How does our brain create memory

How does the brain store memories? Two recent PNAS publications on variability and stability in neural networks

How can our brain reliably store important information and experiences without the number of its cells and connections having to keep growing over the course of a lifetime? Prof. Martin Korte's research group at the Institute of Zoology at the Technical University of Braunschweig has come two steps closer to answering this question. The results of their basic research may later be of importance for clinical research, for example on Alzheimer's disease. You are now in two articles in the prestigious journal PNAS (Proceedings of the National Academy of Sciences of the United States of America) released.

Save and forget

Our brain stores a huge amount of information every day. The storage units for this information can be found in the synapses, i.e. in the fine branches through which the nerve cells in the brain network with one another. Each individual cell has up to 10,000 of these tiny branches. As soon as we process information, it changes. If certain information is now to be overwritten in the long-term memory, this means that the corresponding synapses have to change permanently. A mechanism is set in motion in the cell nucleus that releases certain proteins via the genes there. This process is called transcription.

But how do the centrally produced proteins “know” which synapses should be permanently strengthened? And how do you get to the right place? Martin Korte and Shreedharan Salikumar at the Institute of Zoology at the TU Braunschweig observed how the affected areas of the synapses draw attention to themselves in a sophisticated way for this purpose. They produce a marker ("tag") that ensures that the necessary proteins are only effective at these marked synapses. Thanks to “synaptic tagging”, proteins no longer have to be specifically transported from the cell nucleus to the right place, but can be “sent” to a larger functional unit. They only develop their effect in the right place. “In this way, the brain hangs a pennant with the inscription 'please process and keep' on the incoming signals,” explains Prof. Martin Korte. "The brain can react to signals that do not receive this pennant, but it will later forget them in order to conserve its storage capacity."

For a long time, research assumed that all storage units in the dendrite tree function in a similar way. Korte and Salikumar have now been able to prove that they can actually pick up differently coded signals independently of one another and react very flexibly to requirements. In turn, they can form very complex structures which, in addition to the cell nucleus, can be responsible for long-term memory - or for forgetting.

The memory couples different signals together

The more important the information, the more complex the signals that ensure that these memories find permanent storage locations. This is the only way for the signals to overcome the obstacles that shield the areas in the cells that are responsible for long-term memory.

If we later call up the information that is not stored in the cell nucleus but in the network itself, there may be overlaps or couplings in the memory. Because then a whole system of other signals can be activated at the same time, which are stored in the cells involved. Therefore, one often not only remembers formative events, but also, for example, the exact place where they took place. “This phenomenon can also explain why it is so difficult to learn Spanish and Portuguese at the same time,” explains Korte. For example, one and the same neuron can be involved in processing important terms in the related languages.


Published online before print January 19, 2011, doi: 10.1073 / pnas.1016849108


Permanent structures in the highly flexible brain

A second PNAS publication deals with the question of why we can reliably retrieve important information over a long period of time and differentiate between what has been learned and what is new. Andrea Delekate, Marta Zagrebelsky, Stella Kramer and Prof. Martin Korte from the Institute of Zoology at the Technical University of Braunschweig and the Swiss brain researcher Prof. Martin E. Schwab (University of Zurich and ETH Zurich) helped answer a medical riddle.

NogoA is a protein that inhibits the growth of nerve cells. It occurs in the body only in the central nervous system, i.e. in the brain and spinal cord - and can have a fatal effect. If nerve cords, for example in the hand, are injured, the tissue can usually regenerate. However, if the spinal cord is injured, as is not uncommon in motorcycle accidents, this protein ensures that the nerves do not reconnect with each other. “So far, we knew how NogoA works, mainly because of Martin Schwab's research. But we didn't know why it existed. Above all, the fact that the protein occurs most of all in the hippocampus puzzled us, ”explains Martin Korte. “It is mainly found in the brain region that is responsible for which information is transferred from short to long-term memory.” In order to be able to recognize patterns and retain experience, we not only need a very flexible nervous system. Our brain also has to build up permanent structures in certain areas.

NogoA protects against too much change

The researchers were able to show that the NogoA stabilizes both the function and the structure of nerve networks and in this way helps to store memories. It therefore defines the functionality of neural networks in certain parts of the brain and protects them from further changes. "However, antibodies can bind to the molecule and in turn prevent its effect at certain points," says Korte.

The findings can lead to the development of new drugs in a few years. In the event of damage to the central nervous system, such as occurs in a stroke, the targeted blockade of NogoA can promote plasticity and support rehabilitation (i.e. facilitate the variability of the neural networks). The brain of older people could then also be supported in remembering things better and learning new things.



Published online before print January 24, 2011, doi: 10.1073 / pnas.1013322108



Prof. Martin Korte
Technical University of Braunschweig
Zoological Institute
Cellular Neurobiology Department
Biozentrum, Spielmannstr. 7th
38106 Braunschweig
Tel .: +49 531 391 3220
Email: [email protected]