18 Oct 2001

The Where? Key Role Found for Entorhinal Cortex, Area of Earliest Neuron Death in Alzheimer's

It has become almost a truism in memory research that the seahorse-shaped brain area called the hippocampus is key to memory formation. It helps solidify short-term memories into more permanent ones. Recently, however, areas surrounding the hippocampus have received more attention, especially the entorhinal cortex (EC). For example, studies have shown that lesions including the EC cause more profound memory loss than lesions of the hippocampus only (Kapur and Brooks, Zola-Morgan et al.).

In the November Nature Neuroscience, published online last Monday, Frank Haist at Children's Hospital Research Center in La Jolla, California, and colleagues report that the hippocampus plays only a temporary role in consolidating a recently acquired memory, whereas the EC helps firm up those memories for up to twenty years. This is the first direct evidence from normal humans that the EC contributes to very remote memory recollection, said Haist. (The EC is a gateway into and out of the hippocampus; most of the perceptual input to the hippocampus and most of its output to the cortex flows through the EC.)

The researchers presented eight healthy, elderly volunteers with pictures of faces of famous actresses and politicians dating from the1990s back to the1940s, and then measured activity in various brain areas with fMRI as the volunteers recognized the images. The data showed that the hippocampus was activated when the volunteers recognized the most recent famous faces, for example Madeline Albright, but not upon recognition of people who became famous earlier, such as Mia Farrow. By contrast, the EC lit up with decreasing intensity when presented with pictures dating back to the1950s.

The scientists interpret their findings as suggesting that the EC is one of the brain areas that help strengthen an initial memory every time it is recalled for up to two decades until, eventually, the synaptic efficacy associated with that memory is sufficiently strong. From then on the memory is entrenched in cortical brain regions, and future recall of it no longer requires the EC.

This is intriguing because the EC is also the brain area where the earliest pathology and neuronal loss has been detected in incipient Alzheimer's at a time when the disease cannot yet be diagnosed. "Our paper extends some of the monkey and human work showing importance of the EC for memory," Haist added. "It may be valuable to look at what sort of neurobiological processes happen in other regions surrounding the hippocampus, for example the parahippocampal cortex and perirhinal cortex."-Gabrielle Strobel.

The How? Molecular Changes on Both Sides of Synapse Seen Soon After Learning Stimulus
What might be going on in those neurons as they consolidate a memory? A new study in tomorrow's Science might provide a glimpse of an answer. Irina Antonova, working with Robert Hawkins, Eric Kandel, and others at Columbia University in New York, reports that she sees rapid changes in the distribution of proteins on both sides of the synapse in response to a stimulus causing long-term potentiation (LTP). LTP, of course, is an NMDA receptor-dependent mechanism thought to underlie learning and memory formation.

Previous research had found that the number of synaptic clusters, or puncta, containing AMPA glutamate receptors increases on the postsynaptic membrane after hippocampal synapses are stimulated. However, whatever changes occur on the presynaptic side remained less clear (Luscher C. et al.).

Antonova et al, cultured hippocampal neurons, applied the LTP-inducing stimulus glutamate for one minute, and looked for protein changes 10 minutes later. They found an increase in number not only of the glutamate receptor puncta on the postsynaptic side, but also of puncta containing the presynaptic marker synaptophysin. What's more, both types of puncta co-localized opposite each other at the same spots across the synaptic membrane, suggesting they were forming functional glutamatergic synapses. The same was true for another set of presynaptic-postsynaptic protein markers.

The researchers found essentially the same result in two ways; first by labeling pre-and postsynaptic proteins with antibodies, and second by expressing green fluorescent protein fused to synaptophysin in cultured hippocampal neurons. The changes lasted for at least 30 minutes.

The scientists also studied where those new, LTP-induced puncta might come from. New protein synthesis turned out to be unnecessary, though it does occur in late-phase LTP. Instead, new puncta appeared to emerge as previously dispersed proteins aggregated into discrete dots. The puncta assemble and dissemble in a state of equilibrium, and glutamate shifts the balance toward greater aggregation, the authors write. This process involves the actin cytoskeleton, though the study could not visualize the precise mechanism at work. The aggregation of puncta might underlie a conversion of "silent" presynaptic terminals to functional ones, the authors write.

In summary, the study shows that even in the first minutes of LTP, long-lasting plasticity involves coordinated protein changes on either side of the synaptic cleft.—Gabrielle Strobel

Comments

1. • Mark Gluck
     
   • Posted: 18 Oct 2001

   Comment by Mark Gluck, Kin Ho Chan, M. Todd Allen
   
   Although these results are
   intriguing, the authors' interpretation of them seems to go beyond what
   the data actually might support. Briefly, the pattern of results they
   report is that as you look at activation of older memories, both the
   hippocampal and entorhinal cortex (EC) activation decreases. The
   authors claim this supports the idea that the EC is responsible for
   consolidating older memories. But if this were true, might not the
   older memories show higher EC activations, not lower, since they have
   been strengthened over the years.

   More broadly, it appears the experiment was designed to test the
   consolidation hypothesis against the multiple trace hypothesis. The
   data should be examined with this in mind. The data in this paper show
   that only 1990s faces significantly activated hippocampus. Faces from
   1940s to 1980s were not significantly over baseline activation. One
   other point that makes interpretations difficult is that activations for
   1990s and the rest were not statistically significant. EC activations,
   on the other hand, showed a linear decreasing function from 1990s to
   1970s (then the 3 data points from 70s, 60s, and 50s plateaued with
   another drop below baseline for faces from 1940s).

   The authors concluded that MTL activation (at least EC) was time-limited
   and not time-invariant, thus supporting consolidation but not multiple
   trace. The fact that the EC but not hippocampus showed a linear function
   suggested that hippocampus plays a role in the shorter time frame ( 10 years). But the data
   might also be interpreted as EC playing a longer term and time-dependent
   role in retrieval while hippocampus plays a short term one. If encoding
   (and consolidation) and retrieval are separate processes, it is not
   clear how this data provide clear information about consolidation.

   View all comments by Mark Gluck

Make a Comment

To make a comment you must login or register.

References

No Available References

Further Reading

Papers

1. Kapur N, Brooks DJ. Temporally-specific retrograde amnesia in two cases of discrete bilateral hippocampal pathology. Hippocampus. 1999;9(3):247-54. PubMed.
2. Zola-Morgan S, Squire LR, Amaral DG, Suzuki WA. Lesions of perirhinal and parahippocampal cortex that spare the amygdala and hippocampal formation produce severe memory impairment. J Neurosci. 1989 Dec;9(12):4355-70. PubMed.
3. Lüscher C, Nicoll RA, Malenka RC, Muller D. Synaptic plasticity and dynamic modulation of the postsynaptic membrane. Nat Neurosci. 2000 Jun;3(6):545-50. PubMed.