Making memories last

A protein with prion-like properties may play a key role in transforming short-term memory into long-term memory, enabling us to recall events from our distant past.

As we form memories the connections between neurons in our brain undergo subtle changes. But how these connections, or specialized contact points called synapses, stay strong and keep memories alive for decades has remained elusive. Associate Investigator Kausik Si, PhD, and his team discovered a major clue in the tiny brains of fruit flies: The ability of the synaptic protein Orb2 to form hardy, self-copying protein clusters known as oligomers may be what makes memories stick.

The finding supports a surprising new theory about memory, and may have profound impact on explaining other oligomer-linked functions in the brain, including Alzheimer’s disease and prion diseases. “The idea that prion-like molecules could have a normal physiological function has challenged our perception about prions and proteins as a heritable factor,” says Si.

Prions first made headlines when they were identified as the cause of bovine spongiform encephalopathy, which later became known as “mad cow disease.” During a prion infection, the infectious form of the prion converts the normal version of the protein into a toxic form that clumps together, triggering an out-of-control chain reaction that wreaks havoc on brain cells. Despite their similarities, Orb2 and prions differ in important ways.

“Unlike prions, Orb2 doesn’t convert spontaneously but instead oligomerizes in a controlled fashion in response to a physiological signal,” Si explains. And unlike other known prion-like aggregates, oligomeric Orb2 doesn’t kill nerve cells. Instead it regulates the synthesis of proteins necessary to increased synaptic strength. What’s more, once activated, oligomeric Orb2 can replenish itself without any further input making it a perfect “molecular flag” to designate a synapse for a sustained increase in its efficiency.

Si’s investigations in this area began nearly a decade ago during his doctoral research in the Columbia University laboratory of Nobel-winning neuroscientist Eric Kandel, PhD, in the sea slug Aplysia californica, which has long been favored by neuroscientists for memory experiments because of its large, easily studied neurons. He found that in Aplysia, a protein known as CPEB that maintains an increase in synaptic efficacy, has an unexpected property.

A portion of the structure is self-complementary and—much like empty egg cartons—can easily stack up with copies of itself. CPEB thus exists in neurons partly in the form of oligomers, which increase in number when neuronal synapses strengthen.

CPEB-like proteins exist in all animals, and in brain cells they play a key role in maintaining the production of other synapse-strengthening proteins. Studies by Si and others in the past few years have hinted that CPEB’s tendency to oligomerize is not merely incidental, but is indeed essential to its ability to stabilize longer-term memory. “What we’ve lacked till now are experiments showing this conclusively,” Si says.

The key was to show that the disruption of Orb2 oligomerization on its own impairs fruit flies’ ability to form long-term memories. Yes, fruit flies can learn. They can be trained to associate a chemical odor with a sugary reward. Hungry flies will rely on these odor memories to guide their behavior for several days after training. In a different memory test known as male courtship conditioning, male flies are exposed to an unreceptive female. Lured by the female, male test flies will initiate courtship, but their advances are inevitably rejected by the unreceptive female. After being scorned multiple times over several hours, the fly learns not to make advances when they encounter an unreceptive female again at a later time.

When the researchers interrupted Orb2’s ability to stack up, the genetically modified fruit flies flunked their long-term memory tests. “For the first twenty-four hours after a memory-forming stimulus, the memory was there, but by forty-eight hours it was gone, whereas in flies with normal Orb2 the memory persisted,” recalls Amitabha Majumdar, PhD, a postdoctoral  researcher in Si’s lab who performed most of the fly experiments.

Si and his team are now following up with experiments to determine how long Orb2 oligomers are needed to keep a memory alive. “We suspect that they need to be continuously present, because they are self-sustaining in a way that Orb2 monomers are not,” says Si.