Much as we would like to trust our memories, they are not exactly archival quality. A new memory is fragile and easily changed or lost. It takes some time—known as the consolidation stage—to cement a memory and transfer it from short-term, hippocampal storage to the long-term, cortical vault. And every time we dust off an old memory to relive it, it becomes temporarily amenable to alterations and updates. During this labile period, new information can modify the old memory. A reconsolidation period then solidifies the knowledge once again. Although the exact mechanisms of consolidation and reconsolidation are unknown, scientists know the process requires protein synthesis (Nader et al., 2000; Milekic, 2002).

Loosening a memory every time it is used may sound like an evolutionary blunder, but it is no mistake, said Robert Stickgold of Harvard Medical School, who was not involved in either study. In fact, it is an advantage to allow memories themselves to evolve, incorporating new information. “Whenever you open a document in your mind, it is opened in edit mode,” he said.

But editing can go awry. In some disorders, people fail to consolidate, losing fragile memories. Conversely, in PTSD, a painful recollection seems as fresh years later as the day it was made. New research suggests not only ways people could strengthen desirable memories, but also possibilities for neutralizing unpleasant ones. For example, scientists have found that a person’s reactivated fear memory is liable to disruption by either new information (Schiller et al., 2010) or pharmaceutical treatment (Brunet et al., 2008). Essentially, you have to pull out the distasteful memory in order to subdue it.

Avoid the Dark Side
In the Nature study, first author Dillon Chen and senior author Cristina Alberini at the Mount Sinai School of Medicine in New York tested inhibitory avoidance training, that is, how well rats remember to eschew the dark side of a box. Rats are naturally drawn to dark places, but in this one, they receive an unpleasant shock—definitely something they would want to remember in the future. After training, Chen timed how long it took rats to wander into the dark area as a measure of how well they remembered what would happen in there.

Previously, scientists in Alberini’s lab discovered that expression of a transcription factor, CCAAT enhancer binding protein β (C/EBPβ), rises in the hours and days immediately following a training session in the box (Taubenfeld et al., 2001). Among the targets of C/EBPβ is the growth factor IGF-II, which is expressed highly in the adult hippocampus. Scientists do not know much about what it does there, and Chen hypothesized that it could be involved in consolidating memory.

First, he tested whether learning affected IGF-II levels. He confirmed by Northern blotting, quantitative reverse transcriptase polymerase chain reaction, and Western blotting that like C/EBPβ, IGF-II levels rose in the days following memory training. If the growth factor was involved in memory consolidation, he reasoned, then blocking it via antisense oligonucleotides should interfere with memory. Sure enough, anti-IGF-II-treated rats wandered into the dark side of the box in as few as 100 seconds, while animals that received a scrambled, nonspecific antisense treatment generally took three times as long.

The timing of IGF-II blocking was crucial. Antisense injections within the first two days disrupted the memory, but injections at the four-day point had little effect on memory retention. Thus, there is a narrow window of time when the treatment can influence memory consolidation. “Only younger, more recent memories are sensitive to inhibitors,” Chen said.

If normal amounts of IGF-II help consolidate memory, maybe an extra dose would boost it. Chen and colleagues found that was indeed the case when they injected IGF-II into the hippocampi of rats right after the training session. IGF-II-treated animals took as long as 800 seconds to brave the dark space, and they remembered not to go there for as long as three weeks.

IGF-II represents a new kind of memory molecule that deserves a close look, said Howard Eichenbaum of Boston University in Massachusetts, who was not involved with the study. However, he cautioned that scientists have discovered many memory enhancers, but few work as well in people as a good old cup of Joe or a simple candy bar.

IGF-II has a marketing advantage in that it's a regular human protein. “For those in search of a memory boost, natural enhancers would be an attractive alternative to purely pharmacological agents,” wrote Johannes Gräff and Li-Huei Tsai of the Massachusetts Institute of Technology in a commentary that accompanied the paper in Nature. They also cautioned, of course, that side effects are possible. For example, elevated IGF-II has been linked to some cancers (Feinberg, 2007).

If IGF-II can safely boost memories, Chen speculated on a scenario in which a person with cognitive impairment might take a dose after learning something important, such as his or her home address or the identity of a grandchild. IGF-II readily crosses the blood-brain barrier, he said, so injecting it into the hippocampus would not be necessary. Of course, Eigenbaum noted, such a treatment would only work if the synapses and neurons and connections capable of cementing the memories were still present; the treatment might not work as well in people who are already severely demented. The researchers plan to test whether IGF-II is involved in extinction of memories—for example, when a rat discovers that the dark area is no longer shocking and has to adjust its synapses accordingly. Such a treatment might be useful for PTSD—a person would revisit the memory to loosen it, and then modify it with a boost from IGF-II.

But Take Plenty of Naps and Aromatherapy
At the University of Lübeck, Germany, researchers led by first author Susanne Diekelmann and senior authors Jan Born and Björn Rasch investigated a gentler way to play with memories, simply by dozing. Specifically, they were interested in the loosening of a memory that happens every time it is accessed, and predicted that memories would be as labile after a short nap as they would during waking. They used odor to reactivate memories in people who’d done either.

The researchers trained people on a card-matching task, rather like the children’s game “Concentration,” in which players have to find matching pairs in an array of face-down cards. Throughout the learning session, the subjects smelled a chemical with an unfamiliar, slightly unpleasant odor. Thus, they came to associate the odor with the card-learning task.

After learning the cards, some subjects sat quietly awake for 40 minutes. The researchers also re-exposed some of them to the smell. This led them to revisit the card-pairing memory, thus keeping it loose and labile. After their down time, the participants had to learn a whole new pattern of cards. This dislodged the already loosened memory of the first task. Because of previous research on re-labilization and interfering tasks, Diekelmann and colleagues expected that this experiment would cause waking subjects to forget the first card pairs (Forcato et al., 2007). Sure enough, when asked to remember the initial card pattern, the people exposed to the odor only got about 40 percent right, compared to 60 percent among control waking participants who did not receive the odor during their quiet sit, and thus did not revisit and render labile the first pattern. The odor reminder made the memory worse, presumably because it activated it and made it sensitive to the interfering task.

But what if the subjects took a nap while getting a whiff of the smelly chemical, instead of sitting awake? The researchers previously showed that a full night’s sleep helps consolidate memories (see ARF related news story on Rasch et al., 2007), but they predicted that a shorter snooze would not. Specifically, the researchers hypothesized that, because the brain transfers recent memories from the hippocampus to the cortex during a brief period of slow-wave sleep, memories would be as labile during a nap as during waking.

However, the results suggested that subjects who napped while exposed to the odor and then learned the interfering task had better memories than those who stayed awake. The odor-exposed sleepers remembered approximately 85 percent of the first pairs, compared to 60 percent for subjects who napped odor-free, and 40 percent for those who got the aroma while awake. Thus, revisiting memories while sleeping enhances the same memories that sitting awake destabilizes. Perhaps, Diekelmann suggested, memories have traveled part of the way to the cortex after the slow-wave sleep period. That way, she speculated, new information in the hippocampus would not alter those memories that were already transferred to long-term storage.

For people with memory problems, the work suggests that linking a memory to a cue like odor or music might be useful. Using that same smell or sound during sleep could help consolidate the memory. However, Stickgold noted, scientists have no idea what specifically happens to sleep-based memory processing in Alzheimer’s, or other dementias, so it is uncertain if it would help in those cases.

The researchers only addressed slow-wave sleep in this study. Stickgold suspects that rapid eye movement (REM) sleep will be more labile to disruption because other parts of the brain are active, accessing and examining those memories. Diekelmann said she and her coauthors intend to address other sleep stages such as REM.

There is plenty more research to be done to fully understand how to keep or dislodge memories. If you want to keep these findings at your fingertips, maybe it is time for a quick nap—unless you happen to have a syringe full of IGF-II handy.—Amber Dance.

References:
Chen DY, Stern SA, Garcia-Osta A, Saunier-Rebori B, Pollonini G, Bambah-Mukku D, Blitzer RD, Alberini CM. A critical role for IGF-II in memory consolidation and enhancement. Nature. 2011 Jan 27;469:491-7. Abstract

Diekelmann S, Büchel C, Born J, Rasch B. Labile or stable: opposing consequences for memory when reactivated during waking and sleep. Nat Neurosci. 2011 Jan 23. Abstract

Gräff J, Tsai LH. A molecular memory booster. Nature. 2011 Jan 27;469:474-5. Abstract