ReviewLTP mechanisms: from silence to four-lane traffic
Introduction
There has been much interest in delineating the mechanisms underlying long-term potentiation (LTP). A detailed understanding of LTP may allow us to determine its roles in learning and memory and developmental plasticity. Mechanistic studies over the last three decades have produced contradictory models regarding the synaptic sites that undergo modification during LTP. However, work over the last few years has led to a model that includes delivery of receptors to functional ‘silent’ synapses. This model can account for most results in this field. Here, we review the evidence supporting this model and extend the model to include more recent results on receptor trafficking.
Section snippets
Why study long-term potentiation?
For almost 30 years there has been considerable effort by a number of laboratories to determine if the modification underlying long-term potentiation (LTP) takes place pre- or postsynaptically (reviewed in 1, 2, 3). This issue is important for several reasons. First, once the site of the modification that is responsible for LTP is found, it should be easier to identify the relevant cellular and molecular machinery. Such identification could facilitate determining the relationship between LTP
Mechanistic studies
The initial steps that trigger LTP have long been agreed on. Evoking synaptic transmission at a low frequency activates primarily AMPA-type glutamate receptors and leads to a small rise in intracellular calcium concentration. Intense synaptic activity that triggers LTP activates NMDA receptors and produces a significant rise in the postsynaptic calcium concentration. Events downstream of this are progressively less well understood. The activation of the calcium/calmodulin-dependent protein
Silent synapses: biophysical basis
A postsynaptic mechanism for LTP was proposed a decade ago, based on two observations: first, LTP selectively increases the postsynaptic current that is attributable to AMPA receptor channels 6, 7; second, LTP can increase the sensitivity to exogenously delivered AMPA [8]. This view was seriously challenged a few years later by studies showing that the frequency of failures in synaptic transmission (a classic measure of presynaptic function) changes during LTP 9, 10, 11, 12, 13, 14, 15. This
Silent synapses: role in LTP
The discovery and general acceptance of silent synapses has had two important consequences for LTP research. The first of these consequences is that their presence provides critical evidence for a purely postsynaptic mechanism for LTP that can explain a wide range of previous physiological observations 3, 36. In this model (see Figure 1), LTP results from the delivery of functional AMPA receptors to synapses from non-synaptic sites. These sites could be either intracellular, or on nearby
Focus on AMPA receptors
The second, and arguably more significant, fallout of the exciting discovery of silent synapses is that it has moved the field of LTP research beyond the general pre- versus post-synaptic dichotomy and focused attention on a much more specific molecular question: how are functional AMPA receptors added to synapses? Our understanding of the cellular and molecular mechanisms controlling the organization of these receptors is rapidly expanding. The AMPA receptor is a heteromeric complex made up of
AMPA receptor traffic
How do the molecular interactions between AMPA receptors and their associated proteins regulate synaptic plasticity and stability? Our current understanding is still murky, but a number of recent studies provide interesting clues. NSF is a protein involved in the presynaptic fusion machinery that is also present in postsynaptic sites and interacts with GluR2 54•, 55•, 57•, 61•. Expression of a peptide that interferes with the interaction of NSF and GluR2 causes a decrease in surface AMPA
Trafficking model for plasticity
The dichotomy between the properties of GluR1/4 and GluR2/3 subunits, with respect to carboxy-tails, interaction proteins, and apparent roles in basal transmission and plasticity, leads us to suggest a model for subunit-specific trafficking of AMPA receptors that can account for a wide range of experimental observations regarding LTP. The basic principle of our model is that two distinct regulatory mechanisms govern the local insertion and removal of AMPA receptors from the synapse: a
Conclusion
In the future, we expect that there will be much work examining in detail how signal transduction mechanisms interact with the machinery that regulates the trafficking of AMPA receptors and their associated proteins. This should lead to an elucidation of the roadways used to achieve ‘AMPA-fication’, and ultimately a satisfactory molecular understanding of LTP.
Acknowledgements
We thank S Shi, J Esteban, J Zhu, A Piccini, A Barria and F Kamenetz for helpful discussions.
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest
References (68)
- et al.
The site of expression of NMDA receptor-dependent LTP: new fuel for an old fire
Neuron
(1995) - et al.
Measuring the impact of probabilistic transmission on neuronal output
Neuron
(1993) - et al.
A persistent postsynaptic modification mediates long-term potentiation in the hippocampus
Neuron
(1988) - et al.
Direct measurement of quantal changes underlying long-term potentiation in CA1 hippocampus
Neuron
(1992) - et al.
Evidence for silent synapses: implications for the expression of LTP
Neuron
(1995) - et al.
Silent synapses during development of thalamocortical inputs
Neuron
(1997) - et al.
Cell type and pathway dependence of synaptic AMPA receptor number and variability in the hippocampus
Neuron
(1998) - et al.
Postsynaptically silent synapses in single neuron cultures
Neuron
(1998) - et al.
Extrasynaptic glutamate spillover in the hippocampus: evidence and implications
Trends Neurosci
(1998) Glutamate spillover mediates excitatory transmission in the rat olfactory bulb
Neuron
(1999)