How the Brain Keeps Memories Alive

Wake/sleep cycle may influence storage

by Jim Schnabel

December 15, 2008

While a good night’s sleep helps long-term memories to form in the brain, the 24-hour “circadian” sleep/wake cycle may also be needed to keep these memories from fading away, according to a recent study.

“There are circadian oscillations in the activities of proteins required for memory in the hippocampus,” says Kristin Eckel-Mahan, lead author of the study, published Aug. 10 in Nature Neuroscience. “These are important for the maintenance of memories and so it appears that they should not be disrupted.”

Most neuroscientists think a memory of something is roughly a reactivation—with retouchings here and there—of the same pattern of activity in the brain that occurred when the thing was first perceived. This pattern of activity amounts to a unique constellation of neurochemical couplings at the tiny, sensitive junctions—synapses—through which neurons receive input signals from other neurons.

When the perception of a thing evokes the same pattern of synaptic activity again and again, these activated synapses become increasingly sensitive, until the same pattern of activity can be evoked even in the absence of the original stimulus. This enhancement of synapses with usage represents, for the most part, the phenomenon we call learning.

But how is the sensitivity of a synapse enhanced with usage? And, in the case of humans, how does it stay in this memory-retaining state for as long as a century?

Maintaining  the program

Researchers know that the synaptic enhancements that make up a memory don’t just “stick” by default, as in a magnetic hard drive. They must be maintained by some active process within their respective neurons. Eckel-Mahan’s laboratory, at the University of Washington, Seattle, has focused on a genetic manipulation process known as CRE-mediated transcription (CRE is a class of protein whose acronym stands for “cAMP response element). Neuroscientist Daniel Storm, head of the laboratory and principal investigator on Eckel-Mahan’s study, has shown that the activation of this molecular cascade inside a neuron causes the increased production of key synapse-enhancing proteins.

Other experiments have shown that short-term memories from important perceptions during the day are strengthened into long-term memories at night, during sleep. While working on her Ph.D. in Storm’s laboratory, Eckel-Mahan has conducted a series of experiments to find out whether these two memory-enhancing phenomena—sleep and CRE-mediated transcription—are linked.

The new study reports on the results of that effort. The study shows that in the mouse brain region known as the hippocampus, which is crucial for the formation of most kinds of memory, the activities of molecules in the CRE-mediated transcription cascade do indeed rise and fall according to a 24-hour cycle, reaching their normal peak during the day and a low point at night (mice mainly sleep during the day and are active at night).

Eckel-Mahan and her colleagues also found that this peak coincided with the ability of mice to form new memories. The mice were less successful at forming memories when the researchers disturbed the normal circadian cycling of this signaling cascade—either by injecting a compound that blocks the cascade’s activity or by keeping the lights on over the mouse cages constantly.

Training mice to remember

In these tests, the researchers used a standard “contextual fear conditioning” technique in which mice learn to associate a new environment—a dark box, for example—with electric shocks. Mice conditioned this way show that they have a memory of this association when they “freeze” before entering this environment. Less-frequent freezing among a suitably conditioned group of mice represents greater memory impairment.

Using this method, Eckel-Mahan and her colleagues showed that disturbances in the rhythms of the CRE signaling pathway seemed to do more than interfere with the formation of new memories in the mice: The disturbances also were followed by the impaired recall of trained, existing memories. They saw no such impairment in mice whose CRE pathway cycling had been undisturbed.

Eckel-Mahan says the results need to be confirmed and elaborated by further studies, but “from our work so far it seems that if you disrupt the oscillation of that pathway, either by decreasing the peak of the oscillation or by increasing the trough of the oscillation, then you’re going to interfere with the strength of memory.”

 “During the day we acquire new information, we learn new things,” Storm says. “During the night, these memories are consolidated and made stronger by activation of the [CRE-mediated transcription pathway] in the hippocampus.”

Storm and his colleagues believe that the upstream event triggering this molecular pathway is a signal from the suprachiasmatic nucleus, the brain’s circadian pacemaker. This signal causes the brain to strengthen the synapses whose activity during the day effectively “tagged” them.

Link might explain jet lag, other troubles

“Their discoveries certainly give us a new insight into the molecular basis of memory,” says Tania Roth, a neurobiologist at the University of Alabama who works in the same research area. “But on the grander scale, this gives us a framework to understand why circadian rhythms in general, and sleep cycles in particular, have such a tremendous impact on cognition.”

The precise relationship between sleep and the activation of the synapse-enhancing CRE-mediated transcription pathway is not yet clear, says Storm. But a direct molecular connection between the two could go a long way toward explaining why sleep disturbances and cognitive impairment are so often linked.

For example, airline crews routinely experience the circadian-rhythm disturbance known as jet lag, and in a study in 2001, also published in Nature Neuroscience, Kwangwook Cho at the University of Bristol in the U.K. reported deficits in hippocampus-related memory and even shrinkage of the temporal lobe, which includes the hippocampus, in a group of airline crew members after years of jet lag exposures. Cho attributed these effects to higher stress hormone levels triggered by jet lag; but the results from Eckel-Mahan’s study suggest that disturbances to the normal rhythms of the CRE-mediated transcription pathway could also have been to blame.

“I think that could certainly be an explanation for any cognitive impairments in people who are fatigued or undergoing jet lag symptoms,” says Eckel-Mahan.

Disturbances in both sleep patterns and cognition also occur in sleep apnea and Alzheimer’s disease—as well as in normal aging. “We are looking at the relationship between aging and the decline in signaling events that lead to CRE-mediated transcription,” says Storm.