Late Breaking Abstract

Late Breaking Abstract

7月27日（土）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-509 成体脳特異的Sleepy変異sik3発現マウスの睡眠覚醒行動：新規の誘導性Creマウスを用いた解析 Kanako Iwasaki（岩崎加奈子）¹，Tomoyuki Fujiyama（藤山知之）¹，Seiya Mizuno（水野聖哉）²，Fumihiro Sugiyama（杉山文博）²，Satoru Takahashi（高橋智）²，Hiromasa Funato（船戸弘正）^1,3，Masashi Yanagisawa（柳沢正史）¹ ¹筑波大学　国際総合睡眠医科学研究機構 ²筑波大学　生命科学動物資源センター ³東邦大学　医学部　解剖学講座微細形態学分野 The salt-inducible kinase (SIK) family, a class of the AMP-activated protein kinase (AMPK) related kinases, has 3 paralogs in vertebrate animals. SIK3 is expressed in many organs including liver, skeleton, and brain, and known to be essential for glucose/lipid metabolism and skeletal development. The function of SIK3 in the brain had been virtually unstudied; however, a forward genetic analysis in randomly mutagenized mice recently disclosed the role of SIK3 in regulating sleep amounts. The mutant allele <I>Sleepy</I>, a nucleotide substitution at one of <I>Sik3</I> splice donor sites causing the skipping of exon 13, results in prolonged non-rapid eye movement (NREM) sleep and increased sleep need. The <I>Sleepy</I> mutant mice whose sleep abnormality was investigated in previous studies systemically expressed the <I>Sleepy</I> allele of <I>Sik3</I>. Therefore, it remained to be elucidated whether neurons or other types of cells are responsible for the increased NREM sleep amount. In addition, we could not deny the possibility that <I>Sleepy</I> mutant <I>Sik3</I> expression during the developmental stage decides the time spent in NREM sleep in adult mice. To investigate where and when <I>Sleepy</I> mutant SIK3 increase NREM sleep, spatial and temporal control of the expression is required. Here, we newly developed <I>Synapsin1-CreERT</I>2 mice, in which a tamoxifen inducible <I>Cre</I> sequence was knocked in the <I>Synapsin1</I> locus, to achieve neuronal expression of <I>Sleepy</I> mutant <I>Sik3</I> after the tamoxifen injections. To examine expression patterns of <I>CreERT2</I>, we compared mRNA expression of endogenous <I>synapsin1</I> and knocked-in <I>CreERT2</I> with in situ hybridization. The recombination efficiency was characterized using Cre reporter mice lines. <I>Synapsin1-CreERT2</I> mice were then crossed with the <I>Sik3 Sleepy-flox</I> mice, in which <I>Sik3</I> exon 13 is flanked by loxP sequence to express <I>Sleepy</I> mutant <I>Sik3</I> in a Cre-dependent manner. Sleep/wake behavior was assessed in the <I>Synapsin1-CreERT2; Sleepy-flox</I> mice. These results clarify if <I>Sleepy</I> mutant <I>SIK3</I> increase NREM sleep in neurons of the mice after the tamoxifen injections. In addition, the <I>Synapsin1-CreERT2</I> mouse can be a useful mouse line for Cre-mediated genetic manipulation in the nervous system upon tamoxifen administration.		7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-511 コモンマーモセットにおける明暗サイクルのシフト下での睡眠の解析 Kyoko Matsuzawa（松澤京子）^1,2,4，Daisuke Ono（小野大輔）^1,2,4，Takayuki Yamashita（山下貴之）^1,2,4，Fumitaka Osakada（小坂田文隆）^2,3，Shuntaro Izawa（伊澤俊太郎）^1,2,4，Yasutaka Mukai（向井康敬）^1,2,4，Chowdhury Srikanta（すりかんたちょどり）²，Akihiro Yamanaka（山中章弘）^1,2,4 ¹名古屋大学環境医学研究所　神経性調節学 ²クレスト ³名古屋大学大学院　創薬科学研究科　基盤創薬学 ⁴名古屋大学大学院　医学系研究科　神経性調節学 The common marmoset, <I>Callithrix jacchus</I>, is a diurnal non-human primate. Similarly to humans, common marmosets show monophasic sleep and are rarely active in the dark period. Therefore, unlike to nocturnal animals showing fragmented periods of sleep such as mice, common marmosets can serve as a good experimental model for studying human-like sleep. Here we examined how marmosets are adapted to the light/dark phase shift, either a phase advance by 6 hours or a phase delay by 6 hours, which is reminiscent of jet lag. Using a wireless transmitter implanted under the back skin, we successfully recorded electroencephalogram (EEG), electromyogram (EMG) and body temperature of four marmosets continuously for one month. Their locomotor activity in the home cage was also monitored using an infrared sensor. After recovery from the surgery, these marmosets were active in the day time and showed monophasic sleep, which was accompanied by a rapid shift in body temperature at the onset and offset of the light period. In response to phase delay and phase delay, these marmosets were adapted to the novel light/dark period within approximately 5 days with a gradual shifting of the onset and offset of locomotor activity. We analyzed sleep vigilance states, body temperature and locomotor activity during this adaptation, and found that the time course of adaptation was not constant among these parameters. For example, the locomotor activity was decreased immediately after the offset of the novel light period on the first day of phase delay even though body temperature was still high. Interestingly, rapid eye movement (REM) sleep was specifically elongated in the first half of the dark period after phase delay while non-REM sleep was unchanged. Our results suggest that rapid light/dark phase shifts cause a mismatch between locomotor activity and body temperature and also induce disarrangement of sleep vigilance states at the onset of the dark period.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-513 ランニングホイール回転運動に対するモチベーション維持の脳内メカニズム解明 Yusaku Futami（二見優作），Tong Zhang（張 ?），Satoshi Deyama（出山諭司），Eiichi Hinoi（檜井栄一），Katsuyuki Kaneda（金田勝幸）金沢大医薬保健薬理 Although exercise is important for health in humans, many people hesitate to exercise. On the other hand, it is well known that mice exercise voluntarily; mice are willing to rotate running wheels (RWs) persistently without any rewards. This indicates that there may be neural mechanisms to maintain the motivation for rotating RWs. In the present study, by focusing on this characteristic feature of mice, we investigated neural mechanisms of the maintenance of motivation for exercise. Individual male C57BL/6J mice were trained to rotate RWs voluntarily by giving free access to a RW for 30 min every other day. The number of rotations gradually increased and became stable after 3-4 weeks. Mice were then subjected to pharmacological examinations. Because dopamine (DA) is considered to be involved in motivated behaviors, we firstly examined the effects of DA receptor antagonists on RW rotations. Systemic injection of a D1 receptor antagonist SCH23390 or a D2 receptor antagonist raclopride significantly reduced the number of RW rotations compared with vehicle injection. The open field test (OFT) revealed that both drugs also significantly reduced locomotor activity. These results suggest that the reduced number of rotations by SCH23390 or raclopride were not necessarily caused by decreased motivation for wheel running. We secondly examined the effect of serotonin (5-HT) receptor antagonists on rotation of RWs, given that 5-HT is also suggested to be associated with motivated behaviors. Systemic injection of a 5-HT<SUB>2B/2C</SUB> receptor antagonist SB206553 or a selective 5-HT<SUB>2C</SUB> receptor antagonist SB242084 significantly reduced the number of rotations. Moreover, SB242084 also reduced the total riding time on RWs and the mean rotation speed. Contrary to the effects of the DA receptor antagonists, SB206553 and SB242084 increased locomotor activity measured in the OFT, suggesting that the reduced number of rotations induced by the 5-HT receptor antagonists might not be accounted for by decreased locomotor activity. Taken together, these findings imply that 5-HT<SUB>2C</SUB> receptors may play an important role in the maintenance of motivation for voluntary wheel running in mice.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-514 洞毛運動を指標とした報酬に関連した脳内状態予測 Takashi Nakano（中野高志）¹，Jumpei Ozaki（尾崎隼平）¹，Kohta Mizutani（水谷晃大）^2,3，Takayuki Yamashita（山下貴之）²，Junichiro Yoshimoto（吉本潤一郎）¹ ¹奈良先端大情報 ²名古屋大環境医 ³大阪大院理生物科学 Internal states of the brain can be often reflected as facial expressions. However, how animals show their facial expression is largely unexplored. Here we investigate whether a whisker of mice might convey some information on their internal states related to reward processing. We trained mice for an auditory Go/No-Go task and filmed a whisker during task performance. We found that protraction of the whisker occurred at the reward-associated "Go" cue onset followed by approximately 5-8 Hz periodic whisking which was phase-locked to anticipatory licking during the cue presentation. Such whisker movements rarely occurred in "No-Go" trials or in trials where the mice were not motivated to get a reward. Interestingly, upon acquiring water rewards, the mice showed whisking at around 8-10 Hz while the whisking frequency stayed at the licking-phase-locked frequency (5-8 Hz) when the expected reward was omitted. Moreover, after reward acquisition, the mice often whisked at a more protracted set-point. Using machine learning, we next tested whether these characteristic whisker movements can solely report the internal states of the brain. We trained two binary classifiers: One is for discriminating whether the mice were in reward-anticipating states; and the other is for discriminating whether the mice acquired rewards or not. Leave one-subject out cross-validations revealed that the classifiers accurately indicated reward-anticipating states and reward-acquiring trials only from whisker time plots. Our analyses suggest that mice exhibit stereotypic whisker movements as a part of orofacial movements related to reward anticipation and acquisition.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-515 高プレッシャー状況においてより良い課題遂行を達成するために、腹内側前頭前野は扁桃体の活動を抑制している。 Noriya Watanabe（渡邊言也）^1,2,3,4，Jamil P Bhanji（Bhanji P Jamil）¹，Hiroki C Tanabe（田邊 C 宏樹）³，Mauricio R Delgado（Delgado R Mauricio）¹ ¹Dept Psych, Rutgers Univ, NJ, USA ²高知工科大学　脳コミュニケーション研究センター ³名古屋大学　情報学研究科 ⁴情報通信研究機構　脳情報通信融合研究センター When preparing for a challenging task with high stakes, such as a solo musical performance in front of an audience, potential rewards for success can give rise to physiological arousal that may impair performance. In this case, it is important to control reward-driven arousal while preparing for task execution. We recently examined how the brain controls arousal with a simple time-perception-motor task and simultaneous data acquisition of pupil dynamics and BOLD signal from fMRI (n = 22), and found that performance failure was explained by relatively increased reward representation in the left caudate nucleus and arousal representation in the right amygdala (Watanabe, et al., 2018, Cerebral Cortex). Here we examined how prefrontal cortex influences the amygdala and caudate to control reward-driven arousal. First, ventromedial prefrontal cortex (VMPFC: p < 0.05 FWE correction) exhibited activity that was negatively correlated with trial-wise physiological arousal change, which identified this region as a potential modulator of amygdala and caudate. Next we tested the VMPFC - amygdala - caudate effective network using dynamic causal modeling (Friston et al., 2003). Post-hoc Bayesian model selection (Friston and Penny, 2011) identified a model that best fit data from 2048 possible combinations, in which amygdala activation was suppressed by the VMPFC only in success trials. The ratio of this best and the second best model probability was 87.0 / 5.7 = 15.26, indicating strong evidence for the chosen model being the best explanation. Furthermore, intrinsic connectivity strength from VMPFC to amygdala explained individual task performance (t<SUB>20</SUB> = 5.157, R<SUP>2</SUP> = 0.575, p = 0.0005 Bonferroni correction). These findings highlight the role of effective connectivity from VMPFC to amygdala in order to control arousal during preparation for successful performance.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-516 香りの記憶と呼吸　ー前頭前野との関連ー Yuri Masaoka（政岡ゆり）¹，Keiko Watanabe（渡辺慶子）^1,2，Masahiro Ida（井田正博）³，Masaki Yoshida（吉田正樹）⁴，Nobuyoshi Koiwa（小岩信義）⁵，Akira Yoshikawa（吉川輝）¹，Motoyasu Honma（本間元康）¹，Natsuko Iizuka（飯塚奈都子）^1,2，Kenjirou Ono（小野賢二郎）²，Masahiko Izumizaki（泉崎雅彦）¹ ¹昭和医生体調節 ²昭和医神経内科 ³荏原病院放射線科　総合脳卒中センター ⁴東京慈恵医大院医眼科学 ⁵人間総合科学大人間科学人間科学心身健康科学 Autonomic breathing is not only controlled by metabolic demands but also constantly responds to changes in emotions, such as sadness, happiness, anxiety and fear. Final respiratory output involves a complex interaction between the brainstem and higher centers, including the limbic system and cortical structures. In the limbic area, the amygdala (AMG) and hippocampus (HI) are center of olfactory system, and overlapped with areas related to emotions, memory and respiration. Olfactory information ascend directly to olfactory-related limbic structures, including the piriform cortex (Pir), entorhinal cortex (ENT), AMG, HI, and orbitofrontal cortex (OFC) bypassing the thalamus. This anatomical process in olfaction reflects why olfactory cue rapidly elicit emotions and memory retrieval rather than other sensory modalities. Autobiographical odor memory (AM-odor) accompanied by a sense of realism of a specific memory elicits strong emotions. In this study, we examined the orbitofrontal cortex (OFC), using functional magnetic resonance imaging (fMRI) to determine which OFC subregions are related to AM-odor. Both AM-odor and a control odor successively increased subjective ratings of comfortableness and pleasantness. Importantly, AM-odor also increased arousal levels and the vividness of memories, and was associated with a deep and slow breathing pattern. fMRI analysis indicated robust activation in the left posterior OFC (L-POFC). Connectivity between the POFC and whole brain regions was estimated using psychophysiological interaction analysis (PPI). We detected several trends in connectivity between L-POFC and bilateral precuneus, bilateral rostral dorsal anterior cingulate cortex (rdACC), and left parahippocampus, which will be useful for targeting our hypotheses for future investigations. The slow breathing observed in AM-odor was correlated with rdACC activation. Odor associated with emotionally significant autobiographical memories was accompanied by slow and deep breathing, possibly involving rdACC processing.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-517 中脳ドーパミン神経は報酬に関連した洞毛運動を駆動する Kohta Mizutani（水谷晃大）^1,3，Takashi Nakano（中野高志）²，Takatoshi Hikida（疋田貴俊）¹，Akihiro Yamanaka（山中章弘）³，Junichiro Yoshimoto（吉本潤一郎）²，Takayuki Yamashita（山下貴之）³ ¹大阪大院理生物科学 ²奈良先端大 ³名古屋大環境医 Humans have discrete facial expressions indicating certain emotional states. Recent studies have shown that animals including rodents also have robust body languages correlated with positive and negative behavioral contexts. However, temporal aspects of facial expression of animals remain unclear. In addition, nothing is currently known about the neural mechanisms underlying facial movements reflecting some aspects of internal states. We here repeatedly induced reward-related internal states in mice by training them to learn a simple auditory Go/No-Go task and filmed their face during task performance. We found quantitative patterns of movement sequences of their whisker which include rapid protraction upon reward anticipation and whisking at more protracted set-points after reward acquisition. Whisker movement related to reward anticipation, but not reward acquisition, developed over the course of task learning. Because reward-associated cues would transiently activate dopamine neurons in the ventral tegmental area (VTA), we next tested whether VTA dopamine neurons might be involved in triggering and modulating whisker movements. We found that phasic optogenetic stimulation of dopamine neurons in the VTA drove whisking and whisker protraction. We are now testing whether optogenetic silencing of VTA dopamine neurons attenuates reward-related whisker movements during task performance. Taken together, our preliminary results suggest that reward-related activities of dopamine neurons are reflected to facial movements through induction of whisker movements.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-518 扁桃体中心核・分界条床核へ投射するドパミン作動性ニューロン群の同定 Qin Zhao（Zhao Qin）¹，Ryo Yamamoto（山本亮）¹，Tetsufumi Ito（伊藤哲史）²，Munenori Ono（小野宗範）¹，Nobuo Kato（加藤伸郎）¹ ¹金沢医大院医生理 ²金沢医大院医解剖二 Dopaminergic system in the central nervous system is known to play significant roles in the motor system and affective behaviors. Especially the striatum and the nucleus accumbens receive most dense dopaminergic innervations and the origin of those dopaminergic innervations was revealed as the substantia nigra compacta (SNc) and the ventral tegmental area (VTA), the dopaminergic nucleus A9 and A10, which have been extensively studied. Among the brain region, next to the striatum and the nucleus accumbens, the lateral part of central nucleus of the amygdala (CeL) and the oval nucleus of the bed nucleus of the stria terminalis (BNSTov), the extend amygdala system, receive strong dopaminergic innervations. For long time, it was assumed that the origin of these dopaminergic innervations was SNc and VTA. While, in fact, these regions innervate unto CeL and BNSTov, the dopaminergic neurons in the dorsal raphe (DR) and the periaqueductal gray (PAG) recently have been focused as another source of dopaminergic input unto these regions. To elucidate the origin of dopaminergic projection unto CeL and BNSTov, we injected Fluoro Gold, a retrograde tracer, into the mouse CeA or BNST and examined colocalizations with tyrosine hydroxylase (TH) positive neurons in the DR-PAG area and the SNc-VTA area. First, we found the number of retrogradely stained neurons were more prominent in the DR-PAG area than the SNc-VTA area. Second, in CeA FG injected cases, the retrogradely stained neurons were distributed in the broad DR-PAG area along the rostro-caudal axis, whereas, in BNST FG injected cases, the retrogradely stained neurons were relatively restricted in the caudal part of DR-PAG. These results suggest that (1) DR-PAG dopamine neurons, not SNc-VTA, are the major dopaminergic source for the CeL and BNSTov, and that (2) there are different functional regions in the DR-PAG dopaminergic system along the rostro-caudal axis. Given that the CeL and BNSTov are regarded as the brain regions relating to fear and anxiety, the dopaminergic system in the DR-PAG area must also be involved in controlling the affective behaviors.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-519 Functional Connectivity Neurofeedback for Depression: Variance Matters! Jessica Elizabeth Stewart（Stewart Jessica Elizabeth），Takashi Yamada（Yamada Takashi），Mitsuo Kawato（Kawato Mitsuo） Advanced Telecommunications Research Institute International A recent fMRI real-time functional connectivity neurofeedback (FC-Nef) experiment (partially presented in Yamada et al., 2017) resulted in the successful reduction of participants' depressive symptoms. Functional connectivity (FC) between the left middle frontal gyrus and left precuneus during the resting state has previously been shown to be more negative in healthy than in depressed people (Ichikawa et al., 2017). Therefore, FC between these regions of interest (ROIs) was targeted in this experiment. As expected, the more negative participants' resting state FC became (when pre- versus post- FC-Nef was compared), the more their depressive symptoms were reduced. Surprisingly, changes in average FC during the task itself were not found to relate to changes in depressive symptoms. This begs the question- how specifically did neural activity during FC-Nef training influence depression scores and resting state FC? Here we set out to investigate potential answers to the above question. The data of 14 participants from the aforementioned experiment were re-analyzed. We considered that neural activity within different subsections of the 40s 'induction' (neural activity manipulation) period used on each trial may not have been of equal importance, e.g. participants may have struggled to maintain attention for the whole 40s. We therefore divided the this into smaller time-windows for analysis. However, even changes in average activity in these smaller time-windows did not correlate significantly with changes in depressive symptoms or resting state FC. Interestingly, however, we found that the more participants were able to reduce the variance of their FC in some of these time-windows (over the course of training), the more their depression symptoms were reduced and their resting state FCs became negative. Simple leave-one-subject-out regression analyses confirmed the robustness of these results. These results suggest that a reduction in the variance of FC between these ROIs during FC-Nef training may have resulted in more negative resting state FC and reduced depression scores. While typical FC experiments focus on average FC, our results suggest that the variance in FC may also be of importance. Targeting FC variance in future FC-Nef studies may prove useful. To reduce depression, these results suggest that it may not just be important to have a negative connectivity between these ROIs, but also to be able to maintain this consistently.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-520 不安を介してリスクアセスメント行動を制御する視床下部ニューロン Noriko Horii（堀井謹子）¹，Nomoto Kensaku（野元謙作）²，Kikusui Takefumi（菊水健史）²，Nishi Mayumi（西真弓）¹ ¹奈良医大医第一解剖 ²麻布大獣医伴侶動物 Risk assessment (RA) behavior involves a vigilant exploration of potential threats or uncertainties, which is essential for the survival of animals. In this study, we found that hypothalamic urocortin-3/enkephalin-co-expressing neurons in the perifornical area of the anterior hypothalamus (PeFAH), a brain area that we had recently identified in mice, regulate RA behavior and the associated anxiety evoked by a novel object stimulus. A combined approach of immunohistochemical and fiber photometric experiments revealed that the activity of these neurons was increased during RA of a novel object. Pharmacogenetic activation of these neurons evoked piling of bedding material toward a wall of a homecage, despite the absence of the novel object. Activation of urocortin-3/enkephalin neurons increased both RA behavior to a novel object and burying behavior in marble-burying test, while it had no effects on anxiety level in open field test. In contrast, cell ablation of urocortin-3/enkephalin neurons by selective expression of diphtheria toxin subunit A led to unvigilant behaviors particularly in a homecage, such as excessive contacts with the novel object and gnawing it. When an electrified shock prod was used as a novel objectin a homecage, the probability that mice receive the shock by touching the prod with the nose/the mouth was significantly increased in urocortin-3/enkephalin neurons-ablated group compared with control group. Our findings indicate that hypothalamic perifornical urocortin-3/enkephalin neurons regulate RA behavior to a novel object via anxious emotion and imply that these neuronal peptides and their receptors could be a potential target for anxiety disorder treatments.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-521 ω3多価不飽和脂肪酸摂取による抑うつ様行動の減少には腹側被蓋野－側坐核のドパミン神経系が関与する Eri Takeuchi（竹内絵理）¹，Daisuke Yamada（山田大輔）¹，Akiyoshi Saitoh（斎藤顕宜）²，Masayuki Itoh（伊藤政之）³，Takashi Hayashi（林崇）³，Mitsuhiko Yamada（山田光彦）²，Keiji Wada（和田圭司）¹，Masayuki Sekiguchi（関口正幸）¹ ¹国立精神・神経セ神経研疾病４ ²国立精神・神経セ精神保健研精神薬理 ³国立精神・神経セ神経研病態生化学 It was showed that omega(ω)-3 polyunsaturated fatty acid (PUFA) had beneficial effects on major depression, and that supplementation of ω-3 PUFA exhibited antidepressant-like effects in the rodents. However, the underlying mechanism is unknown. In this study, we examined the involvement of ventral tegmental area (VTA)-nucleus accumbens (NAc) dopaminergic systems in the behavioural changes in mice fed a diet high in ω-3 PUFAs. Mice were fed test diets (a high ω-3 PUFA diet and a control diet) for 6 weeks, after that we performed the forced swim test (FST) on mice. Mice in the high ω-3 diet group showed reduction in immobility time. It suggests that this diet exerts an antidepressant-like action in the FST. The reduction of immobility time was abolished by microinjection of an antagonist for D<SUB>2</SUB>-like receptor into the NAc. An antagonist for D<SUB>1</SUB>-like receptor also abolished the reduction in the FST. Moreover, we observed that the number of tyrosine hydroxylase-positive cells significantly increased in the VTA in mice fed the high ω-3 PUFA diet (<I>p</I> < 0.05, <I>t</I>-test). These results suggest that modulation of the VTA-NAc dopaminergic pathway is one of the mechanisms by which a diet high in ω-3 PUFAs reduces the immobility behaviour in the mouse FST.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-522 強迫的な報酬探索を司る皮質ー線条体投射系の神経可塑性 Masaya Harada（原田征弥）¹，Vincent Pascoli（Pascoli Vincent）¹，Christian Luscher（Luscher Christian）^1,2 ¹Dept of Basic Neurosciences, Faculty of Medicine, University of Geneva ²Clinic of Neurology, Department of Clinical Neurosciences, Geneva University Hospital Drug seeking that continues even when facing negative consequences is a defining feature of addiction. Such compulsive behavior is also observed in a subset of mice trained for optogenetic dopamine neuron self-stimulation (oDASS). However, the neuronal adaptations underlying compulsive behavior or renouncement remain elusive. Here, we developed a mouse model of oDASS seeking, using a heterogeneous seeking-taking schedule. We observed the emgerence of two groups of mice. One, where the animals persevered pressing the seeking lever even when it leads to a punishment in 30 % of the trials (electrical foot shock), the other, where animals suppressed the seeking behavior. Previous work suggests that the dorsal striatum mediates compulsive reward seeking. We therefore quantified the strength of three distinct opsin-tagged cortico-striatal projections ex vivo in acute brain slices. We show that synapses of contro-dorsal striatal (cDS) neurons receiving input from the OFC were potentiated selectively in persevering mice while synapses at mPFC or M1 inputs remained un-altered. Given the spiraling connectivity of striatal circuits, we next characterized the mPFC to ventral striatum pathway. By contrast to the potentation of the OFC to cDS in persevering mice in D1R and D2R-MSNs, we found that mPFC-NAc synapses onto D2R-MSNs were selectively potentiated in renouncing mice. This potentiation may encode aversive events to allow inhibitory control to halt the progression towards compulsive behavior. Since the mPFC-NAc projection is upstream in the spiraling nigro-striatal loop, we will now test for hierarchical organization. Taken together a model of staged forms of drug-evoked synaptic plasticities in direct and indriect pathways is emerging, eventuallaly leading to compulsion in group of addicted animals.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-523 Women with poor social communication skills prefer excessively edited images of themselves Tamami Nakano（Nakano Tamami）^1,2，Yusuke Uesugi（Uesugi Yusuke）² ¹Grad Sch of Frontiers Bioscience, Osaka Univ ²School of Medicine, Osaka University Young women posting edited face photographs of themselves on social networking sites is a popular phenomenon, but an excessively retouched face image sometimes, instead of making the photograph more attractive, gives a strange impression to its viewers. This study investigates what personal characteristics facilitate a bias toward an excessively edited face image. Fifteen young Asian women evaluated the attractiveness and naturalness of their face images, which were edited in eight different levels from mild to excessive by expanding their eyes and thinning their chin. People with higher autism-spectrum quotient (AQ) scores were found to regard their own excessively edited images as more attractive. People with lower self-esteem also preferred higher levels of face edit, and there was a significantly high negative correlation between AQ and self-esteem scores. Moreover, the pupil response for the highly edited self-face images increased with higher AQ scores, but such correlation was not observed for the lightly edited self-face images. This study shows that women with lower AQ scores have lower levels of self-esteem and prefer more extensively edited images of their face, suggesting that people with poor social communication skills are rarely creeped out by excessively retouched faces.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-524 オスマウスにおけるメス個体の社会記憶は交尾によって強化される Akiyuki Watarai（度会晃行），Myung Chung（Chung Myung），Katsutoshi Imanaka（今中克俊），Kentaro Tao（田尾賢太郎），Teruhiro Okuyama（奥山輝大）東京大学定量生命科学研究所行動神経分野 Mice (Mus musculus) naturally spend more time interacting with a novel mouse than a familiar one. So, when social memory is quantified by measuring the interaction durations of a stimulator mouse, the olfactory investigation toward the same mouse is reduced with the formation of social memory. Using this investigation assay for assessing social memory, a large number of studies show various findings about the neural mechanisms of memorizing male individuals in male mice. On the other hand, since male mice can show high level of social interaction with reproductive females even after forming social memories, the neural mechanisms underlying the encoding and retrieval of social memories about females are largely unknown, due to the lack of behavioral assay for females. Here, we succeeded in establishing a novel behavioral assay for estimating the social memory of a female by measuring the number of ultrasonic vocalizations (USVs) generated by males. When male mice encounter a female mouse, they perform the olfactory investigation, sexual behaviors, or emitting USVs to the female. We presented males to a female for five times and another female for once respectively and recorded the vocalizations of males. When we presented males to the female in the five consecutive behavioral sessions, the numbers of USVs to the female declined with encounter frequency. Following the five encounters of the same female, the presentation to another female increased the numbers of USVs of males. The results indicate that the male mice decline the emitting of USVs when he is familiar with a female and emit more USVs to a novel female compare to a familiar female. Therefore, male mice regulate emitting USVs to a female by the social memory of the female. We also investigated the length of the social memory. Generally, the social memory about a male disappears in male mice by 24 hours separation. On the other hand, monogamous prairie voles store the social memory about a mating partner for more than 24 hours. The long-term partner memory in male voles is thought to be formed by copulation. We paired a male mouse with a female and recorded his USVs in the presentation of the mating partner and another female 72 hours after their copulation. Male mice emitted more USVs to the not-mating female than the mating partner even 72 hours after the copulation. This result suggests that the copulation may enable male mice to store the long-term social memory about a mating partner.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-525 Hatanoラットの社会性再認記憶の系統差の検討 Taichi Hatakeyama（畠山太一）^1,2，Kotaro Kawakami（川上光太郎）¹，Toru Okawara（大河原利）¹，Koji Musya（武者晃司）¹，Ryo Ohta（太田亮）³，Maiko Kawaguchi（川口真以子）¹ ¹明治大学農学部 ²研究・知財戦略機構 ³食品薬品安全センター・秦野研究所 Inbred strains of Hatano high- (HAA) and low- (LAA) avoidance rats, derived from the Sprague-Dawley strain, were selectively bred based on their respective avoidance performances in the shuttle-box active avoidance task. In this task, utilization of aversive stimuli as a negative reinforcer can affect the animal's physical, motivational and emotional states, which makes it difficult to determine the specificity of an experimental manipulation. To clarify the behavioral characteristics of the Hatano rats, male HAA and LAA rats were subjected to behavioral tests for the ability to distinguish between novel and familiar stimuli, such as other animals (social recognition test) or objects (spontaneous object recognition test). Thus, the present study investigated the strain differences of HAA and LAA rats in the ability of social recognition and spontaneous object recognition. In the result of social recognition test, HAA, but not LAA rats, spent more time investigating novel animals than familiar animals. In contrast, there was no strain difference between the HAA and LAA rats in the spontaneous object recognition test, and both rats explored equally novel and familiar objects. These results suggest that avoidance performance in the Hatano HAA and LAA rats is almost not resulted in difference of cognitive function. In addition to the fact that the HAA rats showed higher anxiety behavior than LAA, avoidance learning and social recognition are known to be associated with emotion. Therefore, Hatano HAA and LAA rats can be useful models which show clear strain differences in the ability of learning and memory involved with emotional behavior.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-526 Functional segregation along the proximo-distal axis of CA1 Yusuke Ota（Ota Yusuke）^1,2，Marrisa Lafreniere（Lafreniere Marrisa）²，Brian J Wiltgen（Wiltgen Brian J）^1,2 ¹Psychology Dept, University of California, Davis, USA ²Center for Neuroscience, University of California, Davis, USA Recent evidence suggests that some functions of the hippocampus may be segregated along its transverse axis. For instance, the HPC receives spatial input from the medial entorhinal cortex (MEC) and postrhinal cortex while it receives odor/object-related information from the lateral entorhinal cortex (LEC) and perirhinal cortex. Although these inputs are mixed in dentate gyrus and CA3, they remain segregated in the entorhinal cortex projections to CA1. The proximal (next to CA2) and distal (next to subiculum) segments of CA1 receive direct projections from MEC and LEC, respectively. This suggests that information processing may be functionally distinct along the proximo-distal axis of CA1. To test this idea, we quantified c-fos expression in proximal and distal CA1 of mice that were either exposed to a novel context, novel objects in a familiar context, or a familiar context without objects. Consistent with previous reports, we found that novel context exposure increased c-fos activity in proximal CA1 while exposure to novel objects resulted in higher c-fos levels in distal CA1 compared to control mice. Using targeted infusions of halorhodopsin, we are currently in the process of silencing distinct regions during testing of spatial and object-related memory tasks. We predict that silencing proximal CA1 will impair contextual memory retrieval of a familiar environment. In contrast, we predict that silencing distal CA1 will impair performance on novel object location recognition task. This work will further improve our current framework of how the hippocampus forms and retrieves different types of memories.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-527 高脂肪食摂取マウスへの末梢オキシトシン投与による社会的、物体認知機能改善効果 Ryotaro Hayashi（林遼太郎）^1,2，Yoshiyuki Kasahara（笠原好之）³，Shizu Hidema（日出間志寿）⁵，Satoshi Fukumitsu（福光聡）¹，Kiyotaka Nakagawa（仲川清隆）²，Katsuhiko Nishimori（西森克彦）⁴ ¹日本製粉(株)イノベーションセンター ²東北大学大学院農学研究科　機能分子解析学分野 ³東北大学大学院医学系研究科　障害科学専攻　機能医科学講座融合医工学分野 ⁴福島県立医科大学　肥満・体内炎症解析研究講座 ⁵福島県立医科大学　病態制御薬理医学講座 [Background] Excessive intake of fat is a major risk factor for lifestyle-related diseases such as heart disease and furthermore intake of a high-fat diet also affects brain function such as object cognitive function and depression-like behavior. Social recognition is one of the recognition memory and information of "Who you are". Autism patients is in communication disorder and their social memory is decreased. It has been reported that high-fat diet reduces social behavior, but there are few reported cases and the mechanism is unknown. In this study, we analyzed the effect of high-fat diet on social memory and analyzed its mechanism. In addition, we examined the effect of peripheral oxytocin (OXT) administration to recognition memory in high fat diet fed mice. [Method] Six week old C57BL6 / J male mice were treated with a 60 kcal% high-fat diet and a control diet for 10 weeks. Three treatment groups were set up: a control diet + saline administration group, a high fat diet + saline administration group, and a high fat diet + OXT administration group. Recognition memory was evaluated by the object recognition test, object location test and three chamber test. Saline and OXT was intraperitoneally administered at 1 mg / kg BW 45 minutes before each test. Sixty minutes after the three chamber test, mice were sacrificed and brain samples were obtained and subjected to qPCR. [Result] Decreased social recognition, object recognition and object location memory were observed in the high fat diet + saline administration group compared to the control diet group. In addition, recovery of social recognition, object recognition and object location memory were observed by administration of OXT to mice fed high-fat diet. Hippocampal OXT receptor (OXTR) and c-fos expression were reduced in high fat diet fed mice compared to normal-fed mice. In OXT-treated mice, expression of OXT was elevated in the hypothalamus. [Discussion] Our data suggest that oxytocin peripheral administration restores social recognition, object recognition and object location memory in high fat diet fed mice. Decreased cognitive function by high-fat intake was considered to be one of the causes of decreased OXTR and c-fos expression in the hippocampus. The improvement of cognitive function by the administration of OXT could be due to the OXT / OXTR signal by the upregulation of OXT gene in hypothalamus.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-528 ショウジョウバエLIMホメオドメイン型転写因子Apterousは神経ペプチド分泌ニューロンにおけるCl<SUP>-</SUP>流入を制御することで長期記憶を固定化する Sho Inami（井並頌），Shou Inami（井並頌），Tomohito Sato（佐藤智人），Shoma Sato（佐藤翔馬），Naoto Shimada（嶋田直人），Takaomi Sakai（坂井貴臣）首都大学東京, 理学部, 生命科学専攻 New protein synthesis is essential for memory consolidation which is a process required for converting short-term memory (STM) into long-term memory (LTM). Several transcription factors relevant to memory consolidation have been identified. Genetic tools available in <I>Drosophila</I> are useful to identify molecular mechanisms involved in memory processes. Previously, using Ap::GFP, a GFP fusion knock-in allele of <I>ap</I> (<I>ap::GFP</I>), we reported that the LIM homeodomain transcriptional factor Apterous (Ap) are expressed in the Kenyon cells in the mushroom bodies (MBs) and the pigment dispersing factor (Pdf), neuropeptide expressing neurons. To measure <I>Drosophila</I> memory, we used courtship conditioning paradigm, in which previous experience with mated females causes males to reduce their courtship toward virgin females. Conditional knockdown of <I>ap</I> revealed that <I>ap</I> in Pdf neurons is necessary for memory consolidation while <I>ap</I> in MB α/β neurons was required for LTM maintenance. In addition, knockdown of <I>Chi</I> in MB α/β neurons induces defective LTM, suggesting transcriptional regulation of AP/Chi complex is necessary for LTM maintenance in MB α/β neurons. In contrast, knockdown of <I>Chi</I> in Pdf neurons did not impair LTM. Thus, it is possible that AP-dependent memory consolidation is independent from the transcriptional regulation of AP/Chi complex. Furthermore, it is reported that the Resistant to dieldrin (Rdl), GABA<SUB>A</SUB> receptor subunit, are expressed in Pdf neurons and regulate fly behavior. In our <I>ex vivo</I> imaging study using SuperClomeleon, a FRET-based sensor for Cl<SUP>-</SUP> fluxes, we have found that Ap regulate Cl<SUP>-</SUP> influx in Pdf neurons. GABA treatment under <I>ap</I> null mutant background caused significant increasement of Cl<SUP>-</SUP> influx in Pdf neurons. Furthermore, knockdown of Rdl in Pdf neurons under <I>ap</I> null mutant background restored LTM defect, suggesting that Ap may negatively regulate Rdl-dependent Cl<SUP>-</SUP> influx and following signals for memory consolidation. All in all, our results show that Ap in MB α/β neurons and Pdf neurons has different molecular functions to establish and maintain LTM. In Pdf neurons, Cl<SUP>-</SUP> influx regulated by Ap has an important role for memory consolidation.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-529 海馬場所細胞による新しい環境情報の既存の空間地図への組み込み Saichiro Yagi（八木佐一郎）¹，Yuji Ikegaya（池谷裕二）¹，Takuya Sasaki（佐々木拓哉）^1,2 ¹東京大院薬薬品作用 ²JSTさきがけ Hippocampal place cells selectively fire when an animal enters into specific areas, termed as place fields. When animals learn a novel environment, place fields rapidly emerge at new locations within minutes. In most papers, the new emergence of place fields has been studied from novel conditions that are completely independent from animal's previous experiences. However, there is also a condition in which animals explore in an environment of which some surrounding objects and shapes are somewhat similar to those in their previous experiences, here termed a familiar-novel condition. In this case, the hippocampal circuit might utilize pre-existing spatial maps to form new spatial maps when exposed to a familiar-novel condition. In addition, such familiar-novel conditions may induce the remapping of place cells, in which place cells adaptively alter their place fields in response to changes in external contexts. To address this issue, we recorded spiking activity of dorsal hippocampal CA1 neurons as rats run in a familiar U-shaped track (track 1). Subsequently, the rats are exposed to a novel U-shaped track with the same size and shapes but different visual cues (track 2). In track 2, some cells newly exhibited place fields whereas the other cells retained their place fields at the same locations, compared with track 1. These results suggest that pre-existing and new spatial information is cooperatively represented within a single hippocampal spatial map. Further, we analyzed theta phase precession, in which spikes of place cells occur at progressively earlier phases of theta oscillations. In track 2, the place cells, irrespective of the presence of their place fields in track 1, exhibited apparent theta phase precession. Such phase precession emerged at the beginning of periods when the animals started to run on track 2. As theta phase precession is a basis of theta sequence, in which place cell populations sequentially fire at a specific theta phase, our results suggest that place cells in a familiar-novel condition are rapidly incorporated into an organized theta sequence. Our results show that pre-existing and new spatial information are combined by the hippocampal circuits at the early phase of learning.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-530 光依存的な長期記憶維持にかかわる光受容体の同定 Riho Hashimoto（橋本莉穂），Takaomi Sakai（坂井貴臣），Shou Inami（井並頌）首都大学東京　理学研究科　生命科学専攻 Environmental light can modulate animal physiology and behavior in nature. We previously reported that light is necessary for long-term memory (LTM) maintenance in the fruitfly Drosophila melanogaster (The 41th Annual Meeting of the Japan Neuroscience Society). To identify photoreceptors involved in the light-dependent LTM maintenance, courtship conditioning paradigm, in which previous experience with mated females causes males to reduce their courtship toward virgin females, was used. We have identified that two brain photoreceptors, Cryptocrome (Cry) and Rhodopsin7 (Rh7), are important for LTM maintenance. Cry is a blue light-sensing photoreceptor, while Rh7is a UV light-sensing photoreceptor. In this study, we examined whether blue and UV lights are involved in LTM maintenance. When flies were kept under blue and UV lights (450 nm and 405nm, respectively) after conditioning, flies showed LTM for at least 5days, indicating that blue and UV lights contribute to LTM maintenance. These results support our idea that brain photoreceptors regulate light-dependent LTM maintenance in Drosophila. Drosophila has three light-sensing organs: the compound eyes, ocelli, and Hofbauer-Buchner (H-B) eyelets. However, it remains unclear whether these organs are also involved in LTM maintenance. Since Rhodopsin 6 (Rh6) is expressed in the H-B eyelets and axons of H-B eyelet photoreceptors project to the Cry- and Rh7-expressing circadian pacemaker neurons, we focused on Rh6 in this study. Electrical silencing of Rh6-expressing neurons impaired 5d memory after conditioning suggesting that Rh6 is also involved in LTM maintenance.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-531 加齢・空間認識記憶とカルモジュリンキナーゼII活性 Yoko Yamagata（山肩葉子）^1,2，Yuchio Yanagawa（柳川右千夫）³，Keiji Imoto（井本敬二）¹ ¹生理研 ²総研大院生命科学生理 ³群馬大院医遺伝発達行動 Hippocampus-dependent spatial memory declines steadily with aging. However, what kinds of molecules are involved in such a decline of cognitive function is still not clear. Ca<SUP>2+</SUP>/calmodulin-dependent protein kinase IIα (CaMKIIα) is an essential molecule in the brain for the acquisition of hippocampus-dependent memory. Indeed, we showed that homozygous mutants of the kinase-dead CaMKIIα knock-in mouse are severely impaired in hippocampus-dependent spatial and contextual memory. In addition, by using heterozygous mutants, which retain a half of the wild-type level of kinase activity of CaMKIIα, we found that older heterozygotes showed severe impairments in spatial memory, whereas young heterozygotes performed just as well as the wild-type littermate control mice. Combined with previous reports indicating reduced CaMKII activity in aged animals, our results indicated that a reduction in kinase activity of CaMKII might be responsible for age-related declines of hippocampal function. To start systematic analysis of aging, spatial memory and CaMKII activity, we first prepared 1-y-old wild-type mice and control young adult mice of 3-4 m of age, and subjected them in parallel to the Morris water maze tasks. In hippocampus-dependent hidden platform training, 1-y-old mice took longer time to reach the platform than young mice did. They eventually showed spatial memory after 6-day training, but not after 3-day training. On the other hand, young control mice showed clear spatial memory after 3-day training. Besides, spatial memory of 1-y-old mice was not so accurate as that of young control mice even after 6-day training. These results indicate that extensive training may compensate for spatial memory deficits in aged mice, but not be sufficient to acquire accurate spatial memory. We are planning to compare kinase activity of CaMKII in the hippocampus taken from these mice.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-532 古典的条件付けによる行動の変化を担う可塑的変化は摂食コマンドニューロンで起こる Akira Sakurai（櫻井晃）^1,2，Hiroaki Kojima（小嶋寛明）³，Motojirou Yoshihara（吉原基二郎）^1,2 ¹情報通信研未来ICT記憶PJ ²The Picower Inst. for Learning and Memory, MIT, Cambridge, USA, ³情報通信研未来ICT生体物性PJ Classical conditioning demonstrated by Pavlov is a well-defined form of memory formation. To investigate neurophysiological mechanisms underlying Pavlovian conditioning, we identified a pair of command neurons that control feeding behavior ("Feeding neuron") in <I>Drosophila</I> brain. The Feeding neuron functions downstream of sensory and metabolic cues and reside at the top of feeding motor programs. Thus, we can expect that plastic changes in the Feeding neuron will be correlated with behavioral changes induced through Pavlovian conditioning. We have now established a novel conditioning protocol to associate somatosensory stimuli (CS) with a feeding behavior (proboscis extension) induced by sucrose stimulation (US). After repeated pairing of the CS and US, flies extended their proboscis in response to the CS alone. In calcium imaging experiments after conditioning, we found that the CS alone activated the Feeding neuron whereas it did not before the conditioning. These results suggest that a new connection from the CS-conveying circuit to the feeding circuit was created at the Feeding neuron and/or upstream of the Feeding neuron. To pinpoint where the new connection is generated, we inactivated the Feeding neuron through photoactivation of halorhodopsin during conditioning. Observation of behavior after conditioning demonstrated that it led to a large suppression of the conditioned response. These results are consistent with an assumption that the new connection responsible for the conditioned response was formed, at least partly, on the Feeding neuron. Our results open an avenue for the systematic analyses of the molecular and cellular basis of changes in circuit connections underlying memory formation at the single cell level.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-533 系列タッピング課題による昼間睡眠に依存した運動性技能の向上 Mako Ogawa（小川茉子），Takeshi Aihara（相原威），Hiroshi Sasaki（佐々木寛）玉川大学大学院　工学研究科 Daytime nap-dependent improvement in motor skill performance was studied by using the sequential finger tapping task which performance was reported to be improved due to overnight sleep (Walker et al., 2002). 9 healthy subjects, 20 to 22 years of age, were participated in this study. Subjects were trained over 12 times of 30 seconds trials at 11 a.m. and were tested by 2 times of 30 seconds trials just before daytime nap period started at 2 p.m.. After the 2 hours of daytime nap, subjects were retested by 2 times of 30 seconds trials at 4 p.m.. The performance in test phase was estimated by averaged number of successfully performed finger tapping sequences over 2 trials in each test. As a result, the performance in the test before daytime nap was 19.6 whereas that in the test after daytime nap was 21.2. Statistical analysis by paired t-test showed that the difference in performance between 2 test phases was significant (p<0.05), indicating that daytime nap might improve motor skill performance in the sequential finger tapping task. Furthermore, comparing to the overnight effects in Walker's report, the degree of improvement due to daytime nap was less. Mednick reported that even the brief 60 minutes nap containing both SWS and REM sleep similarly improved texture discrimination task performance, which is used to study the perceptual skill learning, compared to overnight sleep effects (Mednick et al., 2003). Taken together, the daytime nap partially improved motor skill performance and the degree of daytime nap-dependent improvement might be task dependent.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-534 加齢性長期記憶障害の原因となる記憶過程の特定 Nozomi Uemura（上村望），Motomi Matsuno（松野元美），Junjiro Horiuchi（堀内純二郎），Minoru Saitoe（齊藤実）東京都医学総合研究所 As animals age, they suffer a decrease in memory, known as age-related memory impairment (AMI). In Drosophila, two types of memory, middle-term memory and long-term memory (LTM), are affected by aging. In this study, we characterize age-related impairments in LTM. LTM requires long-term modification of specific neuronal networks, referred to as memory engrams. These modifications require new transcription and translation, and should be maintained for as long as the memory is retained. In addition, modified networks are thought to encode LTMs, and should be reactivated upon memory recall. The transcription factors, CREB and c-FOS are necessary for proper LTM, and in particular, induction of c-FOS has been used to identify and label memory engram cells. While c-FOS-labeled engram cells have been studied in young animals, they have not been characterized in aged animals. In this study, we are studying engram cells in old Drosophila to determine whether they have defects in formation, maintenance, or reactivation. To examine formation of engram cells, we examined c-FOS-positive cells in brains from young and old Drosophila that were trained to form associations between an odor and painful electrical shocks. We found that training-dependent increases in c-FOS positive cells were the same in young and old flies, suggesting that formation of LTM and memory engrams is relatively normal in old flies. We next examined whether engram cells are reactivated upon memory recall upon subsequent odor exposure. We measured neuronal activity upon odor exposure as an increase in ERK phosphorylation, and observed a similar increase in pERK in c-FOS positive cells in young and old flies, indicating that engram cells are reactivated equally upon memory recall. Interestingly, however, engram cells in old flies were also reactivated by odors not associated with electrical shocks, suggesting that old flies have an inability to distinguish between odors. Consistent with this, we found that after training, old flies have an increased avoidance of odors unpaired with electrical shocks compared to young flies. Currently, I am studying whether memory engram cells are maintained properly in old flies, whether memory forgetting pathways are increased upon aging, and whether memory engram cells are appropriately reactivated at later time points.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-535 概念的知識の学習における脳のアルゴリズム解明 Asuka Yamamoto（山本明日翔），Mitsuo Kawato（川人光男），Aurelio Cortese（コルテーゼアウレリオ）株式会社　国際電気通信基礎技術研究所 Humans can learn new concepts within few trials. It is thought that the prefrontal cortex (PFC) plays an important role in this ability, but how it does so remains largely unknown. We hypothesize that the PFC integrates low-level feature information into abstract concepts if they are tied to value. To verify this hypothesis, we used two tasks. In the first task, subjects learned the fruit preference of an imaginary character through trial-and-error by maximizing their expected rewards. The character had three features: color (red, green), the direction of the mouth direction (right, left), and stripes orientation (vertical, horizontal). The preferred fruit was decided by a random combination of two features, which changed in each block. The second task was based on neural reinforcement of brain activity patterns (Decoded Neurofeedback, or DecNef). This method allows the strengthening of stimulus-specific neural representations using real-time decoding of functional magnetic resonance imaging (fMRI) signals. In our design we hypothesized that DecNef could bias the importance of one target feature of the first task in visual cortex (VC). We found hallmarks of conceptual knowledge learning when subjects performed the first task, and that PFC, and in particular vmPFC, is involved. After the neural manipulation, subjects learned new concepts faster if the targeted feature was relevant. This result shows that we can bias the construction of concepts through integration of features that happen in different brain areas. We conclude that the brain integrates some feature information and creates new concept by weighing the importance of lower level features, depending on how "valuable" they are in the recognition of objects.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-536 The role of prefrontal cortex in bias and shift between visual dimensions Farshad Alizadeh Mansouri（Mansouri Farshad Alizadeh）¹，Daniel J Fehring（Fehring Daniel J）¹，Mark J Buckley（Buckley Mark J）²，Keiji Tanaka（Tanaka Keiji）³ ¹Monash University ²Department of Experimental Psychology, Oxford University, Oxford, OX1 3UD, UK ³Cognitive Brain Mapping Laboratory, RIKEN Center for Brain Science, Wako, Saitama 351-0198, Japan. Imaging and neural activity recording studies have shown activation in the primate prefrontal cortex when shifting attention between visual dimensions is necessary to achieve goals. A fundamental unanswered question is whether representations of these dimensions emerge from top-down attentional processes mediated by prefrontal regions or from bottom-up processes within visual cortical regions. In large cohorts of humans and macaque monkeys, performing the same attention shifting task, we found that both species successfully shifted between visual dimensions but both species also showed a significant behavioural advantage/bias to a particular dimension; however, these biases were in opposite directions in humans (bias to colour) versus monkeys (bias to shape). Monkeys' bias remained after selective bilateral lesions within the anterior cingulate cortex (ACC), frontopolar cortex (FPC), dorsolateral prefrontal cortex (DLPFC), orbitofrontal cortex (OFC), or superior-lateral prefrontal cortex. However, lesions within certain regions (ACC, DLPFC or OFC) impaired monkeys' ability to shift between these dimensions. We conclude that goal-directed processing of a particular dimension for the executive control of behaviour depends on the integrity of prefrontal cortex, however representation of competing dimensions and bias toward them does not depend on top-down prefrontal-mediated processes. Future studies need to examine whether the dimension-based bias in rule-guided behaviour in humans and monkeys appear in other cognitive domains and investigate the underlying neurobiological substrate for any species differences. It is important to examine whether such differences also exist between humans and other anthropoid apes (such a Chimpanzees) because that would bring further insight to the functional milestones in the development of the anthropoid brain.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-537 dlPFC is more than conflict-based regulation: a tDCS study on the congruency sequence effect Nan Li（Li Nan）¹，Ying Wang（Wang Ying）²，Fang Jing（Jing Fang）²，Rujing Zha（Zha Rujing）²，Zhengde Wei（Wei Zhengde）²，Keiji Tanaka（Tanaka Keiji）¹，Xiaochu Zhang（Zhang Xiaochu）² ¹Center for Brain Science, RIKEN ²School of Life Science, University of Science and Technology of China, China Responses are slower when the relevant and irrelevant cues indicate different responses (incongruent condition) than when the two cues indicate the same responses (congruent condition). This difference is referred to as conflict cost. More interestingly, the conflict cost is larger when the previous trial is incongruent than congruent (the congruency sequence effect, CSE). The conflict monitoring hypothesis (CMH) proposes that the anterior cingulate cortex (ACC) detects the conflict in an incongruent trial, and the dorsolateral prefrontal cortex (dlPFC) upregulates cognitive control accordingly so as to decrease the conflict cost in the following trial. However, the CMH does not tell if this is only the process for dlPFC to influence the CSE. In the present study, we used the transcranial direct current stimulation (tDCS) to excite and inhibit the activity of ACC and dlPFC while subjects conducted the Flanker task (press key according to the central arrow while the side arrows interfered with the response). We found that the excitation of ACC increased the CSE, which is consistent with the CMH. However, we also found that the inhibition of dlPFC increased the CSE. Specifically, the intervention of dlPFC induced a larger change in the conflict cost after a congruent trial than after an incongruent trial. These effects cannot be explained by the CMH. To determine the underlying mechanisms of these changes, we examined the effects of the congruency in earlier trials (the trial before the previous trial (current - 2) and current - 3 trials) on the CSE. While the congruency in the earlier trials had significant effects on the CSE, the intervention of dlPFC did not significantly change the effects. Therefore, the effects of dlPFC inhibition on the CSE were not related with the maintenance of the cognitive control level raised by experiencing the conflict. We propose that dlPFC is involved in CSE by actively maintaining the task rule. A degradation of this function of dlPFC increases the spontaneous activation of the irrelevant response set (responding to the irrelevant cues) in congruent trials and the set is carried over to the following trial, while the irrelevant response set is suppressed, regardless of the task rule maintenance, in incongruent trials to produce a correct response.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-538 ワーキングメモリ課題遂行中のサル前頭前野皮質の機能構造 Hiroshi Abe（阿部央）^1,5，Weibin Song（Song Weibin）^1,2,3,4，Niansheng Ju（Ju Niansheng）^1,2,3,4，Noritaka Ichinohe（一戸紀孝）^5,6，Shiming Tang（Tang Shiming）^1,2,3,4 ¹Peking University School of Life Sciences ²Peking-Tsinghua Center for Life Sciences ³IDG/McGovern Institute for Brain Research at Peking University ⁴Key Laboratory of Machine Perception (Ministry of Education), Peking University ⁵理研CBS 高次脳機能分子解析 ⁶国立精神・神経セ神経研微細構造 The brain performs computation in the network of populations of neurons. Organizing principles of such populations are well studied in the sensory and motor cortices. It remains, however, unclear in the association cortex such as the prefrontal cortex (PFC) important for higher-level cognition. Several studies suggested columnar organizations in the PFC neural activity during working memory, which is a cognitive system temporally holding information for processing and a hallmark of PFC functions. However, fine functional structures have never been visualized in the PFC. To examine if functional structure exists in the PFC, we performed two-photon imaging. Two adult rhesus monkeys were trained to perform the standard oculomotor delayed response (ODR) task where they had to remember a spatial position. Neural activity was measured using a calcium sensor, GCaMP6s, which was expressed in PFC neurons by injecting an adeno-associated virus vector into multiple sites around the principal sulcus of the PFC in the left hemisphere. While performing the task, calcium signals were imaged from neurons in the superficial layers. We found that functional clustering of neurons was absent in examined 8 sites near the principal sulcus in two monkeys. One exception was a cluster located near the caudolateral end of the principal sulcus. In this cluster, approximately 40% of neurons had sustained activity representing spatial information while the monkey was remembering an instructed spatial location. The activity of nearby neurons represented similar spatial positions. At more rostral 8 sites, only a minority of neurons (~10%) encoded spatial information and those neurons were located sparsely. Thus, PFC neurons with similar functional properties were clustered at the caudoralteral site, but not at other rostral sites, suggesting that columnar organizations do not exist ubiquitously in the dorsolateral PFC as a p

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-539 ブロック毎に変動する価値を符号化した神経活動と酔歩を見分ける統計検定 Takayuki Hiwatashi（樋渡峰透）¹，Alice Y Wang（Wang Y Alice）²，Naoshige Uchida（内田直滋）²，Keiji Miura（三浦佳二）¹ ¹関西学院大院理工生命科学 ²ハーバード大学ケンブリッジ　アメリカ Neurons in the striatum were reported to code action values, or the expected amounts of reward associated with a specific action, using decision-making paradigms in which reward conditions were altered across blocks of trials (Samejima, 2005, Wang, 2013). However, a recent study showed that even the simulated random walk neurons can be falsely judged to be action value-coding neurons by a conventional test (e.g. t-test) even if they are not correlated with values (Elber-Dorozko, 2018). This raises the question of what analysis methods are appropriate to correctly identify value coding neurons. In this study, we sought to explore conditions in which these false detections occur.First, we simulated neural activities using a random walk model, and examined how varying timescales of random walk affect the results. We confirmed that the random walk model can result in false positive value-coding neurons. Interestingly, false-positive identification occurred even if the number of trials is large: the false positive rate (p<0.05 or \|t\|>1.96, t-test) did not converge to the chance level (=0.05) but remained at an asymptotic level with increasing trials. To examine if it is caused by the long temporal autocorrelations of random walks, we considered the refined (AR) model, whose timescales of autocorrelations are very short. Surprisingly, the simulated neural activities were still misjudged as value-coding neurons even with a large number of trials.We next explored the mathematical basis of why false-positive identification can occur. Although the action values (=block structure) and random walks are not correlated on average, they can be incidentally correlated on a session basis. For example, a simulated neuron's activity can be positively correlated with action values in one simulation session (t>0), while it can be negatively correlated in another (t<0). Our analysis showed that the key factor is not the mean (=0) but the variance of the t-value distributions, which turned out to be proportional to the block size (#trials in which the reward amounts are fixed). And, theoretically, false-positive identification occurs minimally at the chance level in the short block size limit, i.e. when the block size is one.In conclusion, the autocorrelations of neural activities, however short, as well as block experimental structures leads to the false detection of value-coding neurons in the conventional t-test. These results provide insights into new statistical tests.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-540 霊長類側坐核ニューロンは接近回避葛藤下において意欲と価値を区別してコードする Jungmin Oh（オジョンミン）¹，Masahiko Takada（高田昌彦）¹，Ken-Ichi Amemori（雨森賢一）^1,2 ¹京都大　霊長研統合脳システム ²京都大?眉センター Approach-Avoidance (Ap-Av) conflict emerges under the circumstances where an individual is forced to decide whether to accept or reject an offer that has both positive and negative aspects (e.g., deciding for a new job offer promising high salary but requesting heavy workload). The Ap-Av conflict thus arises in the process of cost-benefit analysis of the conflicting valences. In contrast, Approach-Approach (Ap-Ap) conflict results from competing motivations toward more than one desirable offers (e.g., which flavor of ice creams to buy). Several studies hypothesized that the Ap-Av and Ap-Ap conflicts differentially influence motivation or valuation processes (Miller, 1944; Hall et al., 2011). The rodent nucleus accumbens (NAc) has been implicated in regulating the task performance, suggesting that the NAc is involved in motivational processing (Wyvell and Berridge, 2000). However, it is still unclear whether the primate NAc has a dissociable mechanism for motivation and valuation. To address this, here we focused on the primate NAc while we monitored the conflict decisions in the macaque monkey performing Ap-Av and Ap-Ap tasks. We first found two different types of fixation break errors: motivation-related and conflict-related ones. Notably, the conflict-related errors only appeared in the Ap-Av task, not in the Ap-Ap task, suggesting that the Ap-Av and Ap-Ap conflicts differentially influenced motivation. Next, we carried out single-unit recordings from the NAc of a monkey during the performance of the Ap-Av and Ap-Ap tasks. We recorded 95 neuronal activities and classified them based on the model selection procedure using Bayesian information criteria. Among 35 task-related activities, the most prominent feature was a significant correlation (all subset regression, P < 0.05) with the motivational state of the monkey (n = 10; 10/35=28.6%). This type of activity would support the hypothesis that the NAc plays a central role in regulating motivation that critically have an influence on task performance. Nevertheless, a considerable proportion of neuronal activity exhibited a significant correlation with the value of choice (n = 7; 7/25=20.0%), raising the possibility of a dissociable role of the NAc in value judgment. These results suggest that the macaque NAc consists of two different groups of neurons representing motivation and value.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-541 音楽は利他性を高めるー行動経済学、行動内分泌学研究 Hajime Fukui（福井一）¹，Kumiko Toyoshima（豊島久美子）²，Anri Hattori（服部安里）⁵，Shuji Morishita（森下修次）³，Yoko Ogawa（小川容子）⁴ ¹奈良教育大学教育学部 ²大阪樟蔭女子大学 ³新潟大学教育学部 ⁴岡山大学大学院 ⁵名古屋芸術大学 1.Preface Recent studies have shown that music increases empathy and increases altruism. On the other hand, listening to music is known to affect the secretion of biochemical substances (steroids hormones and neurotransmitters). In this study, in order to explore the mechanism that music enhances empathy and altruism, the experiment was conducted with junior high school students as subjects (12 to 14 years old). Dictator Game (DG) which is an economical index of altruism and Oxytocin (OXT), Testosterone (T) and Cortisol (C) which are physiological index of empathy were examined. The results of DG and OXT are reported in this paper. 2. Method An experiment was conducted on 11 junior high school students (4 boys and 7 girls) after obtaining written informed consent from the subjects and their parents. Subjects played DG and submitted saliva before and after listening to music. The music used for the experiment was "the favorite chill-induced (Goose bumps) music" and "the disliked music" selected by each subject. DG and O, T, C were used as indicators of altruism. 3. Results DG The amount of distribution increased significantly after listening to the favorite music, regardless of whether the recipient to distribute was in-group or out-group, and the amount of distribution to the recipient decreased after listening to the disliked music. Listening to favorite music is considered to promote altruistic behavior beyond the affiliation. OXT OXT significantly decreased OXT with favorite music in the group with high base value, and OXT significantly increased with favorite music in the group with low base value. Previous studies have shown that the OXT increases after listening to favorite music. In favorite music listening (pleasant stimuli), the amount of allocation increased in both groups, so it can be judged that altruism has increased. However, in OXT, conflicting changes occurred in the two groups. This result is interesting as the increase in OXT is believed to increase altruism. In recent years, it has been reported that there are individual differences in the reactivity of the oxytocin system. This result shows that music is an important factor in understanding the relationship between altruism and OXT.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-542 マカクザル前頭前野ECoG信号を用いた、意思決定関連情報のデコーディング Shingo Tanaka（田中慎吾）^1,3，Yosuke Ito（伊藤陽介）²，Keisuke Kawasaki（川嵜圭祐）³，Isao Hasegawa（長谷川功）³，Takafumi Suzuki（鈴木隆文）⁴，Masamichi Sakagami（坂上雅道）¹ ¹玉川大脳研 ²東京大院総合文化研 ³新潟大院医歯学総合研 ⁴情報通信研究機構脳情報通信融合研究センター Decision making is one of the important mental processes in our daily life. When deciding between multiple options, our brain estimates their values and choose the most desirable one. Wide-ranging areas of the brain are involved in this decision making process. A lot of studies show that areas in the prefrontal cortex (PFC) play different but essential roles for value calculation and comparison. Because of this division of labor, the value signal should be reciprocated within areas in the PFC. Distribution of the value related signals is a key feature to understand the signal transmission among multiple areas of the PFC. However, few studies directly examined how each area of the PFC contributes to this distributive process. Here, we tried to reveal the spatial and temporal distribution of the value and the value-based decision related signals in the wide-ranging areas of the PFC. Two monkeys were trained to perform a free-choice task. During this task, two visual cues which indicated the kinds of juices were presented sequentially with a short blank. After that, these two cues were presented simultaneously and the monkeys chose one of them to obtain juice reward. From the choice behavior between various pairs of juices (in total 5 juices, 10 cues), the values of the juice rewards were estimated. While the monkeys performed the free-choice task, electrocorticographic (ECoG) signals were recorded from the ECoG electrodes implanted on the left lateral prefrontal cortex (LPFC), the left orbitofrontal cortex (OFC), the left medial prefrontal cortex (MPFC) and the left anterior cingulate cortex (ACC). The reward value signals were found in all of the four PFC subareas during the 1st and the 2nd cue presentation. However, the choice related signals were found only during the 2nd cue presentation. The value information of the 2nd cued juices existed more when the monkeys chose 2nd cued juices than when they chose 1st cued juices. Furthermore, in the LPFC and OFC, the value information of the 1st cued juices also existed during the 2nd cue presentation only when the monkeys chose 1st cued juices. In addition to that, the information of the 1st cued value and 2nd cued value showed different spatio-temporal distribution during the 2nd cue presentation. From these results, we suggest that the different population of neurons code the two different value and might compare them to generate the choice related signals in the LPFC and OFC.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-543 背外側と背内側線条体間接路細胞の行動選択における異なる役割 Satoshi Nonomura（野々村聡）^1,2，Shigeki Kato（加藤茂樹）³，Yutaka Sakai（酒井裕）²，Atsushi Nambu（南部篤）⁴，Kazuto Kobayashi（小林和人）³，Yoshikazu Isomura（礒村宜和）^1,2，Minoru Kimura（木村實）^1,2 ¹東京医歯大医歯総合細胞生理 ²玉川大脳研 ³福島県立医大医生体情報伝達研生体機能 ⁴生理研統合生理生体システム We showed previously that direct-pathway neurons in the rat dorsomedial striatum (DMS) signal reward outcome and repeat action selection, while indirect-pathway neurons (iSPNs) signal no-reward outcome and switch action selection, suggesting goal-directed monitoring and updating action selection (Nonomura et al., 2018). In this study, we examined signals of iSPNs in the dorsolateral striatum (DLS) and compared with those in the DMS, because region-specific involvement in action and learning was previously proposed in the dorsal striatum. Two Drd2-Cre rats performed a lever push/pull task for a probabilistic reward in which they chose either push or pull in response to GO signal according to the reward probability (70% vs. 10%) assigned to the two actions to get water reward. Reward probabilities of the two actions were alternated every 40-60 trials. Reward and no-reward outcomes were preceded by different tones. We injected AAV into the DLS to express channelrhodopsin (ChRWR) in iSPNs, recorded single neuron activity through a 32-ch silicon probe, and identified iSPNs by antidromic activation after optical stimulation of external globus pallidus (GPe). We found that: (1) Activity of iSPNs after action selection did not discriminate reward from no-reward outcome (0/26 neurons). This was in sharp contrast to our previous observations of iSPNs in DMS specifically signaling no-reward outcome during outcome tone period (34/47 neurons) (chi-square test, p = 6.3 x 10-5). (2) In both the DLS and DMS, a large number of iSPNs exhibited robust activation after GO signal for action selection than suppression of activity, while smaller number of them exhibited suppression of activity. (3) Most iSPNs (25/26) were not responsive to reward- and no-reward-associated cues during classical conditioning. These results reveal outcome-independent nature of iSPNs in the DLS for action selection in contrast to no-reward outcome-based action update of iSPNs in the DMS.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-544 前頭眼野と補足眼野による視空間性注意制御の神経メカニズム Osamu Yokoyama（横山修），Yukio Nishimura（西村幸男）東京都医学総合研 Two oculomotor areas in the frontal lobe, i.e., the frontal eye field (FEF) and supplementary eye field (SEF) are involved in the control of visual spatial attention. However, specific roles played by each area and how these areas interact to exert the function remain elusive. To address these issues, we recorded and compared the activity of single neurons from the FEF and SEF while monkeys performed a task demanding dynamic control of attention field size. In this task, monkeys were required to direct their attention to either the left or right visual hemifield initially, and to direct their attention to a small target in the hemifield later. We found in both the FEF and SEF, many neurons that encoded dynamically changing attention field, i.e., large attention field to either hemifield transforming into small attention field on a target. The proportion of the neurons representing the large attention field, their selectivity strength, and the timing for the emergence of the representation was comparable between FEF and SEF neurons, indicating that these areas emitted a similar amount of attentional signals in a similar time course. In contrast, FEF and SEF neurons exhibited a clear difference in their spatial preference: in the FEF more neurons exhibited a preferred direction in the hemifield contralateral to the recorded hemisphere, whereas SEF neurons did not exhibit such directional bias. Furthermore, to evaluate the directionality of interactions between the FEF and SEF, we recorded local field potentials (LFPs) simultaneously from these areas of a monkey while it performed the task. Granger causality analysis revealed that when large attention was directed to a visual hemifield, beta-band (~20 Hz) activity of the LFPs in the SEF influenced that in the FEF, especially more when large attention was directed to the hemifield contralateral to the recorded hemisphere. These results indicate that the FEF is involved mainly in directing attention to the contralateral space, whereas the SEF is involved in directing attention to either hemifield, and that the SEF sends information to the FEF when attention is directed to the contralateral visual hemifield in a top-down manner. These findings demonstrated a manner in which the FEF and SEF function distinctively and cooperatively in dynamic control of spatial attention.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-545 Neural Integration of Linguistic Expressions and Sociocultural Conventions in Comprehending Socio-pragmatic Knowledge: The Case of Japanese Honorific Expressions Haining Cui（Cui Haining）¹，Hyeonjeong Jeong（Jeong Hyeonjeong）^1,2，Kiyo Okamoto（Okamoto Kiyo）²，Daiko Takahashi（Takahashi Daiko）¹，Ryuta Kawashima（Kawashima Ryuta）²，Motoaki Sugiura（Sugiura Motoaki）² ¹Graduate School of International Cultural Studies, Tohoku University, Sendai, Japan. ²Institute of Development, Aging and Cancer, Dept. Human Brain Science, Tohoku University, Sendai, Japan. Introduction Although it becomes clear that the left frontal and temporal cortex are highly demanded in the processing of linguistic knowledge, less is known about how different linguistic expressions are integrated with sociocultural norms and conventions to comprehend socio-pragmatic knowledge. Thus, we examined the neural processing of socio-pragmatic knowledge by focusing on Japanese honorific expressions. In Japanese, the honorific expressions are called grammaticalized honorifics since the expressions require speakers of lower social status apply different inflected verb forms (i.e., honorific or humble) toward interlocutors with higher social status. By manipulating conventional and unconventional expressions by both lower and higher social status speakers, we examined the brain mechanisms underlying the processing of socioculturally conventional expressions and the grammaticalized honorifics. Methods Thirty-three native speakers of Japanese were instructed to perform a socio-pragmatic judgment task containing auditory sentence accompanied by interlocutors' images. The auditory sentences were designed with two factors, each at two levels: `Convention' (conventional vs. unconventional) and `Social status' (lower vs. higher). The participants were instructed to judge whether each sentence was a socioculturally conventional expression by pressing the reaction buttons of one (conventional) or two (unconventional) during fMRI scanning. Brain data were analyzed using SPM12. First, to identify how the socioculturally conventional expressions were processed during language comprehension, we compared brain activation in processing conventional and unconventional expressions, second, to investigate which brain regions were involved in the processing of grammaticalized honorifics, we compared the lower with higher social status speakers' expressions (FWE p < 0.05 at the voxel level). Results and Conclusion The results revealed that the bilateral anterior temporal lobe (ATL) showed greater activation in the comparison between conventional with unconventional expressions, and the left inferior frontal gyrus (LIFG) produced significant activation for the processing of grammaticalized honorifics. We argue that the ATL may function to integrate the linguistic expressions along with sociocultural conventions, and the LIFG is taken in the processing of inflected verbs to comprehend the grammaticalized socio-pragmatic knowledge.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-546 fMRIによる1文字連続呈示課題を用いた文理解中の2重構造化処理に関わる神経基盤の解明 Ryutaro Kasedo（加世堂竜太郎）^1,2，Atsuhiko Iijima（飯島淳彦）^1,3，Kiyoshi Nakahara（中原潔）⁴，Yusuke Adachi（足立雄哉）²，Fumitaka Homae（保前文高）⁵，Ryu-Ichiro Hashimoto（橋本龍一郎）⁵，Kazuhide Yamamoto（山本和英）⁶，Masashi Fukuda（福多真史）⁷，Hiroshi Shirozu（白水洋史）⁷，Isao Hasegawa（長谷川功）² ¹新潟大学大学院　自然科学研究科　電気情報工学専攻 ²新潟大学大学院　医歯学総合研究科　神経生理学分野 ³新潟大学　医学部　保健学科 ⁴高知工科大学　情報学群　脳コミュニケーション研究センター ⁵首都大学東京　人文科学研究科　言語科学分野 ⁶長岡技術科学大学　工学部　電気電子情報工学専攻 ⁷国立病院機構西新潟中央病院　機能脳神経外科 Comprehension of a language requires dual-step `merge' processes, which construct a linguistic constituent from a number of lower-level constituents in two distinct stages. Namely, streams of sublexical constituents (letters) are initially combined (the 1st-level merge) into minimal lexical constituents (words), which are further combined (the 2nd-level merge) into a phrase or higher-order constituent. According to previous studies focusing on the 2nd-level merge, build-up neural activations growing with the size of unmerged constituents (or the number of open nodes in the parse tree) were located in the left temporal pole (TP), anterior superior temporal sulcus (STS), and inferior frontal gyrus (IFG), whereas transient activations correlated with the number of node closings were located in the triangular part of IFG. However, neural mechanisms integrating the 1st-level and 2nd-level merge processes remains largely unknown. To address this issue, we developed a self-paced sequential letterstrings reading task and conducted functional magnetic resonance imaging (fMRI) while normal participants performed the task. Brain activations parametrically modulated with the number of open nodes and node closings for the 1st-level merge were found mainly in the temporal regions, especially in the left TP. At the 2nd-level merge, the left superior temporal gyrus (STG) was modulated with the number of open nodes, whereas activations in the supramarginal gyrus and middle frontal gyrus were modulated with the number of node closings. Taken together, our findings indicate that dynamic neural activity in the left TP/STG serves a key to link the sublexical-level and lexical-level merge processes during sentence reading.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-547 瞳孔径の変化に基づく前言語期における語彙獲得過程の予備的検討 Masahiro Hata（秦政寛），Eriko Yamamoto（山本絵里子），Yasuyo Minagawa（皆川泰代）慶應義塾大学 The first year of life is crucial for the development of representations of word meanings in infants (Friedrich & Friederici, 2010; Nazzi & Bertoncini, 2003). The study of neural representation of the so-called "preparation period" before the vocabulary spurt is essential; but there are few studies on the neural substrates underlying the initial acquisition of word meaning. Pupil dilation has been linked to various cognitive processes and is also an indicator of the processing of word meaning (Kuipers & Thierry, 2011). In this study, we investigated whether infant pupil size changed during a picture-sound match/mismatch paradigm in 9 and 12-month-old infants. The infants were presented with picture-sound matched/mismatched stimuli, while they sat in their mothers' laps; dog or cat pictures were presented for 4000 ms and the sound, i.e., inu (dog) or neko (cat), which matched or did not match the presented picture, was presented 500 ms after picture presentation. The infant's eye movement and the pupil size were recorded using Tobii X120 Eye Tracker system (Tobii, Stockholm, Sweden). Data were epoched into time intervals ranging from -300 ms to 1800 ms relative to the sound-stimulus onset. Thirteen 9-month-old infants (6 girls, mean age = 286.4 days, SE = 2.3) and ten 12-month-old infants (5 girls, mean age = 376.2 days, SE = 1.3) were included in the analysis. We used a paired-sample t-test to obtain the mean pupil size in an 800 ms window (800 to 1600 ms), comparing the difference in pupil diameter between stimulus conditions every month. Pupil size in the matched condition was greater than that of the mismatched condition in 12-month-old infants [t(9) = 2.27, p < 0.05], but there was no pupil size difference in 9-month-old infants. Twelve-month-old infants showed greater pupil dilation in the picture-sound matched condition, although adults and children aged 2-3 years exhibited greater pupil dilation in the mismatched condition (Kuipers and Thierry, 2011, 2013). Pupil size is modulated by activity in the locus coeruleus-norepinephrine system and the superior colliculus; the latter receives cognition-related signals from cortical areas (Wang & Douglas, 2015). Since pupillary change could be a marker of involuntary attention in infants (Wetzel et al., 2015), the present findings suggests that 12-month-old infants have a unique attention system during the course of acquisition of word meanings.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-548 オリゴデンドロサイト前駆細胞由来の活性化アストロサイトによるグリア瘢痕形成 Hinata Matsuda（松田ひなた），Haruna Ueda（上田晴菜），Erika Tsuji（辻恵里佳），Mitsuhiro Morita（森田光洋）神戸大学大学院理学研究科生物学専攻 Accumulating evidence indicates the diversity of reactive astrocytes, however the origins and functions of these cells are still elusive. We have developed a closed-head injury model, named photo-injury and found prominent cerebral tissue recovery surrounded by a reactive astrocyte population expressing a neural stem cell marker, nestin (nestin-expressing reactive astrocyte; NRA). Photo-injury created in nestin-CreERT2/Rosa-EGFP mice indicated the GFP-labeling of non-astrocytic cells in lesion core (LC) and the glial scar formation by reactive astrocyte derived from a fraction of these GFP+ cells (LC-NRA). The distribution and morphology of GFP-labeled reactive astrocytes by nestin-CreERT2 were distinct from those by GFAP-CreERT2, especially their localization at the border of LC. Another fraction of GFP-labeled cells by nestin-CreERT2 lacked GFAP expression even after tissue recovery, and their fate is under investigation. By statistical analysis about staining for each cell type markers, it is revealed that all these GFP+ cells were derived from nestin+, NG2+ and PDGFRα+ oligodendrocyte progenitor cells. Furthermore, nestin-CreERT2/floxed-STAT3 mice showed robust reduction of GFP+ cells even at the early stage before generating GFAP+ reactive astrocytes. These results indicates the generation of scar-forming reactive astrocytes from oligodendrocyte progenitor cells. The more detailed analysis of these reactive astrocytes is needed for the development of new treatments for brain injury.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-549 Enriched EnvironmentによるMCAOの回復とNeuritin1の関与 Yumi Fukuzaki（福崎由美）^1,2，Kenichi Kunii（国井賢一）²，Hideki Kawai（川井秀樹）² ¹DPAG, Univ of Oxford, Oxford, UK ²創価大工学研究科 Enriched environment (EE) has been shown to reduce cortical injury areas in the rat model of middle cerebral artery occlusion (MCAO), suggesting that sensory-evoked activities provide their protection (Lay et al., 2014, Eur J Neurosci. 40-9:3413-21). Molecules participating in this protection are not known. Here, we have tested the possibility that neuritin 1 is involved in this using a mouse model of MCAO. MCAO was induced by photo-thrombosis using green laser irradiation (15 mW, 30 min) after rose bengal injection followed by a cut of a clotted MCA. We first confirmed the reproducibility of cortical injury protection by EE by showing decreased injury areas with triphenyltetrazolium chloride (TTC) assay. Western blotting indicated that EE increased the expression of neuritin 1. To test if neuritin 1 alone could reduce injury areas, immediately after MCAO, recombinant human neuritin 1 (1 mg/l) was injected stereotaxically into somatosensory cortex (at 1.5 or 2.5 mm posterior from bregma and 2 mm lateral from sagittal suture), and mice were returned to cages for 3 days. The size of injured areas measured within the GFAP-immunopositive glial scar was significantly reduced by neuritin 1 injection in a posterior part of the cortex near auditory cortex. GFAP-positive reactive astrocytes were apparently reduced at areas distant from penumbra in neuritin 1-treated mice. There were denser NeuN-positive neurons within the injured areas. We also observed macrophage-like NG2 glia inside the penumbra along the glial scar. Reactive NG2 glia were present within penumbra, while regular NG2 glia were affected little outside penumbra. These results suggest that EE-induced protection of MCAO injuries involved neuritin 1.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-550 タウ遺伝子改変マウスrTg4510のタウ蓄積における遺伝的背景の影響 Daijiro Yanagisawa（柳沢大治郎），Hamizah Shahirah Hamezah（Hamezah Hamizah Shahirah），Aslina Pahrudin Arrozi（Pahrudin Arrozi Aslina），Hiroyasu Taguchi（田口弘康），Ikuo Tooyama（遠山育夫）滋賀医科大学神経難病研究センター Tau pathology is characterized by the pathological aggregates of abnormally hyperphosphorylated tau, which is a neuropathological feature in neurodegenerative diseases including Alzheimer's disease. The rTg4510 mouse strain is a bitransgenic mouse model that uses a system of responder and activator transgenes to express human four-repeat tau with the P301L mutation in the forebrain, which induces the accumulation of tau pathology in the forebrain region. In the present study, we compared tau pathology between rTg4510 mice obtained by crossing female CaMKII-tTA mouse line on a C57BL/6J background and male tetO-MAPTP301L mice on a FVB/NJ background (rTg4510_CxF) and rTg4510 mice obtained by crossing female tetO-MAPTP301L mice on the FVB/NJ background and male CaMKII-tTA mouse line on the C57BL/6J background (rTg4510_FxC). Male and female rTg4510_CxF and rTg4510_FxC (n = 6 in each group) at 6 months of age were used in this study. Histological analysis revealed intense immunoreactivity for phosphorylated tau antibody (AT8) and argentophilic structures by Gallyas silver staining in rTg4510_FxC, compared with rTg4510_CxF. Results in the western blotting showed a massive accumulation of insoluble tau in rTg4510_FxC, compared with rTg4510_CxF. The rTg4510_FxC also showed a higher level of soluble tau detected at 64 kDa than rTg4510_CxF, although there was no significant difference in soluble tau level detected at 55 kDa between rTg4510_CxF and rTg4510_FxC. No significant sex differences were found in AT8-immunoreactivity, argentophilic structures, and the levels of soluble and insoluble tau. These results suggested that genetic background induced the difference in the severity of tau accumulation between rTg4510_FxC and rTg4510_CxF. Further study to clarify the cause of the difference would provide important insights into the mechanism underlying tau accumulation, which may help to develop therapeutic approaches aiming to modulate tau accumulation in tauopathy.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-551 加齢に伴い進行する認知機能低下におけるTRPM2の病態生理学的役割 Hisashi Shirakawa（白川久志），Masashi Kakae（抱将史），Jun Miyanohara（宮之原遵），Kazuki Nagayasu（永安一樹），Shuji Kaneko（金子周司）京都大院薬生体機能解析 Aging causes various functional changes, including cognitive decline and inflammatory responses in the brain. Several lines of evidence suggest that age-related vascular diseases are associated with altered inflammatory responses, particularly increased production of pro-inflammatory cytokines. Moreover, a previous report shows that chronic cerebral hypoperfusion-induced cognitive impairment in mice and humans is highly associated with inflammation, suggesting that regulating inflammatory response could be a potential method of arresting cognitive decline during normal aging. Microglia are immune cells resident in the central nervous system. Studies show that dystrophic changes occur in the aged human brain. To date, it has been thought that functional alterations to microglia play important roles in aging-associated pathologies. Transient receptor potential melastatin 2 (TRPM2), a Ca<SUP>2+</SUP>-permeable channel expressed abundantly in immune cells, exacerbates inflammatory responses. Previously, we reported that TRPM2 on resident microglia plays a critical role in exacerbating inflammation, white matter injury, and cognitive impairment during chronic cerebral hypoperfusion; however, the physiological or pathophysiological role of TRPM2 during age-associated inflammatory responses remains unclear. Therefore, we examined the effects of TRPM2 deletion in young (2-3 months) and older (12-24 months) mice. Compared with young wild-type (WT) mice, middle-aged (12-16 months) WT mice showed working and cognitive memory dysfunction and aged (20-24 months) WT mice exhibited impaired spatial memory. However, these characteristics were not seen in TRPM2 knockout (TRPM2-KO) mice. Consistent with the finding of cognitive impairment, aged WT mice exhibited white matter injury and hippocampal damage and an increase in the number of Iba1-positive cells and amounts of pro-inflammatory cytokines in the brain; these characteristics were not seen in TRPM2-KO mice. These findings suggest that TRPM2 plays a critical role in exacerbating inflammatory responses and cognitive dysfunction during aging.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-552 脳内でβ-アミロイド蓄積を誘発する可溶な高分子量Ａβオリゴマーの同定 Mayu Hakozaki（箱崎眞結）^1,2，Yasushi Naka（仲泰史）¹，Masashi Fukayama（深山正久）²，Tetsuo Ushiku（牛久哲男）²，Tadafumi Hashimoto（橋本唯史）^1,3，Takeshi Iwatsubo（岩坪威）¹ ¹東京大院医神経病理 ²東京大院医人体病理 ³東京大院医認知症先進予防治療 Extracellular deposition of amyloid-β (Aβ) peptides as senile plaques is a key pathogenic feature of Alzheimer's disease (AD), in which the critical role of the seeding capacity of aggregated forms of Aβ has been implicated. However, the molecular characteristic of the seed Aβ strains that initiate and spread β-amyloidosis <I> in vivo </I> remains elusive. To identify the seed Aβ strains in brains, we separated the Tris-buffered saline (TBS)-soluble fractions from the Aβ-laden brains of 18-month-old APP transgenic (tg) mice by size-exclusion chromatography using a Superdex 75 column. We detected three peaks of Tris-soluble Aβ by Aβ42 specific ELISA, eluting at ~200-300 kDa (peak 1 Aβ), ~50-60 kDa (peak 2 Aβ) and ~10-20 kDa (peak 3 Aβ). The amount of Aβ42 in peak 1 Aβ fraction was positively correlated with the Aβ-positive areas in the hippocampus, suggesting that peak 1 Aβ is associated with Aβ deposits. To examine whether these Aβ strains are capable of initiating Aβ deposition in vivo, we injected peak 1 or peak 2 Aβ into the unilateral hippocampus of 10-month-old APP tg mice. Four months later, Peak 1 Aβ elicited a unique pattern of Aβ deposition within the specific layers of the ipsilateral hippocampus, whereas peak 2 Aβ and the ~200-300 kDa fraction from the wild-type mice did not. Peak 1 Aβ contained Aβ oligomers as detected by Aβ-oligomer specific ELISA, and immunodepletion of peak 1 Aβ with anti-Aβ antibodies abolished the seeding potency of peak 1 Aβ. These results suggested that the Aβ strains in peak 1 Aβ act as aggregation seeds <I> in vivo </I>. Lastly, we detected peak 1 Aβ from the TBS-soluble fraction in autopsied human AD brains, but not in control brains. Intrahippocampal inoculation of peak 1 Aβ derived from AD brains induced Aβ-deposition similar to that from APP tg mice, whereas peak 3 Aβ did not, suggesting that Aβ strains from the brains of APP tg mice and patients with AD share similar characteristics. We showed that the high-molecular-weight soluble Aβ strains (peak 1 Aβ) derived from Aβ-laden brains plays a critical role in the initiation and propagation of Aβ pathology. The findings may provide new insights to the mechanism of spatiotemporal spreading of β-amyloidosis and therapeutic strategies to prevent Aβ deposition.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-553 ヒト型とマウス型apolipoprotein EがAβの凝集及び蓄積に与える影響の相違 Asuka Kokawa（粉川明日香）¹，Yutaro Ohno（大野裕太郎）¹，Tadafumi Hashimoto（橋本唯史）^1,2，Takeshi Iwatsubo（岩坪威）¹ ¹東京大院医神経病理 ²東京大院医　認知症先進予防治療学講座 Alzheimer's disease (AD) is the most common cause of neurodegenerative dementia, pathologically characterized by massive deposition of amyloid-β peptide (Aβ) as senile plaques in the brain. APOE ε4 allele is the major genetic risk factor for AD, although little is known how apoE impacts the Aβ metabolism in the brain. Because apoE and Aβ bind to each other and co-localize in senile plaques in AD brains, we hypothesized that apoE may affect the process of Aβ fibrillization and deposition. We first conducted <I>in vitro</I> ThioflavinT (ThT) binding assays to examine the effects of apoE on the fibrillization of Aβ. Compared with the incubation of Aβ alone, co-incubation with human apoE3 delayed the initiation of Aβ fibril formation. In contrast, the delay in Aβ fibril formation upon co-incubation with mouse apoE was shorter than with apoE3, suggesting that the inhibitory effect of apoE3 on Aβ fibrillization is stronger than that of mouse apoE. To examine this effect <I>in vivo</I>, we crossed APP transgenic (tg) mice (APP/PS1 line) with human apoE3 knock-in (KI) mice. The amount of Aβ deposition in the cortices of 9-month-old APP tg/apoE3 KI mice was significantly smaller than that in APP tg mice. To further examine the effect of apoE on the propagation of Aβ deposition, we inoculated brain lysates of the Aβ-laden APP tg mice into the hippocampi of 3-month-old APP tg or APP tg/apoE3 KI mice. After incubation for 3 months, Aβ deposition in a characteristic pattern in and around the hippocampus was observed in APP tg mice, whereas it was hardly detectable in APP tg/apoE3 KI mice, suggesting that human apoE3 suppressed Aβ deposition in the brain. In conclusion, our data suggest that human apoE3 suppresses Aβ fibrillization <I>in vitro</I> and Aβ deposition <I>in vivo</I>, whereas mouse apoE has smaller effects. Further understanding of the molecular mechanism of apoE effects on Aβ aggregation may provide important clues to the prevention of Aβ deposition in the brain.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-554 Protective effects of <I>Scolopendra</I> water extract on trimethyltin-induced hippocampal neurodegeneration and seizures in mice Changjong Moon（Moon Changjong）¹，Mary Jasmin Ang（Ang Mary Jasmin）¹，Byeong Cheol Moon（Moon Byeong Cheol）²，Hyo Seon Kim（Kim Hyo Seon）²，Goya Choi（Choi Goya）²，Hye-Sun Lim（Lim Hye-Sun）²，Sohi Kang（Kang Sohi）¹，Mijin Jeon（Jeon Mijin）¹，Sung-Ho Kim（Kim Sung-Ho）¹，Joong Sun Kim（Kim Joong Sun）² ¹College of Veterinary Medicine and BK21 Plus Project Team, Chonnam National University, Gwangju, Republic of Korea ²Herbal Medicine Resources Research Center, Korea Institute of Oriental Medicine, Naju city, Jeollanam-do 58245, Republic of Korea Trimethyltin (TMT) is an organotin compound with potent neurotoxic action, characterized by neuronal degeneration in the hippocampus. This study evaluated the protective effects of <I>Scolopendra</I> water extract (SWE), on hippocampal neurons, against TMT intoxication, using <I>in vitro</I> and <I>in vivo</I> model systems. Specifically, we examined the actions of SWE on TMT (5 mM)-induced cytotoxicity in primary cultures of mouse hippocampal neurons (7 days in vitro), and the effects of SWE on hippocampal degeneration of adult TMT (2.6 mg/kg, intraperitoneal)-treated C57BL/6 mice. We found that SWE pretreatment (0-100 μg/mL) significantly reduced TMT-induced cytotoxicity in cultured hippocampal neurons, in a dose-dependent manner as determined by lactate dehydrogenase and 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assays. Also, this study shows that perioral administration of SWE (5 mg/kg) from days -6 to 0, before TMT injection, significantly attenuated hippocampal cell degeneration and seizures in adult mice. Furthermore, quantitative analysis of Iba-1- and GFAP-immunostained cells, revealed a significant reduction in the levels of Iba-1 and GFAP positive cell bodies in the DG of mice pretreated with SWE, prior to TMT injection. These data suggest that SWE pretreatment significantly protects the hippocampus against the massive activation of microglia and astrocytes elicited by TMT. In addition, our data show that the SWE-induced reduction of immune cells activation was coupled to a significant reduction in cell death and a significant improvement in TMT-induced seizures behavior. Thus, we conclude that SWE ameliorated the detrimental effects of TMT toxicity on hippocampal neurons, both <I>in vitro</I> and <I>in vivo</I>. Altogether, our findings hint at a promising pharmacotherapeutic use for SWE in hippocampal degeneration and dysfunction.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-555 パーキンソン病患者髄液における「睡眠の質の低下」関連蛋白質の同定 Eiko N Minakawa（皆川栄子）¹，Hiroko Yagihara（八木原紘子）¹，Yuji Saito（齊藤勇二）^1,2，Masayuki Miyazaki（宮崎将行）¹，Ayumi Tsuru（都留あゆみ）³，Minori Suzuki（鈴木みのり）³，Yuichi Kamei（亀井雄一）⁴，Koji Ueda（植田幸嗣）⁵，Keiji Wada（和田圭司）¹，Yoshitaka Nagai（永井義隆）^1,6 ¹国立精神・神経セ神経研疾病４ ²国立精神・神経セ病院脳神経内科 ³国立精神・神経セ病院睡眠障害センター ⁴上諏訪病院精神神経科 ⁵がん研がんプレシジョン医療研究センター ⁶大阪大院医神経難病認知症探索治療学 Sleep abnormality, especially impairment of sleep quality due to intermittent nocturnal arousal leading to sleep fragmentation and a decrease in deep non-rapid eye movement (NREM) sleep, is prevalent among patients with Parkinson's disease (PD) from the early course of the disease. Recent epidemiological studies demonstrated an association between impaired sleep quality and PD, suggesting that impaired sleep quality may be a prodromal marker or risk factor for PD pathology. In the present study, we aimed to explore biochemical changes in the components of PD patients' cerebrospinal fluid (CSF) that are associated with impaired sleep quality, which would reflect the biochemical changes in the brain related to PD pathology under impaired sleep quality. We performed an unbiased proteomic analysis by liquid chromatography-tandem mass spectrometry (LC-MS/MS) of the CSF of 20 PD patients, who underwent objective sleep assessment using polysomnography (PSG) within three months of CSF collection. In addition, detailed clinical assessment of PD by expert neurologists was performed at the timing of CSF collection. We identified 1,388 proteins in total in the CSF and categorized them into those that were significantly up- or down-regulated in PD patients with impaired sleep quality. This categorization was performed using multiple parameters of PSG including Arousal Index (ArI), an indicator of sleep fragmentation, or percentage of stage N3 sleep (%N3), the deepest NREM sleep, per total sleep time. Through these analyses, we identified CSF proteins that up- or down-regulated in association with an increase in sleep fragmentation or decrease in stage N3 sleep in patients with PD. Intriguingly, enrichment analyses of these proteins revealed that proteins associated with sleep fragmentation or decreased stage N3 sleep have a different enrichment profile, suggesting that different components of sleep may differently alter proteins in the brain. Further analyses of these CSF proteins would lead to a better understanding of the relationship between impairment of sleep quality and the pathomechanisms underlying PD.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-556 アルファシヌクレインが未熟オリゴデンドロサイト細胞死を誘導し多系統萎縮症における神経変性を加速させる可能性がある Seiji Kaji（梶誠兒）^1,2，Takakuni Maki（眞木崇州）¹，Takashi Ayaki（綾木孝）¹，Ryosuke Takahashi（髙橋良輔）¹ ¹京都大学大学院脳神経内科 ²日本赤十字社和歌山医療センター脳神経内科 <B>Background</B> Fibrillar α-synuclein (α-syn) in oligodendrocytes (OLGs), known as glial cytoplasmic inclusions (GCIs), likely affects physiological function of OLGs in in multiple system atrophy (MSA). In order to clarify the impact of pathological α-syn on differentiating /myelinating premature OLGs (pre-OLGs), the present study shows post-mortem analysis of altered cell population of pre-OLGs due to α-syn accumulation in MSA brains and in vitro investigation of α-syn-induced cytotoxicity against premature OLGs. <B>Method</B> <I>Histopathological analysis of autopsied cases</I> Formalin-fixed, paraffin-embedded sections from patients with MSA (n=9) and Parkinson's disease (PD) / Diffuse Lewy body disease (DLB) (n=9) were deparaffinized and immunostained with primary antibodies for phosphorylated α-syn (p-α-syn) and a marker protein recognizing pre-OLGs. Morphologically, pre-OLGs were either defined as "Early stage" (having less than 4 cellular processes), or "Late stage" (having 4 or more cellular processes or myelin-forming). Emergence of p-α-syn-immunoreactive inclusions in pre-OLGs was assessed. <I>In vitro analysis</I> Primary pre-OLG cell culture was prepared from neonatal rats. Bacterially obtained human recombinant α-syn pre-formed fibrils (PFFs) were applied to pre-OLGs. The uptake of α-syn PFFs was visualized by application of α-syn PFFs labeled with Alexa Fluor 594 protein. The viability of α-syn PFFs-treated pre-OLGs was evaluated by WST assay after incubation with differentiation medium. <B>Results</B> Histopathological analysis of MSA brains revealed a negative correlation between the numbers of late-stage pre-OLGs and p-α-syn density in MSA brains. A portion of pre-OLGs contained p-α-syn-immunoreactive GCI-like inclusions. <I>In vitro</I> analysis showed internalization of extracellularly applied α-syn PFFs into most primary pre-OLGs, resulting in reduced cell viability. <B>Discussion</B> The present study confirms crucial interaction between pathological α-syn and premature OLGs, which is also supported by a few previous reports. These results imply that protection of pre-OLGs from α-syn PFF-induced cytotoxicity and achievement of successful myelination lead to improvement of oligodendroglial neuro-supportive function, preventing the progressive neurodegeneration in MSA.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-557 マーモセット一次視覚野でのCRTC1のノックダウンはてんかん様の神経活動を惹起する Yuki Nakagami（仲神 "友貴 "）^1,2，Akiya Watakabe（渡我部昭哉）¹，Misako Komatsu（小松三佐子）¹，Masafumi Takaji（高司雅史）¹，Hiroaki Mizukami（水上浩明）³，Junichi Hata（畑純一）^4,5，Hideyuki Okano（岡野栄之）^4,5，Ken Nakae（中江健）⁶，Shin Ishii（石井信）⁶，Noritaka Ichinohe（一戸紀孝）^1,7，Tetsuo Yamamori（山森哲雄）¹ ¹理研CBS 高次脳機能分子解析 ²新潟大脳研 ³自治医大 ⁴慶應大医生理 ⁵理研CBS マーモセット神経構造 ⁶京都大院情報システム科学 ⁷国立精神・神経セ神経研微細構造 CREB-relgulated transcriptional coactivator 1 (CRTC1) is a key coactivator of cyclic-AMP response element binding protein (CREB), which is expressed across the entire brain. CRTC1 plays an important role in learning and memory, and its abnormality is known to be involved in various neuropsychiatric disorders such as Huntington's disease. In an attempt to explore the mechanisms of activity-dependent gene expression in the primary visual cortex (V1) of marmoset monkeys, we performed a series of knockdown (KD) experiments of CRTC1 gene expression by AAV-driven RNA interference (RNAi). When a CRTC1 KD AAV vector was injected at one or two sites, we observed abnormally strong expression of immediate early genes (IEGs) around the injection sites, in conjunction with lack of CRTC1 protein expression. To our surprise, when KD vector injections were performed at three or more locations of the same hemisphere, strong IEGs expression expanded to the whole hemisphere, and gliosis and neuronal cell death was observed at the injection sites. We hypothesized that this abnormal IEGs may reflect development of epilepsy caused by CRTC1 KD. To confirm this point directly, we recorded the whole-cortical electrocorticography (ECoG) (in one hemisphere) from KD animals from 10 days after injection. About 3 weeks after injection, we observed the high frequency activities at the knockdown sites. Furthermore, we found the high frequency oscillations (HFOs) from the temporal lobe. Although, the voltage of these abnormal ECoGs were weaker than those of epileptic seizure observed from patients or model animals, the repetition of HFOs of the KD marmosets increased over time, finally leading to facial spasm corresponding to HFOs 2 months after injection. We also found that the correlation among electrodes transiently changed from three weeks after injection. Taken together, we conclude that CRTC1 KD causes abnormal excitatory activities in the marmoset V1 and that repetition of such abnormal neural activity leads to functional circuit degeneration in the brain, resulting in development of epilepsy. The abnormal IEGs expression by CRTC1 knockdown appeared reproducibly, suggesting the potential utility of this system as the model animals for the study of epileptogenic process. Furthermore, our data suggests some unknown molecular mechanism that link CRTC1 gene expression with epileptogenesis.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-558 脊髄小脳失調症42型モデルマウスを用いた神経変性分子病態基盤の解明 Shunta Hashiguchi（橋口俊太）¹，Hiroshi Doi（土井宏）¹，Misako Kunii（國井美紗子）¹，Yukihiro Nakamura（中村行宏）²，Misa Shimuta（志牟田美佐）²，Etsuko Suzuki（鈴木江津子）²，Masaki Okubo（大久保正紀）¹，Toshikuni Sasaoka（笹岡俊邦）³，Hideyuki Takeuchi（竹内英之）¹，Taro Ishikawa（石川太郎）²，Fumiaki Tanaka（田中章景）¹ ¹横浜市大神経内科・脳卒中医学 ²慈恵医大院医薬理 ³新潟大脳研リソース研究セ Spinocerebellar ataxia 42 (SCA42) is a neurodegenerative disorder recently shown to be caused by c.5144G>A (p.Arg1715His) mutation in CACNA1G, which encodes the T-type voltage-gated calcium channel CaV3.1. To determine whether this mutation causes ataxic symptoms and neurodegeneration, we generated knock-in mice harboring c.5168G>A (p.Arg1723His) mutation in Cacna1g, corresponding to the mutation identified in the SCA42 family. Both heterozygous and homozygous mice developed an ataxic phenotype from the age of 11-20 weeks and showed Purkinje cell loss at 50 weeks old. Degenerative change of residual Purkinje cells and atrophic thinning of the molecular layer were seen in homozygous knock-in mice. Electrophysiological analysis of Purkinje cells using acute cerebellar slices from young mice showed that the point mutation altered the voltage dependence of CaV3.1 channel activation and reduced the rebound action potentials after hyperpolarization, although it did not significantly affect the basic properties of synaptic transmission onto Purkinje cells. Finally, we revealed that the resonance of membrane potential of neurons in the inferior olivary nucleus was decreased in knock-in mice, which indicates that p.Arg1723His Cav3.1 mutation affects climbing fiber signaling to Purkinje cells. Altogether, our study first reveals not only that a point mutation in CACNA1G causes an ataxic phenotype and Purkinje cell degeneration in a mouse model, but also that the electrophysiological abnormalities in Purkinje cells and inferior olivary nucleus at an early stage of SCA42 precede the loss of Purkinje cells.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-559 SUMO化による進行性核上性麻痺におけるTau制御 Hironori Takamura（高村明孝）^1,2，Shinsuke Matsuzaki（松崎伸介）^2,3，Yoshiaki Nakayama（中山宜昭）⁴，Hidefumi Ito（伊東秀文）⁴，Taiichi Katayama（片山泰一）²，Paul E. Fraser（Fraser E. Paul）^1,5 ¹Tanz Centre for Research in Neurodegenerative Diseases, University of Toronto, Toronto, Canada ²大阪大学連合小児発達学研究科 ³和歌山県立医科大学薬理学講座 ⁴和歌山県立医科大学神経内科学講座 ⁵Department of Medical Biophysics, University of Toronto, Toronto, Canada Small ubiquitin-like modifiers (SUMOs) conjugated to target proteins regulate a number of roles in cellular events and have been implicated in several neurodegenerative diseases. The microtubule associated protein tau is modified by SUMOylation which has been shown to promote aggregation and regulate phosphorylation and degradation of the tau protein linked to Alzheimer disease (AD). The current study has demonstrated that SUMO1 co-localizes with intraneuronal tau inclusions and contributes to the pathological pathways in tauopathies. Immunoprecipitation of isolated and solubilized tau fibrils from progressive supranuclear palsy (PSP) brain tissue revealed that SUMO1 is covalently conjugated to an N-terminally truncated tau. The effects of SUMOylation were also examined using tau-SUMO1 fusion proteins which showed a higher propensity for tau misfolding and oligomerization that was specific for the PSP-truncated tau as compared to full-length tau protein. Comparable tau-SUMO2 fusion proteins displayed a normal diffuse cytoplasmic distribution and did not result in significantly altered levels of tau aggregation. Blocking proteasome-mediated degradation exacerbated the aggregation of the tau-SUMO1 fusion proteins. The specific SUMO1 modification of truncated tau in PSP may represent a detrimental event that promotes aggregation and impedes the ability of cells to remove the resulting protein deposits. These findings suggest that the combination of tau truncation and SUMO1 conjugation is a contributing factor in PSP pathogenesis.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-560 孤発性パーキンソン病遺伝子MCCC1におけるパーキンソン病発症メカニズム解析 Shunsaku Sogabe（曽我部俊策）¹，Wataru Satake（佐竹渉）²，Hiroko Nakano（中野浩子）¹，Morio Ueyama（上山盛夫）³，Ryusaku Matsumoto（松本隆作）⁴，Tetsushi Hamaguchi（浜口哲矢）¹，Daisuke Doi（土井大輔）⁴，Shigeo Murayama（村山繁雄）⁵，Jun Takahashi（高橋淳）⁴，Takashi Aoi（青井貴之）¹，Yoshitaka Nagai（永井義隆）³，Tatsushi Toda（戸田達史）² ¹神戸大院医 ²東京大院医神経内科 ³大阪大院医 ⁴京都大iPS細胞研 ⁵都健康長寿医療セ Parkinson's disease (PD) is a neurodegenerative disorder characterized by dopaminergic neuron loss. Recent genome-wide association studies (GWAS) have identified numerous genetic variants associated with multiple risk loci for sporadic PD, but mechanistic insights are impeded by a lack of understanding of how specific risk variants functionally contribute to the underlying pathogenesis. To address the possibility that variation in MCCC1 could play a role in PD pathogenesis, here we investigated MCCC1 function in a neuronal model system. We focused on a candidate single nucleotide polymorphism (SNP) in MCCC1 gene, and found that the SNP changes the gene expression of MCCC1 in human brain. To identify the mechanisms, we used human induced pluripotent stem cells (iPSCs) with CRISPR-Cas9 genome editing method and differentiated them into dopaminergic neuron. Our data showed that the PD-associated SNP of MCCC1 decreased oxygen consumption rate in mitochondria. Moreover, in vivo experiments using human Alpha-synuclein transgenic flies showed that Drosophila homolog of MCCC1 (CG2118) induced rough eye phenotype and age-dependent locomotor impairment. Taken together, these studies indicate that the PD-associated SNP in MCCC1 plays direct and critical role in the MCCC1 gene expression. These changes might affect mitochondrial function and PD pathogenesis, highlighting that this SNP can be a potential therapeutic target for PD.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-561 microRNAを用いた遺伝性パーキンソン病患者iPS細胞由来成熟神経細胞の作製 Etsuro Ohta（太田悦朗）^1,2,3,4，Akito Hattori（服部精人）⁵，Makiko Nagai（永井真貴子）⁶，Hiroyuki Miyoshi（三好浩之）⁴，Takefumi Sone（曽根岳史）⁴，Ryuto Nitta（新田龍人）³，Andrew Yoo（Yoo Andrew）⁷，Kazuya Iwabuchi（岩渕和也）⁵，Hideyuki Okano（岡野栄之）⁴ ¹北里大・医療衛生・細胞デザイン研究開発センター ²北里大・医療衛生・免疫学 ³北里大院・医療系・臨床免疫学 ⁴慶應義塾大・医・生理学 ⁵北里大院・医療系・細胞免疫学 ⁶北里大・医・神経内科学 ⁷Dept Dev Biol, Washington Univ Sch of Med, St. Louis, USA Parkinson's disease (PD) is one of the neurodegenerative diseases caused by progressive loss of midbrain dopaminergic neurons. Our university has reported that induced pluripotent stem cells (iPSC) derived from a patient with I2020T LRRK2 in a Japanese family (the Sagamihara family) replicate the pathologic phenotype evident in the PD brain to some extent. Recently, it has been reported that microRNA (miR) an induction directly convert fibroblasts into specific subtype neurons. It also has been reported that miR-based neuronal conversion of fibroblasts exhibited age-associated cellular properties such as DNA damage, telomere length, and DNA methylation. In this study, to examine whether miR-based induction generate PD iPSC-derived mature neurons, we performed induction of iPSC-derived neurons using miR9/9*-124-BclxL-KO1-lentiviral vector. After miR induction, we found that KO1-expressing iPSC-derived neurons showed neurite extension. At 21 days after miR induction, the numbers of MAP2-positive cells among miR-treated PD-iPSC-derived neurons were higher than those among untreated iPSC-derived neurons. Also, miR-based iPSC-derived neurons increased SA-β-gal (a cellular senescence marker)-positive cells relative to conventional iPSC-derived neurons. In addition, miR-based iPSC-derived neurons increased CellROX (a cellular senescence marker)-positive cells relative to conventional PD-iPSC- and isogenic-iPSC-derived neurons. These results indicate that miR-based iPSC-derived neurons facilitate abnormal phenotype and may serve as a useful tool to elucidate PD pathogenesis.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-562 知的障害を伴う自閉症スペクトラム障害に見られるGluA1遺伝子変異がAMPA受容体に与える影響 Taku Uchida（内田琢），Ryousuke Midorikawa（緑川良介），Kougo Takamiya（髙宮考悟）宮崎大学　医学部　機能制御学講座　統合生理学分野 Autism spectrum disorders (ASDs) are neurodevelopmental disorders affect 0.5 -1 % of children. They are characterized deficits in language and communication, impaired or abnormal social interactions, and restricted interests or repetitive behaviors. Some patients also have severe intellectual disability. Recent studies have identified gene mutations implicated in synaptic dysfunction in patients with ASD including <I>GRIA1</I>, encoding GluA1 subunit of AMPA receptor. AMPA receptor plays a primary role in major fast excitatory synaptic transmission and many neuronal functions including synaptic plasticity that underlie learning and memory. Thus, alternation of the AMPA receptor function caused by GluA1 gene mutations could contribute to the pathologies of ASDs through the synaptic dysfunction. To investigate, we prepared human GluA1 gene variants expression vectors (I627T, A636T, V640L), reported in ASDs patients with severe intellectual disability (Geisheker MR <I>et al </I>., 2017). We have investigated electrophysiological properties of AMPA receptors with the mutant GluA1s expressed on HEK293T cells. One mutant presented prolonged current duration and another one exhibited no current by 1mM L-Glutamate application. Especially in the A636T mutation has the property of the tonic inward current without agonist may possibly cause cytotoxicity. These results indicate that the gene mutation found in GluA1 could contribute to the pathomechanisms of ASD via modulating the AMPA receptor functions.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-563 メチマゾール誘発一過性甲状腺機能低下症マウスの行動、脳波、及び脳血流動態の解析 Norihiro Katayama（片山統裕），Yuto Yoshida（吉田侑冬），Ryusei Uchida（内田竜生），Mao Sasaki（佐々木真生），Ryunosuke Togawa（外川龍之介），Mitsuyuki Nakao（中尾光之），Kei-Ichi Itoi（井樋慶一），Katsuya Uchida（内田克哉）東北大学大学院情報科学研究科 Decreased thyroid function in the fetal and juvenile period causes mental retardation and motor function decline. It was found in juvenile mice that hypothyroidism causes a decrease in the density of parvalbumin-positive cells and abnormal formation of lamellar structures in the cerebral cortex and hippocampus. However, it remains unclear whether there are any significant abnormalities in physiological properties of the cortex, such as electroencephalogram and hemodynamics, or not. In this study, we performed behavioral testing, measurements of spontaneous electroencephalographic (EEG) activity, and optical imaging of cerebral hemodynamics of the hypothyroid mice. C57BL/6 mice (Male, n = 52) were used for the experiment. Hypothyroid mice were prepared by adding methimazole and potassium periodate to the drinking water of mother and offspring mice from 17 days of pregnancy to 14 days of parturition to suppress fetal/infantile thyroid hormones. This study was carried out under the approval of Tohoku Univ. animal experiment committee. The open field test revealed decreased activity in the hypothyroid mice. During non-REM sleep, the power of delta-band (0.5-4 Hz) EEG in the motor cortex of the hypothyroid mice was significantly larger than the wild-type mice. There was no significant difference between them in the resting-state functional connectivity (rsFC) in the cerebral cortex based on the hemodynamics. However, cross-correlation function analysis revealed that propagation of infraslow activity (0.01-0.1 Hz) from the motor to the visual cortex was slower in the hypothyroid mice. These results suggest that a histological abnormality of the cortex caused by the hypothyroidism would be expressed as abnormal dynamical properties of the spontaneous brain activity.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-564 自閉症責任領域15q11-q13モデルマウスにおける原因遺伝子の探索 Kota Tamada（玉田紘太）¹，Keita Fukumoto（福本景太）^1,2，Tsuyoshi Toya（戸谷豪志）^1,3，Nobuhiro Nakai（中井信裕）¹，Awasthi Janak（Janak Awasthi）^1,4，Sandra Ruf（Ruf Sandra）⁵，Francois Spitz（Spitz Francois）⁵，Toru Takumi（内匠透）^1,2 ¹理研CBS 精神生物学研究チーム ²広島大学　大学院医歯薬保健学研究科 ³慶応義塾大学薬学部 ⁴埼玉大学　大学院理工学研究科 ⁵EMBL, Heidelberg, Germany Autism spectrum disorders (ASDs) are neurodevelopmental disorders that exhibit poor social interaction and restricted and repetitive behaviors. One of the major causes is thought to be abnormal copy number variations (CNVs) such as chromosome 15q11-q13 duplication. Our previous studies revealed that mice with paternally inherited duplication of chromosome 7B-C (15q dup), corresponding to human chromosome 15q11-q13, represented ASDs-like abnormal behaviors, hypo-activity in the dorsal raphe nucleus, enhanced dendritic spine turnover and cortical excitatory/inhibitory imbalance. In the duplicated locus, there are 11 protein coding genes, several of snoRNAs and non-protein coding genes. Among them, Ube3a gene is a strong candidate gene for 15q11-q13 duplication syndrome because the duplicated chromosome is maternally derived in most affected cases and this gene expresses only from maternal allele. However, paternal duplications also lead to an increased risk of ASDs. Therefore, we hypothesized there are other ASD-linked genes in 15q11-q13 in addition to Ube3a. Here, we performed in vivo screening to identify responsible genes for ASDs related behaviors and altered spine dynamics. We generated 1.4 and 3 Mb duplication mice which contain the truncated chromosome 7B-C excluding several paternally expressed genes (Ndn, Magel2 and Mkrn3). These mice did not replicate the behavioral abnormalities found in 15q dup mice, indicating Ndn, Magel2 and Mkrn3 are the candidates. We then examined the effect of over expression of these 3 genes on spine dynamics of cortical pyramidal neurons by two-photon in vivo imaging. As a result, over expression of Ndn drastically enhanced the spine formation rate. Furthermore, electrophysiological analysis revealed Ndn overexpressed neurons had hyper-excitability, similar to 15q dup mice. Finally, we removed single copy of Ndn from 15q dup mice by CRISPR-Cas9 system (15q dupΔNdn) and validate whether this mouse shows similar abnormality found in 15q dup mice or not. 15q dupΔNdn mice did not exhibit abnormal social behaviors, reversal learning disability, cortical excitatory/inhibitory (E/I) imbalance, abnormal micro structure in the dorsal raphe nuclei, enhanced dendritic spine turnover found in 15q dup mice. Taken together, Ndn is a critical role for multiple abnormal phenotypes found in 15q dup mice and a possible risk factor for ASDs.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-565 Therapeutic potential and underlying mechanism of sarcosine (N-methylglycine) in N-methyl-D-aspartate (NMDA) receptor hypofunction models of schizophrenia Wen-Sung Lai（Lai Wen-Sung）¹，Ju-Chun Pei（Pei Ju-Chun）¹，Wei-Li Hung（Hung Wei-Li）¹，Bei-Xuan Lin（Lin Bei-Xuan）¹，Min-Han Shih（Shih Min-Han）¹，Liang-Yin Lu（Lu Liang-Yin）¹，Da-Zhong Luo（Luo Da-Zhong）¹，Hwan-Ching Tai（Tai Hwan-Ching）¹，Vincent Studer（Studer Vincent）^2,3，Ming-Yuan Min（Min Ming-Yuan）¹ ¹National Taiwan University ²Interdisciplinary Institute for Neuroscience, University of Bordeaux, F-33077, Bordeaux, France ³French National Center for Scientific Research (CNRS), UMR 5297, F-33077, Bordeaux, France Compelling animal and clinical studies support the N-methyl-d-aspartate receptor (NMDAR) hypofunction hypothesis of schizophrenia and suggest promising pharmacological agents to ameliorate negative and cognitive symptoms of schizophrenia. Sarcosine, a glycine transporter-1 inhibitor, was found to improve schizophrenic symptoms in add-on and monotherapy clinical studies. It is imperative to evaluate the therapeutic effect of sarcosine in animal models, which provide indispensable tools for testing drug effects in detail and elucidating the underlying mechanisms. In this study, a series of 7 experiments was conducted to investigate the therapeutic effect of sarcosine in the amelioration of schizophrenia-related behavioral deficits and the underlying mechanism in pharmacological (i.e., MK-801-induced) and genetic (i.e., serine racemase-null mutant (SR-/-) mice) NMDAR hypofunction models of schizophrenia. In Experiments 1, our results indicated that the acute administration of 500/1000 mg/kg sarcosine (i.p.) had no adverse effects on motor function and serum biochemical responses. In Experiments 2-4, sarcosine significantly alleviated MK-801-induced (0.2 mg/kg) brain abnormalities and schizophrenia-related behavioral deficits in MK-801-induced and SR-/- mouse models of schizophreniaNMDAR hypofunction. In Experiment 5, the injection of sarcosine enhanced CSF levels of glycine and D-serine in rat brain. In Experiments 6-7, we show for the first time that sarcosine facilitated NMDAR-mediated hippocampal field excitatory postsynaptic potentials (fEPSPs) and influenced the movement of surface NMDARs at extrasynaptic sites. Collectively, sarcosine effectively regulated surface trafficking of NMDARs, NMDAR-evoked electrophysiological activity, brain glycine levels, and MK-801-induced abnormalities in the brain, which contributed to the amelioration of schizophrenia-related behavioral and cognitive deficits in NMDAR hypofunction models of schizophrenia.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-566 Optimization of the parameters of magnetic resonance imaging for the intracranial human U87 glioblastoma orthotopic xenograft mouse model Wen Hsin Chang（Chang Wen Hsin），Jayaseema Dasan Mary（Dasan Mary Jayaseema），Ming-Huang Lin（Lin Ming-Huang） Institute of Biomedical Sciences, Academia Sinica, Taiwan Glioblastoma is the most frequent and malignant primary brain tumor in adults. Magnetic resonance imaging (MRI) is used for monitoring the growth of gliomas and has been applied to murine models of glioma to evaluate the treatment effects of new drugs. In this study, the intracranial human U87 glioblastoma mouse orthotopic xenograft model was used with 6-week-old male Nu/Nu mice and 8-to-10 week-old male NOD-SCID mice. Many conditions of the mouse model were optimized, including injected cell volume, injection speed, syringe gauge, and the speeds of the insertion and withdrawal of the syringe into and out of the brain. For the injections of male Nu/Nu mice (6 weeks old), the U87 cells were harvested from a culture dish and then resuspended with more than 200 μL of 4°C phosphate buffered saline (PBS) in a 2-mL Eppendorf tube with a cell density of 2 × 10e5 cells in 1 μL of PBS. Thereafter, we kept this tube of cells on a shaker at 5 rpm and 360° rotation in a 4°C cold room until the intracranial injection of a mouse. The cell survival ratio was monitored using trypan blue stain just before injection, and results were as follows: Mouse 1 (99%), Mouse 2 (88%), and Mouse 3 (84%). The cell injection amount for each mouse was adjusted to 5 × 10e5 cells per mouse, according to the cell survival ratio at that time. For the injections of male NOD-SCID mice (9 weeks old), the U87 cells were freshly harvested from each mouse and resuspended using 10 μL of room temperature (RT) PBS. The cell survival ratio was also monitored using trypan blue stain just before injection and the results were as follows: Mouse 1 (96%) and Mouse 2 (97.5%). Then, the total volume of 5 × 10e5 cells in 2.0 μL of RT PBS was injected into each NOD-SCID mouse with an injection rate of 500 nL/min. In Nu/Nu mice, the tumor cells exhibited a growth burst between Day 24 and Day 36 after U87 cells were injected under the following parameters: field of view = 20 mm, slice thickness = 0.80 mm, 20 slices, repetition time (TR) = 4000 ms, echo time (TE) = 60 ms, average = 8, refocusing flip angle = 180.0 deg., and matrix size = 256 × 256. Additionally, we compared two sets of parameters for MRI scanning in the U87 injected NOD-SCID mice and found that the modality and some inside content of the tumor microenviroment could show at Day 7 vividly with the TR 2482.4 ms and TE 37.8 ms but not with the aforementioned TR 4000 ms and TE 60 ms.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-567 Roles of rat glutamate decarboxylase isoforms in the perinatal period Weiru Jiang（Jiang Weiru）¹，Toshikazu Kakizaki（Kakizaki Toshikazu）¹，Kazuyuki Fujihara（Fujihara Kazuyuki）¹，Shigeo Miyata（Miyata Shigeo）¹，Yue Zhang（Zhang Yue）¹，Takashi Suto（Suto Takashi）²，Daiki Kato（Kato Daiki）²，Shigeru Saito（Saito Shigeru）²，Koji Shibasaki（Shibasaki Koji）³，Yasuki Ishizaki（Ishizaki Yasuki）³，Yoshiki Miyasaka（Miyasaka Yoshiki）⁴，Tomoji Mashimo（Mashimo Tomoji）⁴，Yuchio Yanagawa（Yanagawa Yuchio）¹ ¹Department of Genetic and Behavioral Neuroscience, Gunma University Graduate School of Medicine ²Department of Anesthesiology, Gunma University Graduate School of Medicine ³Department of Molecular and Cellular Neurobiology, Gunma University Graduate School of Medicine ⁴Institute of Laboratory Animals, Graduate School of Medicine, Osaka University, Suita, GABA is a major neurotransmitter in the mammalian CNS. Two GABA synthetic pathways are known to originate from glutamate and putrescine in the brain and retina. Glutamate decarboxylase (GAD) synthesizes GABA from glutamate, and two isoforms of GAD, GAD65 and GAD67 are encoded by separate genes. In recent years, we have been exploring a novel function of GAD isoforms by comparing the knockout phenotypes between mice and rats. GAD65 knockout (GAD65-/-) mice showed an increase in susceptibility to seizures and an elevated level of anxiety. Approximately 25% of the knockout mice died till 6 months of age. On the other hand, GAD67-/- mice showed cleft palate and died at postnatal day 0 (P0). By contrast, GAD65-/- rats showed spontaneous seizures during the third postnatal week and 80% of them died till P23, while a considerable number of GAD67-/- rats survived after 8 weeks of age and had lower body weight than GAD67+/+ rats at weaning. These results indicate the phenotypes are different in severity between GAD knockout mice and rats. In the present study, we investigated the relationship between rat GAD genotypes and their phenotypes. All GAD65-/-; GAD67-/- rats and GAD65+/-; GAD67-/- rats died at P0. All GAD65-/-; GAD67-/- rats and 12.5% of GAD65+/-; GAD67-/- rats showed cleft palate at embryonic day 20 (E20), but no rats of the other genotypes showed it. These results indicate both GAD65 and GAD67 play an important role in the rat palate formation. The body weight of E20 GAD65-/-; GAD67-/- (n = 4) rats was lower than that of GAD65+/+; GAD67+/+ (n = 6) rats, while the body weight of E20 GAD65+/+; GAD67-/- (n = 2) rats and GAD65-/-; GAD67+/+ (n = 6) rats was similar to that of GAD65+/+; GAD67+/+ (n = 6) rats. These results suggest that both GAD65 and GAD67 are important for fetal growth. We examined the response of E20 rats to mechanical stimuli. None of the GAD65-/-; GAD67-/- rats responded to a pinch of the tail by forceps, but all GAD65+/+; GAD67+/+ rats responded with a twisting of the trunk. These results suggest that GAD65-/-; GAD67-/- rats suffer from severe impairments in motor function. GABA content in GAD65-/-; GAD67-/- rat brains and retinas at E20 is extremely low, indicating that almost all of GABA is synthesized from glutamate by GADs in the perinatal period. Based on these results, molecular and morphological phenotypes in GAD mutant rat brains will be further addressed.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-568 <I>Scn1a</I> KOマウス海馬由来抑制性神経細胞の機能解析 Yasuyoshi Tanaka（田中泰圭）¹，Kevin M Moreira（モレイラケビン）¹，Shinichi Hirose（廣瀬伸一）^1,2 ¹福岡大学てんかん分子病態研究所 ²福岡大学医学部小児科 Dravet syndrome (DS), a devastating type of infantile-onset epilepsy that presents with cognitive deficits and autistic traits, is caused by a mutation in <I>SCN1A</I>, which encodes the α-subunit of the voltage-gated sodium channel, Na<SUB>v</SUB>1.1. Several types of mutations, including nonsense, frame-shift, and missense mutations, located at different sites in <I>SCN1A</I> have been identified in patients with DS. DS, or Severe Myoclonic Epilepsy of Infancy (SMEI), is a severe and rare form of intractable childhood epilepsy, caused mainly by a mutation in the <I>SCN1A</I> gene that impairs the hippocampal inhibitory neuron excitability while the striatum excitatory neuron function is maintained causing an imbalance resulting in the epileptic activity. By use of Axion Biosystems's Maestro MEA (Multi Electrodes Array) systems, we captured, in vitro, the neuronal activity of both Wild Type (WT) and Knockout (KO) mice Hippocampus Inhibitory Neurons in the form of Spikes (NS), Bursts and Network Bursts (NB). Compared to WT mice, KO mice had a maturement delay and took longer to start emitting NS and NB. Also, WT Hippocampus Neurons' NB would start earlier but less than a month later would mostly be absent while in the KO Neurons it had a delayed start but grew exponentially. We theorized this is possibly due to NB blockage caused by unimpaired GABA activity in the WT mice as Hippocampus are known to have a significant percentage (10-20%) GABA neurons. When adding Biccuculine, a known GABA-Blocker to the WT cells we were able to notice a decrease in the number of NB, but that they lasted longer compared to the non-Biccuculine cultures. These findings provide compelling evidence that the inhibitory neurons are affected by <I>SCN1A</I> mutations and that this system might be applicable for drug development as it opens up a new potential path for the treatment of Dravet Syndrome.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-569 正常及び癲癇原性獲得後の海馬介在神経細胞サブタイプ間におけるGAD65の発現 Yuki Kajita（梶田裕貴），Yuki Fukuda（福田湧希），Takanori Oyanagi（大柳貴紀），Hajime Mushiake（虫明元）東北大院医生体システム <B>Introduction:</B> Gamma-aminobutyric acid (GABA) is widely distributed inhibitory neurotransmitter in the central nervous system. GABA is synthesized from glutamate by two isoforms of glutamic acid decarboxylase (GAD) 67 and GAD65. These GAD67 and GAD65 are necessary for regulating normal neural activity and regarded to play distinct roles. GAD65 is related to the activity-dependent rapid GABA synthesis and GAD65-expression decreases in seizure-prone mice hippocampus, suggesting that GAD65 has crucial role to suppress the epileptic seizure. The GABAergic neurons have many subpopulation that can be classified using several chemical makers. However, it is unclear which subtypes of interneuron contribute to the acquisition of vulnerability to epileptic seizure. To investigate this issue we examined the distribution of GAD65-expression in each interneuron subtypes in normal and seizure-prone hippocampus. <B>Methods:</B> To investigate seizure-prone hippocampus, we made chemical-kindling rats using GABA-A antagonist, pentylenetetrazole (PTZ). We injected PTZ in Long-Evans rats repeatedly (40 mg/kg, i.p., 20 times, one shot/day) and monitored the daily development of seizure stage. The control (CTL) group were injected equal volume of saline. Rats were sacrificed and their brain were removed 24h after last injection. We calculated the GAD65-intensity (expression level) in the cell bodies merged with several GABAergic subtype markers across the hippocampus. <B>Results:</B> The GAD65-expression level is significantly differenct among interneuron subtypes in the CTL (normal) hippocampus. Interneuron subtypes were divided two groups in terms of GAD65-expression level, high-expression group: cholecystokinin-positive (CCK+), neuropeptide Y (NPY)+ or neuronal nitric oxide synthase (nNOS)+ neuron; low-expression group: parvalbumin (PV)+, somatostatin (SOM)+ or calretinin (CR)+ neurons. In seizure-prone hippocampus, the GAD65-expression decreased in every GABAergic subtype, but in particular, CCK+ neuron showed dramatically decrease compared with other subtypes. <B>Summary:</B> GAD65 is heterogeneously expressed among multiple-interneuron subtypes. Additionally, it is suggested that hippocampal vulnerability to epileptic seizure is closely related to selective changes in GAD65-expression among interneuron subtypes.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-570 カイニン酸誘発痙攣と熱性痙攣におけるプロスタグランジンE<SUB>2</SUB> EP3受容体の関与 Yuri Ikeda-Matsuo（松尾由理）^1,2，Ryouma Yamamoto（山本稜真）¹，Shino Nakamura（中村柴乃）¹，Sachie Hirano（平野幸恵）²，Hiroto Ishikawa（石川弘人）²，Yuji Kubo（久保祐二）²，Syuh Narumiya（成宮周）³，Mitsuo Tanabe（田辺光男）² ¹北陸大薬薬学臨床薬理 ²北里大薬薬理 ³京都大院医創薬医学 We have recently demonstrated that prostaglandin E<SUB>2</SUB> EP3 receptors are involved in stroke-reperfusion injury and glutamate-induced excitotoxicity. In this study, we investigated the involvement of EP3 receptors in seizures of two acute models, kainic acid (KA)-induced seizures and febrile seizures. In a KA-induced seizure model, KA was injected (ip) to wild-type (WT) or EP3 knockout (EP3KO) mice. In a febrile seizure model, seizures were induced by pretreatment of lipopolysaccharide (ip) at 2 h before increasing body temperature using infrared radiation lamp in 10 day-old WT mice. The seizure score for 2 h after KA-injection in EP3KO mice was significantly lower than that in WT mice. The increases in numbers of c-fos-positive cells at 4 h after KA-injection and GFAP- and Iba-1-positive cells at 3 days after injection were significantly less in EP3KO mice as compared with WT mice. The increases in mRNA expressions of COX-2, mPGES-1 and EP3 receptors, as well as GFAP and Iba-1, and production of prostaglandin E<SUB>2</SUB> observed in WT mice was also significantly less in EP3KO mice. Furthermore, KA induced severe neuronal loss in hippocampal CA3 region of WT mice, but not in that of EP3KO mice. In behavioral test, KA was injected repeatedly every hour until induction of status epilepticus. The total dose of KA in EP3KO mice was higher than that in WT mice, however, the seizure score in EP3KO mice was significantly less than that in WT mice. The spatial memories measured by Y-maze test were impaired by KA, and there were no significant differences between genotypes. Although total migratory distance in open field test was similar in all groups, the time spent in center area in WT mice, but not EP3KO mice, was significantly prolonged by KA. On the other hand, expression of EP3 receptors after febrile seizures were decreased, even those of IL-1β and TNF-α were increased. These results suggest that activation of EP3 receptors after KA injection contributes to seizure susceptibility, glial activation, hippocampal neuronal loss, and reduction in anxiety, and that involvement of EP3 receptors are absolutely different between two seizure models, KA-induced seizures in adult mice and febrile seizures in infant mice. Although the role of EP3 receptors in febrile seizures should be determined using EP3KO infants, inhibition of EP3 receptors will be a valuable therapeutic option in treatment of temporal lobe epilepsy.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-571 学習による脳の情報表現の変化 Yoshiki Yoshii（吉井誉揮）^1,2，Mitsuo Kawato（川人光男）¹，Cortese Aurelio（Aurelio Cortese）¹ ¹国際電気通信基礎技術研究所 ²奈良先端科学技術大学院大学 Humans have the ability to generalize quickly their behavior to new tasks by utilizing knowledge gained through past experience. Previous studies suggested that the interaction between the prefrontal cortex (learning abstract rules) and the basal ganglia (learning specific task aspects) underlies generalization. However, the dynamics of this interaction mechanism between the two brain regions has yet to be fully elucidated. In this study, we measured how information transfer between these regions changed with learning by applying multivariate analysis to human functional magnetic resonance imaging (fMRI) data. We found that the interaction changes at a specific time point as learning progresses. In addition, we found that interaction increased rapidly before the subject selects an action related to reward. Furthermore, we estimated the functional dimensionality of the brain region by a method combining singular value decomposition and cross-validation. We found that the dimensionality of the brain region changes by learning in specific frontal subregions. Our findings suggest that as learning progresses, the recurrent loops connecting the prefrontal cortex and the basal ganglia are optimized to emphasize task-specific loops and to increase abstraction level and reduce dimensionality.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-572 深層学習ネットワークを用いた予測符号化説による錯視再現の一般性の検証 Taisuke Kobayashi（小林汰輔）¹，Akiyoshi Kitaoka（北岡明佳）²，Kiwako Sakamoto（坂本貴和子）³，Kenta Tanaka（田中健太）⁴，Eiji Watanabe（渡辺英治）^1,5 ¹基生研神経生理学 ²立命館大総合心理学部 ³自然科学研究機構研究力強化推進本部 ⁴サクラ研究事務所 ⁵総研大 Neural networks, which have been proposed as simple mathematical models of the brain, have been widely applied. Our research group has introduced mathematical models of the brain into deep neural networks (DNN) and utilizes them as a brain simulator to clarify the human visual information processing mechanism. The predictive coding is one of theories for computer science, and explains the visual information processing performed in the cerebral cortex. The theory assumes that the brain predicts what happens in the world and how change the world in real time, and then optimizes the internal predictive system to minimize the error between the observation and the prediction. The theory has been verified actively by both experiments and numerical computations. In 2018, we had presented an interest result that when the DNN incorporating the predictive coding was input one of the moving illusion images, the predictive image reproduced the illusory phenomena in a manner similar to human perception. These results suggested the effectiveness of the predictive coding theory. Optical illusion is a perception phenomenon that makes a difference between physical parameters of the real world and psychological parameters that the brain causes. By analyzing the mechanism that produces the difference, we could understand the visual information processing performed by neural circuits. In this study, DNN incorporating predictive coding theory was used as a cerebrum simulator to show the generality of the illusion phenomenon appearance by DNN. The test stimuli were prepared as 300 kinds of moving illusion images, and the same number of general images from photographs and pictures without illusion. Each predicted image was generated from the trained DNN, and the temporal change of the predicted image was calculated as an optical flow to obtain statistical data. As a result, a clear difference appeared between the illusion image and the general image, and the result supported the validity of the predictive coding theory. In this presentation, we will also discuss the effectiveness of brain simulation research by DNN.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-573 吸息性ニューロン間の因果的連結性の推定とその実験的検証 Fumikazu Miwakeichi（三分一史和）^1,2，Yoshihiko Oke（尾家慶彦）³，Yoshitaka Oku（越久仁敬）³，Swen Hulsmann（Hulsmann Swen）^4,5，Johannes Hirrlinger（Hirrlinger Johannes）⁶ ¹統計数理研究所モデリング研究系 ²総合研究大学院大学統計科学専攻 ³兵庫医科大学生理学生体機能部門 ⁴Center for Nanoscale Microscopy and Molecular Physiology of the Brain (CNMPB), Gottingen, Germany ⁵The Clinic for Anesthesiology, Univ Medical Center, Gottingen, Germany ⁶Carl-Ludwig-Institute for Physiology, University of Leipzig, Leipzig, Germany Recent advances in bioimaging technology have enabled neuroscientists to examine aspects of brain activities at various levels from single neuron to region. Since imaging data can be continuously recorded at multiple measurement points arranged in a plane grid, it became possible to observe the spatiotemporal activities of several hundred of neurons simultaneously. Using two-photon microscopy and calcium imaging, we have investigated the properties of neurons involved in respiratory rhythm generation in the brainstem. For the sake of data analysis, General linear Model (GLM) has become a popular approach. Using Cross-correlation analysis, which is one of a family of GLM, temporal relation of a pairs of inspiratory neurons can be statistically evaluated. However, causality of the activity among inspiratory neurons cannot be evaluated, because cross-correlation analysis can evaluate only morphological resemblance of the form of two signals. In this study, we attempted to evaluate causal connectivity among inspiratory neurons based on statistical time series analysis. We applied two models, autoregressive (AR) mode and autoregressive model with exogenous inputs (ARX), which considers inputs from other inspiratory neurons, to the same time series of intracellular calcium fluctuation obtained from an inspiratory neuron. These models were compared using Akaike Information Criteria (AIC) procedure to choose the model that is best supported by the data. In the case ARX model is chosen, exogenous input from other neurons is significant for the inspiratory neuron. On the contrary, if AR model is chosen, the inspiratory neuron is likely to be spontaneously activated and/or affect other neurons. Moreover, in this study, we conducted experiments that blocked synaptic connections among neurons by the addition of antagonists (Strychnine, Bicuculline and CNQX). Then, we developed a strategy to verify the existence of the network by quantitatively comparing the parameters of the time series model applied to the data before and after the blockades.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-574 時空間的な遺伝子/細胞ターゲティングを目的としたオールインワン型二重支配shRNA発現ベクターの構築と導入 Hideto Oyamada（小山田英人）¹，Yoko Ao（青暢子）^1,2,8，Yoshiko Kudo（工藤芳子）^1,7,8，Tomoyuki Matsuoka（松岡朋之）¹，Reiko Okamoto（岡本怜子）^1,8，Takashi Murayama（村山尚）⁴，Toshiko Yamazawa（山澤徳志子）⁵，Sho Kakizawa（柿澤昌）⁶，Tatsunori Oguchi（小口達敬）^1,8，Atsushi Michael Kimura（木村篤史）^1,3,8，Taro Yasumoto（安本太郎）^1,3,8，Yukiko Mori（森友紀子）^1,3,8，Mayumi Tsuji（辻まゆみ）^1,8，Katsuji Oguchi（小口勝司）^1,8，Yuji Kiuchi（木内祐二）^1,8 ¹昭和大・医・薬理学 ²昭和大・電顕研 ³昭和大・医・神経内科 ⁴順天堂大・医・薬理 ⁵慈恵医大・医・分子生理 ⁶京都大・院薬・生体分子認識 ⁷国際医療福祉大・成田保医・医学検査 ⁸昭和大・薬理科学研究センター <B>Purpose</B>: Gene silencing by RNAi is widely used for assessing gene function. An important advance in the RNAi field was the discovery that plasmid-based RNAi can substitute for synthetic small interfering RNA in vitro and in vivo. However, constitutive and ubiquitous knockdown of gene expression by RNAi in mice can limit the scope of experiments because this process can lead to embryonic lethality, or result in compensatory overexpression of other genes such that no phenotypic abnormalities occur. Either way, analyses of the physiological roles of the gene of interest in adult mice are not possible. To overcome these limitations, we previously constructed a double-conditional short-hairpin RNA (shRNA) expression vector that can regulate shRNA expression in a spatio-temporal manner with a tetracycline (Tet)-inducible floxed stuffer sequence selectively excised by application of Cre recombinases(Matsuoka T et al., 2013). In this study, we aimed to modify this vector to create an all-in-one vector that produces double-conditional transgenic mice through a single round of gene transfer to fertilized eggs. <B>Methods</B>: We added the coding region for one of the four kinds of Cre(Cre, Cre-MODC, DD-Cre and Cre-ERT2) having a multi-cloning site for a cell-specific promoter between the pair of modified loxp-like TATA-lox as the floxed sequences (Ventura A., et al., 2004) of the double-conditional shRNA expression vector originally from the commercially available (?) pSingle-tTS-shRNA vector. For the integration of the core region of this all-in-one vector into the genome of the mammalian cells, we carried out the Transposon plus Insulators and Super PiggyBac Transposase systems (Transposagen Inc,), which have been reported to be capable of mobilizing 100kb DNA fragments in mouse embryonic stem cells. <B>Findings</B>: We confirmed successful construction of the vector. First, functional recombination of the floxed sequence TATA-lox was confirmed by analyzing restriction enzyme-digested fragments. Second, the expression of four kinds of Cre was confirmed by western blotting. Third, we also confirmed the genetic integration of the core region of the all-in-one vector into the cultured cells. <B>Conclusions</B>: This all-in-one double-conditional shRNA expression vector will be useful for reversible in vitro and in vivo knockdown of target gene expression, in target cells via promoter-specific expression of Cre, and at specific times by Tet application.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-575 AAV-PHP.BのBBB透過性は種々の近交系マウス間で共有されている Yasunori Matsuzaki（松﨑泰教），Masami Tanaka（田中真実），Sachiko Hakoda（箱田祥子），Tatsuki Masuda（増田龍樹），Ryota Miyata（宮田遼太），Hirokazu Hirai（平井宏和）群馬大学院医脳神経再生医学 Adeno-associated virus (AAV) vector is a powerful tool for transgene expression in the central nervous system (CNS) of mammals. In 2016, Deverman et al. reported a capsid variant of AAV serotype 9 (AAV-PHP.B) highly permeable to the blood-brain barrier (BBB) in C57BL/6J mice, which enabled us to express transgenes throughout the CNS simply by the intravenous injection of AAV-PHP.B. Meanwhile, a different group found failure of the BBB penetration when they used a different mouse strain (BALB/c), and suggested that BBB transmission of AAV-PHP.B might be limited to a C57BL/6 strain. Here we further characterized the AAV-PHP.B in terms of the BBB permeability in different mouse strains. The BBB permeability was evaluated by GFP fluorescence intensity of the whole brains 2 weeks after intravenous injection of AAV-PHP.B expressing GFP. As the results, AAV-PHP.B showed great BBB permeability in different mouse strains such as C57BL/6J, C57BL/6N, DBA/2, SJL/J and FVB/N. On the other hand, we found that the F1 hybrid obtained by crossing C57BL/6N and BALB/c became entirely impermeable to AAV-PHP.B. Intriguingly, ICR mice, a closed colony, showed variable BBB permeability. These results indicate that great BBB permeability of AAV-PHP.B is not restricted to C57BL/6 strain, and rather, the impermeability is likely limited to BALB/c and the filial generation. Comparison of the BBB structure between BALB/c and other strains may provide a key insight for the mechanism that defines the BBB transmission efficacy of AAV-PHP.B.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-576 Wide field surface emission deep ultraviolet based fluorescence microscope for whole brain imaging Deepa Kamath Kasaragod（Kasaragod Deepa Kamath），Zhu Meina（Meina Zhu），Hidenori Aizawa（Aizawa Hidenori） Hiroshima Univ Three-dimensional cellular scale imaging techniques that improve our understanding of the brain architecture is crucial for structural and functional integration and analysis of normal and pathological conditions in the brain. We have developed a wide-field fluorescent microscope using deep ultra violet (UV) light emitting diode (285 nm) as the illumination source. This microscope employs oblique illumination of deep UV light and the optical sectioning is obtained on the tissue surface over a few micron thickness; largely attributed to the large absorption and thus low tissue penetration of the deep UV light. The fluorescence emissions are obtained in the visible range using dye or a combination of dyes that stain only the nucleus and the cytoplasm such as DAPI, Hoechst, Rhodamine-B, etc. The fluorescence signal is captured using water immersion objectives (Olympus UMPLFLN 10XW; Olympus UMPLFLN 20XW) and a color camera (Point Grey; GS3-U3-51S5M-C). Arduino Mega 2560 controlled 3-axis motorised microscope stage is developed for obtaining the wide-field imaging of the entire animal brain section. The data acquisition along with the camera and the microscope stage is then controlled using Python programming language. To enable 3D imaging, the microscope set-up is integrated with Compresstome (VF-700-0Z) vibrating microtome that can slice thin sections lower than the UV excitation depth. In this study, we specifically discuss the serial-sectioning images obtained for the whole block Hoechst 33258-stained mouse brain.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-577 新しい広視野２光子励起顕微鏡Cosmoscopyによる刺激符号化ニューロンの同定 Keisuke Ota（太田桂輔）¹，Yasuhiro Oisi（大石康博）¹，Takayuki Suzuki（鈴木崇之）¹，Tsubasa Ito（伊東翼）^1,2，Yoshiki Ito（伊藤圭基）^1,3，Kenta Kobayashi（小林憲太）⁴，Midori Kobayashi（小林碧）¹，Maya Odagawa（小田川摩耶）¹，Chie Matsubara（松原智恵）¹，Junya Matsushita（松下純也）⁵，Hiroyuki Hioki（日置寛之）⁶，Masamichi Ohkura（大倉正道）⁷，Junichi Nakai（中井淳一）⁸，Masafumi Oizumi（大泉匡史）^1,3，Atsushi Miyawaki（宮脇敦史）¹，Toru Aonishi（青西亨）²，Takahiro Ode（大出孝博）^1,9，Masanori Murayama（村山正宜）¹ ¹理研脳神経科学研究センター ²東工大院 ³東京大学 ⁴生理研 ⁵浜松ホトニクス ⁶順天堂大学医学部 ⁷九州保健福祉大学 ⁸東北大学 ⁹株式会社フォブ In vivo two-photon calcium imaging with a single cell resolution over a wide field of view (FOV) is one of the most attractive techniques in neuroscience. This technique has propelled us to attempt to answer several basic questions, related to the extent of the map of the neurons coding similar information, the difference in the maps between excitatory and inhibitory neurons, and the distribution of neurons with a strong correlation across different areas, among other. To answer these questions, we have previously reported the wide-field two-photon microscopy, Cosmoscopy, that is equipped with a novel objective lens, tube lens and scan lens with large pupil diameters and low aberrations over the FOV. We also developed a low computational cost cell detection (LCCD) algorithm that has a high ability to detect neurons from our megapixel data (Ito et al., bioRxiv 502153). Finally, we succeeded in extracting more than 16,000 neural activities of L2/3 excitatory neurons in an awake mouse. As the next step, we identified the functional map of the neurons. A linear support vector machine was applied to classify the neurons into stimulation-related and stimulation non-related neurons (p < 0.001, compared to null dataset). As expected in previous studies, the neurons in the primary somatosensory cortex of the hindlimb region were activated, and the neurons in the secondary somatosensory cortex were subsequently activated. Although most of the activated neurons were localized in the somatosensory areas, some were scattered in other areas. Additionally, we identified the functional map of movement-related neurons using the Chi-squared test (p < 0.005). These neurons were not as localized as the ones related to the hindlimb stimulation.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-578 ex vivo マンガン造影MRI法の改変による脳構造の高解像度イメージング Chika Sato（佐藤千佳）^1,2，Kazuhiko Sawada（澤田和彦）³，Wright K David（David K Wright）⁴，Tatsuya Higashi（東達也）¹，Ichio Aoki（青木伊知男）⁵ ¹量子科学技術研究開発機構　放射線医学総合研究所　分子イメージング診断治療研究部 ²量子科学技術研究開発機構　量子生命科学領域　量子生命情報科学グループ ³つくば国際大・医療保健・栄養 ⁴Monash University, Melbourne, Australia ⁵量子科学技術研究開発機構　量子生命科学領域　量子制御MRIグループ In vivo manganese-enhanced magnetic resonance imaging (MEMRI) is a neuroimaging method for visualizing brain structures related to neuronal function that exploits the mobilization of Mn2+ through Ca2+ channels to enhance tissue contrast in living animals (Lin, et al., Magn Reson Med, 1997, Pautler, et al., Magn Reson Med, 1998, Aoki, et al., NeuroImage, 2004). Although in vivo MEMRI is useful for functional and kinetic detections, it is difficult to obtain high-resolution 3D images due to physiological motions including cardiac and respiratory motions. Previous ex vivo micro-imaging studies succeeded in an improvement of contrast and shorten acquisition times in gadolinium-enhanced MRI (Sharief, et al., Magn Reson Med, 2006, Kamsu, et al., PLoS One, 2013). Here, we attempted to optimize an ex vivo approach based on MEMRI for visualizing detailed 3D microstructures of the manganese-bound mouse brain. Mice received 50 mM MnCl2 (osmotic pressure-controlled, 100 mg/kg B.W., 250 ul/h) via tail veins. T1-weighted images were acquired in vivo using a 7T preclinical MRI (Avance-III, 20 cm bore, Bruker Biospin) with a cooled RF coil 24hr after the manganese administration. Just after the perfusion fixation by 4% paraformaldehyde, ex vivo T1-weighted images were achieved by the same MRI procedures. In vivo MEMRI visualized brain microstructures, such as layer structures of several brain regions, i.e., the olfactory bulb, cerebral cortex, hippocampal formation, and cerebellum consisting with previous reports (Silva, et al., Schizophr Bull. 2008), although attenuating T1-weighted signals in the gray matter compared to in vivo MEMRI. Ex vivo MEMRI enabled us to visualize more detailed brain structures by a clear dissociation of white matter structures due to the higher spatial resolution. The ex vivo MEMRI method introduced in this study may not only used as a micro-imaging method but can also be used for depolarization-dependent functional MEMRI (AIM-MEMRI) and tract tracing with high-resolution imaging in disease models.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-579 多様体学習によるマウス脳質量顕微鏡イメージングデータの解析 Hiromichi Suetani（末谷大道）^1,2，Fumihiro Eto（衞藤史博）^3,4，Ikuko Yao（矢尾育子）^3,4,5 ¹大分大理工 ²理研CBS 脳リズム情報処理連携（BTCC） ³浜松医科大学光尖端医学教育研究センター ⁴関西学院大理工 ⁵浜松医科大学国際マスイメージングセンター Recently, the methodologies of imaging mass spectrometry (IMS) [1], which is the combinations of mass spectrometry that ionizes chemical species and sorts the ions based on their mass-to-charge ratio with spatial visualizations, are receiving attention in the various fields of science and technology. When the IMS is applied to biological tissues such as brain slices, we obtain the information on particles and molecules as the ion signals for each spatial point as a function of the mass-to-charge ratio, which is called the mass spectrum. Therefore, data in association with IMS is composed of a huge number of mass spectra according to spatial observation points, which brings difficulty in applications of conventional statistical methods based on linear analysis. In this study, we propose an approach based on manifold learning [2], which is a framework of unsupervised nonlinear dimensionality reduction for visualizing high-dimensional data into a low-dimensional space, for analyzing the IMS data of mice brains. To this end, we first introduce several distance measures between two mass spectra based on information divergences including Kullback-Leibler and Itakura-Saito divergences [3] as well as log-det divergence [4] in an appropriate way. Then, we employ conventional algorithms of manifold learning including such as LLE, ISOMAP and t-SNE using the information-geometrical distance measure. Here, we also extend these conventional manifold learning methods to treat embedding of test samples that are not used in the training processes using the notion of kernel methods [5]. We apply our proposed approach to slices of the mice brains and show that our prosed approach is useful for visualizing a set of the mass spectra as well-separated islands of points in the low-dimensional space, where each island reflects a specific character of chemical species in the brain. We also discuss how our results can be interpreted in terms of neurophysiology. [1] M. Setou, Imaging Mass Spectrometry: Protocols for Mass Microscopy, Springer (2010). [2] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition, Springer (2008). [3] S. Amari and H. Nagaoka, Methods of Information Geometry, American Mathematical Society (2007). [4] A. Cichocki, S. Cruces and S. Amari, Entropy 17, 2988-3034 (2015). [5] J. Shawe-Taylor and N. Cristianini, Kernel Methods for Pattern Analysis, Cambridge University Press (2004).		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-580 標準化自発活動を用いたマウス個体間共通神経活動パターンの検出 Kazunori Ohashi（大橋一徳），Kazuhiro Sohya（惣谷和広），Hitomi Matsuno（松野仁美），Shoko Tsuchimine（土嶺章子），Hiroshi Kunugi（功刀浩）国立精神・神経セ神経研疾病３ The cerebral cortex is divided into functional modules referred to as cortical areas. In general, the spatial organization of the cortical areas in the brain is almost identical across animals of the same species. Thus, in human fMRI studies, the standardization procedure for the parcellated brain in terms of function and microanatomy has been devised to define a representative (i.e., standard) cortical areas observed across individuals. In contrast, due to the lack of such a common cortical space, the areal information of animal brains is not quantitatively comparable, or not applicable to inter-animal averaging. To solve this problem, in this study, we developed a novel method to spatially normalize spontaneous brain activity from transgenic mice cortices expressing G-CaMP7. This method was achieved by a wide-field optical imaging system based on the evidence that the activity of cortical areas spontaneously emerges during the anesthetized condition and that the basic configuration of the areas is common across animals. First, we calculated the functional connectivity (FC) maps for a wide range of cortical areas by seed-pixel correlation of spontaneous activity, and then combined them to generate the jointed-FC maps for simultaneously representing many cortical areas. Second, we structurally matched functionally homologous regions in the generated map across different mice with geometric transformations. Finally, we assigned all the recorded signals into the transformed space, to make spatially normalized signals in the common cortical space. The most beneficial aspect of this method is that it facilitates the comparison of activity patterns in cortical areas across individuals, canceling out the individual structural differences. We applied this method to extract the recursive common activity patterns across animals. They were more complicated than the cortical map representing the particular functional modality. Therefore, our method is able to contribute to identification of novel brain activity patterns, which are intrinsically shared inter-animal brain. In conclusion, the presented approach will be imperative to establish the common cortical space and to contribute to identification of commonalities or differences of brain activities across different mice. In addition to signals of the genetically-encoded calcium indicator, this method could be applicable to all kind of wide-field functional brain imaging.

7月26日（金）17:15～18:35　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-581 理研バイオリソース研究センターに収集された神経科学研究用マウスリソース Toshiaki Nakashiba（仲柴俊昭），Shinya Ayabe（綾部信哉），Atsushi Yoshiki（吉木淳）理研バイオリソース研究センター RIKEN BioResource Research Center (BRC) is the core facility of the National BioResource Project (NBRP) by the Japan Agency for Medical Research and Development (AMED) for the collection, preservation and distribution of mouse resources developed mainly by the Japanese scientists. We have collected over 8,800 mouse resources including fluorescent reporters, tissue-specific Cre-drivers, floxed alleles and mouse strains using the optogenetics and/or tetracycline-regulatable TET systems. These are useful for in vivo experiments in dissecting neuronal circuits in higher brain functions as well as in elucidating etiological mechanisms in psychiatric disorders. Of particular interest, we have collected over hundred human disease models including Alzheimer's disease, Amyotrophic lateral sclerosis, bipolar disorder models, and so on. Through an international collaborative effort, we have registered the deposited strains in the International Mouse Strain Resource (IMSR) to facilitate for their distribution to investigators worldwide and participated in the International Mouse Phenotyping Consortium (IMPC) for production, systematic broad-based phenotyping and distribution of knockout mice and their phenotypic data for global scientists. These mouse resources and data will be critical resources for our understanding of the higher brain functions and pathogenesis of neuronal and psychiatric disorders and should contribute for the global neuroscience community. I will update our mouse resources available at RIKEN BioResource Research Center for neuroscience research.		7月27日（土）12:55～14:15　ポスター会場（朱鷺メッセ　1F　展示ホール） PB-L-582 マウス用に開発された再構成可能な迷路における移動軌跡の解析 Riku Takahashi（高橋陸），Yuta Tamatsu（玉津裕太），Kaoru Ide（井出薫），Susumu Takahashi（高橋晋）同志社大院脳科学研究科 It is well known that the hippocampus is critically involved in spatial navigation and episodic memory. In the hippocampus, place cells that maximally fire, when animals are in a specific location in their environment, have been found. To understand the neuronal underpinning of spatial navigation and episodic memory, several studies on hippocampal place cells have been conducted using experimental animals such as rats and mice performing various tasks on several shapes of mazes. However, each maze used in conventional studies has its own disadvantages as tests of spatial navigation and episodic memory. We therefore have developed a novel maze, called reconfigurable maze, in which the parts including paths, movable walls, treadmills, and dispensers can be configured to form any existing mazes: the W maze, O-shaped maze, plus maze, radial-arm maze and figure-8 shaped maze (Takahashi, S. et al., PA-496). Here, we develop a mouse version of the reconfigurable maze. The size of the parts is designed to fits the body size of standard mice so that the running trajectory was undisturbed. Since the size of the entire maze is much smaller than the rat version, it could be installed in the soundproofing room to remove the influence of an unanticipated noise. Similar to the rat version, we could configure the parts including paths, movable walls, treadmills and dispensers to form the existing mazes: the O-shaped maze, plus maze, figure-8 shaped maze. The action of the parts is controlled by a custom-made software on Arduino mega 2560 through a control box. The reconfigurable parts may affect the mouse behavior and neuronal activity. To evaluate such influences, we trained mice to perform traversing a linear track and analyzed the running trajectory and neuronal activity. Our mouse version of the reconfigurable maze would be an efficient tool to investigate the neuronal correlates of spatial navigational behavior of genetically modified mice such as Alzheimer's disease model.