“Use it or lose it,” a common mantra in the Alzheimer disease community, stems from studies that suggest the brain is analogous to muscle—it needs to be exercised to stay in top shape. But if we take that analogy further, does it suggest that some brains are naturally brawny or that we could boost our intellectual game with a little mind medicine? Two recent papers hint at the possibilities.

Dominique de Quervain and Andreas Papassotiropoulos, at the University of Zurich, Switzerland, have tackled the natural endowment question. Writing in the March 14 issue of PNAS, available as an open access article online, the authors identify a cluster of genetic polymorphisms and haplotypes that boosts memory and hippocampal activity in humans. AD researchers will readily recognize the genes because they have been implicated in learning and memory, and many of them have been tied to pathological activity of amyloid-β. The genetic cluster includes adenylate cyclase (see ARF related news story), cAMP-dependent and protein kinases A and C (see ARF related news story), and NMDA and metabotrobic glutamate receptors (see ARF related news story).

De Quervain and Papassotiropoulos used a two-stage, multilocus search to find the genetic variations among carefully selected human genes. The 47 examined were all homologs of genes that have been implicated in learning and memory processes in animals. In the first stage of the search, 16 genetic variations were identified in 304 healthy humans who were tested in an episodic memory task. In the second stage, these variations were put through a more stringent evaluation to look for permutations among the 16 loci that might have a statistically significant impact on memory test scores. In this second stage, the authors identified one cluster of seven variations—five single nucleotide polymorphisms and two multilocus haplotypes—that significantly improved recall.

The study suggests that individual genetic variation partly explains why some of us have better memories than others. In support of this, the authors found that gene cluster genotypes correlate with functional MRI measurements of brain activity. To make this connection, they first developed what they call an IMAGS score (Individual Memory-Associated Genetic Score). This is based on how many of the seven variations are in a person’s genome, weighted by the effect sizes of those variations—a higher IMAGS number should equate to better episodic memory. Then they tested a second set of 32 volunteers in an episodic memory test while measuring brain activity. They found a direct and positive correlation between IMAGS scores in these individuals and activity recorded in the medial temporal lobe, particularly the hippocampus and parahippocampal gyrus, regions known to be involved in episodic memory.

The authors conclude that these seven genes form a cluster that has a strong impact on human memory performance, but they caution that other genes that did not emerge in this study are undoubtedly important for memory, too, both episodic and other forms. In fact, the authors found that the IMAGS score did not correlate with any neuropsychological measurement of memory, intelligence, or spatial cognition. It appears, therefore, that this particular cluster of variance may not have any influence on other brain areas, or that perhaps there are other, yet to be detected variances that negate the impact of this cluster in some brain regions. In this regard, de Quervain and Papassotiropoulos emphasize that their study does not enable them to draw conclusions about molecular interactions among the seven gene products.

Two other genes that have been implicated many times in learning and memory are CREB (cyclic AMP response element-binding protein) and ICER (inducible cAMP early repressor). In a second PNAS paper due out shortly, Matthew During and colleagues at the University of Auckland, New Zealand, report that doubling expression of CREB in the hippocampus improves learning and memory in elderly rats, while having no effect in the exact same animals at three months of age. The finding suggests that boosting CREB levels might help with age-related memory loss. Not surprisingly, these authors also found that increasing ICER, which cools the heels of the CREB pathway, has the opposite effect.

First author Alexandre Mouraviev and colleagues used recombinant adeno-associated virus (rAAV) to achieve long-term expression of the proteins in the hippocampi of rats. The author injected viral particles laced with genetically engineered CREB or ICER genes into the dorsal hippocampus of 8-week-old animals, then tested them at 3 and 15 months in a variety of behavioral tasks. The expression of genes driven by these vectors remains stable for at least 15 months, making them ideally suited for this type of experiment.

The authors found that at 3 months, animals injected with rAAV-CREB performed no better or worse than age-matched controls. However, at 15 months the rAAV-CREB rats performed considerably better in passive-avoidance tests and in the Barnes circular table test, where the animals must remember where to find an escape hole. In this latter test, the CREB-boosted animals found sanctuary faster than control littermates, while animals injected with the rAAV-ICER particles were slower than wild-type animals. Results for passive avoidance were similar, in that animals treated with CREB took longer than wild-type rats to enter a chamber where they had previously received a mild shock, while the ICER animals were quicker to go where all other animals feared to tread.

It turns out that these vectors boost CREB by about twofold, as judged by Western blotting of hippocampal tissue. The experiments suggest, therefore, that even this modest increase in CREB improves learning and memory in elderly animals while having no obvious effects on younger animals. Curiously though, even ICER had no effect on the 3-month-old animals, leading the authors to suggest that perhaps there are enough untransfected neurons to take up the slack when the animals are young. “However, in old animals, which likely experience global deficits in activities of CREB signaling, the expression of ICER interferes with long-term memory because the non-transduced cells are no longer able to compensate, consistent with the theory of a reduced memory reserve or capacity in aged animals,” write the authors. Reduced cognitive reserve has been touted as a reason why some people succumb to dementia earlier than others (see ARF related news story).

Whatever the reason for ICER's failure to affect the young animals, the authors suggest that strategies that increase CREB expression might turn out to be useful for treating neurodegenerative disorders of learning and memory.—Tom Fagan


  1. If G protein-linked glutamate receptors are an important component through neural cell excitation, and transpositioning of said receptors is an important finding (such as in cerebellar ataxia), then a cognitive behavioral approach in conjunction with appropriate drug treatment would offer some neural protection. Keep in mind that synapse-associated polyribosome complexes (SPRCs) discovered by Oswald Steward are also important variables here.

    Without proper dendritic spine synthesis, cognitive decline is guaranteed. We witness this severe deficiency in reasoning ability in persons with Down syndrome. It is now well known that stimulating thought and problem-solving over the long term greatly enhances long-term potentiation. It is also known that SPRCs are enhanced to facilitate this increase in neuronal communication through greater protein synthesis at the base of synapses in dendrites at both ionotropic signal-gated neurons and the metabotropic receptors mentioned in the previous paper.

    We see much similarity between Alzheimer disease risk and Down syndrome: Presenilin genes 1 and 2 handle important enzyme manufacture and still-unknown protein modification processes. What we do know is that problems in these genes can and do increase amyloid-β 42 and cause conformational changes at both synapses (dendritic spines, varicose axons). A host of intracellular cascade modifications occur at G protein-linked receptors. Glutamate overexcitation is seen, as well as damaged transport mediums, which do not permit enough glutamate transport to critical areas of cognition and motor movement such as cerebellum, frontal lobe, Broca's, etc.)

    Neuroscience Exploring the Brain, second edition.
    Mark F. Bear, Barry W, Connors, Michael A. Paradiso,
    2001 Lippincott Williams & Wilkins. pp.42-43; 44-45;154-158.

    (see )Ataxia—Rare Spectrin Mutations and a Rarer Pedigree

  2. Sampling and expression of amyloid-β (Aβ) peptides in vascular dementia (VD) brains and Alzheimer disease (AD) brains may explain the difference in learning and memory and not genetic differences, entirely. Since the effects of VD and AD on learning and memory are similar, but Aβ extractions differ between VD and AD, the so called IMAG scores will differ accordingly. However, this difference may be related to extraction variation as a function of VD.


    . Quantification of Alzheimer pathology in ageing and dementia: age-related accumulation of amyloid-beta(42) peptide in vascular dementia. Neuropathol Appl Neurobiol. 2006 Apr;32(2):103-18. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Filling in Signaling Steps Between Aβ and Long-Term Potentiation
  2. Sharpen Your Synapses with Rolipram!
  3. Amyloid-β Zaps Synapses by Downregulating Glutamate Receptors
  4. Add Mental Exercise to Potential AD Protection

External Citations

  1. open access article online

Further Reading

No Available Further Reading

Primary Papers

  1. . Identification of a genetic cluster influencing memory performance and hippocampal activity in humans. Proc Natl Acad Sci U S A. 2006 Mar 14;103(11):4270-4. PubMed.
  2. . Somatic gene transfer of cAMP response element-binding protein attenuates memory impairment in aging rats. Proc Natl Acad Sci U S A. 2006 Mar 21;103(12):4705-10. PubMed.