A new study by researchers from the Medical College of Georgia and the Shanghai Institute of Brain Functional Genomics shows that an enzyme called alpha calcium/calmodulin-dependent protein kinase II (αCaMKII) [this is a type of CaM Kinase] is essential for the formation and retrieval of memories. By briefly altering levels of αCaMKII activity at different stages of the memory process, they were able to prevent the transfer of new memories from short-term to long-term storage and to selectively erase specific memories as they were being recalled.
CaMKII is found only in the brain, and constitutes more than 1 percent of the proteins in that organ. The enzyme appears to be a key mediator of long-term potentiation (LTP), the process by which connections between neurons are strengthened, and which is thought to be the cellular basis for learning and memory. It is attached to the NMDA receptor, which is known to be involved in synaptic plasticity, and is activated by calcium currents entering through it. Upon activation, CaMKII dissociates itself from the receptor, and goes on to regulate the activity of dozens of proteins in multiple signaling pathways. The end result is that a single enzyme modulates a wide variety of crucial cellular activities.
To better understand the function of αCaMKII, neuroscientist Joe Tsien at the Medical College of Georgia and his colleagues generated a strain of transgenic mice expressing increased levels of αCaMKII in the forebrain. But the mice weren’t expressing a standard version of the enzyme: the scientists had engineered the molecule so that it could be selectively turned off via an inhibitor molecule for 40 minutes at a time.
The researchers trained this strain of mice (along with a control group) in three different memory tasks. When the recall tests were performed an hour later, the mice with increased αCaMKII levels were found to be severely impaired on all three tasks. When they were treated with the αCaMKII inhibitor 15 minutes before the recall tests, however, their performance was comparable to that of the controls, suggesting that the earlier poor performance was due to deficits in recall and not memory acquisition or storage.
In yet another set of experiments, the recall tests were carried out one month after training. Those transgenics treated with the αCaMKII inhibitor 15 minutes before recall performed normally. However, when the inhibitor was administered two days before training and then continuously for 28 days, so that it was removed two days before the mice were asked to recall the memory, the transgenic mice again exhibited severe memory deficits. The mice were then made to perform two recall tests one month after training, with aCaMKII activity inhibited only during the second. If the memory impairments were due to recall deficits, the mice would successfully retrieve the memories in the second trial. If, on the other hand, they were due to erasure of the memories, then recall should fail in both trials. Sure enough, the transgenics treated with the αCaMKII inhibitor during the second trial still performed very badly in both trials, suggesting that the memories are in fact erased at the time of recall.
Finally, a series of sequential retrieval tasks showed that the memory erasure was highly selective. Mice were again trained in two of three previous memory tasks, which involved fear conditioning. They were placed in a compartmentalized chamber and given several mild electric shocks; after a few trials, they learned to associate the shock with the compartment in which it had been given and with a loud sound that had been presented along with the shocks. One month later, when placed back in the same part of the box, the control mice exhibited fear behavior—they quickly froze when returned to the chamber—but the transgenics did not.
Later, outside the chamber, the animals were presented with the sound they had previously learned to associate with an electric shock. This time, those transgenics treated with the αCaMKII inhibitor 15 minutes before the recall test behaved just like the controls—they froze when they heard the sound. This finding suggests that, while their fear memory of the box had been erased, they were still able to recall the connection between the electrical shock and the sound. The memory erasure was also found to occur very rapidly. In all the recall tests, the untreated transgenics initially performed well during the first minute of testing, but the memory of the association quickly decayed, so that performance grew progressively worse after minutes.
A Cure for PTSD?
This study shows that αCaMKII is crucial for short-term and long-term memory, and that both types of memory share the same molecular mechanisms. The contextual and cued fear memories examined in the study are thought to be encoded by overlapping neuronal circuits, so it seems that the selective memory erasure occurs at specific subsets of synapses within those circuits, because the researchers were able to erase one of these memories while sparing the other.
Theoretically, αCaMKII inhibitors would be useful as treatments for conditions such as post-traumatic stress disorder (PTSD), in which people are unable to forget traumatic memories. (The movie Eternal Sunshine of the Spotless Mind, for instance, imagined a scenario in which people were able to selectively erase memories.) In practice, though, there are enormous difficulties in developing such compounds for use in humans, and any such treatments would case raise serious ethical questions. Nevertheless, this paper marks an important advancement in understanding how chemical pathways in the brain are able to encode and recall events and experiences.
Are you a scientist? Have you recently read a peer-reviewed paper that you want to write about? Then contact Mind Matters editor Jonah Lehrer, the science writer behind the blog The Frontal Cortex and the book Proust Was a Neuroscientist.