小脳 Cerebellum
P1-2-75 ラットのリズムに基づく行動中または時間間隔に基づく行動中に見られる小脳活動 Cerebellar activity during rhythm-based or duration-based behavior in rats ○山口健治1, 高橋晋2, 櫻井芳雄1 ○Kenji Yamaguchi1, Susumu Takahashi2, Yoshio Sakurai1 京都大学大学院　文学研究科　心理学専修1, 同志社大院脳科学　神経回路形態2 Dept Psycho, Kyoto Univ, Kyoto1, Dept Neural Circuitry, Doshisya Univ, Kyoto2 It is said that cerebellum relates to time perception that counts millisecond-to-second time intervals. The cerebellar time-perception mechanism has often been studied in imaging experiments using human subjects. Grube et al. (2010) shows, for example, that cerebellar obligatory function relating to time perception is absolute (duration-based) timing of single intervals rather than relative (beat-based) timing of rhythmic sequences by using continuous transcranial magnetic theta-burst stimulation (cTBS). Meanwhile, there are little studies reporting relation between timing behavior and cerebellar neuronal spikes that organize behavioral timing. In the present study, we explore this relation by recording cerebellar spikes during rat's foreleg tapping behavior. The tapping task requires the rats to touch a switch continuously with regular intervals of a few hundreds millisecond. When fixed numbers of successive touch responses are conducted in a trial, the rats get reward of food pellet. When the rats fail to wait a fixed interval between behavioral responses, the current trial is canceled and starts again. Hence the rats need to perceive interval times between the touch responses to succeed in trials. In this task, if the required response number is 2, then the rats are required to generate only one IRT (inter-response time). But if the criterion was up to 2 times, the rats are required multiple IRTs and the tapping become to yield some rhythms. The former is duration-based task, the latter is rhythm-based task. We compared these two types of behavior and patterns of cerebellar Purkinje cell activity during these behaviors. In the rhythm-based task, we found that Purkinje cell's simple spikes paused just when the rats touched the switch. It is possibly that milliseconds temporal processing is encoded by this simple spike activity pattern.	P1-2-76 小脳運動学習における長期トレーニングの実世界時間シミュレーション Computer simulation of long-term cerebellar learning in real-life time ○山崎匡1, 五十嵐潤2, 永雄総一3 ○Tadashi Yamazaki1, Jun Igarashi2, Soichi Nagao3 電気通信大学 大学院 情報理工学研究科 情報・通信工学専攻1, 理化学研究所 次世代計算科学研究開発プログラム2, 理化学研究所 脳科学総合研究センター 運動学習制御研究チーム3 Graduate School of Informatics and Engineering, The University of Electro-Communications, Tokyo1, Computational Science Research Program, RIKEN, Saitama2, Lab for Motor Learning Control, RIKEN Brain Science Institute, Saitama3 Computer simulation of detailed spiking network models with a fine temporal resolution needs long computational time. For example, to simulate just 1 second of the network activity, the simulation could consume a few hours in physical time. Due to this limitation, simulating long-term adaptation in cerebellar motor learning that lasts for hours or days by a detailed spiking network model was almost impossible to date.We have been developing Realtime Cerebellum, a large-scale spiking network model of the cerebellum composed of more than 100,000 spiking neuronal units and is implemented on graphics processing units (GPUs). Owing to the parallel computing capability of GPUs, Realtime Cerebellum runs in realtime with the temporal resolution of 1 millisecond, allowing us to conduct computer simulation of long-term adaptation of cerebellar motor learning.To confirm this, we adopted Realtime Cerebellum to simulate the experiments of long-term adaptation of optokinetic response eye movements (OKR) conducted by Okamoto et al. (Okamoto T, Endo S, Shirao T, Nagao S. Role of cerebellar cortical protein synthesis in transfer of memory trace of cerebellum-dependent motor learning. J Neurosci 31:8958-66, 2011). The experiment is composed of four, 15-minutes training sessions of OKR adaptation with 1 hour break between sessions. The OKR gain is increased slightly during each training session, and is further increased during each 1 hour break. Thus, the short-term gain increase is induced by each training session, and the long-term gain increase is observed throughout the whole four sessions. We were able to reproduce the short-term and long-term gain increase by the computer simulation. We confirmed that the simulation finished within 4 hours in physical time. These results suggest that now we are ready to simulate completely the detailed cerebellar activity during whole training periods that could last for hours or days.
P1-2-77 小脳における8-nitro-cGMPとS-グアニル化タンパク質の分布 Distribution of 8-nitro-cGMP and S-guanylated proteins in cerebellum ○遠藤昌吾1, 中村真菜美1, 新崎智子1, 赤池孝章2 ○Shogo Endo1, Manami Nakamura1, Tomoko Arasaki1, Takaaki Akaike2 東京都老人総合研究所1, 熊本大学大学院生命化学研究部2 Tokyo Metropolitan Institute of Gerontology, Tokyo1, Kumamoto University Graduate School of Medical Sciences, Kumamoto2 cGMP plays essential roles in a variety of physiological responses in organs and cells. In the brain, however, the distribution of cGMP and its downstream signal transduction molecules are restricted in several areas. The recent discovery of 8-nitro-cGMP, a new component of cGMP pathway, is attracting attentions for its role as a long-lasting cGMP signal. Furthermore, the role of 8-nitro-cGMP for S-guanylation of the Cys residue in the proteins to modify the activity of the proteins may play important physiological roles in the cell regulation. In this report we have characterized the distribution of 8-nitro-cGMP and guanylated proteins in the brain. We have characterized the immunological distribution of 8-nitro-cGMP and guanylated proteins using specific antibodies. Mouse monoclonal antibody against 8-nitro-cGMP and rabbit polyclonal antibody against guanylated protein were used to stain mouse (C57BL/6) slices. The immunoreactivity was visualized using secondary antibodies labeled with Alexa Fluor. The fluorescence images of the immunoreaction were obtained by laser scanning confocal microscope. The 8-nitro-cGMP staining was observed in cerebellum but the staining in other area was quite low compared with that in the cerebellum. In the cerebellum, 8-nitro-cGMP staining was restricted in Purkinje cells, co-localized with G-substrate, an excellent PKG substrate. Most of Purkinje cells in cerebellum including flocculus were positive for the 8-nitro-cGMP staining with various degree of staining intensity. Furthermore, the staining for the S-guanylated protein was also concentrated in cerebellar Purkinje cells, co-localized with calbindin. The Purkinje cells express guanylate cyclase, PKG and G-substrate, a PKG. The intense localizations of 8-nitro-cGMP and S-guanylated proteins indicate the important roles of these molecules in cerebellar Purkinje cells.	P1-2-78 知覚的予測における小脳の役割 Contribution of the cerebellum to perceptual prediction ○松嶋藻乃1, 伊藤さやか2, 吉田篤司1, クルキンセルゲイ1, 矢部一郎2, 佐々木秀直2, 田中真樹1 ○Ayano Matsushima1, Sayaka Ito2, Atsushi Yoshida1, Sergey Kurkin1, Ichiro Yabe2, Hidenao Sasaki2, Masaki Tanaka1 北海道大学　医学研究科　神経生理学分野1, 北海道大学　医学研究科　神経内科学分野2 Dept Physiol, Hokkaido Univ Schl Med, Sapporo1, Dept Neurol, Hokkaido Univ Schl Med, Sapporo2 The cerebellum has long been implicated in the forward internal models that predict outcomes of movements. However, it remains unknown whether the cerebellum is also involved in the prediction of upcoming sensory events in the absence of actions.To explore the role of the cerebellum in perceptual prediction, we conducted three psychophysical experiments in subjects with genetically-diagnosed spinocerebellar ataxia types 6/31 (SCA6/31, pure cerebellar types) and matched controls. In Experiment 1, the subjects covertly tracked three moving targets in the presence of identical distractors (Multiple object tracking). When all stimuli were briefly (300 ms) extinguished during the motion period, the subjects with SCA lost track of the targets more often than controls, indicating their difficulty in the extrapolation of object trajectories. In Experiment 2, the subjects were presented with a gradually color-changing ring and were asked to judge the relative color of a disk flashed (150 ms) at the center of the ring (Color discrimination). Although the control subjects often compared with the colors presented slightly after the disk appearance, the SCAs tended to compare with those presented previously. These results suggest that the cerebellum might be needed to predict the stimulus color ahead of the actual change. In Experiment 3, the subjects were asked to judge the relative timing of an auditory tone and the collision of two bars moving smoothly in the opposite directions (Multimodal simultaneity). Although neither the accuracy nor precision of judgment were different between groups, the latter correlated with the severity of cerebellar ataxia (scaled by SARA and BBS). Thus, our results suggest that the cerebellum might play a role in sensory prediction emerged from purely sensory experience without movement. The cerebellar functions in patients could be evaluated more precisely by assessing the deficits in sensory prediction along with motor controls.
P1-2-79 小脳におけるnetrin-G1、G2の機能 Functions of netrin-G1 and -G2 in cerebellum ○安島綾子1, 矢口邦雄1, 後藤大道1, 糸原重美1 ○Ayako Ajima1, Kunio Yaguchi1, Hiromichi Goto1, Shigeyoshi Itohara1 理化学研究所　脳科学総合研究センター　行動遺伝学技術開発チーム1 Behavioral Genetics, BSI, RIKEN, Wako Saitama1 Netrin-G1 and -G2, coded by Ntng1 and Ntng2, respectively, are synaptic adhesion molecules that are expressed in a complementary manner in various brain areas. It is demonstrated that netrin-G1 and -G2 locate at presynaptic and their ligands NGL1 and 2 at postsynaptic sites. Knockout mice for these genes exhibit distinct behavioral phenotypes, and also show related physiological abnormalities in the hippocampal circuits so far analyzed. In the cerebellum, Ntng1 is expressed in Purkinje cells with zebrin-like stripes, and Ntng2 in Bergmann glia. Their ligand genes Ngl1 and Ngl2 are expressed in Purkinje cell and in granular cell, respectively. The functions of these molecules in the cerebellum are totally unknown. In this study, we focused on the roles of netrin-G1 and -G2 in the cerebellar function. Ntng1 KO mice showed no clear deficiencies in motor behaviors, while we found that Ntng1 and Ngl1 KO mice tended to have smaller and denser spines of Purkinje neurons revealed by Calbindin immunostaining. Ntng2 KO mice showed impaired motor behaviors in rotarod and hanging wire tests. They also exhibited tremor, and it was enhanced by either swimming in water pool or by running on treadmill. We found no histological abnormalities in Ntng2 KO cerebella, but found slightly higher c-fos immunoreactivity in the inferior olive (IO) of Ntng2 KO mice than that of WT mice after either swimming or running for enhancing tremor. Ntng2 was expressed in IO, whose hyperactivity is known to cause tremor. Thus, we speculate that netrin-G2 has a role in modulating neuronal activity in IO and altered outputs to Purkinje cells through climbing fibers has a crucial role in the tremor behavior. However, we do not rule out the possibility that local netrin-G2/NGL2 interaction in the cerebellum play a role in the tremor and other motor behaviors.	P1-2-80 小脳の発生過程におけるkirrel3の発現 Expression of kirrel3 in the cerebellum during development ○久岡朋子1, 形部裕昭1, 北村俊雄2, 仙波恵美子1, 森川吉博1 ○Tomoko Hisaoka1, Hiroaki Gyobu1, Toshio Kitamura2, Emiko Senba1, Yoshihiro Morikawa1 和歌山県立医大・医・第二解剖1, 東京大・医科研・先端医療研究センター・細胞療法分野2 Dept. of Anatomy & Neurobiology, Wakayama Medical Univ., Wakayama, Japan1, Div. of Cellular Therapy, Advanced Clinical Research Center, Inst. of Medical Science, Univ. of Tokyo, Tokyo, Japan2 We have reported that a member of the immunoglobulin superfamily, kirrel3, is expressed in the brain, including the cerebellum using in situ hybridization histochemistry. It has been reported that kirrel3 is involved in the neuronal migration of pontine nucleus in the developing hindbrain. However, little is known about the function of kirrel3 in the developing cerebellum. In order to gain insights into the role of kirrel3 in the cerebellum, we characterized kirrel3-expressing cells in the embryonic and postnatal cerebellum using kirrel3-lacZ knockin mice. The expression of β-gal was first observed at embryonic day (E) 13.5 in the differentiating cells of the nuclear transitory zone, from which deep cerebellar nuclei (DCN) projection neurons develop. In addition, a small population of β-gal-positive cells was observed in the cortical transitory zone, which contains DCN interneurons and Purkinje cells (PCs) at E13.5. On the other hand, no expression of β-gal was observed in the proliferating cells of the cerebellar ventricular zone and rhombic lip at E13.5. From E15.5 to E17.5, β-gal was widely expressed in the DCN neurons, whereas the subpopulations of PCs and external granule cells (GCs) expressed β-gal. From postnatal day (P) 0 to P21, β-gal-positive cells were also observed in both DCN and cerebellar cortex. In the cerebellar cortex, the expression of β-gal was found in the internal granule cell layer (GCL) and Purkinje cell layer (PCL) from P0 to P21, whereas β-gal-expressing cells were observed in the external GCL at P0 and in the molecular layer from P7 to P21. Within the PCL, clusters of β-gal-positive PCs were localized to specific parasagittal stripes. A unique spatio-temporal pattern of kirrel3 expression suggests that kirrel3 may be involved in some developmental processes of DCN neurons, PCs, GCs, and interneurons, including the neuronal migration, axonal pathfinding, and synapse formation.
P1-2-81 小脳核ニューロンはどの座標系によって情報をコードしているのか Coordinate frame of movement-related activity of cerebellar nuclear cells ○石川享宏1, 戸松彩花2, 角田吉昭3, 筧慎治1 ○Takahiro Ishikawa1, Saeka Tomatsu2, Yoshiaki Tsunoda3, Shinji Kakei1 東京都医学総合研究所1, 国立精神・神経医療研究センター　神経研究所2, 理研BSI　運動学習制御ラボ3 Tokyo Metropolitan Inst of Med Sci, Tokyo1, Inst of Neurosci, National Center of Neurology and Psychiatry, Tokyo2, Motor Learning Control Lab, RIKEN BSI, Wako3 Cortical motor areas and a part of the cerebellum are interconnected through cerebro-cerebellar communication loop. The primary motor area (M1) and premotor area (PM) project to a hemispheric part of the cerebellar cortex via precerebellar nuclei, and the cerebellar nuclei return output to M1 and PM via thalamus (Kelly and Strick 2003). A number of models have been postulated to explain the functional role of this loop structure. According to the forward model hypothesis, the cerebellum receives a copy of motor command and generates an estimation of the next state of body parts. On the other hand, the inverse model hypothesis requires the cerebellum to receive target information and generate necessary motor command. Each model predict output of the cerebellum in a specific coordinate frame: an extrinsic and/or intrinsic (somatosensory) coordinate frame for a forward model and an intrinsic (muscle) coordinate frame for an inverse model. So far, little physiological evidence is available in terms of the coordinate frame for the output from the cerebro-cerebellum. To address this issue, we recorded unit activity of neurons in the cerebellar nuclei (NC) during a step-tracking movement of the wrist with different forearm postures in two monkeys. We examined coordinate frames for the task-related activity of NCs by calculating shift of preferred direction (PD) for a change in forearm posture. If a NC show little PD shift, activity of the NC may encode information in an extrinsic coordinate frame. Or if a NC show significant PD shift, its activity may encode information in an intrinsic coordinate frame. As a result, we found both types of NCs. Moreover, onset of movement-related activity of NCs was about 60 msec before movement onset and significantly lagged behind that of M1/PM neurons. Although coordinate frame of NCs is not yet clear enough to identify or exclude a specific model, the time lag of NCs seem prefer the forward model hypothesis.	P1-2-82 橋核における周期的シナプス活動 Periodic synaptic activity in the pontine nuclei ○石川太郎1, 志牟田美佐1 ○Taro Ishikawa1, Misa Shimuta1 東京慈恵会医科大学　薬理学講座1 Dept Pharmacol, Jikei Univ Sch of Med, Tokyo1 Principal neurons in the pontine nuclei receive excitatory synaptic connection from the cerebral cortex and project mossy fibres to the cerebellar cortex. It has been reported that the in vivo cerebellar mossy fibres fire low-frequency spontaneous action potentials in the resting state and high-frequency bursting action potentials upon sensory stimulation. It has not been known, however, how those firings are generated and modulated in the pontine principal neurons. In this study we made whole-cell path-clamp recordings from the pontine neurons in rats anaesthetized by ketamine/xylazine to characterize the synaptic integration in these neurons. Spontaneous excitatory postsynaptic currents (sEPSCs) recorded in the voltage-clamp mode had a large variation in the amplitude, with an overall average amplitude being 127 ± 171 pA (mean ± s.d., n = 5 cells). Interestingly, the sEPSCs occurred in a form of short bursts that take place in slow oscillatory cycle (4.2 ± 0.3 Hz, n = 5). Simultaneous cortical field potential recordings indicated that the oscillatory synaptic activity in the pontine neurons reflected periodic activity in the cerebral cortex. Spontaneous inhibitory postsynaptic currents were detected only in a subset of cells (2/6 cells). In the current clamp mode, each burst of synaptic inputs generated one or a few action potentials. These findings suggest that the pontine principal neurons integrate multiple synaptic inputs from cerebral cortex to transmit signals to the cerebellum.
P1-2-83 Statistical estimation of gap-junctional and inhibitory conductance in inferior olive neurons from the spike trains by network model simulation ○Hoang T. Huu1,3, Miho Onizuka2,3, Mitsuo Kawato2,4, Isao T. Tokuda1, Nicolas Schweighofer5, Yuichi Katori6,7, Kazuyuki Aihara7, Eric J. Lang8, Keisuke Toyama3 Department of Mechanical Engineering, Ritsumeikan University, Japan1, Graduate School of Information Science, Nara Advanced Institute of Science and Technology, Ikoma, Nara, Japan2, ATR Brain Information Communication Research Laboratories, Soraku-gun, Kyoto, Japan3, ATR Computational Neuroscience Laboratories, Soraku-gun, Kyoto, Japan4, Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA, USA5, FIRST Aihara Innovative Mathematical Modelling Project, JST, Tokyo Japan6, Institute of Industrial Science, University of Tokyo, Tokyo, Japan7, Department of Physiology & Neuroscience, School of Medicine, New York University, New York, USA8 The inferior olive (IO) possesses synaptic glomeruli, which are thought to be important for determining the oscillatory and synchronized activity displayed by IO neurons. Indeed, the tendency to display such activity patterns is enhanced or reduced by the local administration of the GABA-A receptor blocker picrotoxin (PIX) or the gap junction blocker carbenoxolone (CBX), respectively. We studied the functional roles of the glomeruli by solving the inverse problem of estimating the inhibitory (gi) and gap-junctional conductance (gc) using an IO network model. gc and gi were determined as the solution to the inverse problem such that the simulation and experimental spike data closely matched in the PCA space. In the PIX condition, gi was found to decrease to approximately half its control value. CBX caused an about 30% decrease in gc from control levels. These results support the hypothesis that the glomeruli are control points for determining the spatiotemporal characteristics of olivocerebellar activity. Our future work is to introduce a Bayesian framework, in which the posterior probability of the synaptic conductance is given as the product of the likelihood and the prior probability of the synaptic conductance. Functional approximation is utilized to derive the posterior probability.	P1-2-84 Distinct transgene expression profiles in the cerebellum by the direct cortical, intrathecal and intravenous injection of AAV9: Implication of gene therapy for cerebellar diseases ○Fathul Huda1,2, Ayumu Konno1, Yasunori Matsuzaki1, Hanna Goenawan1,2, Hirokazu Hirai1 Dept Neurophysiology, Gunma University Graduate School of Medicine1, Dept of Physiology, Faculty of Medicine Universitas Padjadjaran, Bandung, West Java, Indonesia.2 Route of viral vector administration is important not just for the efficiency of transduction and easiness of application but also for deciding transduction areas and cell types. We compared three routes of viral vector administration, between invasive procedures with less number of vectors, direct cortical (DC), and intrathecal (IT) and non-invasive, but large number of vectors, intravenous (IV) injections. We injected adeno-associated viral vectors serotype 9 (AAV9) that express GFP under control of the synapsin I promoter into mice, and examined the transduction profiles in the CNS 2 weeks after injection by immunohistochemistry. The AAV9 transduced only neuronal cells, but the transduced areas and cell types differed substantially depending on viral administration routes. DC and IT injection resulted in strong transgene expression in areas closed to the injection sites, while IV injection caused diffuse and homogenous transgene expression throughout the CNS. The spinal cord efficiently expressed GFP by IT and IV injection, but not by DC injection. In the cerebellum, projection neurons in the deep cerebellar nuclei (DCN) were efficiently transduced by all 3 different injections. In the cerebellar cortex, PCs were selectively transduced, with some Golgi cell transduction, by IV injection, whereas basically all types of cortical neurons were transduced by DC and IT injection. Thus, AAV9 with the Synapsin I promoter are useful tools that efficiently and widely transduce neurons in the CNS. However, the transduced areas and cell types significantly differ depending on routes of viral administration.
P1-2-85 Purkinje neuron translatome at sub-cellular resolution ○Thomas Launey1, Anton Kratz2, Pascal Beguin1,2, Megumi Kaneko1, Takahiko Chimura1, Ana Maria Suzuki2, Sachi Kato2, Nicolas Bertin2, Rejan Vigot1, Piero Carninci2, Charles Plessy2 RIKEN Brain Science Institute, Launey Research Unit forMolecular Neurocybernetics, Wako-shi, Japan1, RIKEN Omics Science Center, Yokohama, Japan2 Brain neurons often display an extensive dendrites playing key roles in receiving and processing the signals that ultimately constitute the basis of consciousness. While the cell body is typically 10-20 μm, these dendrites can reach length of several 100s μm, creating a challenge for neurons to transport and properly address newly synthesized proteins. It is now established that some mRNA are present in neuronal dendrites and that proteins are locally synthesized. Elucidating the mechanisms and functions of dendritic protein synthesis has become a very active research frontier, with roles demonstrated in development, learning and disease. The cerebellar Purkinje cells (PC) present one of the most elaborate dendrite morphology among all brain neurons. Each PC receives up to 200,000 synapses that can be modulated to enhance motor and cognitive abilities. The ability to learn new skills depends on protein synthesis. To clarify the mechanisms controlling protein synthesis in rat PCs and in particular in their dendrites, we succeeded in specifically extracting ribosomes from this neuron. As the main component of the protein synthesis machinery, the ribosome is transiently bound to the messenger RNAs being translated into proteins. By encoding a ribosome-binding probe ("TRAP") into a virus vector targeted to PCs, we succeeded to capture several thousand ribosome-bound PC RNAs (translatome). High-sensitivity NanoCAGE and deep-sequencing were then used to identify all the isolated RNAs, either as protein-coding sequence or as regulatory RNA. Using microdissections to separate the PC dendrites from its cell body, we could identify a group of several hundred RNAs that appear to be transported into dendrites for remote translation. Interestingly, in addition t transcripts of protein known to be involved in synaptic plasticity we also found that these ribosome-bound RNAs include transcripts with no known function, representing either novel genes or regulatory RNA sequences.

上部に戻る

前に戻る