Protein clumps like a prion, but proves crucial for long-term memory

Sea slugs make memories with a twist. Screwing a normal nerve cell protein into a distorted shape helps slugs, and possibly people, lock in memories, new research shows.

Notably, the shape change also brings a shift in the protein’s behavior, leading it to form clumps. That kind of behavior is the sort seen in prions, the misshapen, infectious proteins that cause mad cow disease, scrapie and other disorders (SN: 7/31/04, p. 67). But the new study, published February 5 in Cell, shows a possible normal function for the shape-shifting, suggesting that twists and clumps don’t necessarily make prions monsters.

In one sense, prions are machines of “molecular memory,” says Yury Chernoff, a biologist at the Georgia Institute of Technology in Atlanta and editor in chief of the journal Prion. The proteins remember what happened to them — changing shapes — and then transmit that change to other proteins. “But the notion of these machines being used for cellular, and therefore organismal, memory is truly amazing,” he says.

If further research shows the process works the same way in humans as it does in sea slugs, prionlike proteins might eventually be used in memory-enhancing treatments, Chernoff says.

Prions have a bad reputation due to the most famous of the shape-changing proteins, called prion protein or PrP. When PrP switches from its harmless form, which is normally present in nerve cells, into a prion form, it corrupts other PrP molecules that then assemble themselves into nearly indestructible plaques known as amyloids. Scientists don’t know for sure what the protein’s everyday role is, although recent work has shown that the normal form of PrP may help maintain the insulation that nerve cells need for effective electrical communication (SN: 2/13/10, p. 17). What is certain is that the prion form is a neuron killer that can be transmitted from one animal or person to another if infected tissue is eaten. Other noninfectious, amyloid-forming proteins, such as those involved in Alzheimer’s disease and Parkinson’s disease, also kill brain cells (SN: 8/16/08, p. 20).

But the new study suggests that some proteins may change shape and form amyloids in ways that are actually helpful for neurons.

Working with large Aplysia sea slugs, scientists had previously discovered that a protein called CPEB is needed for the slugs to form a long-term memory of a mild electric shock paired with a touch on the nose. CPEB, short for cytoplasmic polyadenylation element binding protein, normally helps regulate production of other proteins in neurons, but breaks down quickly. The first clue to how the short-lived protein could help produce long-lasting memories came when Kausik Si, a neuroscientist at the Stowers Institute for Medical Research in Kansas City, Mo., discovered that CPEB could behave like a prion in yeast, changing shape and triggering the formation of durable and self-assembling amyloid structures.

In the new study, Si, Eric Kandel of Columbia University in New York City and colleagues show that CPEB behaves much like a prion in sea slug neurons too. But instead of killing the nerve cells, CPEB amyloids help strengthen connections, called synapses, between neurons.

Serotonin, an important chemical messenger in the brain, signals CPEB to change shape and assemble into amyloid clumps at the synapses, the team found. When the researchers blocked CPEB’s ability to form amyloid, the slugs couldn’t remember for more than a few hours that a poke in the nose would be followed by a shock on the tail. When CPEB can change into its prionlike amyloid form, the slugs remember the painful pairing for months. The research adds to previous work by an Australian group that found CPEB was important for long-term memory in fruit flies. Humans also have a version of this protein.

“It’s super exciting that what they have here is good evidence for functional amyloid in the brain,” says John C. Price, a biochemist at the University of California, San Francisco.

Because the CPEB protein isn’t infectious, it doesn’t meet the strictest definition of a prion, Price says. “Some people don’t really truly believe the yeast proteins are prions either because they don’t show this infectivity.”

CPEB’s prion form might not be infectious because the transformation is tightly controlled by serotonin, Price says. “One potential question this raises is whether there is a nonpathogenic prion form of PrP,” he says. The prion may cause disease only when amyloid formation runs amok.

CPEB’s involvement in memory might suggest that cells use aggregates of proteins to store information. “This could be a whole new repository for biological information,” Price says. “I really like the implications.”