Cells maintain a balance between LTP and LDP synapses.


Long-term potentiation (LTP), a phenomenon by which previously stimulated synapses become increasingly sensitive to stimulation such that the same level of presynaptic input induces a larger postsynaptic output, and long-term depression (LTD), which conversely reduces efficacy at such synapses, have been implicated in memory formation and storage. But computer models predict that without a balancing force of some kind, LTP and LTD will cause neural circuits to go haywire. Now, Sébastien Royer and Denis Paré have identified just such a force that can maintain balance in a network—and might imply that humans have a limited memory capacity.

“There have been lots of studies about what happens in LTP and LTD, but little attention has been paid to the synapses that are not stimulated when you induce LTP,” says Paré, despite the fact that each cell has thousands of synapses, only a few of which gain LTP or LTD.

To find out what was happening at the distant synapses, the authors induced LTP or LTD at a known location within a single neuron in brain slices from the amygdala of guinea pigs and then measured the response to stimuli at physically remote synapses in the same cell. They found that if LTP was induced in one locale, the distant synapses showed a slight depression in efficacy. Thus, there appeared to be an overall compensation for the synaptic activity in the cell, so that it remained constant.

When the group blocked calcium-induced calcium release in the neurons, the distant synapses failed to change their behavior in response to localized LTP or LTD induction. Therefore, a self-propagating wave of calcium released from internal stores appears to be the mechanism of communication between distant synapses.

These results imply that the formation of one memory affects others, since a single neural network is involved in many memories. “I guess one of the depressing implications of this is that there is a limit to how much you can accommodate,” says Paré. ▪


Royer and Paré. 2003. Nature. 422:518–522.