![リガクル[RIGAKU-RU] Exploring Science](/ja/rigakuru/images/top/title_RIGAKURU.png)
Whole-brain imaging to see between inputs and outputs
Organisms receive information (input), process that information appropriately, and then take action (output). At this point, there is no one-to-one correspondence between input and output. For example, if an organism receives information about the smell of food, it will go to the source of the smell if it is hungry, but it may not respond if it is not. In such cases, how is the information processed? What is information for living organisms? Yu Toyoshima, an associate professor in the Department of Biological Sciences at the School of Science, hopes to clarify this question through “whole-brain imaging.”
Toyoshima explains what exactly he means by whole-brain imaging.
“The brain processes information through the skillful use of neural circuits, stable connections of nerve cells. However, the human brain with its complex web of billions of cells is difficult to study. Therefore, I conduct research using the roundworm C. elegans as a model organism. C. elegans has 302 neurons, all of which have names. All the connections between the neurons are also known. So, it becomes possible to study every stage of information processing: observe which neurons process environmental information, how they process it, and if and how that leads to behavior.”
In other words, his approach is to observe everything happening between inputs and outputs. Toshima often uses AM radio as an example to explain the purpose of this approach.
“In AM radio, the input, radio waves, is converted into the output, sound. In practice, first, the antenna receives the radio waves and converts them into an electric current. Next, a tuning circuit made up of a coil and capacitor extracts a signal of a specific frequency from the current, and a diode regulates the signal. The speaker then converts the envelope, the result of amplitude modulation, into an audio signal. If you did not have this knowledge and were given an AM radio and asked to investigate its operating principles, what would you do? You might first remove the parts one by one to see if the radio works regardless, or you might observe the characteristics of the individual parts. Once you have finished, you may want to investigate the types of signals flowing through each component. It would be good to know how the signal changes between the input and output of a single component and how the changes lead to the final output. This is true for the study of neural circuits. Whole-brain imaging is an effort to study the components and follow the dynamics of neural circuits.”
The “intangible” revealed “in-between”
It is not a simple task to observe what happens between inputs and outputs in a radio, let alone in a living organism. Developing techniques to identify each neuron by its name in the roundworms used for experiments and to track the same cells in the image data is needed. Another challenge is to capture the head, where the neurons are concentrated. To do this, the roundworm’s head has to be kept in the microscope's field of view without its movements being hindered. Toyoshima's ability to quickly acquire the necessary knowledge and skills and to take a multifaceted approach are his strengths in tackling these challenges. He has many tricks up his sleeve: improving microscopes, developing image analysis techniques, designing microfluid chips that restrain roundworms, creating genetically engineered strains of roundworms, and more.
“For example, when observing neurons, the nuclei of the cells are so close together that it is difficult to determine which cell has fired. So, we developed image processing methods and genetically engineered roundworms in which each cell has a different fluorescent color. These approaches helped to tell apart each neuron.”
As a result of various efforts, Toyoshima successfully created an experimental system for time-series whole-brain imaging in which neural activity is captured while following a moving nematode. What emerged “between” input and output diverged greatly from what the researchers had initially anticipated.
“We managed to observe the firing of each named cell in chronological order, although they were not always responding to stimuli in an orderly manner. At first, we had no idea where to look for patterns. However, we were sure that there had to be a pattern somewhere, so we made even more detailed observations. Using statistical techniques such as mathematical modeling and independent component analysis, we found that many neurons were spontaneously active almost independently of stimuli and that only a limited number of neurons responded to stimuli in an orderly manner. We now know that there is an inherent pattern of activity in the C. elegans neural network and that without considering it, the input-output relationship cannot be understood or predicted.”
Bioinformatics: the biology of the “intangible”
Whole-brain imaging indeed helps to visualize the spontaneous activity of many neurons. But, according to Toyoshima, what is important is not the neural activity itself but the information it carries.
“When bioinformatics is mentioned, the first thing that usually comes up is information in genome sequences. But that is only one field of bioinformatics. Obtaining information from activity and imaging data of organisms is also a part of bioinformatics. In the radio example, the amplitude of radio waves encoded the audio information. But what happens in neural circuits? I want to explore how information for living organisms, such as information about the surrounding environment, is encoded in neural activity and how neural circuits extract this information to trigger appropriate behavior. Clarifying what information means to organisms should also be considered a field of bioinformatics. Much of biological research focuses on the “tangible,” such as genes, molecules, and cells. I was strongly attracted to research on the “intangible,” such as how organisms retain and retrieve information. We might call it the “biology of the intangible.”
Understanding life phenomena from start to finish
To take on the challenge of the “biology of the intangible,” Toyoshima ingeniously combines various methods of biology and information science: microscopy with image analysis and genetic engineering with microfluidic chip design. How did he arrive at this research theme?
One factor, he recalls, is his desire to understand mechanisms by taking them apart and putting them back together again. In other words, a desire to know the “in-between.” As a memorable anecdote, he cites a junior high school biology class in which he was taught the mechanisms of photosynthesis in college-level detail. He found it interesting that the class went beyond explaining merely the start and finish points, where water, carbon dioxide, and the energy in light are converted into sugar and oxygen. The class broke the process down to the cycles “between” those points.
The other factor, he says, is that, surprisingly, there was nothing he was especially good at or particularly wanted to do.
“When I enrolled in the Faculty of Science as an undergraduate student, I did not know what I wanted to do. Then, by chance, a friend and I had the opportunity to go to a talk by Professor Satoru Miyano, then Vice Dean of the Institute of Medical Science. There, he said, “systems biology is the future.” He also said that “the life sciences will need strategies that mobilize knowledge from various fields.” I did not have a firm idea of systems biology, but studying living organisms using all the available knowledge appealed to me.”
“After that, I learned that I could study systems biology in a special undergraduate education program that served as the preparation for those intending to major in bioinformatics and systems biology. The program was run by the Department of Biophysics and Biochemistry, and I managed to slip in just over the cut-off point. While studying systems biology as a master's and doctoral student and researching intracellular signal transduction, I realized that some phenomena are difficult to explain without looking at temporal patterns and fluctuations in the amounts of signaling molecules. Thus, I realized the importance and fun of research that looks at life phenomena from start to finish.”
Toward a virtual roundworm
What is the ultimate goal of Toyoshima's research?
“One of our goals is to create a highly accurate virtual roundworm that properly captures the nervous system and its activity. The virtual worm is a quantitative mathematical model that can reproduce and predict what kind of information is received, what kind of information processing takes place, and what kind of behavior is triggered. Simple models that can explain the principles are vital, but they tend to be top-down and ignore areas where the model does not match the experimental results. However, using machine and deep learning makes it possible to create models that can reproduce experimental results well. These detailed and complex models also have high predictive power and can predict the results of experiments that have not yet been performed. We believe that creating these highly accurate virtual roundworms and then testing and correcting the model whenever necessary is important to understand properly how neural circuits process information.”
This naturally leads to the question: when a detailed and complex model is constructed, can we truly say that we understand the phenomenon called life? In other words, is looking at a complex model the same as looking at the roundworm itself? Toyoshima's response: some things can only be achieved using models.
“There are experiments in which specific neurons are stimulated to study downstream effects, but to examine differences with and without stimulation, the conditions at the time of stimulation need to be the same. However, because the neural activity in the whole brain switches stochastically, it is difficult to control for baseline conditions outside of models. Other research areas would also benefit from models, such as exploring combinations of cells to be stimulated, broadening the scope of possible research. I expect this hybrid way of conducting research will spread to other fields as well.”
※Year of interview:2024
Interview/Text: HORIBE Naoto
Photography: KAIZUKA Junichi
Related articles
