In 2001, Eric R. Kandel walked into Susan L. Lindquist's office and asked her a question out of left field. "Do you think prions could be involved in storing memory?"

Normally considered deformed renegades that mercilessly rob the brain of functions, prions might seem like unlikely candidates for safeguarding memories. "I nearly jumped out of my chair because I was speculating the same thing," recalls Lindquist, a noted prion researcher who was then an HHMI investigator at the University of Chicago and has just stepped down after 3 years as director of the Whitehead Institute for Biomedical Research in Cambridge, Massachusetts.

Kandel, an HHMI investigator at Columbia University College of Physicians and Surgeons and a 2000 Nobel laureate, told her about a puzzling prion-like property in a protein that seemed to play an essential role in long-term memory storage. Kandel's previous research had described a memory-forming cascade set in motion by CREB (cAMP response element binding protein). CREB functions as a molecular switch that turns short-term memory to long-term memory. A learning experience flips that switch, allowing proteins and untranslated messenger RNAs (mRNAs) to be dispatched throughout the cell body. However, he found that many mRNAs remain dormant until another learning experience acts as an alarm clock, arousing the mRNAs to build the proteins that strengthen just those synapses where long-term memories form.

Kandel was intent on discovering what molecular mechanism acted as the alarm clock for dormant mRNAs. He further wanted to understand why that mechanism was set in motion at just those synapses where a specific, enduring memory forms. Like many neuroscientists, his lab used the foot-long snail Aplysia as a model organism because its exceptionally large neurons are relatively easy to study and manipulate.

Kausik Si, a postdoc in his laboratory, had pinpointed CPEB (cytoplasmic polyadenylation element binding protein) as the selective trigger that woke up the dormant mRNAs at those preloaded synapses. Kandel's team wanted to know how CPEB maintains the continuing protein synthesis that stores a memory long after the learning experience has passed.

While investigating that question, Si noticed that the CPEB found in neurons differs from that of other cell types. Normally, CPEB has phosphorylation sites that activate mRNAs, which increases their levels of protein expression. In snails as well as in flies and mammals, however, the CPEB in neurons does not have those sites. Instead, it has a strange functional domain that is unusually rich in the amino acid glutamine and lacks any secondary structure.

Si knew of only one other protein type with a similar domain—prions.

Prions cause a protein to alter its shape and then enforce a similar, self-perpetuating conformation on similar proteins. Si and Kandel wondered whether prion-like, self-perpetuating qualities in CPEB could be what keep a memory vivid over time. That's why Kandel popped his question to Lindquist and later collaborated with her to pursue it.

Lindquist explains that the "funky" prion domains attach to one another, forming immobile, insoluble clumps. The rest of the proteins dangle along the side of the clump, like charms on a bracelet. This state can cause them to lose their beneficial function and gain a toxic function that leads to neurological disorders such as mad cow disease.

However, Lindquist had shown that the dangle sections of yeast prions can remain functional—and behave beneficially—in cultures. Theoretically, she reasoned, prion qualities could enhance a protein's function if, for instance, it needs to be anchored to a site and work cooperatively to sustain a process—such as storing a memory at a synapse.

To test whether the Aplysia neuronal CPEB does act like a prion, Si attached its weird domain to a yeast protein. Using a color assay that Lindquist had developed, he watched the protein morph into a self-perpetuating prion before his very eyes. Then Si devised an assay for the whole Aplysia CPEB when expressed in yeast. It, too, appeared to behave like a prion.

Moreover, while Aplysia's CPEB was very active in the prion clusters, it barely functioned in the nonprion state. Perhaps, the researchers hypothesized, CPEB needs to be in a prion-like state to sustain the perpetual protein synthesis necessary for storing memories. And perhaps that is why certain events—like repeating the times tables, practicing piano, and crashing a car—can become so unforgettable. They trigger enough CPEB production so that some copies of the proteins convert into prions to perpetuate themselves at the synapses where the long-term memory forms.

Kandel, Lindquist, and Si proposed this model of memory storage in the December 26, 2003, issue of Cell, accompanied by a Kandel and Si article on the role of CPEB in Aplysia long-term memory storage. "It's a very nice finding," Kandel says. "We're now going back to the Aplysia nervous system, where we've shown that this protein is required for the maintenance of long-term memory. We want to find out if the prion domain causes self-perpetuation in neurons as it does in yeast, and, if so, if it is the mechanism that maintains memory in Aplysia neurons."

Lindquist and Kandel speculate further that a similar prion mechanism might be involved in other contexts, such as developmental processes and cancers, where cells maintain a continuing function. This research therefore might lead to improved treatments for a range of disorders.

—Cathryn M. Delude

Download this story in Acrobat PDF format.
(requires Acrobat Reader)

Reprinted from the HHMI Bulletin,
Fall 2004, pages 14-23.
©2004 Howard Hughes Medical Institute