Where are the memories hidden?
Whether it's a computer's RAM, or our brain, all memory storage devices store information by changing their physical properties.
What physical changes occur in the brain when memories are formed? More than 130 years ago, neuroscientist pioneer Santiago Ramóny Cajal first proposed that the brain stores information by rearranging synapses (connections between neurons). Since then, neuroscientists have tried to reveal physiological changes related to memory formation.
However, it is very difficult to image and map synapses. First of all, because synapses are so small, their size is about one billionth of the smallest object that a standard clinical MRI can see, these tiny synapses are always tightly packed together. The second difficulty is that in laboratory studies, scientists often use mice to study brain function, and there are about 1 billion synapses in the brains of mice, all of which are opaque or translucent in color like the tissues that surround them.
In some related studies in the past, scientists have focused on recording the electrical signals produced by neurons. While those studies have confirmed that neurons change their response to certain stimuli after memory formation, it is still impossible to determine exactly what drives these changes.
Now, a new study uses innovative imaging methods to exhaustively map the synapses of zebrafish before and after memory formation over a period of six years. Over the past 40 years, it has been widely believed that memory is achieved by changing the intensity of synapses. New research suggests that what changes memory when it is formed is not the strength of existing synapses, but rather the creation of more synapses in some areas of the brain, while others lose synapses.
Induce new memories with classical conditioning
Zebrafish are an ideal test subject for neuroscience research because they are both large enough to have brain function like humans and small and transparent enough to provide a window into the living brain. In the new study, the researchers tried to induce new memories for zebrafish fry that are only 12 days old. They used a learning process known as classical conditioning to train these zebrafish.
Zebrafish are particularly suitable as animal models for neuroscience research. | Image credit: Zhuowei Du and Don B. Arnold, CC BY-NC-ND
Classical conditioning involves exposing subject animals to two different types of stimuli at the same time—one that is neutral and does not cause a reaction; the other that is uncomfortable and will avoid as much as possible. When these two stimuli appear in pairs enough, the animals respond to neutral stimuli in the same way as they do to uncomfortable stimuli. This suggests that they have combined the two stimuli to form an associative memory.
In the experiment, they used an infrared laser to gently heat the zebrafish's head, using this stimulation as an uncomfortable stimulus. Zebrafish will try to swim away to avoid this irritation. Therefore, when the fish begins to flick its tail, it can be seen as a fish trying to escape. At the same time, they turn on a light as a neutral stimulus, and if the zebrafish is exposed to this neutral stimulus, it also shows a tail flick, which means that it recalls the previous situation when it encountered uncomfortable stimuli.
After 5 hours of training, the researchers were able to observe and capture changes in the brains of the zebrafish.
In this study, to map synapses, the researchers invented a way to alter the ZEBRAfish's DNA so that their neurons could produce fluorescent proteins that could bind to synapses. In this way, when a laser is swept by, the intensity and position of the synapse are marked by fluorescent proteins.
This picture shows neurons in the brains of a live zebrafish, where synapses are painted green. | Image credit: Zhuowei Du and Don B. Arnold, CC BY-NC-ND
They then imaged the synapses with a specialized microscope that uses a much lower laser dose and much less damage to the neurons than a standard microscope, allowing the researchers to image the synapses without losing their structure and function.
In this way, the researchers were able to directly observe the changes in live animals and obtain pictures of the changes before and after the changes in the same sample. Instead of experimenting on dead samples, as in past studies, only two different samples can be compared.
Finally, the researchers used newly developed algorithms to process and analyze hundreds of images and experiments, resulting in clear three-dimensional synaptic maps.
When comparing three-dimensional synaptic diagrams before and after memory formation, the researchers clearly found that neurons in one region of the brain produced new synapses, while neurons in another region lost synapses. In addition, they were surprised to find that the intensity of existing synapses related to memory formation was very small, which meant that the formation of associative memories was more about the formation and disappearance of synapses than the obvious intensity changes previously thought.
The brain synapses created by the scientists compared the size and location of synapses before and after memory formation and learning, and identified the synapses that were produced and eliminated during the process. | Image credit: Don Arnold
The researchers wanted the results to be as transparent and reproducible as possible, so they put all the data related to the paper on a public website for everyone to search and access.
Does removing synapses remove memories?
The new study sheds light on the role synaptic connections may play in memory and explains why associative memories are more persistent and vivid than other types of memories.
Associative memory induced by classical conditioning is thought to be similar to traumatic memory that causes post-traumatic stress disorder (PTSD). So the methods used in the new study to look at brain cell function will not only give us a deeper understanding of how memory works, but also open up potential new avenues for treating neuropsychiatric disorders such as PTSD and addiction.
Currently, the most common treatment for PTSD is exposure therapy, which repeatedly exposes patients to a harmless but stimulating environment to suppress recollection of traumatic events. Theoretically, this could indirectly modify the synapses of the brain to make memories less painful. Despite some success with exposure therapy, patients often have a tendency to relapse. This suggests that the underlying memories that led to the traumatic response were not erased.
While it's unclear whether synapse generation and loss actually drive memory formation, they have developed a technique that can quickly and precisely remove synapses without damaging neurons. They plan to next remove synapses from zebrafish or mice in a similar way to see if this changes associative memory.
Using these methods may be possible to erase the associative memory of devastating diseases such as PTSD and addiction from the body. However, before this treatment can be formally considered, synaptic changes that encode associative memory need to be more precisely defined. And, clearly, it also involves serious ethical and technical obstacles that need to be addressed urgently.
If, one day in the future, scientists develop synaptic surgery that can be used to modify memories, would you be willing to erase those bad memories?
#创作团队:
Don Arnold (Professor of Biological Sciences, University of Southern California)
Compilation: Sugar Beast
Typography: Wenwen
#参考来源:
https://theconversation.com/where-are-memories-stored-in-the-brain-new-research-suggests-they-may-be-in-the-connections-between-your-brain-cells-174578
https://dornsife.usc.edu/news/stories/3613/how-memories-are-stored/
#图片来源:
Cover image: Fakurian Design/Unsplash
First image: Don Arnold