Memories are made of molecular motors

Learning and memory are widely thought to involve long-term potentiation (LTP), a form of synaptic plasticity in which a neuron's response to the chemical signals it receives is enhanced. This leads to a strengthening of the neuronal circuit, so that the memory encoded in the circuit can persist for long periods of time.

One of the mechanisms by which this synaptic strengthening occurs is an increase in the density of receptors in the membrane of the neuron receiving the signals. This process, which involves trafficking of receptors within cells, is implicated not only in learning and memory but also in conditions such as chronic pain and drug addiction. 

In the current issue of Cell, a group of researchers led by Michael Ehlers of Duke University Medical Center report that receptor trafficking is mediated by a protein which is closely related to the molecule which generates the force needed for the contraction of muscle cells.

When LTP was discovered in the early 1970s, a debate ensued about whether the mechanisms underlying it were taking place in the presynaptic or postsynaptic cell (that is, the neuron sending the chemical signal or the neuron receiving it). One possibility was that enhanced synaptic transmission occurs because of changes in the presynaptic cell: an increase in the amount of neurotransmitter molecules released into the synapse would enhance the response of the postsynaptic cell.

However, research carried out in the past decade suggests that LTP primarily involves changes in the postsynaptic membrane. A key event in the induction of LTP is the activation of NMDA receptors in the postsynaptic membrane. This leads to activation of the enzyme CaMKII, which then triggers, among other things, the insertion of new AMPA receptors into the membrane of the postsynaptic cell. AMPA receptor insertion causes morphological changes in the neuron - it increases the surface area of the membrane, and consequently to an enlargement of dendritic spines, the mushroom-shaped protruberances at which much of the inter-neuronal signalling in the brain takes place, leading to strengthening of the synapse.  

More recently, it has been shown that membrane-bound structures called recycling endosomes, which are located at the base of the dendritic spines, contain reservoirs of inactive AMPA receptors which are rapidly delivered to the cell membrane in response to neuronal activity. The receptors are inserted into the membrane by a process called exocytosis; a part of the endosome membrane, which has AMPA receptors embedded in it, buds off and fuses with the cell membrane. (Hence the increase in the surface area of the cell membrane, which can be measured by an increase in membrane capacitance, or the capacity of the membrane to store electrical charge).

Until now, the mechanisms by which AMPA receptors are trafficked towards the membrane were poorly understood. However, a protein called myosin Vb had previously been implicated. Myosin is closely associated with another protein called actin, which forms microfilaments throughout the interior of neurons, including in the dendritic spines. Myosin is the prototypical molecular motor - it generates force by converting chemical energy into mechanical energy, and produces the cellular movements in processes such as muscle, by using that force to slide along actin microfilaments. Myosin Vb is enriched in the hippocampus, a region in the medial temporal lobe known to be critical for memory, and is therefore a prime candidate for a role in receptor trafficking.

Ehlers and his colleagues generated transgenic mice in which myosin Vb was tagged with green fluorescent protein in order to visualize the movements of the protein in response to neuronal activity. They first isolated neurons from the hippocampus, a brain structure known to be critical to memory formation, and  used antibody staining to localize the distribution and movements of myosin Vb and the recycling endosomes. Consistent with earlier studies, this showed that the endosomes were often located at the base of dendritic spines, and sometimes within the main body of the spines. They also found that the endosomes at the base of the spines, but not those in the spines, were colocalized with myosin Vb.  

Time-lapse microscopy was then used to monitor the movements of myosin Vb and the endosomes in response to NMDA receptor activation. In cultured hippocampal neurons, the protein was found in spines, but its distribution rarely overlapped with the endosomes, which remained at the base of the spines. When, however, LTP was induced by addition of glycine to the culture medium, which activates the NMDA receptors, there was a significant increase in the amount of myosin Vb associated with endosomes, but no change in the total level of the protein. Upon addition of the glycine, the fluorescent mysoin Vb molecules were seen to rapidly associate with endosomes at the base of the spines; both the protein and the endomsomes were then trafficked together into the head of the spine.

Further experiments showed that myosin Vb is required for the LTP-induced insertion of AMPA receptors by exocyotsis, and that this process is dependent upon NMDA receptor activation. Endosomes were labelled with a protein which could be used to monitor exocytic events. This protein is pH-sensitive, so fusion of the endosome with the cell membrane causes it to change colour, as it becomes exposed to the exterior of the cell, and so to a change in acidity, during exocytosis. In cells containing this protein, exocytosis was prevented by inhibiting myosin Vb synthesis by RNA interference, or by inhibiting NDMA receptor activation with the drug APV. Inhibiting myosin Vb synthesis was also found to prevent growth of dendritic spines in response to glycine application.

Finally, the researchers genetically dissected the myosin Vb molecule; by generating mutated forms of the protein they determined its mode of action. This showed that myosin Vb is highly sensitive to the small calcium currents which flow into the cell through the activated NMDA receptor. Calcium induces a conformational change in the structure of myosin Vb, during which the molecule increases in length. This enables it to bind to a small "adaptor" protein called Rab11-FIP2, whose other end is bound to the endosomes. Subsequently, myosin Vb binds to actin microfilaments, and trafficks the endosome into the body of the dendritic spine.


Wang, Z. et al. (2008). Myosin Vb Mobilizes Recycling Endosomes and AMPA Receptors for Postsynaptic Plasticity. Cell 135: 535-548 DOI: 10.1016/j.cell.2008.09.057

More like this

One of the mechanisms -- perhaps even the primary mechanism -- by which synapses in the brain are potentiated -- made more sensitive to activation -- is the insertion of more AMPA receptors (AMPA-R) into the synapse. AMPA-R are glutamate-activated, cation (Na and Ca) channels that are really the…
[Introduction|Part 2] It is well established that synaptic strengthening involves the recruitment of AMPARs to the postsynaptic membrane. However, the subunit composition of the receptors has not been investigated closely, so the studies discussed here allow for refinement of this basic model,…
This is a repost from my old blog, about a famous scientist, Dr. George Augustine, who came to UM to give a talk about LTP and LTD. Tho occasion was "NeuroDay," a seminar series where us Neurokids get to invite our favorite scientists to come talk to us about brain stuff. :) The third and last…
In my second coursework essay, I discuss a number of recent studies which demonstrate that synaptic strengthening in different regions of the mammalian brain requires the incorporation of Ca2+-permeable GluR1-lacking AMPA receptors into the postsynaptic membrane of active or newly-potentiated…

Far, far afield from my area of expertise, but fascinating and well written. Science rocks, keep posting!

Great stuff; thanks for the excellent post.

By Greg Peterson (not verified) on 04 Nov 2008 #permalink

Why do they choose glycine?

The NMDA receptor has a binding site for glycine. But I'm not sure why they used glycine instead of glutamate to activate the receptors.

Glycine is required in order to activate NMDA.
I mean the glycine produces NMDAR-dependent LTP, what about other kinds of chemically induced NMDA-dependent/independent LTP?