Intrasulcal electrocorticography in macaque monkeys

Abstracts / Neuroscience Research 71S (2011) e108–e415

P4-v03 Establishment of new approaches to understanding the brain glycogen metabolism—By NMR spectroscopic technique using 13 C-labeled glucose and lactate Hirokazu Mizokawa Tomoyuki Kanamatsu

, George

Kokubu, Youichi

Kobayashi,

Department of Environmental Engineering for Symbiosis, Faculty of Engineering, Soka University, Tokyo, Japan It has been reported that the concentration of glycogen in brain is one tenth of that in liver and the localized-distribution of glycogen is found only in astrocytes. Magistretti et al. reported that one possible neuro-protective mechanism of glial glycogen might be to provide lactate as fuel for neurons. However, the physiological function and metabolic role of brain glycogen are still unclear. Due to postmortem degradation, it is difficult to determine the concentration of brain glycogen and lactate. So, we established a new approach to know the brain glycogen metabolism. After 24 h starvation, the rat were raised only [1–13 C]-glucose solution (10%, w/v) for 6, 12, 24 h, and then collect blood under anesthesia and sacrificed by microwave treatment (5.0 kW, 1.2 s) to the head. The brain, liver and blood was homogenized with 6% PCA and divided into two portions respectively. One portion was used for glucose assay. The other portion was added NaBH4 to change the NMR chemical shift of glucose, and then incubated for 16 h after adding amyloglucosidase for measuring the glycogen concentration. 1 H NMR shows the fractional enrichment (FE) of glucose and lactate (which indicates the % of glucose used for synthesis glycogen and lactate). 13 C NMR also shows the amount of 13 C incorporated into glycogen and lactate from [1–13 C]-glucose. The blood glucose level was almost normal from 6 to 24 h after the onset of [1–13 C]-glucose solution and the values of FE of blood glucose and lactate was about 80% and 35%, respectably. The FE of glucose and lactate in the brain was also detected by 1 H NMR. These methods for knowing the metabolism of glucose, lactate and glycogen may be useful for understanding the brain energy metabolism.

e413

P4-v05 The selective injection system into hippocampus CA1 utilizing theta oscillation Kogo Takamiya , Jyoji Tsutajima, Takato Kunitake Dep. of Integrative Physiology, Faculty of Med., Unv. of Miyazaki, Miyazaki, Japan Hippocampus CA1 has been intensively studied due to its important roles in synaptic plasticity underlying learning/memory. In particular, much effort has been devoted to clarify its molecular mechanism of in vitro synaptic plasticity paradigm, LTP (long-term potentiation). Hippocampus CA1 LTP has been studied using acute brain slices. In contrast, it is difficult to mimic LTP in dissociated cultured neurons, suggesting that hippocampus LTP is expressed by unknown mechanisms including a network connection. In addition, several different forms of synaptic plasticity are reported in various brain regions and ages, presumably expressed in individual mechanisms. These complicated natures of LTP provoke difficulty in creating a detailed analysis of the molecular mechanism. One approach to this study is directly injecting viruses expressing cDNA/shRNA into hippocampus CA1 and analyzing neuronal functions in normal neuronal circuitry. In general, virus is injected via pipette that is inserted from the top of the brain cortex. However, by this approach, it is difficult to locate CA1 accurately using a pipette, since CA1 pyramidal cell layer is only 0.2 mm in width. In the present study, we have established a system to accurately introduce viral vectors in the stereotactic injection system into mouse hippocampus CA1 by monitoring theta oscillation. We decided the correct depth of the pipette tip based on the integrated values of theta oscillation at each depth. This approach allows us to inject virus accurately and provide an efficient method to transfer genes into hippocampus CA1 neurons, in the studies of molecular mechanisms of synaptic plasticity and other neuronal functions. Research fund: The Naito Foundation Natural Science Scholarship, Uehara Memorial Foundation, KAKENHI 22300122. doi:10.1016/j.neures.2011.07.1810

doi:10.1016/j.neures.2011.07.1808

P4-v04 Neuromodulation of prefrontal background oscillatory activities with high-frequency repetitive transcranial magnetic stimulation Yoshihiro Noda 1 , Motoaki Nakamura 1 , Takashi Saeki 1 , Hideo Iwanari 1 , Kiyoto Kasai 2 1

Kinko Hospital, Kanagawa Psychiatric Center, Yokohama, Japan Neuropshychi., Grad. Sch. of Med., Univ of Tokyo, Tokyo, Japan

2

Div. of

Objectives: Repetitive transcranial magnetic stimulation (rTMS) over primary motor cortex has been reported to induce an alternation of the background oscillatory activity around the stimulation site, especially for alpha band spectral power. However, modulating effects on background activities of prefrontal stimulation have little been known so far. In the present study, electroencephalogram (EEG) spectral power at prefrontal region were investigated before and after consecutive 10 rTMS sessions over left dorsolateral prefrontal cortex (DLPFC), to explore any longitudinal changes in spectral power at theta, alpha, and beta band oscillations. Methods: Twenty-four patients with major depression underwent 10 daily rTMS sessions over two weeks. The stimulation site was left DLPFC determined by ultrasound-based navigation system. Stimulation frequency was 20 Hz and intensity was 90-110% resting motor threshold, with total pulses per week of 5000. Scalp EEG recording was performed before and after 10 consecutive rTMS sessions. In order to focus on the prefrontal region, Fast Fourier Transformation (FFT) analyses were performed on EEG data derived from Fp1, Fp2, F3, F4 electrodes using EEGLAB software. Spectral power values of each frequency band were compared applying paired T test using SPSS software. Results: The paired T test revealed a significant increase in alpha spectral power and an increasing trend of theta spectral power at Fp1 electrode after the 10 serial rTMS sessions. Conclusions: The present study suggests that the power of alpha and theta band oscillations recorded at Fp1 electrode could be enhanced by high frequency rTMS sessions over left DLPFC, which may be related to the neuromodulation on the stimulated region. doi:10.1016/j.neures.2011.07.1809

P4-v06 Examing the calcium response of white matter glia cells with calcium imaging Tsunehiro Takeuchi , Tatsuhiro Hisatsune Dep. of Integrated Biosci., Grad. Sch. of Frontier Sciences, The University of Tokyo, Kashiwa, Japan The objective of this study is to examine the calcium responses of white matter glia cells with calcium imaging technique. It is known that glia cells play an important role in transmitting information in addition to neurons. Using SR101 as a specific marker of glia cells, we applied various drugs to brain slices of mice. We have demonstrated that glia cells responded to neurotransmitters such as ATP and P2Y1 agonist (N)-methanocarba-2MeSADP. Furthermore, we have demonstrated that glutamate evokes calcium signal, but mGluR5 agonist DHPG did not produce the same result. These results mean that whitematter glia cells do not always have all glutamate receptor subtypes. We need to use agonists and antagonists of glutamate for investigating what kind of receptor subtypes are involved. And, we demonstrated that glia cells responded to anti-inflammatory substances such as adenosine A2A agonist CGS21680, but did not respond to GPR40/GPR120 agonist GW9508. We need to examine the change of the response to antiinflammatory substances under inflammatory conditions. Research fund: 22120505. doi:10.1016/j.neures.2011.07.1811

P4-v07 Intrasulcal electrocorticography in macaque monkeys Keisuke Kawasaki 1 , Takeshi Matsuo 1,2 , Takahiro Osada 3 , Hirohito Sawahata 1 , Takafumi Suzuki 4 , Masahiro Shibata 5 , Naohisa Miyakawa 1 , Kiyoshi Nakahara 6 , Noboru Sato 5 , Kensuke Kawai 2 , Nobuhito Saito 2 , Isao Hasegawa 1 1

Dept Physiol, Niigata Univ Sch Med 2 Dept Neurosurg, Univ Tokyo Sch Med Dept Physiol, Univ Tokyo Sch Med 4 Grad Sch Info Sci and Tech, Univ Tokyo 5 Dept Anat, Niigata Univ Sch Med 6 National Center of Neurol and Psych 3

The highly evolved primate’s brain has deep cerebral sulci, and both gyral and intrasulcal cortical regions have been implicated in important functional processes. Electrocorticography (ECoG) has become a potent methodology for large area recording and stimulation. However, direct experimental access is typically limited to gyral regions, since placing probes into sulci is difficult without damaging the surrounding tissues. Here we describe a novel method for intrasulcal ECoG in macaques by combining minimally invasive

e414

Abstracts / Neuroscience Research 71S (2011) e108–e415

neurosurgical protocols and flexible probes fabricated with micro-electromechanical systems. To evaluate the feasibility of intrasulcal ECoG, we first inserted parts of the probe into the superior temporal sulcus (STS) to compare ECoG responses from the ventral bank of STS with those from the inferior temporal gyrus (ITG). We found no location effect (STS or ITG) on the power spectral density (PSD) of the intertrial interval response (p = 0.66, two-way ANOVA) and no significant interaction between frequency and location (p = 0.82). There was a significant effect of frequency on the PSD (p < 10–16). Signal-to-noise ratio of the visually evoked ECoG responses in STS (1.55 ± 0.62, mean ± SD, n = 34) was not significantly different from those in ITG (1.59 ± 0.62, n = 187) (p = 0.73, t-test). Histological examination revealed no obvious physical damage in the implanted cortical areas. Second, we placed an ECoG probe into the central sulcus to stimulate the anterior bank of the sulcus and the surface of the precentral gyrus. Thresholds for muscle twitching were significantly lower during intrasulcal stimulation (1.0 ± 0.2 mA, n = 43) compared to gyral stimulation (3.6 ± 0.4, n = 50) (p < 0.001, t-test). These results demonstrate the feasibility of intrasulcal ECoG in macaque monkeys. This work was supported by SRPBS from MEXT. Research fund: SRPBS.

neuronal networks constructed from particular cells. Moreover the physiological mechanism such as imprinting period or stability information for each individual cell can be investigated through neurotransmission manipulation. Furthermore, these techniques allow us to evaluate interactions of neurons and glia within created neuronal networks. Then, we constructed artificial neuronal networks and characterized the electrophysiological features using primary neurons. Such artificial neuronal networks showed electrophysiological activities displaying various activity patterns. We intend to elucidate the operating principles of neuronal networks by investigating the hysteretic changes of individual neurons and neuronal plasticity at a neuronal network level.

doi:10.1016/j.neures.2011.07.1812

1

P4-v08 Measurement of action potentials of dissociated cultured neurons by high density CMOS array and network analysis

doi:10.1016/j.neures.2011.07.1814

P4-v10 Spatially arranged microelectrode probes for multi-site electroretinogram recordings in the isolated mouse retina Shun Taga 1,2 , Wataru Konishi 5 , Chieko Koike 2,6

Tonomura 3 , Akira

Amano 4 , Satoshi

Grad. Sch Sci Eng, Ritsumeikan Univ, Shiga, Japan 2 Lab Sys Neurosci Del Biol, Col Pharm Sci, Ritsumeikan Univ, Shiga, Japan 3 R-GIRO, Ritsumeikan Univ, Shiga, Japan 4 Organ Function Simulation Lab, Dept Bioinfo, Col Life Sci, Ritsumeikan Univ, Shiga, Japan 5 Dept Micro Sys Tech, Fuc Sci And Eng, Ritsumeikan Univ, Shiga, Japan 6 PRESTO, JST, Saitama, Japan

We study network of dissociated cultured neurons by taking advantage of high density CMOS array. High density CMOS array is a kind of extracellular electrophysiological device like multi-electrode arrays and it has about ten thousand electrodes in about 4 mm2 area. With this high spatial resolution, we are able to figure out the position of neurons without an imaging technique and also able to observe propagation of action potential within the network. High density CMOS array allows us to investigate dynamics of neuronal network in vitro precisely. We propose a demonstrative result obtained by this device. A result of our preliminary analysis on dissociated cultured network showed the possibility that spontaneous network might not be small world. To examine this result more exact, we are studying the factors that affect network characteristics. The factors might be density of plated neurons, density of electrodes, the cultured day in vitro, connectivity estimation method and effects of stimulation. First, we compared connectivity estimation methods to know how methods are critical to detect network. We employed two methods: one is based on cross-correlation and the other is based on mutual information. The networks resulted from these methods were approximately the same; the detected core networks were not small world. We show this result as a step of the series of analysis.

The visual information is primarily processed in the retina and transmitted into brain by the ganglion cell axons. The mammalian retina consists of five major neural cell types (photoreceptors, bipolar cells, horizontal cells, amacrine cells and ganglion cells) and a type of glia (muller glia). The retinal photoreceptors convert light into electrical signals that are processed by the sequential retinal layers and the ganglion cells. Electroretinogram (ERG) procedure is useful to obtain neuronal electrical activities from the retina. Microelectrode arrays (MEAs) can detect light responses from the isolated retina that have advantages to analyze ganglion cell functions in addition to micro ERG. However, the present planar electrodes are not able to detect signals from the inside of the retina. In order to directly and spatially stimulate and record multisite electrical responses of the inside retina, we have developed spatially arranged microelectrodes using the wire-bondingbased probe technology. Both 2D and 3D microelectrodes were designed in a chip, because we can make a comparison between ERG recorded by the fabricated microelectrode probes and conventional planar microelectrodes. Light-evoked ERG responses successfully recorded from both 2D and 3D microelectrodes at multiple points. The typical signals were successfully detected in 3D microelectrode. These signals were similar to the wave forms of ERG responses recorded by the typical planar electrodes. These results suggest that the gold microelectrode probe arrays are applicable to retina experiments as a new analysis tool, and it will allow us direct and spatial multi-recordings of micro ERG.

doi:10.1016/j.neures.2011.07.1813

doi:10.1016/j.neures.2011.07.1815

P4-v09 Construction of an artificial neuronal network and electrophysiological measurement with a selective collection method of cultured primary neurons

P4-v11 Toward the best parameter for spatial-filter source estimation of MEG data

Takeshi Mita 1 , Douglas Bakkum 2 , Urs Frey 2 , Andreas Hierlemann 2 , Ryohei Kanzaki 1 , Hirokazu Takahashi 1,3 1

The University of Tokyo, Japan 2 ETH Zurich, Switzerland 3 JST PRESTO, Japan

Terazono 1

Hideyuki , Hyonchol hiro Hattori 2 , Kenji Yasuda 1,2 1

Kim 1 ,

Masahito

Kanagawa Academy of Sci. Tech., Kawasaki, Japan Med. Dent. University, Tokyo

2

Hayashi 1 ,

Aki-

Masao Matsuhashi 1 , Tetsuo Kobayashi 2 , Yohei Yokoyama 1,3 , Nobuhiro Mikuni 3,5 , Riki Matsumoto 4 , Akio Ikeda 4 , Hidenao Fukuyama 1 1

Biomed. Info., Tokyo

We have been studying living systems by creating artificial tissues composed of individual cells as a minimum functional unit. More specifically, we have been studying cell-to-cell communications by controlling the spatial pattern of cell networks. In the course of research, it has emerged that cells form acceptable or resistible spatial patterns capable of processing external information such as long term potentiation (LTP). Therefore, we hypothesize that individual cells within a network retain their respective characters even then the network as a whole comes under the influence of external stimulus. To clarify this phenomenon, we have developed three techniques to make artificial neuronal networks: (1) an agarose-microprocessing technique to create artificial neuronal networks on a culture dish through manipulation of neurotransmission direction, (2) a non-invasive technique for picking cells from a group of primary cultured neurons and depositing the selected cells in a micropattern, (3) a multi-electrode array system for measurement of multi-point extracellular potential of neurons while simultaneously stimulating these cells. Three techniques allow us to create and evaluate artificial

HBRC, Kyoto University, Kyoto, Japan 2 Dept Electr. Eng., Kyoto University, Kyoto, Japan 3 Dept Neurosurgery, Kyoto University, Kyoto, Japan 4 Dept Neurology, Kyoto University, Kyoto, Japan 5 Dept Neurosurgery, Sapporo Med. University, Sapporo, Japan Objective: Various spatial filtering methods are proposed and used trying to solve the Magnetoencephalographic (MEG) inverse problem, but even the conventional techniques such as minimum norm (MN) and minimum variance (MV) beamformers are seldom checked or compared for the efficacy with the live human brain. Using electrocorticography (ECoG) data directly recorded from Human Cerebral Cortices as reference, our goal is to see how the various MEG solutions and models reconstruct the cortical activities, and to find the most effective parameter for the clinical application. Methods: Two patients with medically intractable partial epilepsy had MEG and ECoG recording simultaneously in the course of presurgical evaluation. Source distribution estimated from MEG data was used to calculate electrical activities at the ECoG electrode location and correlation coefficient in the spatial domain between true and reconstructed ECoG activities. Following factors were used for analysis: Conductor model (single sphere model (SS) and realistic single layer boundary element model (BEM)), source distribu-