A research group including Assistant Professor Kazumasa Tanaka, Graduate Student Yu-Ju Lin (at the time of research), and Technical Staff Ai Takahashi at the Memory Research Unit, Okinawa Institute of Science and Technology (OIST), alongside Deputy Director Takeshi Sakurai at the International Institute for Integrative Sleep Medicine, University of Tsukuba, in joint research with the Institute of Physiology, Charité - Universitätsmedizin Berlin (Germany), and the Scripps Research Institute (USA), clarified that for maintaining long-term memory, not synaptic potentiation but connection patterns of engram synapses are important. It was found that long-term memory remains even after artificial hibernation where many strengthened synapses are lost, and engram synapses selectively survive during hibernation. The findings were published in the August 13 issue of Science.
In current neuroscience, synaptic potentiation where connection strengths of synapses in the brain are maintained strong is considered essential for maintaining long-term memory. In fact, inhibiting synaptic plasticity inhibits memory acquisition.
On the other hand, it has been reported that synapses in the brain are unstable long-term and undergo constant change. The research group utilized synthetic hibernation to investigate how memories are stably carved in a constantly changing brain.
Until now, Sakurai and colleagues at the University of Tsukuba had clarified that hibernation is widely conserved evolutionarily, and manipulating circuits involved in hibernation conserved in the hypothalamus enables artificially inducing hibernation (artificial hibernation) in mice and rats that do not naturally hibernate.
The hippocampus of mice induced into artificial hibernation was investigated before hibernation, during hibernation, and after hibernation. Neural activity lowered remarkably in the mouse hippocampus during hibernation (for 2 days), dendritic spines of neurons almost disappeared, and synapse amounts also decreased by more than half. Such large-scale changes in neurons are not observed in brains of animals under normal states.
Therefore, performing various memory experiments on mice, they investigated whether long-term memory is maintained. As a result, mice waking from hibernation were healthy, retaining long-term memory. They confirmed that performance was comparable to mice without hibernation, rather being slightly higher.
Utilizing these brain changes accompanying hibernation, they removed structures unrelated to memory maintenance through hibernation, searching for structures for memory maintenance. Using genetic methods, they labeled only neurons activated specifically during memory acquisition, and furthermore fluorescently labeled only synapses between activated neurons (engram synapses).
Inducing hibernation in mice subjected to this operation, they verified whether synapses remain. It was found that while the amount of engram synapses during hibernation lowered by about half, dense, clustered engram synapses were left behind.
Next, using correlative light-electron microscopy (CLEM) combining optical microscopes and electron microscopes, they analyzed how clustered engram synapses connect with nerve fibers and other structures. As a result, it was clarified that clustered engram synapses surviving during hibernation preferentially bind to structures called multisynaptic boutons (MSBs). MSBs possess the characteristic that terminals of input-side neurons connect with multiple neurons simultaneously. It was also found that each spine within clusters binds to different input partners.
With this in mind it was theorized that geometric connection patterns across wide ranges such as MSBs and clustered engram synapses are important for memory maintenance. The possibility exists that memories are maintained in a compressed manner. It was clarified that maintaining strengths of synapses strengthened during memory acquisition is not essential for memory maintenance.
Tanaka commented, "We consider that the major significance of this study is that we were able to bring out the minimum unit of memory traces in terms of structure. In the future, if memory traces remain even when unnecessary synapses are pruned by hibernation, we plan to advance research toward elucidating how necessary and unnecessary ones are selected. In addition, when returning to original states after hibernation, we are also interested in how that origin is defined inside the brain."
Journal Information
Publication: Science
Title: Artificial hibernation reveals synaptic engram architecture associated with memory retention
DOI: 10.1126/science.aee7004
This article has been translated by JST with permission from The Science News Ltd. (https://sci-news.co.jp/). Unauthorized reproduction of the article and photographs is prohibited.