研究内容
Research Goals
The rat cerebral cortex has a six-layered structure, in which glutamatergic (excitatory) pyramidal cells and GABAergic (inhibitory) interneurons combine to form a "microcircuit". Furthermore, the cortical areas are interconnected and each forms a "cortico-basal ganglia loop" via the striatum, globus pallidus, and thalamus. These brain regions play very important roles in perceptual cognition and behavioral expression.
Since the 1970s, the firing activity (units) of neurons in the cortex and basal ganglia related to behavioral tasks has been actively investigated using the single-unit recording. However, it has been technically difficult to determine the cell subtype, location, and axonal projections of the recording cells using this method. Therefore, we introduced a new experimental technique and started research to explore the circuit principles of the cerebrum, which is responsible for brain functions, by investigating how functional information is processed within the neural circuits of the cerebral cortex and basal ganglia.
Methods
1. Operant conditioning
Traditionally, it has taken weeks to months to train small animals such as rats to perform operant tasks in which rats are rewarded for pressing a lever with their forelimbs. We devised the "spout lever," which unifies the lever and spout (drinking spout), and successfully trained rats to perform operant tasks (e.g., pushing, pulling, and holding) of manipulating the spout lever with their forelimbs in only a few days (Kimura et al., 2012; patent 5692681, 5935221). Utilizing this behavioral experimental system, rats performing the task can be supplied for physiological experiments in an extremely efficient manner. In fact, it is even able to stably record whole-cell membrane potential changes in motor cortex cells related to forelimb movements. It can also be widely applied to go/no-go discrimination and stop-signal tasks, depending on the purpose of the experiment (Yoshida et al., 2018). This is an original behavioral experiment system that our laboratory is proud of.
2．Juxtacellular recording
This is a revolutionary experimental technique that allows us to record the firing activity of single neurons and visualize the morphology of the recording cells. Visualization of recording cells using this technology allows us to identify their cell subtypes, determine the location of cell bodies, follow axonal connections, and examine the expression of various molecular markers. We were the first in the world to try juxtacellular recording experiments on behaving animals and succeeded in recording paracells from rat motor cortex cells and striatum cells (Isomura et al., 2009; 2013). For example, they were the first in the world to find that basket cells (a type of inhibitory cell) in the motor cortex rather increase their activity during the onset of movement.
3．Multineuron (+ local field potential) recording
This is an experimental technique that can record the firing activity of a large number of neurons at once via silicon probes (multiple point electrodes) (Commentary: Isomura, 2011). The signals obtained from each electrode are separated into firing activity (units) derived from individual neurons using an analysis technique called spike sorting. This spike sorting uses a highly accurate software called EToS developed by our collaborators (Takekawa et al., 2010; 2012). Although multi-neuron recordings do not allow visual identification of the recording cells, the cortex can be classified into RS cells (presumably mainly excitatory cells) and FS cells (mainly inhibitory cells) by spike shape. We have shown that there is strong synchronous firing between RS and FS cells in the motor cortex associated with motor expression (Isomura et al., 2009; Kimura et al., 2017). In the basal ganglia, we also recorded and identified projection cells in the direct and indirect tracts of the striatum and found that they exhibit distinct functional activities (Nonomura et al., 2018).
The multi-neuron recording method can also simultaneously record local field potentials (LFPs), which are electroencephalograms within brain tissue. Local field potentials in the cerebral cortex and hippocampus are thought to closely reflect synaptic interactions between neurons. We have previously elucidated the mechanisms of synchronous cortical and hippocampal activity during sleep and epileptic seizures (e.g., Isomura et al., 2006; F.-Tsukamoto et al., 2010). We have also investigated in detail the mechanisms and functions of synchronous activities such as gamma oscillations in the cortex and hippocampus related to motor expression and reward anticipation (e.g., Igarashi et al., 2013). Recently, we have also begun using Neuropixels probes (JJ Jun et al., 2017).
4．Optogenetics
Conventional electrophysiological methods can only observe the "correlation" of various neural activities, no matter how much recorded data is collected. To understand the mechanism of information processing in neural circuits, it is necessary to artificially manipulate the signals flowing through them to show the "causality" of neural activity. Optogenetics (optogenetics) technology, which has been rapidly developing in recent years, makes it possible to verify the "causality" of neural activity. We have introduced transgenic rats expressing channelrhodopsin 2 (membrane potential depolarized by blue light) and pathway-specific viral vectors (Saiki et al., 2018; Soma et al., 2017, 2019; Nonomura et al., 2018; Rios et al., 2019). Furthermore, we have developed a new method to automate collision testing by combining multi-neuron recordings and optogenetics in real time, the "multi-link method" (Mitani et al., iScience 2022). This method enables efficient identification of the projection sites of individual neurons.
5．Modeling and simulation
In our laboratory, a huge amount of recorded data can be obtained in a single multi-neuron recording experiment. The recorded data are automatically subjected to primary analysis (spike sorting, etc.) and stored in data storage, and in principle, the experimenters themselves proceed with secondary analysis in accordance with their research objectives. In doing so, with the cooperation of Professor Yutaka Sakai (Neural Computation Theory) and others at Tamagawa University, advanced theoretical analysis methods are incorporated throughout to achieve a high degree of completeness that leads to more robust conclusions. Theoretical researchers themselves are also making full use of simulation and modeling methods to verify interpretations obtained from experiments, and furthermore, to predict the results of the next experiment and establish new concepts, aiming to realize the "fusion of experiment and theory".
6．Two photon imaging and fiber photometry
Our laboratory has installed a two-photon microscope with the largest field of view in the world (Yu et al., 2021). Calcium imaging using this microscope allows us to observe tens of thousands of cell activities from many areas of the cerebrum simultaneously. We have also developed a new fiber photometry technique that allows us to observe temporal changes in extracellular dopamine concentrations in the striatum and other deep brain regions. While it is difficult to capture fast neural activities such as action potentials, these light-based physiological recording methods, in combination with various molecular biological tools, will contribute to the description of new physiological phenomena.
7．Transcriptome
The transcriptome is one of the molecular biological methods that has exploded in recent years. It has made it possible to identify many gene expression states of individual cells simultaneously. Our laboratory has recently begun to integrate this method with the neurophysiological recording methods described above. For example, what gene expression types do the groups of neurons involved in a particular learning process have, and how is the learning process related to the changes in gene expression? Such approaches provide new avenues for cerebral physiology and will be greatly developed in the future.
Research Direction 〜Persuit of Originality〜
In order to understand the essential principles of the brain, our research focuses on the mechanisms of brain circuits responsible for perceptual cognition and behavioral expression in rats. Most brain science research has so far been directed toward revealing the "localization" of brain functions by "averaging" brain activity. However, there is no doubt now that brain activity changes dynamically from moment to moment and that the entire multiregional network, not just a single region, is responsible for information processing. With this perspective of "from static to dynamic" and "from point to line" in mind, we hope to refine our research methods, advance into interdisciplinary fields, and pursue true originality without fear of failure.