Director’s Blog: How Does Memory Work? The Plot Thickens
By Thomas Insel on
NIMH-funded researchers recently reported that a little-known growth factor boosted the strength and staying power of a fear memory. It’s the latest in a flurry of discoveries that are revising the script for what we might think about as a play in 5 acts: “How Memory Works.” Act 1 would be perception and encoding – when the information first comes in – and Act 5 would be the retrieval and expression of the stored information. There is much ferment about Acts 2, 3 and 4. What’s happening as that initial trace gets processed in the brain? Where does it get stored? How does it get stored? What sustains it? Do different kinds of memory have different molecular and cellular mechanisms?
How this story plays out has implications not only for our ability to stem aging-related memory loss and dementia, but also for mental disorders, such as PTSD, in which the challenge is to help patients selectively manage disabling emotional memories. Or for schizophrenia, which impairs the ability to organize memory. Or for depression, in which memory becomes selective for negative events.
To fathom the workings of such processes, researchers have been drilling-down to the molecules, cells and circuits underlying specific forms of memories – e.g., for places, faces, fears, and events – and unearthing some surprising characters.
Enter the newest actor in the drama of fear memory: insulin-like growth factor II (IGF-II). When injected into a rat’s memory circuit, this heretofore obscure player helped the animal learn to avoid a place where it had previously experienced a mild shock.1 Notably, it only worked if given during time windows when memories become temporarily fragile and changeable – soon after learning, when memories are first consolidated; or during retrieval, when they are reconsolidated. Exactly how IGF-II works is not clear, but the study, by Cristina Alberini and colleagues, suggests that this growth factor promotes protein synthesis and structural changes at connections between brain cells (synapses) during these limited windows of opportunity.
This new report builds on recent work on reconsolidation.2 The concept is simple: memories are not fixed, they are periodically retrieved, and modified each time they are retrieved. This process of strengthening a memory by retrieval is called reconsolidation. One profound implication of this concept is that what you recall is not only a reflection of what you first learned, but also a product of each time you have recalled the original information. Another implication is that the periods of reconsolidation are critical opportunities for modifying memories. In the past two years, various NIMH-funded investigators have shown that conditioned fear memories in humans (and rodents) can be blocked by interfering with reconsolidation. 3,4
Other studies have mapped fear learning to the specific synapses involved in the memory. In genetically engineered mice, Mark Mayford and his colleagues showed that the same synapses are activated by learning and retrieval.5 Adding to our understanding of how these molecular events can extend the staying power of memories are studies by Todd Sacktor and colleagues revealing how memories can be either erased or enhanced by tweaking expression of an enzyme in the cortex even long after the memory is formed. For example, even weeks after rats learned to associate a nauseating taste with saccharin and shunned it, their sweet tooth returned within a couple of hours after chemical blockade of a protein kinase, PKMzeta.6
One of the most exciting areas of memory research has been epigenetics. Learning, by definition, involves an enduring change in behavior. What molecular mechanisms underlie long-term changes in behavior? One involves the modifications of proteins on DNA (chromatin) that alter the way DNA is read. These long-term changes in the proteins binding to DNA are called epigenetic changes because they don’t change the text of the genome but they change its expression.7 Until recently, epigenetic changes were thought to be limited to early development when thousands of genes are turned on or off. Studies initially from David Sweatt’s lab at the University of Alabama at Birmingham demonstrated epigenetic changes in adulthood associated with the formation of a fear memory.8 Are these epigenetic modifications the molecular basis for long-term memories? Much evidence supports this idea, but the big question in front of us is whether we can target these epigenetic changes as a useful treatment.
While these threads inject new twists into the learning and memory plot, it’s still too early to predict if they may lead to improved treatments for the memory components of mental disorders. But the current excitement in this field is already re-writing the narrative of how memories are built from the molecular to the behavioral level.9
1A critical role for IGF-II in memory consolidation and enhancement. Chen DY, Stern SA, Garcia-Osta A, Saunier-Rebori B, Pollonini G, Bambah-Mukku D, Blitzer RD, Alberini CM. Nature. 2011 Jan 27;469(7331):491-7.PMID: 21270887
2Preventing the return of fear in humans using reconsolidation update mechanisms. Schiller D, Monfils MH, Raio CM, Johnson DC, Ledoux JE, Phelps EA. Nature. 2010 Jan 7;463(7277):49-53. Epub 2009 Dec 9.PMID: 20010606
3Induction of fear extinction with hippocampal-infralimbic BDNF. Peters J, Dieppa-Perea LM, Melendez LM, Quirk GJ. Science. 2010 Jun 4;328(5983):1288-90.PMID: 20522777
4Extinction-reconsolidation boundaries: key to persistent attenuation of fear memories. Monfils MH, Cowansage KK, Klann E, LeDoux JE. Science. 2009 May 15;324(5929):951-5. Epub 2009 Apr 2.PMID: 19342552
5Spine-type-specific recruitment of newly synthesized AMPA receptors with learning. Matsuo N, Reijmers L, Mayford M. Science. 2008 Feb 22;319(5866):1104-7.PMID: 18292343
6Rapid erasure of long-term memory associations in the cortex by an inhibitor of PKM zeta. Shema R, Sacktor TC, Dudai Y. Science. 2007 Aug 17;317(5840):951-3.PMID: 17702943
7HDAC3 is a critical negative regulator of long-term memory formation. McQuown SC, Barrett RM, Matheos DP, Post RJ, Rogge GA, Alenghat T, Mullican SE, Jones S, Rusche JR, Lazar MA, Wood MA. J Neurosci. 2011 Jan 12;31(2):764-74.PMID: 21228185
8The IkappaB kinase regulates chromatin structure during reconsolidation of conditioned fear memories. Lubin FD, Sweatt JD. Neuron. 2007 Sep 20;55(6):942-57.PMID: 17880897
9Molecular and cellular approaches to memory allocation in neural circuits. Silva AJ, Zhou Y, Rogerson T, Shobe J, Balaji J. Science. 2009 Oct 16;326(5951):391-5. Review.PMID: 19833959