Session Spotlight: How Do Short-Lived Proteins Keep Long-Term Memories?

All memories evolve in the same way, according to Todd Sacktor, M.D., distinguished professor of anesthesiology, neurology, and physiology and pharmacology at SUNY Downstate Health Sciences University. The process involves four steps: encoding, consolidation, storage, and retrieval. Encoding a memory takes milliseconds, while consolidating a short-term memory into a long-term one takes around an hour.

Then comes the rub: How is it that consolidated memories are securely stored for potentially 90 years when all proteins made in neurons last for a few days at best before degrading? Sacktor—a protégé of Nobel-winning psychiatrist and memory scientist Eric Kandel, M.D.—discussed his lab’s efforts to unravel that mystery during a special session at the Annual Meeting.

He noted that the answer lies within two neuronal proteins. The first is an enzyme called protein kinase M zeta (PKM-Z). When a memory is being learned, neurons briefly increase production of PKM-Z, which travels to target synapses where AMPA receptors (a type of glutamate receptor) have congregated to begin the encoding process. PKM-Z is persistently active, which enables it to significantly boost AMPA activity and strengthen the synaptic connection, but the protein only lasts for a few hours. This is enough to ensure consolidation of the memory but not long-term storage.

Help is nearby in the form of a second protein called KIBRA—a scaffolding protein that attaches to the neuron membrane near AMPA receptors. Sacktor’s team found that three KIBRA proteins and three PKM-Z proteins can merge into a larger, barrel-shaped structure; these hexamers create a stable structure that could last several months. What’s more, if one of the individual components breaks down, the remaining proteins can absorb a nearby free-floating KIBRA or PKM-Z like a sponge. Essentially, these molecular barrels are near-permanent structures located on memory-encoding synapses.

Sacktor said that his work has led to some interesting philosophical thoughts. As time passes, for example, none of the original proteins that helped make a memory are still there, though the synaptic connection endures. Is that the original memory or just a replica?

Perhaps more pertinent to the APA audience were the clinical ramifications. Once a memory is retrieved, it becomes malleable and needs to be reconsolidated. Certain psychotherapies take advantage of this window of reconsolidation to make a stored memory less anxiety-inducing or traumatic, but with a better molecular understanding there might be pharmacological ways to erase that memory completely.

Obviously, pharmacological targeting must be precise or else one risks erasing a boatload of memories. Sacktor suggested the best strategy would be to block the synthesis of new PKM-Z proteins right after a memory has been recalled, rather than trying to degrade or sequester existing proteins in the neuron. ■

(Image: Getty Images/iStock/BlackJack3D)