Involvement of GABAA Receptors in the Neuroprotective Effect of Theanine on Focal Cerebral Ischemia in Mice

Home , GABA reuptake inhibitor, Theanine

Nobuaki Egashira1,2, Kazuhide Hayakawa1, Megumi Osajima1, Kenichi Mishima1, Katsunori Iwasaki1, Rhozo Oishi2, and Michihiro Fujiwara1,* 1Department of Neuropharmacology, Faculty of Pharmaceutical Sciences, Fukuoka University, Fukuoka 814-0180, Japan 2Department of Pharmacy, Kyushu University Hospital, Fukuoka 812-8582, Japan

Received May 21, 2007; Accepted August 10, 2007

Abstract. We investigated the involvement of γ-aminobutyric acidA (GABAA) receptors in the neuroprotective effect of γ-glutamylethylamide (theanine), a component of Japanese green tea, following a 4-h middle cerebral artery (MCA) occlusion in mice. Theanine (1 mg/kg) reduced the size of the cerebral infarct and alterations of NeuN, GFAP, and Iba 1 expression levels at 24 h after MCA occlusion. This neuroprotective effect of theanine was prevented by bicuculline (GABAA-receptor antagonist, 10 mg/kg) but not 3-mercaptopropionic acid (glutamate decarboxylase inhibitor). These results suggest that the neuroprotective effect of theanine is mediated, at least in part, by GABAA receptors.

Keywords: theanine, focal cerebral ischemia, γ-aminobutyric acidA (GABAA) receptor

γ-Glutamylethylamide (theanine), a component of located postsynaptically and mediate the majority of Japanese green tea (Camellia sinensis), is a natural inhibitory synaptic transmission in the central nervous glutamate analog. We previously demonstrated that system (CNS). GABA synthesis and GABAA-receptor theanine reduced the size of cerebral infarcts following expression are reduced following occlusion of the MCA, occlusion of the middle cerebral artery (MCA) for 4 h suggesting that these receptors play a role in cerebral in mice (1). We showed that theanine did not affect ischemia (4, 5). It has been shown that tiagabine, a cerebral blood flow (CBF), brain temperature, or physio- GABA-reuptake inhibitor, and muscimol, a GABAA- logical variables such as pH, pCO2, pO2, and hematocrit receptor agonist, have a neuroprotective action on of plasma. Moreover, intracerebroventricular injection cerebral ischemia in rats (6, 7). Moreover, the of theanine has been shown to prevent cerebral global anesthetics propofol and isoflurane have a protective ischemia-induced delayed neuronal death in the hippo- effect on cerebral ischemia, which can be reversed by campal CA1 region of gerbils (2). These findings sug- bicuculline, a GABAA-receptor antagonist (8, 9). These gest that theanine directly provides neuroprotection findings suggest that activation of the GABAA receptor against cerebral ischemia. might play a role in inhibiting neuronal death induced Following an intraperitoneal (i.p.) injection of by cerebral ischemia. However, the mechanism by theanine into mice, 14C-labeled theanine was incorpo- which theanine exerts its neuroprotective action remains rated into the brain, without metabolic change, and unknown. Therefore, we investigated whether the resulted in elevation of intracerebral levels of γ-amino- GABAA receptor was involved in neuroprotection by butyric acid (GABA) within 30 min (3), suggesting that theanine following ischemic brain damage in the MCA theanine is transported into the brain through the blood- occlusion mouse model. brain barrier (BBB) and alter intracerebral GABA levels. Male ddY mice (25 – 35 g; Kiwa Experimental GABA is the principal inhibitory neurotransmitter in Animal Laboratory, Wakayama) were kept under a 12-h the mammalian brain. GABAA receptors are mainly light-dark cycle (light on 7:00 to 19:00) in an air-condi- tioned room (23 ± 2°C) with food (CE-2; Clea Japan, *Corresponding author. [email protected] Tokyo) and water available ad libitum. All procedures Published online in J-STAGE: regarding animal care and use were performed in doi: 10.1254/jphs.SCZ070901 compliance with the regulations established by the

1 2 N Egashira et al

Experimental Animal Care and Use Committee of Fukuoka University. Focal cerebral ischemia was induced according to the method described in our previous study (1). The mice were anesthetized with 2% halothane and maintained thereafter by 1% halothane (Flosen; Takeda Chemical Industries, Osaka). After a midline neck incision, the left common and external carotid arteries were isolated and ligated. Nylon monofilament (8-0; Ethilon, Johnson & Johnson, Tokyo) coated with silicon resin (Xanto- pren; Heleus Dental Material, Osaka) was introduced through a small incision into the common carotid artery and advanced to a position 9-mm distal from the carotid bifurcation for occlusion of the MCA. At 4 h after occlusion, the mice were re-anesthetized with halothane Fig. 1. Effect of theanine on NeuN-, GFAP-, and Iba1-immuno- and reperfusion was established by withdrawal of the labeled cells in the cerebral cortex following MCA occlusion in mice. filament. Theanine (1 mg/kg, i.p.) was injected i.p. 3 h after the occlusion. At 24 h after MCA occlusion, animals were sacrificed Scale bar = 20 µm. by decapitation. The brains were removed and sectioned coronally into four 2-mm slices using a mouse brain matrix. Slices were immediately stained with 2% 2,3,5- The results are expressed as the mean ± S.E.M. A triphenyltetrazolium chloride (TTC; Sigma-Aldrich, Tukey-Kramer post-hoc test was performed after one- St. Louis, MO, USA). The border between the infarcted way ANOVA to evaluate any change in the size of and noninfarcted tissue was outlined with an image infarct volume. A P-value of less than 0.05 was analysis system (NIH Image, version 1.62), and the area considered to be significant. of infarction was measured to allow calculation of the The number of NeuN (neuron)- and GFAP (astro- infarction volume. On the other hand, mice were cyte)-positive cells was decreased, while the number of sacrificed by decapitation after perfusion using saline Iba1 (microglia)-positive cells was increased in the and 4% paraformaldehyde. The brains were removed of cerebral cortex at 24 h after MCA occlusion, and fat and water by an auto-degreasing unit (RH-12; Sakura theanine (1 mg/kg, i.p.) inhibited these alterations Seiko Co., Tokyo) and then embedded in paraffin. (Fig. 1). Theanine at the same dose also reduced the Subsequently, 5-µm-thick sections were mounted on size of the infarct volume following MCA occlusion in slides and dried at 37°C for a day. After deparaffiniza- mice [F(3, 33) = 8.492, P<0.001 by one-way ANOVA; tion and rehydration, the sections were incubated in a P<0.01 by Tukey-Kramer post-hoc test, Fig. 2]. Further- 1:100 dilution of rabbit polyclonal anti-GFAP primary more, bicuculline (10 mg/kg, i.p.) prevented the inhibi- antibody (Ventana Japan, Kanagawa), 1:100 dilution of tion of infarct by theanine (P<0.05 by Tukey-Kramer biotinylated mouse anti-NeuN (Chemicon Internatinal, post-hoc test), although when administered alone, Temecula, CA, USA) and a 1:200 dilution of rabbit bicuculline had no effect on the size of infarct. On the polyclonal anti-Iba1 primary antibody (Wako Pure other hand, the glutamate decarboxylase (GAD) inhibi- Chemical Industries Ltd., Osaka) overnight at 4°C. The tor 3-MPA had no effect on the protective effect of sections were then incubated with a 1:200 dilution of theanine [F(3, 20) = 8.150, P<0.01 by one-way goat anti-rabbit IgG-Texas Red secondary antibody ANOVA; P>0.05 by Tukey-Kramer post-hoc test, Fig. 3]. (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) Our findings showed that the neuroprotective effect and UltraAvidin Texas Red (Leinco Technologies Inc., of theanine is mediated, at least in part, by GABAA St. Louis, MO, USA) for 1 h. The coverslips were receptors. In our previous (1) and present studies, mounted and then analyzed by fluorescence microscopy theanine was effective by injection 1 h before reperfu- (software: Leica QFISH, ×400 Leica DMRA; Solms, sion in the MCA model. Moreover, theanine-treatment Deutschland). 30 min before reperfusion is also effective (N Egashira L-Theanine (Tokyo Chemical industries, Tokyo), (−)- et al., unpublished data). However, we found that bicuculline methiodide (Sigma-Aldrich, St. Louis, MO, theanine-treatment immediately after reperfusion had no USA), and 3-mercaptopropionic acid (3-MPA, Sigma- effect on the focal cerebral ischemia in the same model Aldrich) were dissolved in physiological saline and were (N Egashira et al., unpublished data). An i.p. injection of administered i.p. 3 h after occlusion. theanine was reported to increase the intracerebral level Theanine and Focal Cerebral Ischemia 3

Fig. 2. Effects of theanine and bicuculline on infarct volume Fig. 3. Effects of theanine and 3-mercaptopropionic acid (3-MPA) following MCA occlusion in mice. Theanine and bicuculline were on infarct volume following MCA occlusion in mice. Theanine and injected i.p. 3 h after the occlusion. Slices were immediately stained 3-MPA were injected i.p. 3 h after the occlusion. Slices were imme- with 2% 2,3,5-triphenyltetrazolium chloride 24 h after MCA occlu- diately stained with 2% 2,3,5-triphenyltetrazolium chloride 24 h after sion. Values are expressed as the mean ± S.E.M. (n = 6–12). MCA occlusion. Values are expressed as the mean ± S.E.M. (n = 6). **P<0.01, compared with vehicle; #P<0.05, compared with theanine *P<0.05, compared with vehicle (one-way ANOVA followed by alone (one-way ANOVA followed by Tukey-Kramer post-hoc test). Tukey-Kramer post-hoc test). of GABA 30 min after injection of theanine (3). These the neuroprotective effect of theanine on focal cerebral findings suggest that the neuroprotective effect of ischemia. theanine is exerted through an indirect action on GABAA In conclusion, the results of this study imply that receptors during reperfusion. Therefore, theanine could GABAA receptors are involved in the neuroprotective prevent the focal cerebral ischemia by injection at 30 – mechanism of theanine following focal cerebral is- 60 min before reperfusion. On the other hand, theanine chemia. inhibited increase of Iba1 expression levels at 24 h after reperfusion. Iba1 immunoreactive cells rapidly appear at 3.5 h after reperfusion (10). Therefore, the decrease of References Iba1 expression levels is a result of neuroprotection by 1 Egashira N, Hayakawa K, Mishima K, Kimura H, Iwasaki K, theanine. Fujiwara M. Neuroprotective effect of γ-glutamylethylamide In general, glutamic acid is metabolized into GABA (theanine) on cerebral infarction in mice. Neurosci Lett. by GAD. When theanine was administered intragastri- 2004;363:58–61. cally to rats, the concentration of theanine in the brain 2 Kakuda T, Yanase H, Utsunomiya K, Nozawa A, Unno T, was increased, but the concentration of glutamic acid Kataoka K. Protective effect of γ-glutamylethylamide (theanine) in the brain remained unchanged (11). These findings on ischemic delayed neuronal death in gerbils. Neurosci Lett. suggest that theanine increases the intracerebral level of 2000;289:189–192. 3 Kimura R, Murata T. Influence of alkylamides of glutamic acid GABA through a mechanism other than metabolism by and related compounds on the central nervous system I. Central GAD. We also found that the GAD inhibitor 3-MPA had depressant effect of theanine. Chem Pharm Bull. 1971;19:1257– no effect on the protective effect of theanine on focal 1261. cerebral ischemia. Therefore, GAD is not involved in 4 Sieghart W. Structure and pharmacology of gamma-amino- 4 N Egashira et al

butyric acid A receptor subtypes. Pharmacol Rev. 1995;47:181– neuronal damage induced by forebrain ischemia, based on the 234. GABAA receptors. Acta Anaesthesiol Scand. 1999;43:153–162. 5 Sommer C, Kollmar R, Schwab S, Kiessling M, Schäbitz WR. 9 Elsersy H, Mixco J, Sheng H, Pearlstein RD, Warner DS. Exogenous brain-derived neurotrophic factor prevents postis- Selective gamma-aminobutyric acid type A receptor antagonism 3 chemic downregulation of [ H]muscimol binding to GABAA reverses isoflurane ischemic neuroprotection. Anesthesiology. receptors in the cortical penumbra. Mol Brain Res. 2003;111: 2006;105:81–90. 24–30. 10 Ito D, Tanaka K, Suzuki S, Dembo T, Fukuuchi Y. Enhanced 6 Lyden PD, Hedges B. Protective effect of synaptic inhibition expression of Iba1, ionized calcium-binding adapter molecule 1, during cerebral ischemia in rats and rabbits. Stroke. 1992;23: after transient focal cerebral ischemia in rat brain. Stroke. 1463–1469. 2001;32:1208–1215. 7 Yang Y, Li Q, Wang CX, Jeerakathil T, Shuaib A. Dose-depen- 11 Terashima T, Takido J, Yokogoshi H. Time-dependent changes dent neuroprotection with tiagabine in a focal cerebral ischemia of amino acids in the serum, liver, brain and urine of rats model in rat. Neuroreport. 2000;11:2307–2311. administered with theanine. Biosci Biotechnol Biochem. 1999; 8 Ito H, Watanabe Y, Isshiki A, Uchino H. Neuroprotective 63:615–618. properties of propofol and midazolam, but not pentobarbital, on