TOP一般口演(Oral)
 
Oral
Motor Pattern Control
一般口演
運動パターン制御
7月25日(木)17:50~18:05 第9会場(朱鷺メッセ 3F 306+307)
1O-09e2-1
ショウジョウバエ幼虫における逃避行動中の筋弛緩を制御する神経回路について
Atsuki Hiramoto(平本 篤紀)1,Julius Jonaitis(Jonaitis Julius)2,Sawako Niki(二木 佐和子)1,Richard Fetter(Fetter Richard)4,Albert Cardona(Cardona Albert)4,Akinao Nose(能瀬 聡直)1,3
1東京大院新領域創成科学複雑理工
2University of St Andrews, Scotland, United Kingdom
3東京大院理物理
4HHMI Janelia Research Campus, Ashburn, VA, UAS

Typical patterned locomotion is achieved as periodical and sequential contraction/relaxation of muscles. While contraction of muscles is essential for locomotion, timely relaxation of muscles is also important for the regulation of locomotion. Each muscle involved in locomotion is innervated by corresponding motor neuron(s), whose activity in turn is regulated by the central pattern generators (CPGs). Locomotion is often elicited by a stimulus from the environment and regulated by the command neurons, which link a specific sensory stimulus to appropriate behavioral output(s) coded by the CPGs. We study how the command system controls the downstream CPGs and regulates the timely contraction/relaxation of individual muscles during locomotion by using the Drosophila larval backward locomotion as a model.
Larval backward locomotion is an escape behavior, which the animal uses to escape away from the blue light or noxious touch stimulus in the head, and is generated by a propagation of muscular contraction from the anterior to posterior segments. Here, we report on the characterization of a novel class of segmental ascending interneurons, named Canons, which we show regulate the timing of muscular relaxation in larval backward locomotion. We identified Canons as segmental interneurons whose activity is correlated with fictive backward but not forward locomotion in calcium imaging of isolated nerve cord. The activity of Canons lagged behind that of aCC motor neurons in each neuromere. This phase delay implied that the function of Canons was required after excitation of motor neurons in backward locomotion. Canon is a cholinergic neuron with an ascending axon that arborizes presynaptic terminals in one and two neuromeres anterior to the cell body. Optogenetic activation of Canons induced relaxation of body wall muscles, whereas their genetic inhibition prolonged muscular contraction, suggesting that Canons inhibit muscular contraction. Connectomics analysis revealed that Canons receive synaptic inputs from MDN, a command neuron known to induce backward locomotion, and send outputs to inhibitory premotor neurons. Taken together, our results suggest that Canons networks regulate timely muscle relaxation during backward locomotion via late phase motor inhibition. This study reveals a motor regulatory circuit that spans from command neurons to muscles.
7月25日(木)18:05~18:20 第9会場(朱鷺メッセ 3F 306+307)
1O-09e2-2
ショウジョウバエ神経系において異なる運動パターン生成を担う動的モジュール構造の集団的シナプスイメージングによる解明
Kazushi Fukumasu(福益 一司)1,Akinao Nose(能瀬 聡直)1,2,Hiroshi Kohsaka(高坂 洋史)2
1東京大院理物理
2東京大院新領域創成科学複雑理工

Animals show multiple behaviors by operating motor neurons in a coordinated manner, which is achieved by the central pattern generator (CPG). The CPG contains numerous neurons and synapses. Though populational activities of the CPG are thought to underlie animal behaviors, the spatiotemporal dynamics in neural circuits for motor control remain unclear. Especially, how distinct behaviors are realized by a single circuit is not fully understood.
In this study, we use the Drosophila larvae as a model system to examine how distinct motor patterns are generated in the CNS in a resolution of single synapses. Fly larvae show crawling behaviors in both forward (FW) and backward (BW) directions. The neural activity corresponding to these behaviors can be observed in an isolated CNS. By expressing membrane localized Ca2+ indicator in all neurons, we recorded the calcium signal from synapses in the CNS. To quantify the signal, we had to define region of each synapse. Unlike cell bodies, each single synapse is difficult to identify only by its morphology because it is densely packed in the CNS. Accordingly, we applied a clustering method based on temporal correlations of intensity between the pixels of the movie, which is developed in the statistical physics. We succeeded in extracting the characteristic morphology and activity of synapses (3,600 synapses x 6,700 frames).
By using the spatiotemporal activity data of the CNS in a single synapse resolution, we analyzed the temporal order of recruitment of each synapse in FW and BW. In a single neuromere, which is a circuitry unit controlling single body wall segment, the recruitment of synapses shows a clear temporal structure: a majority of the synapses are activated almost simultaneously, whereas the remaining synapses are activated gradually after the activation of the major group. This pattern was observed in both FW and BW. Interestingly, although the major groups in FW and BW largely overlap, subsets of them are recruited as the major group only in FW or BW. This observation implies the existence of shared and specialized sub-circuits in the motor circuits that generate distinct behaviors. To further analyze the neural network in detail, we constructed a graph network showing information flow and a causality between synapses based on transfer entropy. By comparing graph structures between FW and BW, we will discuss the operational mechanisms of neural circuits that generate distinct behaviors.
7月25日(木)18:20~18:35 第9会場(朱鷺メッセ 3F 306+307)
1O-09e2-3
分子遺伝学的手法による細胞タイプレベルのショウジョウバエ歩行回路の解析
Ryo Minegishi(峯岸 諒)1,Kai Feng(Feng Kai)2,Barry J. Dickson(Dickson J. Barry)1,2
1Janelia Research Campus, HHMI, Virginia, USA
2Queensland Brain Institute, University of Queensland, Brisbane, Australia

Legged locomotion is one of the common behaviors among terrestrial animals. Locomotion is controlled by rhythmic neural activities, command information from the brain, and sensory feedback. Many neurons relating to the locomotor control have been studied for decades, and the recent progress of microscopic technology and data science is trying to make a whole central nervous system connectivity map including these neurons. To ascertain the specific functions of individual neurons, and how they are integrated into motor control circuits, we need methods to monitor and perturb their activity at cellular resolution.
We aim to analyze the locomotor circuits in the ventral nerve cord (VNC) of Drosophila melanogaster, which is analogous to the spinal cord in vertebrates. To this end, we seek to acquire genetic reagents to express activity reporters and modulators in specific cell types in the VNC. We use the split-Gal4 system to target the intersection of two enhancer elements, each of which drives expression of one of the two split halves (AD and DBD) of the GAL4 transcription factors. In cells that express both enhancers, a functional GAL4 is reconstituted, thereby activating any transgenes under GAL4/UAS control. Candidate AD and DBD pairs were selected by examining the expression patterns of several thousand GMR and VT lines. We have screened over 6000 intersections of the split-Gal4 pairs and have generated ~1000 stable lines which have specific expression in a single or very few cell types.
Using the split-Gal4 lines, we analyzed the anatomical features of cell types and identified their neurotransmitters. We will further analyze their anatomical and functional connectivity in the electron microscopic resolution map and calcium imaging. We also performed behavioral experiments to test the functional role by optogenetic activation of each cell type. Characteristic locomotion phenotypes including initiation and stopping of locomotion as well as specific leg movement were readily observed in flies with cell type specific activation. For selected lines, we then analyzed their leg movement or leg joint angle kinematics and stepping patterns in both activation and silencing experiments.
These genetic reagents, and the initial anatomical and behavioral analysis, provide the foundation for a detailed investigation of the circuit mechanisms underlying Drosophila locomotion.
7月25日(木)18:35~18:50 第9会場(朱鷺メッセ 3F 306+307)
1O-09e2-4
全汎性運動抑制を誘発する脊髄介在細胞の神経伝達物質プロファイル
Kaoru Takakusaki(高草木 薫),Mirai Takahashi(高橋 未来),Toshi Nakajika(中島 敏),Ryosuke Chiba(千葉 龍介),Kazuhiro Obara(小原 和宏)
旭川医科大学 脳機能医工学研究センター

The reticulospinal tract descending from the pontomedullary reticular formation (PMRF) in cats mediates generalized motor inhibition or muscular atonia via inhibitory interneurons in spinal cord. Our studies so far suggest that they are located in the lamina VI and VII of Lexed. The present study was designed to understand neurotransmitter profiles of the inhibitory interneurons whether they utilized gamma-aminobutyric acid (GABA) and/or glycine as inhibitory neurotransmitters. Experiments were performed in 6 cats which were decerebrated at precollicular-postmammillary level. Single or short train pulses of stimuli (1-3 pulses, 30-50 μA) which were applied to the inhibitory region in the PMRF evoked primary afferent depolarization (PAD) and inhibitory population potentials (IPP) in the dorsal and ventral roots of lower lumber segments (L6-L7), respectively. An intravenous administration of bicuculline (0.05-0.1 mg/kg), a GABA-A receptor antagonist, completely abolished dorsal root PAD and reduced the size of ventral root IPP. On the other hand, an intravenous infusion of strychnine (0.05-0.1 mg/kg), a glycine receptor antagonist, did not affect the dorsal root PAD but greatly reduced the size of the ventral root IPP. Subtraction of the dorsal and ventral root potentials before and after application of bicuculline revealed that GABA produced PAD in dorsal roots with a latency of 16.6 ± 5.8 ms (mean ± SD, n=9) and IPP in ventral roots with a latency of 16.9 ± 6.3 ms (n=9). There was no statistical difference between these latencies (p<0.01), indicating that both effects were mediated by the same group of interneurons. In contrast, strychnine produced IPP with a latency of 13.7 ± 4.6 ms (n=7), and did not produced PAD. Because latency of IPP evoked by GABA was longer than that by glycine (p<0.05), the GABAergic and glycinergic effects were mediated by different groups of interneurons. These results suggest that the reticulospinal inhibition arising from the PMRF produce both presynaptic inhibition of sensory afferents and postsynaptic inhibition of motoneurons, and these inhibitory effects are mediated by an activation of different groups of inhibitory interneurons with separate neurotransmitter profiles; GABAergic interneurons induce both presynaptic and postsynaptic inhibition, and glycinergic interneurons induce postsynaptic inhibition. We propose that GABA and glycinergic interneurons may be located in the lamina VI and VII.