 h:12 h light-dark cycle. All animals were fed and watered adlibitum.

2.2. Intravitreal injection

Intravitreal injection was performed as previously described(Sakamoto et al., 2009, 2010a, 2010b, 2011). Briefly, animals wereanesthetized with intraperitoneal ketamine (90 mg/kg, Daiichi-Sankyo, Tokyo, Japan) and xylazine (10 mg/kg, Tokyo Kasei, Tokyo,Japan). We confirmed that ketamine, an NMDA receptor antago-nist, did not affect the NMDA-induced retinal damage at the doseused in the present study by comparing the present results withthose in our previous report in which pentobarbital was used asanesthesia (Sakamoto et al., 2009). Intravitreal injection wasperformed with a 33-gauge needle connected to a 25-μL micro-syringe (MS-N25, Ito Seisakujo, Fuji, Japan). The tip of the needlewas inserted �1 mm behind the corneal limbus. One (for mice) or5 (for rats) microliters of the drug solution described below wasadministered into 1 eye, and a vehicle was administered into thecontralateral eye, which served as the control. Core temperature,

measured by a rectal thermometer, was maintained at 37 1Cduring experiments by using a heating pad and a heating lamp.

2.3. Drug preparations

Capsaicin (Wako Pure Chemical, Osaka, Japan) and RP67580(Santa Cruz Biotechnology, Dallas, TX) were dissolved in ethanol.Capsazepine (Sigma) was dissolved in DMSO. The final ethanol andDMSO concentrations in the drug solutions were 10% and 1%,respectively. NMDA (Nacalai Tesque, Kyoto, Japan), CGRP (PeptideInstitute, Minoh, Japan), CGRP (8–37) (Peptide Institute), andsubstance P (Peptide Institute) were completely dissolved in thevehicle solution (10% ethanol and 1% DMSO in saline). These drugswere given simultaneously with NMDA. SA13353 (a gift fromSanten Pharmaceutical Co., Ltd., Osaka, Japan) was dissolved in3% DMSO and 3% CREMOPHOR EL in saline, and intraperitoneallyadministered 15 min before NMDA injection.

2.4. Histological evaluation

Histological evaluation methods have been described pre-viously (Sakamoto et al., 2009, 2010a, 2010b, 2011). Briefly,animals were euthanized with an overdose of sodium pentobarbi-tal 7 days after intravitreal NMDA injection or the ischemicepisode. Both eyes were enucleated and fixed with a Davidsonsolution (37.5% ethanol, 9.3% paraformaldehyde, 12.5% acetic acid)for 24 h at room temperature. Fixed eyes were dissected throughthe optic nerve head in the vertical meridian with a microtomeblade (PATH BLADEþPRO by Kai, Matsunami Glass, Kishiwada,Japan) and embedded in paraffin after the lens had been removed.We used a microtome (HM325, Microm International, Walldorf,Germany) and a microtome blade (PATH BLADEþPRO by Kai,Matsunami Glass) to make 4-μm thickness, horizontal sectionsthrough the optic nerve head. Sections were made along thevertical meridian so that they contained the entire retina fromthe ora serrata in the superior hemisphere to the ora serrata in theinferior hemisphere. Sections were stained with hematoxylin andeosin and examined for morphometry. Sections from obliqueregions were excluded to avoid tissue artifacts. Using a lightmicroscope (Optiphot-2, Nicon, Tokyo, Japan), the total numberof RGCs in the retinal ganglion cell layer (GCL) was manuallycounted in a region beginning 1 mm from the center of the opticnerve head and ending 1.25 mm from the center of the optic nervehead (for a retinal length of 0.25 mm). No attempts were made todistinguish between cell types in the GCL; displaced amacrine cellswere included in RGC counts. Thickness measurements of theinner plexiform layer (IPL), the inner nuclear layer (INL), the outerplexiform layer (OPL), and the outer nuclear layer (ONL) were alsoperformed. Digital photographs (digital camera [DP11, Olympus,Tokyo, Japan] connected to a light microscope) were taken so that�0.25 mm of retina appeared in each photograph, with sections�1 mm from the center of the optic nerve head chosen.The thickness of IPL, INL, OPL and ONL was then measured. Thenumber of the cells in GCL and the thickness of retinal layers of theNMDA-injected eyes were normalized to those of the contralateraleye and are presented as percentages.

2.5. Retinal ganglion cell count in B6.Cg-TgN(Thy1-CFP)23Jrs/J mice

Seven days after NMDA injection, the mice were sacrificed. Theeyes were enucleated and immediately fixed in 4% paraformalde-hyde in 0.1 M phosphate buffered saline for 1 h at 4 1C. Four radialrelaxing incisions were made and the retina was flattened on a glassslide. The whole mount retina was sealed with fluoromount G(Southern Biotech, Birmingham, AL) and a cover glass (MatsunamiGlass). Images were obtained using a confocal laser microscope

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–2214

(LSM510META, Carl Zeiss, Oberkochen, Germany). The number ofRGC expressing CFP was manually counted using image-processingsoftware (Adobe Photoshop CS5, San Jose, CA) in the area that was500 μm away from the optic disc. The area in which RGC numberwas counted was 0.2 m2. The number of RGC in the NMDA-injectedeye was divided by that in the contralateral eye, and a percentage ofRGC density was calculated.

2.6. Statistical analyses

All data are presented as mean7S.E.M. One-way analysis ofvariance, followed by a Tukey–Kramer test, was used for multiplecomparisons. Differences were considered to be statistically sig-nificant if Po0.05.

3. Results

3.1. Effect of TRPV1 agonists on NMDA-induced retinal injury in rats

We first investigated the effects of TRPV1 agonists on NMDA-induced retinal injury in rats. Typical photomicrographs of the retinataken 7 days after NMDA intravitreal injection (200 nmol/eye)are shown in Figs. 1 and 3. In the NMDA-injected control eyes(no capsaicin), degenerative changes were observed in the GCL

and the IPL, but these changes did not occur in eyes that hadreceived intravitreal NMDA plus intravitreal capsaicin, or intraper-itoneal SA13353. Morphometric measurements from 5 independent

Fig. 1. Photomicrographs showing retinal histology of the rat eyes treated with the vehicle (A), 0.5 nmol/eye capsaicin (B), 5.0 nmol/eye capsaicin (C), 0.5 nmol/eyecapsazepine (D), 5.0 nmol/eye capsaicinþ0.5 nmol/eye capsazepine (E), 200 nmol/eye NMDA (F), 200 nmol/eye NMDAþ0.5 nmol/eye capsaicin (G), 200 nmol/eyeNMDAþ5.0 nmol/eye capsaicin (H), 200 nmol/eye NMDAþ0.5 nmol/eye capsazepine (I), and 200 nmol/eye NMDAþ5.0 nmol/eye capsaicinþ0.5 nmol/eye capsazepine (J).All images were taken 7 days after intravitreal NMDA injection and all drugs were administered intravitreally. Loss of the cells in the ganglion cell layer (labeled witharrowhead) is apparent in eyes given NMDA (F), NMDAþcapsazepine (I), and NMDAþcapsaicinþcapsazepine (J). Retinal cytoarchitecture remains preserved in eyes given200 nmol/eye NMDAþ5.0 nmol/eye capsaicin (H). Scale bar¼50 mm; Original magnification, �200.

Fig. 2. Effect of intravitreal capsaicin on NMDA-induced histological damage, examined7 days after NMDA injection. Histological parameters examined were cell density in theganglion cell layer (GCL) and thickness of the inner plexiform layer (IPL), the inner nuclearlayer (INL), the outer plexiform layer (OPL), and the outer nuclear layer (ONL). Allmeasurements in NMDA-injected eyes were normalized to the vehicle-treated contral-ateral eye and are presented as percentages. Data are presented asmean7S.E.M. nPo0.05.

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–22 15

experiments are summarized in Figs. 2 and 4. Seven days after NMDAintravitreal injection, the cell number in GCL and the thickness of IPLof the vehicle-treated group was smaller than in the eye that hadreceived capsaicin or SA13353. The protective effects of capsaicin andSA13353 were abolished when capsazepine (0.5 nmol/eye), a TRPV1antagonist, was injected. Intravitreal injection of capsaicin or capsa-zepine, and intraperitoneal injection of SA13353 in absence of NMDAdid not affect the retinal morphology.

3.2. Effect of capsaicin on NMDA-induced retinal injury in B6.Cg-TgN(Thy1-CFP)23Jrs/J mice

Not only RGC but also other types of the cell such as amacrinecell exist in GCL. In the present study, to assess the protectiveeffect of capsaicin on RGC directly, we examined the effect ofcapsaicin on the NMDA (40 nmol/eye)-induced loss of RGC in B6.Cg-TgN(Thy1-CFP)23Jrs/J transgenic mice. The mice express ECFPin RGCs in the retina. In the NMDA-injected control eyes (nocapsaicin), the cell number of RGC was dramatically decreased,whereas the deleterious effect of NMDA was reduced in eyes thathad received intravitreal NMDA plus capsaicin (1 nmol/eye). Theprotective effect of capsaicin was abolished when capsazepine

(0.1 nmol/eye) was simultaneously injected (Fig. 5A–H). Cellcounts from 14 (vehicle), 14 (capsaicin), 6 (capsazepine), or 7(capsaicinþcapsazepine) independent experiments are summar-ized in Fig. 5I. Intravitreal injection of capsaicin or capsazepine inabsence of NMDA did not affect the morphology and the numberof RGC. The numbers of the ECFP-positive cells were as follows:358710 (vehicle), 331724 (capsaicin), 364724 (capsazepine),and 325735 (capsaicin plus capsazepine). We found that theintensities of ECFP in the capsaicin-treated groups were stronger(Figs. 5C and D) than those in other groups. This effect tended tobe reduced by capsazepine partially. Although the mechanism ofthis phenomenon is not clearly understood at the present, it ispossible that capsaicin alters the Thy1 promoter activity.

3.3. Involvement of CGRP and substance P in neuroprotectionby capsaicin

To determine whether CGRP and substance P were involved inthe protective effect of capsaicin, we investigated the effects of aCGRP receptor antagonist (0.5 pmol/eye CGRP (8–37)) and antachykinin NK1 receptor antagonist (0.5 nmol/eye RP67580) oncapsaicin-related neuroprotection. Typical photomicrographs of

Fig. 3. Photomicrographs showing retinal histology of the rat eyes treated with the vehicle (A), 10 mg/kg SA13353 (B), 10 mg/kg SA13353þ0.5 nmol/eye capsazepine (C),200 nmol/eye NMDA (D), 200 nmol/eye NMDAþ10 mg/kg SA13353 (E), and 200 nmol/eye NMDAþ10 mg/kg SA13353þ0.5 nmol/eye capsazepine (F). All images were taken7 days after intravitreal NMDA injection and SA13353 was intraperitoneally administered. Retinal damage is apparent in eyes given NMDA (D) andNMDAþSA13353þcapsazepine (F). Retinal cytoarchitecture remains preserved in eyes given NMDAþSA13353 (E). Scale bar¼50 μm; Original magnification, �200.

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–2216

the retina taken 7 days after NMDA intravitreal injection(200 nmol/eye) are shown in Fig. 6. Morphometric measurementsfrom 5–7 independent experiments are also shown in Fig. 7. WhenCGRP (8–37) or RP67580 was intravitreally administered withcapsaicin, the neuroprotective effect against NMDA-induced ret-inal injury was diminished. Eyes treated with CGRP (8–37) orRP67580 in absence of NMDA showed no changes in retinalmorphology.

In addition, we investigated the effects of intravitreal CGRP(50 fmol/eye) and substance P (5 nmol/eye) on the NMDA-inducedretinal injury. Typical photomicrographs of the retina taken 7 days

after NMDA intravitreal injection (200 nmol/eye) are shown inFig. 8. Intravitreal CGRP (50 fmol/eye) and substance P (5 nmol/eye) reduced the NMDA-induced loss of the cells in GCL. Theprotective effect of intravitreal CGRP was blocked by RP67580,whereas the protective effect of intravitreal substance P was notblocked by CGRP (8–37). Morphometric measurements from 5–6independent experiments are summarized in Fig. 9.

4. Discussion

In the present study, we first demonstrated that capsaicin andSA13353 had histologically protective effects against NMDA-induced retinal injury in rats, in vivo. These effects were blockedby capsazepine, a TRPV1 antagonist, suggesting that capsaicin andSA13353 prevented retinal injury via TRPV1 activation. In addition,we directly showed that capsaicin protected RGCs against NMDA-induced cell death in B6.Cg-TgN(Thy1-CFP)23Jrs/J transgenic mice.

Because the neuronal retina does not have nerve terminals thattransmit pain, the drugs spread to the retina will not cause a pain.However, intravitreal capsaicin may also spread to unmyelinatedC-fibers that innervate the cornea. TRPV1 channel is expressed onthe C-fiber, which is part of the ocular branch of the trigeminalnerve, and intravitreal capsaicin may cause a corneal pain. How-ever, no autonomic reflex was found when intravitreal treatmentwas done, and after the anesthetic wore off. This is becauseintravitreal treatment was done under anesthesia, and theanesthesia was enough to relieve the pain induced by the experi-mental procedures.

In most tissues, CGRP and substance P are located in sensoryC-fibers within epithelial linings and around small blood vessels(Lundberg et al., 1992). Activation of TRPV1 evokes CGRP andsubstance P release from sensory nerve terminals (Hoover, 1987;Manzini et al., 1989). Although the retina is not governed bysensory C-fibers, CGRP and substance P levels have been shown to

Fig. 4. Effect of intraperitoneal SA13353 on NMDA-induced histological damage,examined 7 days after NMDA injection in the rat. Histological parameters examinedwere cell density in the ganglion cell layer (GCL) and thickness of the innerplexiform layer (IPL), the inner nuclear layer (INL), the outer plexiform layer (OPL),and the outer nuclear layer (ONL). All measurements in NMDA-injected eyes werenormalized to the vehicle-treated fellow eye and are presented as percentages.Data are presented as mean7S.E.M. nPo0.05.

Fig. 5. Retinal ganglion cell images expressing the enhanced cyan fluoresceint protein (ECFP) in the retina of the B6.Cg-TgN(Thy1-CFP)23Jrs/J transgenic mouse. Wholemount retinal images of the eyes treated with the vehicle (A), 1 nmol/eye capsaicin (B), 1 nmol/eye capsazepine (C), 1 nmol/eye capsaicinþ0.1 nmol/eye capsazepine (D),40 nmol/eye NMDA (E), 40 nmol/eye NMDAþ1 nmol/eye capsaicin (F), 40 nmol/eye NMDAþ0.1 nmol/eye capsazepine (G), and 40 nmol/eye NMDAþ1 nmol/eyecapsaicinþ0.1 nmol/eye capsazepine (H) were taken with the confocal laser microscope. The numbers of RGC expressing ECFP decreased in the NMDA-injected eye. Scalebar is 100 μm. Original magnification, �200. The number of RGC in the NMDA-injected eye was divided by that in the contralareral eye, and a percentage of RGC density wascalculated (I). Data are presented as mean7S.E.M. *Po0.05, between the indicated groups.

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–22 17

be increased in electrically stimulated retina (Bronzetti et al.,2007), and in the retina during myocardial infarction (Yang et al.,2011). These findings indicated that CGRP and substance P arereleased from retinal cells under metabolic stress, either from

ischemia or from prolonged membrane depolarization. In thepresent study, the protective effect of capsaicin treatment wasreduced when CGRP (8–37) and RP67580 were given with capsai-cin. Additionally, exogenous CGRP and substance P were protective

Fig. 6. Photomicrographs showing retinal histology of the rat eyes treated with the vehicle (A), 5.0 nmol/eye capsaicin (B), 0.5 pmol/eye CGRP (8 -3 7) (C), 0.5 nmol/eyeRP67580 (D), 5.0 nmol/eye capsaicinþ0.5 pmol/eye CGRP (8–37) (E), 5 nmol/eye capsaicinþ0.5 nmol/eye RP67580 (F), 200 nmol/eye NMDA (G), 200 nmol/eyeNMDAþ5.0 nmol/eye capsaicin (H), 200 nmol/eye NMDAþ0.5 pmol/eye CGRP (8–37) (I), 200 nmol/eye NMDAþ0.5 nmol/eye RP67580 (J), 200 nmol/eye NMDAþ5.0nmol/eye capsaicinþ0.5 pmol/eye CGRP (8–37) (K), and 200 nmol/eye NMDAþ5 nmol/eye capsaicinþ0.5 nmol/eye RP67580 (L). All images were taken 7 days afterintravitreal NMDA injection and all drugs were administered intravitreally. Retinal damage is apparent in eyes treated with NMDA (G), 200 nmol/eye NMDAþ0.5 pmol/eyeCGRP (8–37) (I), 200 nmol/eye NMDAþ0.5 nmol/eye RP67580 (J), NMDAþcapsaicinþCGRP (8–37) (K), and NMDAþcapsaicinþRP67580 (L). Retinal cytoarchitectureremains preserved in eyes treated with NMDAþcapsaicin (H). Scale bar¼50 μm; Original magnification, �200.

Fig. 7. Effects of intravitreal CGRP (8–37) (A) and RP67580 (B) on neuroprotection by capsaicin, as viewed histologically in the rat. Retinal damage was examined 7 days afterintravitreal NMDA injection. Histological parameters examined were cell density in the ganglion cell layer (GCL) and thickness of inner plexiform layer (IPL), the innernuclear layer (INL), the outer plexiform layer (OPL), and the outer nuclear layer (ONL). All measurements in NMDA-injected eyes were normalized to the vehicle-treatedcontralateral eye and are presented as percentages. Data are presented as mean7S.E.M. nPo0.05.

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–2218

against NMDA-induced retinal injury. These data suggest thatCGRP and tachykinin NK1 receptor stimulation by endogenousCGRP and substance P play a role in capsaicin neuroprotectivemechanisms. Our results are consistent with the previous reports,which indicated that endogenous CGRP protects retinal cellsagainst apoptosis induced by permanent coronary artery occlusionin rats (Yang et al., 2011). They also agree with studies showingthat exogenous CGRP and substance P result in significant post-ischemic recovery improvements in perfused murine hearts(Wang and Wang, 2005). However, protective mechanisms under-lying the protective effect of CGRP and substance P on themyocardium remain unknown. In the present study, the protectiveeffect of intravitreal CGRP was blocked by RP67580, whereas theprotective effect of intravitreal substance P was not blocked byCGRP (8–37). There are several possibilities to explain this result.The first possibility is that substance P is a downstream moleculeof CGRP. Julia and Buéno (1997) suggested that CGRP acted on theterminals of C-fiber to facilitate the release of substance P in thespinal dorsal horn. It is possible that the similar mechanism maybe involved in the protective effect of capsaicin in the retina. Thesecond one is that CGRP retards the metabolism of substance P,and then increases the amount of substance P. Saleh et al. (1996)showed that CGRP indirectly act on tachykinin NK1 receptorsthrough an inhibition of the substance P endopeptidase. Gontijo

et al. (1999) showed that CGRP activated renal pelvic sensorynerves by retarding the metabolism of substance P. The third oneis that RP67580 may block CGRP receptors. However, there is noreport to show that the drug blocks CGRP receptors, as far as weknow. Unfortunately, the present study could not clarify the rolefor CGRP and substance P in the protective effect of capsaicincompletely. Further experiments are clearly needed to solvethis issue.

In the present study, we investigated the effects SA13353, anorally active and potent TRPV1 agonist, on NMDA-induced retinalinjury. Recently, downregulation of TNF-α and upregulation ofIL-10, an anti-inflammatory cytokine, were involved in the pro-tective effect of SA13353 on ischemic kidney injury (Ueda et al.,2009). Although it is unclear whether NMDA elicits inflammatoryreaction, pitavastatin was reported to prevent NMDA-induced RGCdeath by suppressing leukocyte recruitment (Nakazawa et al.,2007). Relationship between the retinal protective effects ofTRPV1 agonists and anti-inflammatory effects is an interestingtheme to clarify in further investigations.

On the other hand, it is possible that the release of SP and CGRPinduced by TRPV1 agonists causes neurogenic inflammation in theeye. At the present, we cannot deny the possibility that the releaseof CGRP and substance P by TRPV1 agonists induces undesirableleakage of the retinal-blood barrier and increased vascular

Fig. 8. Photomicrographs showing retinal histology of the rat eyes treated with vehicle (A), 50 fmol/eye CGRP (B), 50 fmol/eye CGRPþ0.5 nmol/eye RP67580 (C), 5 nmol/eyesubstance P (D), 5 nmol/eye substance Pþ0.5 pmol/eye CGRP (8–37) (E), 200 nmol/eye NMDA (F), 200 nmol/eye NMDAþ50 fmol/eye CGRP (G), 200 nmol/eyeNMDAþ50 fmol/eye CGRPþ0.5 nmol/eye RP67580 (H), 200 nmol/eye NMDAþ5 nmol/eye substance P (I), and 200 nmol/eye NMDAþ5 nmol/eye substance Pþ0.5 pmol/eye CGRP (8–37) (J). All images were taken 7 days after intravitreal NMDA injection and all drugs were administered intravitreally. Scale bar¼50 μm; Original magnification,�200.

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–22 19

permeability in the retina. However, TRPV1 agonists have beenreported to reduce inflammation and the severity of the symptomsof kidney injury, arthritis, lung injury and encephalomyelitismodels (Murai et al., 2008; Ueda et al., 2009; Tsuji et al., 2010a,2010b). In fact, we did not find inflammation in the retina treatedwith capsaicin, CGRP, or substance P at the dose used in thepresent study. Further studies are needed to clarify whethertreatment with TRPV1 agonists induces neurogenic inflammationin the retina and/or undesirable leakage of the retinal-bloodbarrier.

It is possible that intraperitoneal treatment of SA13353 maycause pain, intense bronchoconstriction and a rash causing a dropof blood pressure. Because we treated SA13353 to the anesthetizedrats, the rats did not have pain during anesthesia. Although we didnot assess pain reactions, bronchoconstriction and blood pressurein the present study, we did not find any special symptoms, such

as writhing reaction, difficulty breathing and rash, after theanesthetic wore off, as far as we watched the rats used in theexperiments. In previous reports (Murai et al., 2008; Ueda et al.,2009; Tsuji et al., 2010a, 2010b) 30–100 mg/kg SA13353 was orallytreated to the rats. They did not note any special symptomsinduced by the treatment of SA13353. However, it is possible thatSA13353 induces a decrease of arterial pressure, because SA13353was reported to increase the serum CGRP concentration (Murai etal., 2008). Further studies are needed to clarify the effect ofSA13353 on the retinal blood flow.

Leonelli et al. (2013) also reported that intravitreal capsaicinreduced the expression of neuronal nitric oxide synthase andincreased the expression of inducible nitric oxide synthase. Acti-vation of neuronal nitric oxide synthase was reported to beinvolved in the mechanism of NMDA-induced retinal damage(Adachi et al., 1998). In addition, we previously reported thatactivation of inducible nitric oxide synthase was necessary for theprotective effect of late ischemic preconditioning on the ischemia-reperfusion injury in the rat retina (Sakamoto et al., 2006).Therefore, it is possible that capsaicin protects the retina againstthe NMDA-induced damage through modulation of the expressionof nitric oxide synthases.

The present study suggested that CGRP and tachykinin NK1

receptor activation is important in the neuroprotective effect ofcapsaicin in the retina. Because CGRP and substance P are knownvasodilators, regional retinal-blood–flow improvements in theirpresence may be beneficial. Recently we reported that a prosta-noid EP2 receptor agonistand a β3-adrenergic receptor agonist,potent vasodilators in the retinal vessels (Mori et al., 2007, 2011),were protective against NMDA-induced retinal injury (Mori et al.,2009, Oikawa et al., 2012). Further studies are needed to clarify theinvolvement of regional retinal-blood-flow improvement.

Capsaicin was previously reported to induce RGC degenerationin pre-weanling rats (Ritter and Dinh, 1990) and apoptosis viaextracellular Ca2þ influx in isolated RGCs (Sappington et al., 2009).However, the present study clearly demonstrated that capsaicindid not induce RGC degeneration, but instead reduced RGC deathin adult rat retinas, in vivo. Unfortunately, the present study couldnot clarify this discrepancy. Because capsaicin was reported toinduce cell death in the isolated RGC, other types of cells may alsobe important in the neuroprotective effect of capsaicin. It is knownthat glial cells express CGRP receptors (Priller et al., 1995), and it ispossible that TRPV1 activation, crucial in the capsaicin neuropro-tective effect, occurs in amacrine (Zimov and Yazulla, 2007) andmicroglial cells (Sappington and Calkins, 2008). In addition, it ispossible that the expression of CGRP, substance P, and/or theirreceptors may be insufficient in the immature retina. Leonelli et al.(2010, 2013) reported that capsaicin induced upregulation of3-nitrotyrosine and 4-hydroxynonenal, oxidative stress markers,and apoptosis of the cells in GCL in the rat retinal explants. Theseresults also contradict those in the present study. Although it ishard to clarify the discrepancy at the present, it is possible thatintravitreal capsaicin may spread to unmyelinated C-fibers thatinnervate the cornea. Because TRPV1 channel is expressed on theC-fiber (Murata and Masuko, 2006; Nakamura et al., 2007),intravitreal capsaicin may evoke the release of other mediators,such as CGRP, from the C-fiber, and then elicit the protective effecton the NMDA-induced damage in the retina. Further studies areneeded to clarify these issues.

In conclusion, this is the first study to demonstrate that TRPV1agonists have neuroprotective effects against retinal injuryinduced by intravitreal NMDA. Activation of CGRP and tachykininNK1 receptors is possibly involved in the underlying protectivemechanisms. Intraperitoneal treatment with SA13353, a potentTRPV1 agonist, also elicited the neuroprotective effect in theretina. Although further studies are needed, the present study

Fig. 9. Effect of RP67580 on neuroprotection by CGRP (A), and effect of CGRP (8–37) on neuroprotection by substance P (B), as viewed histologically in the rat. CGRP,RP67580, substance P, and CGRP (8–37) were treated intravitreally. Retinal damagewas examined 7 days after intravitreal NMDA injection. Histological parametersexamined were cell density in the ganglion cell layer (GCL) and thickness of innerplexiform layer (IPL), the inner nuclear layer (INL), the outer plexiform layer (OPL),and the outer nuclear layer (ONL). All measurements in NMDA-injected eyes werenormalized to the vehicle-treated contralateral eye and are presented as percen-tages. Data are presented as mean7S.E.M. nPo0.05.

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–2220

shows the possibility that TRPV1 agonists are effective to preventretinal diseases caused by glutamate excitotoxicity, such as glau-coma and retinal artery occlusion.

Acknowledgments

SA13353 was kindly gifted by Santen Pharmaceutical Co., Ltd.,Osaka, Japan. This work is supported by Ministry of Education,Culture, Sports, Science and Technology in Japan, a Grant-in-Aidfor Scientific Research (C) #21790249 (KS), and Suzuken MemorialFoundation #09-57, Nagoya, Japan (KS).

References

Adachi, K., Fujita, Y., Morizane, C., Akaike, A., Ueda, M., Satoh, M., Masai, H., Kashii,S., Honda, Y., 1998. Inhibition of NMDA receptors and nitric oxide synthasereduces ischemic injury of the retina. Eur. J. Pharmacol. 350, 53–57.

Barja, F., Mathison, R., Huggel, H., 1983. Substance P-containing nerve fibers in largeperipheral blood vessels of the rat. Cell Tissue Res. 229, 411–422.

Bronzetti, E., Artico, M., Kovacs, I., Felici, L.M., Magliulo, G., Vignone, D., D'Ambrosio,A., Forte, F., Di Liddo, R., Feher, J., 2007. Expression of neurotransmitters andneurotrophins in neurogenic inflammation of the rat retina. Eur. J. Histochem.51, 251–260.

Caterina, M.J., Schumacher, M.A., Tominaga, M., Rosen, T.A., Levine, J.D., Julius, D.,1997. The capsaicin receptor: a heat-activated ion channel in the pain pathway.Nature 389, 816–824.

Choi, D.W., 1987. Ionic dependence of glutamate neurotoxicity in cortical cellculture. J. Neurosci. 7, 369–379.

Choi, D.W., 1988. Calcium-mediated neurotoxicity: relationship to specific channeltypes and role in ischemic damage. Trends Neurosci. 11, 465–469.

David, P., Lusky, M., Teichberg, V.I., 1988. Involvement of excitatory neurotransmit-ters in the damage produced in chick embryo retinas by anoxia and extra-cellular high potassium. Exp. Eye Res. 46, 657–662.

Dratviman-Storobinsky, O., Hasanreisoglu, M., Offen, D., Barhum, Y., Weinberger, D.,Goldenberg-Cohen, N., 2008. Progressive damage along the optic nervefollowing induction of crush injury or rodent anterior ischemic optic neuro-pathy in transgenic mice. Mol. Vis. 14, 2171–2179.

Feng, G., Mellor, R.H., Bernstein, M., Keller-Peck, C., Nguyen, Q.T., Wallace, M.,Nerbonne, J.M., Lichtman, J.W., Sanes, J.R., 2000. Imaging neuronal subsets intransgenic mice expressing multiple spectral variants of GFP. Neuron 28, 41–51.

Gontijo, J.R., Smith, L.A., Kopp, U.C., 1999. CGRP activates renal pelvic substance Preceptors by retarding substance P metabolism. Hypertension 33, 493–498.

Guo, S.Y., Yang, G.P., Jiang, D.J., Wang, F., Song, T., Tan, X.H., Sun, Z.Q., 2008.Protection of capsaicin against hypoxia-reoxygenation-induced apoptosis of rathippocampal neurons. Can. J. Physiol. Pharmacol. 86, 785–792.

Harada, N., Okajima, K., Uchiba, M., Katsuragi, T., 2002. Ischemia/reperfusion-induced increase in the hepatic level of prostacyclin is mainly mediated byactivation of capsaicin-sensitive sensory neurons in rats. J. Lab. Clin. Med. 139,218–226.

Hoover, D.B., 1987. Effects of capsaicin on release of substance P-like immunor-eactivity and physiological parameters in isolated perfused guinea-pig heart.Eur. J. Pharmacol. 141, 489–492.

Julia, V., Buéno, L., 1997. Tachykininergic mediation of viscerosensitive responses toacute inflammation in rats: role of CGRP. Am. J. Physiol. 272, G141–G146.

Kuehn, M.H., Fingert, J.H., Kwon, Y.H., 2005. Retinal ganglion cell death inglaucoma: mechanisms and neuroprotective strategies. Ophthalmol. Clin.North. Am. 18, 383–395.

Leonelli, M., Martins, D.O., Britto, L.R., 2010. TRPV1 receptors are involved in proteinnitration and Müller cell reaction in the acutely axotomized rat retina. Exp. EyeRes. 91, 755–768.

Leonelli, M., Martins, D.O., Britto, L.R., 2013. Retinal cell death induced by TRPV1activation involves NMDA signaling and upregulation of nitric oxide synthases.Cell. Mol. Neurobiol. 33, 379–392.

Louzada-Júnior, P., Dias, J.J., Santos, W.F., Lachat, J.J., Bradford, H.F., Coutinho-Netto,J., 1992. Glutamate release in experimental ischaemia of the retina: anapproach using microdialysis. J. Neurochem. 59, 358–363.

Lundberg, J.M., Franco-Cereceda, A., Alving, K., Delay-Goyet, P., Lou, Y.P., 1992.Release of calcitonin gene-related peptide from sensory neurons. Ann. N. Y.Acad. Sci. 657, 187–193.

Manzini, S., Perretti, F., De Benedetti, L., Pradelles, P., Maggi, C.A., Geppetti, P., 1989.A comparison of bradykinin- and capsaicin-induced myocardial and coronaryeffects in isolated perfused heart of guinea-pig: involvement of substance P andcalcitonin gene-related peptide release. Br. J. Pharmacol. 97, 303–312.

Mori, A., Saito, M., Sakamoto, K., Narita, M., Nakahara, T., Ishii, K., 2007. Stimulationof prostanoid IP and EP2 receptors dilates retinal arterioles and increases retinaland choroidal blood flow in rats. Eur. J. Pharmacol. 570, 135–141.

Mori, A., Ishii, T., Kuroki, T., Shigeta, N., Sakamoto, K., Nakahara, T., Ishii, K., 2009.The prostanoid EP2 receptor agonist ONO-AE1-259-01 protects againstglutamate-induced neurotoxicity in rat retina. Eur. J. Pharmacol. 616, 64–67.

Mori, A., Nakahara, T., Sakamoto, K., Ishii, K., 2011. Role of β3-adrenoceptors inregulation of retinal vascular tone in rats. Naunyn Schmiedebergs Arch.Pharmacol. 384, 603–608.

Murai, M., Tsuji, F., Nose, M., Seki, I., Oki, K., Setoguchi, C., Suhara, H., Sasano, M.,Aono, H., 2008. SA13353 (1-[2-(1-adamantyl)ethyl]-1-pentyl-3-[3-(4-pyridyl)propyl]urea) inhibits TNF-alpha production through the activation of capsaicin-sensitive afferent neurons mediated via transient receptor potential vanilloid1 in vivo. Eur. J. Pharmacol. 588, 309–315.

Murata, Y., Masuko, S., 2006. Peripheral and central distribution of TRPV1,substance P and CGRP of rat corneal neurons. Brain Res. 1085, 87–94.

Nakamura, A., Hayakawa, T., Kuwahara, S., Maeda, S., Tanaka, K., Seki, M., Mimura,O., 2007. Morphological and immunohistochemical characterization of thetrigeminal ganglion neurons innervating the cornea and upper eyelid of therat. J. Chem. Neuroanat. 34, 95–101.

Nakazawa, T., Takahashi, H., Nishijima, K., Shimura, M., Fuse, N., Tamai, M., Hafezi-Moghadam, A., Nishida, K., 2007. Pitavastatin prevents NMDA-induced retinalganglion cell death by suppressing leukocyte recruitment. J. Neurochem. 100,1018–1031.

Nucci, C., Gasperi, V., Tartaglione, R., Cerulli., A., Terrinoni, A., Bari, M., De Simone, C.,Agrò, A.F., Morrone, L.A., Corasaniti, M.T., Bagetta, G., Maccarrone, M., 2007.Involvement of the endocannabinoid system in retinal damage after high intrao-cular pressure-induced ischemia in rats. Invest. Ophthalmol. Vis. Sci. 48, 2997–3004.

Oikawa, F., Nakahara, T., Akanuma, K., Ueda, K., Mori, A., Sakamoto, K., Ishii, K., 2012.Protective effects of the β3-adrenoceptor agonist CL316243 against N-methyl-D-aspartate-induced retinal neurotoxicity. Naunyn Schmiedebergs Arch. Pharma-col. 385, 1077–1081.

Pegorini, S., Braida, D., Verzoni, C., Guerini-Rocco, C., Consalez, G.G., Croci, L., Sala,M., 2005. Capsaicin exhibits neuroprotective effects in a model of transientglobal cerebral ischemia in Mongolian gerbils. Br. J. Pharmacol. 144, 727–735.

Priller, J., Haas, C.A., Reddington, M., Kreutzberg, G.W., 1995. Calcitonin gene-related peptide and ATP induce immediate early gene expression in cultured ratmicroglial cells. Glia 15, 447–457.

Ritter, S., Dinh, T.T., 1990. Capsaicin-induced neuronal degeneration in the brainand retina of preweanling rats. J. Comp. Neurol. 296, 447–461.

Sakamoto, K., Yonoki, Y., Kubota, Y., Kuwagata, M., Saito, M., Nakahara, T., Ishii, K.,2006. Inducible nitric oxide synthase inhibitors abolished histological protec-tion by late ischemic preconditioning in rat retina. Exp. Eye Res. 82, 512–518.

Sakamoto, K., Kawakami, T., Shimada, M., Yamaguchi, A., Kuwagata, M., Saito, M.,Nakahara, T., Ishii, K., 2009. Histological protection by cilnidipine, a dual L/N-type Ca2þ channel blocker, against neurotoxicity induced by ischemia-reperfusion in rat retina. Exp. Eye Res. 88, 974–982.

Sakamoto, K., Hiraiwa, M., Saito, M., Nakahara, T., Sato, Y., Nagao, T., Ishii, K., 2010a.Protective effect of all-trans retinoic acid on NMDA-induced neuronal cell deathin rat retina. Eur. J. Pharmacol. 635, 56–61.

Sakamoto, K., Ohki, K., Saito, M., Nakahara, T., Ishii, K., 2010b. Histological protectionby donepezil against neurodegeneration induced by ischemia-reperfusion inthe rat retina. J. Pharmacol. Sci. 112, 327–335.

Sakamoto, K., Ohki, K., Saito, M., Nakahara, T., Ishii, K., 2011. Small molecule cyclin-dependent kinase inhibitors protect against neuronal cell death in theischemic-reperfused rat retina. J. Ocul. Pharmacol. Ther. 27, 419–425.

Saleh, T.M., Kombian, S.B., Zidichouski, J.A., Pittman, Q.J., 1996. Peptidergic modula-tion of synaptic transmission in the parabrachial nucleus in vitro: importance ofdegradative enzymes in regulating synaptic efficacy. J. Neurosci. 16, 6046–6055.

Sappington, R.M., Calkins, D.J., 2008. Contribution of TRPV1 to microglia-derived IL-6 and NF kappa B translocation with elevated hydrostatic pressure. Invest.Ophthalmol. Vis. Sci. 49, 3004–3017.

Sappington, R.M., Sidorova, T., Long, D.J., Calkins, D.J., 2009. TRPV1: contribution toretinal ganglion cell apoptosis and increased intracellular Ca2þ with exposureto hydrostatic pressure. Invest. Ophthalmol. Vis. Sci. 50, 717–728.

Schomacher, M., Müller, H.D., Sommer, C., Schwab, S., Schäbitz, W.R., 2008.Endocannabinoids mediate neuroprotection after transient focal cerebralischemia. Brain Res. 1240, 213–220.

Szallasi, A., Nilsson, S., Farkas-Szallasi, T., Blumberg, P.M., Hokfelt, T., Lundberg, J.M., 1995. Vanilloid (capsaicin) receptors in the rat: distribution in the brain,regional differences in the spinal cord, axonal transport to the periphery, anddepletion by systemic vanilloid treatment. Brain Res. 703, 175–183.

Tsuji, F., Murai, M., Oki, K., Inoue, H., Sasano, M., Tanaka, H., Inagaki, N., Aono, H., 2010a.Effects of SA13353, a transient receptor potential vanilloid 1 agonist, on leukocyteinfiltration in lipopolysaccharide-induced acute lung injury and ovalbumin-induced allergic airway inflammation. J. Pharmacol. Sci. 112, 487–490.

Tsuji, F., Murai, M., Oki, K., Seki, I., Ueda, K., Inoue, H., Nagelkerken, L., Sasano, M.,Aono, H., 2010b. Transient receptor potential vanilloid 1 agonists as candidatesfor anti-inflammatory and immunomodulatory agents. Eur. J. Pharmacol. 627,332–339.

Ueda, K., Tsuji, F., Hirata, T., Takaoka, M., Matsumura, Y., 2008. Preventive effect ofTRPV1 agonists capsaicin and resiniferatoxin on ischemia/reperfusion-inducedrenal injury in rats. J. Cardiovasc. Pharmacol. 51, 513–520.

Ueda, K., Tsuji, F., Hirata, T., Ueda, K., Murai, M., Aono, H., Takaoka, M., Matsumura,Y., 2009. Preventive effect of SA13353 [1-[2-(1-adamantyl)ethyl]-1-pentyl-3-[3-(4-pyridyl)propyl]urea], a novel transient receptor potential vanilloid 1 agonist,on ischemia/reperfusion-induced renal injury in rats. J. Pharmacol. Exp. Ther.329, 202–209.

Wang, L., Wang, D.H., 2005. TRPV1 gene knockout impairs post-ischemic recoveryin isolated perfused heart in mice. Circulation 112, 3617–3623.

Wang, M., Ji, P., Wang, R., Zhao, L., Xia, Z., 2012. TRPV1 agonist capsaicin attenuateslung ischemia-reperfusion injury in rabbits. J. Surg. Res. 173, 153–160.

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–22 21

Wharton, J., Gulbenkian, S., Mulderry, P.K., Ghatei, M.A., McGregor, G.P., Bloom, S.R.,Polak, J.M., 1986. Capsaicin induces a depletion of calcitonin gene-relatedpeptide (CGRP)-immunoreactive nerves in the cardiovascular system of theguinea pig and rat. J. Auton. Nerv. Syst. 16, 289–309.

Yang, J.H., Zhang, Y.Q., Guo, Z., 2011. Endogenous CGRP protects retinal cells againststress induced apoptosis in rats. Neurosci. Lett. 501, 83–85.

Zimov, S., Yazulla, S., 2007. Vanilloid receptor 1 (TRPV1/VR1) co-localizes with fatty acidamide hydrolase (FAAH) in retinal amacrine cells. Vis. Neurosci. 24, 581–591.

K. Sakamoto et al. / European Journal of Pharmacology 733 (2014) 13–2222