Memories Selectively, Safely Erased In Mice

"While memories are great teachers and obviously crucial for survival and adaptation, selectively removing incapacitating memories, such as traumatic war memories or an unwanted fear, could help many people live better lives," says Dr. Joe Z. Tsien, brain scientist and co-director of the Brain & Behavior Discovery Institute at the Medical College of Georgia School of Medicine.

"Our work reveals a molecular mechanism of how that can be done quickly and without doing damage to brain cells," says the Georgia Research Alliance Eminent Scholar in Cognitive and Systems Neurobiology.

Dr. Tsien's research team, in collaboration with scientists at East China Normal University in Shanghai, were able to eliminate new and old memories alike by over-expressing a protein critical to brain cell communication just as the memory was recalled, according to research featured on the cover of the Oct. 23 issue of Neuron.

Dr. Tsien had already created a mouse that couldn't form memories by eliminating the NMDA receptor, which receives messages from other neurons. He then garnered international acclaim by making "Doogie," a smart mouse in which a subunit of the NMDA receptor is over-expressed. Younger brains have higher amounts of this NR2B subunit which leaves communication channels between brain cells open longer. That is why young people can learn faster than older adults.

This time he was examining downstream cascades of the NMDA receptor to learn more about memory formation. An abundant protein found only in the brain, called αCaMKII, was a logical place to look because it's a major signaling molecule for the NMDA receptor. He found that when he over-expressed αCaMKII while a memory was being recalled, that single memory was eliminated.

Receptors such as the NMDA receptor are like front doors to cells, providing an opening for signaling molecules such as calcium. Synapses are the point of communication between two cells, and NMDA receptors are on the receiving end of the message. Like people, neurons change with the signals they receive. "Learning changes the way cells connect to each other," says Dr. Tsien. To form a memory, the NMDA receptor is activated, which results in the insertion of AMPA receptors into those synapses and subsequent strengthening of the synaptic connections among hundreds of thousands of neurons. Scientists believe that αCaMKII plays an important role in the insertion of AMPA receptors into synapses during learning and subsequent strengthening of connections between neurons to create a memory.

Memory has four distinct stages: learning, consolidation, storage and recall. It has been difficult to dissect the molecular mechanisms of these stages because researchers lacked techniques to manipulate proteins quickly. For example, when researchers disable a gene suspected to play a role in the memory process, the deletion typically occurred throughout the entire period so it was impossible to tell which parts of processes were impaired. Previous technology would take several days to switch off a protein, which is the product of a gene.

So Dr. Tsien's team developed a powerful chemical-genetic method that allows him to use a pharmacologic inhibitor to instantly turn αCaMKII off and on in a mouse that he genetically engineered to over express this signaling molecule. That enabled him to study exactly what happened if he threw off the natural balance during the retrieval stage.

Much as a war veteran remembers a fateful patrol when he was fired upon, mice can establish a very long-lasting emotional memory about a place if, for example, they receive a mild shock to the paws while there. The researchers showed if they over-expressed αCaMKII, this powerful memory was rapidly erased as the animals tried to retrieve them while other memories remained intact.

A similar approach was taken with object recognition memory, giving mice a couple of toys to play with then erasing their memory of one of them. "You will feel like every time, it's a new toy," says Dr. Tsien.

While the ability to rapidly erase a selective memory is exciting, he cautions that its translation to humans would be difficult at this stage. "We are barely at the foot of a huge mountain," says Dr. Tsien. A possible strategy for humans would be a drug that mimics the αCaMKII over expression that researchers accomplished through genetic manipulation. Or, further downstream substrates that αCaMKII acts upon could become possible drug targets.

The research was funded by the National Institute of Mental Health, the National Institute on Aging and the Georgia Research Alliance.