J Nutr Sci Vitaminol, 65, 66–71, 2019

Auraptene/-Rich Juice of kawachiensis (Kawachi Bankan) Prevents Ischemia-Induced Neuronal Cell Death in Mouse Brain through Anti-Inflammatory Responses

Satoshi Okuyama1, Mako Katoh1, Tomoko Kanzaki1, Yoshimi Kotani1, Yoshiaki Amakura2, Morio Yoshimura2, Naohiro Fukuda3, Takahisa Tamai3, Atsushi Sawamoto1, Mitsunari Nakajima1 and Yoshiko Furukawa1,*

1 Department of Pharmaceutical Pharmacology, and 2 Department of Pharmacognosy, College of Pharmaceutical Sciences, Matsuyama University, 4–2 Bunkyo-cho, Matsuyama, Ehime 790–8578, Japan 3 Ehime Institute of Industrial Technology, 487–2 Kumekubota, Matsuyama, Ehime 791–1101, Japan (Received July 23, 2018)

Summary Cerebral ischemia/reperfusion leads to delayed neuronal cell death, result- ing in brain damage. Auraptene (AUR) and naringin (NGIN), which exert neuroprotective effects in ischemic brain, are abundant in the of Citrus kawachiensis. Although parts of AUR/NGIN are transited from the peel to the juice during the squeezing of this fruit, these amounts in juice might be too low to exert effects. We thus prepared the AUR/NGIN-rich fruit juice of C. kawachiensis by addition of peel paste to the raw juice. The present study revealed that orally administration of the dried powder of this AUR/NGIN-rich fruit juice (2.5 g/kg/d) for 7 d to ischemic mice significantly suppressed the ischemia-induced neuro- nal cell death in the hippocampus, which was coincidently with the reduction of hyper- activation of and astrocytes. These results suggest that AUR/NGIN-rich juice of C. kawachiensis may possess therapeutic potential for the prevention of neurodegenerative diseases via inhibition of inflammatory processes. Key Words ischemic brain injury, anti-inflammation, auraptene, naringin, fruit juice

Our studies have previously shown that the peel of peel powder of C. kawachiensis (the content of AUR, Citrus kawachiensis (Kawachi Bankan) is exceptionally 4.0760.033 mg/g; of HMF, 0.2760.0039 mg/g; and abundant in auraptene (AUR) compared with that of of NGIN, 44.0260.491 mg/g) and orally (p.o.) admin- other citrus (1). The amount of AUR in C. kawachiensis istered it at the dose of 1.2 or 2.4 g/kg/d to lipopolysac- was about four times more than that in Citrus hassaku charide (LPS)-induced systemic inflammatory model (Hassaku) which was known until then to be rich in mice (4) and a transient global cerebral ischemia mouse AUR among citrus . AUR, a citrus coumarin deriv- model (18). As a result, we succeeded in revealing ative, was reported to have potent anti-cancer and anti- that p.o. administration of the dried peel powder of C. inflammatory activities in peripheral tissues (2). Our kawachiensis exerts potent neuroprotective activities in studies also revealed that it exerts anti-inflammatory the brain of these model mice (4, 18). C. kawachiensis effects not only in peripheral tissues but also in the cen- is a citrus fruit that is mainly eaten as raw fruit and tral nervous system (CNS) (3–6). This peel is also a rich drunk as a juice. During the squeezing of this fruit to source of 3,5,6,7,8,3′,4′-heptametoxyflavone (HMF) obtain the juice, parts of some bioactive constituents and naringin (NGIN). HMF, a citrus polymethoxyfla- including AUR and NGIN are transited from the peel vone, was reported to have both anti-inflammatory (7) to the juice. When we prepared the dried juice powder and immune-modulatory (8) effects in peripheral tis- of C. kawachiensis, the amounts of AUR and NGIN in sues. We also showed that it has the neuroprotective the dried juice powder were 0.3260.0003 mg/g and effects in the CNS which result from its ability to sup- 1.5860.0285 mg/g, respectively, and the HMF content press inflammation (9) and to increase the expression of was below the detection limit (4). We thus prepared test brain-derived neurotrophic factor (BDNF) in the hippo- juice enriched in AUR/NGIN by adding peel paste to the campus (10–13). As for NGIN, which is a bioflavonoid raw juice. The aim of this study was to assess whether commonly found in citrus fruit, it exerts its neuropro- this AUR/NGIN-rich test juice would have a neuropro- tective effects against various brain injuries by reduc- tective effect against the damage caused to the mouse ing oxidation- and inflammation-mediated alterations brain by transient cerebral global ischemia. (14–17). As a model animal of transient cerebral global isch- On the basis of these findings, we prepared dried emia, C57BL/6 strain mice were subjected to bilateral common carotid occlusion (2-vessel occlusion: 2VO), * To whom correspondence should be addressed. because this strain is most susceptible to cerebral isch- E-mail: [email protected] emia following 2VO among seven common mouse

66 Neuroprotective Effect of Fruit Juice of Citrus kawachiensis 67 strains (C57BL/6, ICR, BALB/c, C3H, CBA, ddY and counterstained with 0.1% cresyl violet (Nissl staining). DBA/2) (19). The stained sections (more than two sections per each experimental mice) were examined under an optical MATERIALS AND METHODS microscope (CX21; Olympus, Tokyo, Japan). For microg- Preparation of AUR/NGIN-rich fruit juice. Fruits of lial staining for optical microscopy, the primary anti- C. kawachiensis were harvested in Yawatahama (Ehime, body was a rabbit polyclonal antibody against ionized Japan) and squeezed to obtain the juice, to which was calcium-binding adaptor molecule 1 (Iba1, 1 : 1,000; added the pasty peel residue (1/5 volume of juice) in Wako Pure Chemical Industries, Ltd., Osaka, Japan) and order to increase the AUR/NGIN content. This fruit juice the secondary antibody was EnVision-plus system HRP- enriched in AUR/NGIN was then freeze-dried. This dried labeled polymer (anti-rabbit; Dako). Images captured test sample was suspended in distilled water for p.o. with a microscope (more than two sections per each administration. experimental mice) were analysed by using Image J soft- Animals. Nine-week-old male C57BL/6 strain mice ware (NIH, Bethesda, MD, USA). were purchased from Japan SLC, Inc. (Hamamatsu, In order to stain glial fibrillary acidic protein (GFAP), Japan). All groups of mice were kept at 2361˚C under an intermediate filament protein (50 kDa) of mature a 12-h light/dark cycle (light on 8:00–20:00). During astrocytes, for confocal fluorescence microscopy, mouse the experimental period, the mice were given free access anti-GFAP antibody (1 : 200; Sigma-Aldrich) and to tap water and food until 08:30 and then deprived of Alexa Fluor 568-labeled goat anti-mouse IgG (H1L) food until the time of administration of sample or vehi- was used, as the primary antibody and the secondary cle (16:00). All animal experiments were carried out in antibody, respectively. The mounting medium used was accordance with the Guidelines for Animal Experimen- VECTASHIELD® (Vector Laboratories). Images of the tation specified by the Animal Care and Use Committee hippocampus were captured with a confocal fluores- of Matsuyama University, and the experiments were cence microscopy system (LSM510; Zeiss, Oberkochen, implemented by the approved protocol (No. 9002; 2, Germany). More than two sections were taken from September 2009). each experimental mice to analyze their intensities of Procedures for ischemic surgery. On Day 5 (4 d after immunoreactivity. the start of administration of citrus sample), mice Statistical analysis. Data were expressed as the were subjected to 2VO as previously described (3, 10). mean6SE. Significant differences in values obtained in Throughout surgery, body temperature and brain sur- experiments involving 2 groups were analyzed by Stu- face temperature were maintained at 3760.5˚C and dent’s t-test. Significance was defined asp ,0.05. 36.560.2˚C, respectively. After surgery, all animals RESULTS were placed in a recovery cage under a heat lamp and had free access to drinking water. Effect of the dried AUR/NGIN-rich juice of C. kawachiensis Administration of citrus samples. For experiments, on 2VO-induced neuronal cell death in the hippocampus mice weighing about 25 g were randomized into 3 The hippocampal region is vulnerable to transient groups: sham-control group (Sham; n511), 2VO-con- global cerebral ischemia, and the neurons in it easily trol group (2VO; n514), and 2VO surgery and sample- undergo delayed neuronal cell death within a few days treatment group (2VO1Juice; n514). In the test-sample after the start of ischemia (21). We thus estimated the group (2VO1Juice), mice were p.o. administered sam- effect of the test sample on the neuronal cell death in ple solution (0.75 mL) once a day for 7 d (from Day 1 CA1, CA2, and CA3 regions of the hippocampus at Day to Day 7). Mice in the other 2 groups (Sham and 2VO) 8 (namely 3 d after 2VO) by using immunohistochemi- were administered vehicle (distilled water) in the same cal methods (MAP-2/Nissl staining). Figure 1A shows way. The dose of citrus sample was set at 2.5 g/kg/d. representative photographs of each region of the hip- In this powder, the amounts of AUR, HMF, and NGIN pocampus. In the Sham group, the pyramidal cells with were 0.94660.0096 mg/g, 0.07160.0006 mg/g, and dendrites gave positive staining with both cresyl vio- 3.1760.0449 mg/g, respectively. let (blue signal) and the anti-MAP-2 antibody (brown Immunohistochemistry. At Day 8 (1 d after the last signal). In the 2VO group, the number of intact vivid administration of test sample or vehicle), the mice were neurons with defined cell membranes and nuclei was anesthetized and transcardially perfused with ice-cold apparently reduced. In the 2VO1fortified Juice group, phosphate-buffered saline (PBS). Their brains were then this neuronal loss was effectively suppressed. Figure 1B removed and sagittally sectioned at 30 mm as previously shows the results of a quantitative analysis of the cell reported (3, 20). In order to stain neuronal cell dendrites number of Nissl staining-positive intact cells, indicat- for optical microscopy, the sections were incubated with ing that 1) the number of viable cells in the CA1, CA2, mouse monoclonal antibody against microtubule- and CA3 regions of the 2VO group was significantly associated protein-2 (MAP-2, dilution 1 : 500; Sigma- (*** p,0.001) lower than that of the Sham group; 2) the Aldrich, St. Louis, MO, USA), followed by EnVision- most severe neuronal cell death occurs in the CA2 region plus system HRP-labeled polymer (anti-mouse; Dako, as previously reported (3, 21, 22); 3) the administration Glostrup, Denmark). Immunoreactivity was developed of the dried powder of AUR/NGIN-rich C. kawachiensis and visualized by use of DAB substrate (SK-4100; Vec- juice markedly and significantly (### p,0.001) sup- tor Laboratories, Burlingame, CA, USA). Nuclei were pressed 2VO-induced neuronal loss in the CA1,CA2 and 68 Okuyama S et al.

Fig. 1. Effects of dried AUR/NGIN-rich juice of C. kawachiensis on 2VO-induced neuronal cell loss in the hippocampus. A: Photomicrographs of MAP-2/Nissl staining in the hippocampus. The black scale bar indicates 50 mm. B: Manual cell count data of the Nissl staining-positive intact cells. Values are the means6SE (12–18 sections in each group). Symbols show signifi cant differences between the following groups: Sham vs. 2VO (*** p,0.001), 2VO vs. 2VO1Juice (### p,0.001).

CA3 region. cells indicated the inactive ramifi ed form. Figure 2B Effect of the dried AUR/NGIN-rich juice of C. kawachiensis shows the density of Iba1-positive cells. That of the 2VO on 2VO-induced infl ammation group (5.2260.163/mm2) was about twice that of the It is well known that cerebral ischemia induces a Sham group (2.7860.0646/mm2), and there was a sig- robust neuroinfl ammatory response and oxidative stress nifi cant difference (*** p,0.001) between them. That of that include marked changes in the gene-expression 2VO1Juice group (4.0360.336/mm2) was signifi cantly profi le and phenotype of a variety of endogenous CNS (### p,0.001) lower than that of 2VO group, suggest- cell types (microglia, astrocytes, and neurons) (23, ing that the 2VO-induced activation of microglial cells 24). Regarding the estimation of the anti-infl amma- was suppressed by the dried AUR/NGIN-rich juice of C. tory action of this dried juice enriched in AUR/NGIN kawachiensis. on the 2VO-injured brain, at fi rst we stained microglia We then stained astrocytes in the hippocampal in the hippocampal region of each group of mice with region of each group of mice with antibody against antibody against Iba1 (a microglial marker). Figure GFAP (an astrocytic marker). Representative photo- 2A shows representative photographs of the stratum graphs of the stratum lacunosum-moleculare/stra- lacunosum-moleculare/stratum radiatum in the hip- tum radiatum in the hippocampal region (Fig. 3A) pocampal region. In the Sham group, only a few Iba1- showed that the number and size of GFAP-positive positive cells were observed as being in the ramifi ed cells in the 2VO group was obviously larger than those form (inactivated form). In the 2VO group, the shape for the Sham group or 2VO1Juice group. As shown of Iba1-positive cells changed to the hypertrophied in Fig. 3B, the density of GFAP-positive cells in the form (an activated form called ameboid microglia). In 2VO group (42.1310566.843105/mm2) was signifi - the 2VO1Juice group, the shape of most Iba1-positive cantly (** p,0.01) higher than that in the Sham group Neuroprotective Effect of Fruit Juice of Citrus kawachiensis 69

Fig. 2. Effects of the dried AUR/NGIN-rich juice of C. kawachiensis on 2VO-induced microglial activation. A: Photomicrographs of hippocampus immuno-stained for Iba1. The black scale bar indicates 100 mm. B: Quanti- Fig. 3. Effects of the dried AUR/NGIN-rich juice of C. tative analysis of the densities of Iba1 staining signals. kawachiensis on 2VO-induced astroglial activation. A: Values are the means6SE (12–18 sections for each Photomicrographs of hippocampus immuno-stained for group). Symbols show significant differences between GFAP. The white scale bar indicates 100 mm. B: Quan- the following groups: Sham vs. 2VO (*** p,0.001), titative analysis of the densities of GFAP staining sig- 2VO vs. 2VO1Juice (### p,0.001). nals. Values are the means6SE (12–18 sections in each group). Symbols show significant differences between the following groups: Sham vs. 2VO (** p,0.01), 2VO vs. 2VO1Juice (### p,0.001). (13.6310562.063105/mm2); and this 2VO-induced increase in GFAP-positive cell density was significantly (### p,0.001) suppressed in the 2VO1Juice group (8.37310561.493105/mm2), suggesting that the and MAPK signaling pathways in LPS-induced RAW 2VO-induced activation of astrocytes was suppressed 264.7 macrophages (25). When AUR alone was subcu- by the dried AUR/NGIN-rich juice of C. kawachiensis. taneously (s.c.) injected into the same model of ischemic The above results indicated that the dried AUR/NGIN- mice, the effective dose of AUR was 25 mg/kg/d (3, 5). fortified juice had the ability to suppress the ischemia- The dose of AUR in the present study is 2.37 mg/kg/d, induced inflammation in the CNS. which revealed that about one-tenth dose of AUR was sufficient to exert similar effects, when the dried juice DISCUSSION powder was p.o. administered. When NGIN alone was The data presented here suggested that the dried orally p.o. injected into the same model of ischemic powder of AUR/NGIN-rich C. kawachiensis juice had mice, the effective dose of NGIN was 106 mg/kg/d (18). a neuro-protective effect against ischemic brain via its The dose of NGIN in the present study is 7.93 mg/kg/d, anti-inflammatory action. As for the mechanism of indicating that the intake of juice enriched in AUR/ anti-inflammatory effects of AUR, we previously showed NGIN might be more effective than that the p.o. admin- that AUR suppressed the LPS-induced the expression istration of sole NGIN. These results indicated that the of the mRNA of pro-inflammatory enzyme (COX-2) intake of juice enriched in AUR/NGIN might be more and pro-inflammatory cytokines (IL-1b and TNF-a) in effective than the administration of sole AUR or sole cultured astrocytes (5). As for the mechanism of anti- NGIN. We will investigate in near future why the peel of inflammatory effects of NGIN, it was reported to reduce C. kawachiensis has more potent neuroprotective ability IL-8, MCP-1, and MIP-1a secretion and mRNA expres- than AUR or NGIN alone does. sion, possibly by blocking the activation of the NF-kB Anyway, we are convinced that the peel of C. 70 Okuyama S et al. kawachiensis has neuroprotective ability. Because our the mouse brain. Molecules 20: 20230–20239. recent studies revealed that the experimental diet con- 6) Okuyama S, Semba T, Toyoda N, Epifano F, Genovese taining the peel of C. kawachiensis has the ability 1) to S, Fiorito S, Taddeo VA, Sawamoto A, Nakajima M, ameliorate microglial activation, tau hyperphosphory- Furukawa Y. 2016. Auraptene and other prenyloxy- phenylpropanoids suppress microglial activation and lation, and suppression of neurogenesis in the hippo- dopaminergic neuronal cell death in a lipopolysaccha- campus of senescense-accelerated mice (26); and 2) to ride-induced model of Parkinson’s disease. Int J Mol Sci suppress astroglial activation, tau phosphorylation, and 17: 1716. inhibition of neurogenesis in the hippocampus of type 7) Manthey JA, Bendele P. 2008. Anti-inflammatory 2 diabetic db/db mice (27), in addition to the aforemen- activity of an peel polymethoxylated flavone, tioned papers (4, 18). The accumulation of evidence has 3′,4′,3,5,6,7,8-heptamethoxyflavone, in the rat car- confirmed that inflammation, a physiological response to rageenan/paw edema and mouse lipopolysaccharide- infection/injury, is the key contributor to the pathophys- challenge assays. J Agric Food Chem 56: 9399–9403. iological process of chronic diseases including asthma 8) Nakajima M, Ogawa Y, Amakura Y, Yoshimura M, and various intracerebral diseases such as age-related Okuyama S, Furukawa Y. 2016. 3,5,6,7,8,3′,4′- Hepta- neurological disorders, Alzheimer’s disease, multiple methoxyflavone reduces interleukin-4 production in the spleen cells of mice. Biomed Res 37: 95–99. sclerosis, and brain tumors (28). The present findings 9) Okuyama S, Miyoshi K, Tsumura Y, Amakura Y, taken together with the above findings and the results of Yoshimura M, Yoshida T, Nakajima M, Furukawa Y. a clinical trial (29) indicate that this AUR/NGIN-rich C. 2015. 3,5,6,7,8,3′,4′-Heptamethoxyflavone, a citrus kawachiensis juice might be a functional food. polymethoxylated flavone, attenuates inflammation in the mouse hippocampus. Brain Sci 5: 118–129. Acknowledgments 10) Okuyama S, Shimada N, Kaji M, Morita M, Miyoshi K, We would like to thank Ehime Beverage Inc. (Ehime, Minami S, Amakura Y, Yoshimura M, Yoshida T, Wata- Japan) for supplying juice of C. kawachiensis and its peel nabe S, Nakajima M, Furukawa Y. 2012. Heptame- residue for this study. thoxyflavone, a citrus flavonoid, enhances brain-derived neurotrophic factor production and neurogenesis in Author contributions the hippocampus following cerebral global ischemia in mice. Neurosci Lett 528: 190–195. S.O. and Y.F. conceived and designed the experiments. 11) Okuyama S, Morita M, Miyoshi K, Nishigawa Y, Kaji M, S.O., M.K., T.K., Y.K., A.S., and M.N. performed the Sawamoto A, Terugo T, Toyoda N, Makihata N, Ama­ experiments. Y.A. and M.Y. contributed to the analysis kura Y, Yoshimura M, Nakajima M, Furukawa Y. 2014. of materials. N.F. and T.T. contributed to the preparation 3,5,6,7,8,3′,4′-Heptamethoxyflavone, a citrus flavo- of materials. S.O. and Y.F. wrote the paper. All authors noid, on protection against memory impairment and read and approved the final manuscript. neuronal cell death in a global cerebral ischemia mouse model. Neurochem Int 70: 30–38. Funding 12) Sawamoto A, Okuyama S, Yamamoto K, Amakura This work was supported by the Strategic Research Y, Yoshimura M, Nakajima M, Furukawa Y. 2016. and Developmental Project of Ehime Prefecture. 3,5,6,7,8,3′,4′-Heptamethoxyflavone, a citrus flavo- noid, ameliorates corticosterone-induced depression-like REFERENCES behavior and restores brain-derived neurotrophic factor 1) Amakura Y, Yoshimura M, Ouchi K, Okuyama S, Furu- expression, neurogenesis, and neuroplasticity in the hip- kawa Y, Yoshida T. 2013. Characterization of constitu- pocampus. Molecules 21: 541. ents in peel of Citrus kawachiensis (Kawachibankan). 13) Sawamoto A, Okuyama S, Amakura Y, Yoshimura M, Biosci Biotechnol Biochem 77: 1977–1980. Yamada T, Yokogoshi H, Nakajima M, Furukawa Y. 2) Murakami A, Nakamura Y, Tanaka T, Kawabata K, Taka- 2017. 3,5,6,7,8,3′,4′-Heptamethoxyflavone ameliorates hashi D, Koshimizu K, Ohigashi H. 2000. Suppression depressive-like behavior and hippocampal neurochemi- by citrus auraptene of phorbol ester- and endotoxin- cal changes in chronic unpredictable mild stressed mice induced inflammatory responses: role of attenuation of by regulating the brain-derived neurotrophic factor: leukocyte activation. Carcinogenesis 21: 1843–1850. requirement for ERK activation. Int J Mol Sci 18: 2133. 3) Okuyama S, Minami S, Shimada N, Makihata N, Naka- 14) Gaur V, Aggarwal A, Kumar A. 2009. Protective effect jima M, Furukawa Y. 2013. Anti-inflammatory and of naringin against ischemic reperfusion cerebral injury: neuroprotective effect of auraptene, a citrus coumarin, possible neurobehavioral, biochemical and cellular following cerebral global ischemia in mice. Eur J Phar- alterations in rat brain. Eur J Pharmacol 616: 147–154. macol 699: 118–123. 15) Golechha M, Chaudhry U, Bhatia J, Saluja D, Arya DS. 4) Okuyama S, Yamamoto K, Mori H, Toyoda N, Yoshimura 2011. Naringin protects against kainic acid-induced M, Amakura Y, Yoshida T, Sugawara K, Sudo M, Naka- status epilepticus in rats: Evidence for an antioxidant, jima M, Furukawa Y. 2014. Auraptene in the peels anti-inflammatory and neuroprotective intervention. of Citrus kawachiensis (Kawachi Bankan) ameliorates Biol Pharm Bull 34: 360–365. lipopolysaccharide (LPS)-induced inflammation in the 16) Cui QJ, Wang LY, Wei ZX, Qu WS. 2014. Continual nar- mouse brain. Evid Based Complement Altern Med 2014: ingin treatment benefits the recovery of traumatic brain 408503. injury in rats through reducing oxidative and inflamma- 5) Okuyama S, Morita M, Kaji M, Amakura Y, Yoshimura tory alterations. Neurochem Res 39: 1254–1262. M, Shimamoto K, Ookido Y, Nakajima M, Furukawa Y. 17) Han Y, Su J, Liu X, Zhao Y, Wang C, Li X. 2017. Narin- 2015. Auraptene acts as an anti-inflammatory agent in gin alleviates early brain injury after experimental sub- Neuroprotective Effect of Fruit Juice of Citrus kawachiensis 71

arachnoid hemorrhage by reducing oxidative stress and J. 2017. Astrocyte-derived interleukin-15 exacerbates inhibiting apoptosis. Brain Res Bull 133: 42–50. ischemic brain injury via propagation of cellular immu- 18) Okuyama S, Yamamoto K, Mori H, Sawamoto A, Ama­ nity. Proc Nat Acad Sci USA 114: E396–E405. kura A, Yoshimura M, Sugawara K, Sudo M, Nakajima 25) Liu Y, Su W-W, Wang S, Li P-B. 2012. Naringin inhibits M, Furukawa Y. 2018. Neuroprotective effect of Citrus chemokine production in an LPS‑induced RAW 264.7 kawachiensis (Kawachi Bankan) peels, a rich source macrophage cell line. Mol Med Rep 6: 1343–1350. of naringin, against global cerebral ischemia/reper- 26) Okuyama S, Kotani Y, Yamamoto K, Sawamoto A, Suga- fusion injury in mice. Biosci Biotechnol Biochem 82: wara K, Sudo M, Ohkubo Y, Tamanaha A, Nakajima 1216–1224. M, Furukawa Y. 2018. The peel of Citrus kawachiensis 19) Yang G, Kitagawa K, Matsushita K, Mabuchi T, Yagita Y, (Kawachi Bankan) ameliorates microglial activation, Yanagihara T, Matsumoto M. 1997. C57BL/6 strain is tau hyper-phosphorylation, and suppression of neuro- most susceptible to cerebral ischemia following bilateral genesis in the hippocampus of senescense-accelerated common carotid occlusion among seven mouse strains: mice. Biosci Biotechnol Biochem 82: 869–878. selective neuronal death in the murine transient fore- 27) Okuyama S, Shinoka W, Nakamura K, Kotani M, Sawa- brain ischemia. Brain Res 752: 209–218. moto A, Sugawara K, Sudo M, Nakajima M, Furu- 20) Okuyama S, Makihata N, Yoshimura M, Amakura Y, kawa Y. 2018. Suppressive effects of the peel of Citrus Yoshida T, Nakajima M, Furukawa Y. 2013. Oenothein B kawachiensis (Kawachi Bankan) on astroglial activation, suppresses lipopolysaccharide (LPS)-induced inflamma- tau phosphorylation, and inhibition of neurogenesis in tion in the mouse brain. Int J Mol Sci 14: 9767–9778. the hippocampus of type 2 diabetic db/db mice. Biosci 21) Olsson T, Wieloch T, Smith ML. 2003. Brain damage Biotechnol Biochem 82: 1384–1395. in a mouse model of global cerebral ischemia. Effect of 28) Bramlett HM, Dietrich WD. 2004. Pathophysiology of NMDA receptor blockade. Brain Res 982: 260–269. cerebral ischemia and brain trauma: Similarities and 22) Cho KO, Kim SK, Cho YJ, Sung KW, Kim SY. 2007. differences. J Cereb Blood Flow Metab 24: 133–150. Regional differences in the neuroprotective effect of 29) Igase M, Okada Y, Ochi M, Igase K, Ochi H, Okuyama minocycline in a mouse model of global forebrain ische­ S, Furukawa Y, Ohyagi Y. 2017. Auraptene in the peels mia. Life Sci 80: 2030–2035. of Citrus kawachiensis (Kawachibankan) contributes to 23) Weinstein JR, Koerner IP, Möller T. 2010. Microglia in the preservation of cognitive function: a randomized, ischemic brain injury. Future Neurol 5: 227–246. placebo-controlled, double-blind study in healthy volun- 24) Li M, Li Z, Yao Y, Jin W-N, Wood K, Liu Q, Shi F-D, Hao teers. J Prev Alz Dis 5: 197–201.