New Link Uncovered in Nerve Cell Mechanism Thought to Power Learning and Memory

Cold Spring Harbor, NY - Researchers at Cold Spring Harbor Laboratory have uncovered a significant new link in the molecular chain of events thought to underlie learning and memory. By using a novel electrophysiological method for measuring synaptic activity, Roberto Malinow and his colleagues have demonstrated that strengthening of nerve cell connections in the brain-believed to occur during learning and memory consolidation-can be largely explained by the movement of proteins called AMPA receptors into synapses. The scientists describe the role of two other proteins, one (a protein kinase) known previously to be critical to AMPA receptor-mediated synaptic strengthening, and another (the new link) whose existence is strongly implicated by the recent findings. The study is reported in today's issue of Science.

One seat of learning and memory in the brain is a region deep within the cortex called the hippocampus. Malinow and his colleagues study how neurons in the hippocampus of rats respond to electrical stimulation like that which occurs during learning. In particular, they explore the mechanisms involved in long-term potentiation (LTP), the synaptic strengthening believed to power learning and memory consolidation.

In one of two studies they published in Science last year, Malinow and his colleagues reported how the distribution of AMPA receptors within dendrites changes upon induction of LTP (dendrites are the branch-like arms of neurons that receive synaptic inputs.) This study demonstrated that during LTP induction, some AMPA receptors move rapidly from interior locations within dendrites into regions that contain synapses called dendritic spines. The increased presence of AMPA receptors on the cell surface at synapses enables neurons to respond more strongly to the key neurotransmitter glutamate. Thus, the scientists concluded that the movement of AMPA receptors into spines could be an important mechanism that contributes to the synaptic strengthening believed to underlie learning and memory. But additional evidence to support this view was needed.

"What we really wanted to know next was whether the AMPA receptors we observed moving into spines are actually incorporated into synapses and participate in synaptic transmission," says Malinow. "And we wanted to learn more about the mechanism that mediates this trafficking of AMPA receptors within dendrites."

The new study provides these answers. Malinow and his colleagues took advantage of the fact that most of the neurons they study contain AMPA receptors with both "GluR1" and "GluR2" subunits. Receptors with these subunits allow sodium (Na+) ions to flow into cells and potassium (K+) ions to flow out of cells. To follow with precision the fate of AMPA receptor subunits, the scientists overexpressed just the GluR1 subunit. AMPA receptors that contain GluR1 but lack GluR2 only allow the inward flow of sodium ions. This property provided Malinow and his colleagues the sensitive electrophysiological assay they needed to determine whether GluR1 AMPA receptor subunits are newly-incorporated into synapses following LTP induction, and if so, whether they are functional (i.e. able to increase synaptic transmission.) The answer on both scores was yes.

In these experiments, LTP was mimicked by co-expressing with GluR1 a constitutively active form of a protein kinase (CaMKII) known to be sufficient to elicit the synaptic strengthening that is the hallmark of LTP. To investigate the mechanistic relationship between AMPA receptors and CaMKII, Malinow and his colleagues mutated a site within GluR1 that had been shown by others to be phosphorylated by CaMKII during LTP. This mutation did not appear to affect the regulated delivery of GluR1 to synapses.

Instead, Malinow and his colleagues identified a different site within GluR1 that they believe is required for the CaMKII-mediated delivery of GluR1 to synapses during LTP. This site attracted the attention of the scientists because it resembles motifs found in other membrane proteins that control where those proteins are localized within cells. The motif enables those membrane proteins to bind to a class of so-called PDZ proteins which act to target the membrane proteins to the correct location in the cell. Malinow and his colleagues showed that mutations within the putative PDZ-interaction motif of GluR1 blocked its delivery to synapses following LTP induction. Therefore, a PDZ protein is likely to represent a new link in the regulated delivery of functional AMPA receptors to synapses during learning and memory consolidation.

The study was carried out at Cold Spring Harbor Laboratory by Yasunori Hayashi, Song-Hai Shi, José A. Esteban, Antonella Piccini, Jean-Christophe Poncer (currently at the Pasteur Institute, Paris), and Malinow. Cold Spring Harbor Laboratory is a private, non-profit basic research and educational institution with programs focusing on cancer, neuroscience, and plant biology. Its other areas of research expertise include molecular and cellular biology, genetics, structural biology, and bioinformatics. For more information, visit the Laboratory's website ( or call the Department of Public Affairs at 516-367-8455.