Other news this week involves researchers using our furry friends to understand the fundamentals of memory formation. One study, from the lab of Eric Klann at New York University in New York, finds that FKBP12, a receptor for immunosuppressive drugs, modulates memory and behavior in mice. In their paper in the December 11 Neuron, Klann and coworkers show that FKBP12 affects memory by regulating mTOR, a molecule that controls protein synthesis and autophagy, a pathway of interest to Alzheimer researchers. Finally, a study reported in the December 19 Science from Edvard Moser and colleagues at the Norwegian University of Science and Technology in Trondheim, reports on a new class of place neurons that allow rats to learn and remember the boundaries of their environment, an essential skill for navigation.

Forgetful Flies
The γ-secretase substrate Notch is important in development but has also been implicated in synaptic plasticity, learning, and memory in adult mice (see ARF related news story and ARF news story). The territory between Notch and memory was uncharted, however, until first author Motomi Matsuno started studying a fly mutant that lacked the ability to make long-term memories. The reason, Matsuno found, was that the flies had a mutation in the klingon gene, which encodes a cell adhesion molecule obviously named by a Star Trek fan. The klingon mutants had normal learning and short-term memory, but had defective long-term memory (LTM) measured 1 day after odor-aversion training.

Because klingon mutants had a similar phenotype to Notch mutant flies, the investigators looked for a possible interaction between the two proteins. They found that the Klg protein was required for LTM. Expression of the protein increased after flies were trained to avoid an unpleasant odor, and blocking that induction with RNAi also blocked memory formation. Klingon expression appeared to be regulated by Notch. Flies expressing a dominant-negative Notch mutant show lower long-term memory induction, and, the researchers found, displayed less Klg protein induction after training. Flies overexpressing WT notch had both higher induction of Klg and better long-term memory. The increase in Klg protein required the processing of Notch and production of its intracellular domain fragment. Finally, the researchers showed that Notch-dependent memory formation required Klg: Flies that lacked Klg did not experience Notch-dependent long-term memory formation. They concluded that regulation of klingon is an essential step for Notch-dependent LTM formation.

The need for Notch processing means that proper long-term memory may require presenilin-dependent γ-secretase activity. This might raise another obstacle on the road to the use of γ-secretase inhibitors aimed at shutting down Abeta production. With the advent of inhibitors that apparently prevent APP processing while allowing Notch cleavage to proceed, these worries may be lessened (see ARF related news story). On the flip side, presenilin mutations associated with familial AD not only alter processing of amyloid precursor protein, but can also severely impair Notch processing. That has led to the idea that loss of Notch signaling could contribute to memory impairment in AD (reviewed in Costa et al., 2005).

How Notch regulates klingon is not clear. A direct effect on gene transcription is unlikely, since Klg mRNA levels were unchanged by the training protocol. Protein synthesis is a key step in consolidation of memories, and the authors speculate that Notch intracellular domain induces other factors that regulate protein synthesis or turnover of Klg.

Mice That Never Forget?
Protein synthesis by local translation of mRNAs is regulated by the mammalian target of rapamycin (mTOR), a kinase that controls translation initiation factors. mTOR action is inhibited by the immunosuppressive drugs FK-506 and rapamycin via the small binding protein FKBP12. A recent study showed that the mTOR pathway is involved in memory formation. In that work, rapamycin restored normal long-term potentiation, memory and behavior to mice that had elevated mTOR activity due a mutation in the tuberous sclerosis gene (see ARF related news story).

The new work from Klann and colleagues suggests that FKBP12 by itself puts a physiological damper on mTOR activity. First author Charles Hoeffer found that deleting the FKBP12 gene in the forebrain and hippocampus caused an elevation in mTOR signaling, as evidenced by increased phosphorylation of mTOR and downstream kinases, and enhanced interaction with a partner protein, Raptor. Along with that, the animals showed enhanced LTP and memory. The mice also displayed an increase in repetitive behaviors and perseveration, which suggests they have trouble forgetting previously acquired information in order to learn something new.

The work suggests a physiological role in memory formation for FKBP12. It had not been previously reported that FKBP12 binds to or regulates Raptor or mTOR in absence of rapamycin, Klann says. However, he said, he and his postdoc “felt it was unlikely that FKBP12 evolved to bind things like rapamycin, and that’s why we started the project.”

Interestingly, the kind of perseverative behavior seen in FKBP12-lacking mice is quite common in AD, but it is not clear whether alterations in mTOR, or protein synthesis, are involved. The Akt-GSK3 kinase cascade, altered in AD (see Griffin et al., 2005 and ARF related news story), helps control mTOR activity, but the few published studies that have looked at mTOR in AD have come up with conflicting results, Klann says, “Given the fundamental role that new protein synthesis plays in memory, it would seem surprising to me if some of the translation control pathways are not altered in AD. That’s something we’re interested in and are working on right now,” he said.

The kinase has other downstream effects, too. Besides protein syntheses, mTOR regulates autophagy, which is reduced in AD brain (see ARF related news story). In Huntington disease, inhibiting mTOR with rapamycin boosts autophagy, and represents a possible therapeutic approach (see ARF related news story). To sort out the role of mTOR in neurodegeneration, Klann says, “It will be important for people to really pinpoint which of the downstream effectors of mTOR are impacted in their particular disease in order to make more specific attempts to interfere with those pathways. We need a good systematic examination of the pathway.”

Rats on the Edge
For navigational purposes, our brains use specialized neurons in the hippocampus and surrounding cortex that fire to signal place, direction and movement. This neuronal network helps us to learn and remember signposts in our surroundings. Navigation is affected early on in AD, and in an AD mouse model, a loss of spatial memory and navigation was tied to loss of function of hippocampal place cells (see ARF related news story).

Place cells work together with head direction cells and grid cells to help animals locate themselves in space. Now, researchers have described a new group of cells that fire when an animal gets close to the edge of its environment. By recording neuronal activity in rats roaming in cages of different sizes and shapes, first author Trygve Solstad and colleagues detected what they call border cells scattered throughout the entorhinal cortex and adjacent parasubiculum where the grid cells and head direction cells are also found. The firing of the border cells helps the brain define the perimeter of the environment and provide a frame of reference for other places inside it. Like hippocampal place cells, disruption of these other navigational neurons might lead to the wandering problems associated with AD.—Pat McCaffrey.

References:
Matsuno M, Horiuchi J, Tully T, Saitoe M. The Drosophila cell adhesion molecule klingon is required for long-term memory formation and is regulated by Notch. Proc Natl Acad Sci U S A. Epub 2008 December 15.

Hoeffer CA, Tang W, Wong H, Santillan A, Patterson RJ, Martinez LA, Tejada-Simon MV, Paylor R, Hamilton SL, Klann E. Removal of FKBP12 Enhances mTOR-Raptor Interactions, LTP, Memory, and Perseverative/Repetitive Behavior. Neuron. 2008 Dec 10;60(5):832-45. Abstract

Solstad T, Boccara CN, Kropff E, Moser M, Moser EI. Representation of geometric borders in the entorhinal cortex. Science. 2008 Dec 19; 322:1865-1868. Abstract