(LTP) underlies memory
formation in the hippocampus, according to two new studies, adding support to a contentious hypothesis. The research, appearing in this week's Science
, reports that learning
induces LTP, and blocking
LTP can erase memories already stored in the hippocampus.
That LTP underlies hippocampal learning is "an assumption that we've had for 30 years
," said Mark Bear
of The Picower Institute for Learning and Memory at MIT in Cambridge, Mass., senior author of one of the new papers. "It's almost to the point of embarrassing that it hadn't been demonstrated."
Both studies may help researchers stop arguing
about whether or not LTP is the mechanism of learning, said Michel Baudry
of the University of Southern California in Los Angeles, who was not involved in the research. "I think now, it is. I'm glad to see we are coming to this closure."
It's well-established that LTP -- a long-lasting increase in synaptic transmission -- can be induced
in hippocampal slices through high-frequency stimulation, and it's also been shown
that blocking LTP impairs spatial learning in animals, said Mayank Mehta
of Brown University in Providence, R.I., also not a co-author. But no one had been able to show directly that learning induces LTP or that blocking LTP can erase previous learning. "The evidence has been indirect for quite some time," Mehta said.
In the first Science
paper, researchers led by Jonathan Whitlock
at MIT trained rats to avoid one side of a chamber by giving them a mild food shock whenever they enter that side. The animals can learn this task within one training session, and the task induces gene expression changes in the hippocampus, according to the authors. After the rats learned this task, the researchers found an increase of a known marker of LTP
in the hippocampus: phosphorylation of a particular amino acid in glutamate receptor proteins. Upregulation of this marker appeared 30 minutes after the rats learned the avoidance task. When the researchers blocked glutamate receptors called NMDA (N-methyl-D-aspartate) receptors -- known
to be essential for both LTP and memory -- the increase in the LTP marker disappeared.
The scientists next measured synaptic transmission in hippocampal neurons in the rats before and after learning. By recording with a multielectrode array, they discovered that most sites showed no significant change in transmission after learning, but a few recording sites experienced significantly increased excitatory transmission, a sign of LTP.
Because only some sites revealed LTP, previous studies may have missed this effect by averaging over all recording areas, said Bear, senior author of the paper. "We had a lot of additional insight by recording simultaneously from multiple electrodes," he said.
It's a bit mysterious why the authors were able to detect biochemical markers of LTP, since they found increased synaptic transmission at only a small percentage of synapses, said Baudry. Such sparsely distributed changes "should not be detectable biochemically," he told The Scientist
. But the study "satisfies several of the criteria that people would say are necessary and sufficient to equate LTP and learning," he added.
In the second study, researchers led by Eva Pastalkova and Peter Serrano of SUNY Downstate Medical Center in Brooklyn investigated if they could erase an animal's hippocampal memory by blocking LTP. Serrano and others in Todd Sackton's lab at SUNY Downstate recently discovered
that an unusual kinase called PKMzeta, which is persistently expressed and found only in the brain, is necessary and sufficient for LTP maintenance.
The researchers trained rats in a shock-avoidance task and then injected some of the animals with a PKMzeta inhibitor. While control animals remembered which areas of their environment to avoid, inhibitor-injected animals explored the entire environment, appearing to have no memory of previous avoidance training. This effect was persistent: One week after being injected with the inhibitor, the rats still had no memory of the shock task. However, if the rats were retrained after the drug was eliminated from their systems, they could learn the task again.
"We haven't damaged the brain or even functionally impaired memory, other than wiping out previous long-term memory," Sacktor told The Scientist
. Even older memories could be wiped out: When the researchers waited 30 days after training before injecting the inhibitor, the rats' month-old memories vanished.
Sacktor's studies "push the envelope of long-term learning to not just hours but days," Mehta told The Scientist
. "It's an important contribution."
Melissa Lee Phillips
Links within this article
K. Heyman, "Neurophysiology: Dust clearing on the long-term potentiation debate," The Scientist
, May 23, 2005.
K. Lee, "New neurons are involved in memory formation," The Scientist
, March 19, 2001.
J.R. Whitlock et al., "Learning induces long-term potentiation in the hippocampus," Science
313:1093-1097, August 25, 2006.
E. Pastalkova et al., "Storage of spatial information by the
maintenance mechanism of LTP," Science
313:1141-1144, August 25, 2006.
T.V. Bliss, T. Lomo, "Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path," Journal of Physiology
, July 1973.
E. Russo, "Controversy surrounds memory mechanism," The Scientist
, March 1, 1999.
T.V. Bliss, G.L. Collingridge, "A synaptic model of memory: long-term potentiation in the hippocampus," Nature
, January 7, 1993.
A. Gruart et al., "Involvement of the CA3-CA1 synapse in the acquisition of associative learning in behaving mice," Journal of Neuroscience
, January 25, 2006.
H.K. Lee et al., "Regulation of distinct AMPA receptor phosphorylation sites during bidirectional synaptic plasticity," Nature
, June 22, 2000.
D. Pare et al., "Presynaptic induction and expression of NMDA-dependent LTP," Trends in Neurosciences
, August 2004.
D.S. Ling et al., "Protein kinase Mzeta is necessary and sufficient for LTP maintenance," Nature Neuroscience
, April 2002.