Remembrance Of Things Past

Convincing new evidence backs the premise that the brain uses long-term potentiation (LTP) to store memories over the long haul.

The LTP hypothesis was first published in 1973 by Terje Lømo of the University of Oslo, in Norway, and Tim V. P. Bliss of the U.K.'s National Institute for Medical Research. The pair proposed that a memory is recorded through a lasting increase in the responsiveness of synapses that connect neurons in the brain.

Although this model has been widely accepted, researchers have lacked two key pieces of evidence, according to Bliss. First, scientists needed to demonstrate that LTP occurs in an animal that is learning a task. Second, they needed to show that undoing LTP in an animal that has learned a task erases the memory.

Now, Bliss says, two papers in Science have supplied these missing links. One paper reports on recordings of electrical activity in the brain of rats before and after they learned how to avoid an electric shock in an enclosure. The recordings show that LTP indeed occurs during this learning task, according to Massachusetts Institute of Technology neuroscientist Mark F. Bear and colleagues (2006, 313, 1093). The work extends related findings reported by a Spanish team earlier this year (J. Neurosci. 2006, 26, 1077), Bliss notes.

The second paper shows that reversing LTP after rats acquire a memory wipes out the memory. This type of experiment hadn't been possible previously because no drug was available to undo LTP, Bliss says. But in 2002, neuroscientist Todd C. Sacktor of State University of New York Downstate Medical Center, in Brooklyn, found that protein kinase M zeta (PKM) is required to maintain LTP. Using brain slices, he also determined that a synthetic peptide dubbed ZIP can erase LTP by blocking PKM activity.

In the current work, Sacktor, SUNY Downstate neuroscientist André A. Fenton, and coworkers injected ZIP into the brains of rats after the animals had learned a shock-avoidance technique. The rats forgot the new skill, but once the peptide was eliminated, they could be retrained. The results indicate that ZIP had not damaged their brains (Science 2006, 313, 1141).

Bliss says he is delighted with the "dramatic" new findings. "I'm retiring in six weeks," he adds, "and it's good to be bowing out at a time of striking new support for the LTP and memory hypothesis."

The work could have relevance for suppressing memories in people with conditions like post-traumatic stress disorder (PTSD). "There's a great deal of interest in chemical ways of manipulating information storage" in the brain, Bliss says.

LTP inhibitors could also be designed to treat conditions such as "phantom" pain from a missing limb. Like PTSD, this condition is attributed to excessive excitation of synapses by an LTP-like process. ZIP, Sacktor notes, "is the first agent that can erase that process without disturbing the ability of the synapses to store new information."