Downloaded from http://cshperspectives.cshlp.

org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent Long-Term Potentiation and Long-Term

Depression (LTP/LTD)
Christian Lüscher and Robert C. Malenka

Cold Spring Harb Perspect Biol 2012; doi: 10.1101/cshperspect.a005710 originally published online April
17, 2012

Subject Collection The Synapse

Studying Signal Transduction in Single Dendritic Synaptic Vesicle Endocytosis

Spines Yasunori Saheki and Pietro De Camilli
Ryohei Yasuda
Synaptic Vesicle Pools and Dynamics Short-Term Presynaptic Plasticity
AbdulRasheed A. Alabi and Richard W. Tsien Wade G. Regehr
Synapses and Memory Storage NMDA Receptor-Dependent Long-Term
Mark Mayford, Steven A. Siegelbaum and Eric R. Potentiation and Long-Term Depression
Kandel (LTP/LTD)
Christian Lüscher and Robert C. Malenka
Synapses and Alzheimer's Disease Ultrastructure of Synapses in the Mammalian
Morgan Sheng, Bernardo L. Sabatini and Thomas Brain
C. Südhof Kristen M. Harris and Richard J. Weinberg
Synaptic Cell Adhesion Calcium Signaling in Dendritic Spines
Markus Missler, Thomas C. Südhof and Thomas Michael J. Higley and Bernardo L. Sabatini
Biederer
Synaptic Dysfunction in Neurodevelopmental Synaptic Neurotransmitter-Gated Receptors
Disorders Associated with Autism and Intellectual Trevor G. Smart and Pierre Paoletti
Disabilities
Huda Y. Zoghbi and Mark F. Bear
The Postsynaptic Organization of Synapses Synaptic Vesicle Exocytosis
Morgan Sheng and Eunjoon Kim Thomas C. Südhof and Josep Rizo
Presynaptic LTP and LTD of Excitatory and Vesicular and Plasma Membrane Transporters for
Inhibitory Synapses Neurotransmitters
Pablo E. Castillo Randy D. Blakely and Robert H. Edwards

For additional articles in this collection, see http://cshperspectives.cshlp.org/cgi/collection/

Copyright © 2012 Cold Spring Harbor Laboratory Press; all rights reserved
Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent Long-Term

Potentiation and Long-Term Depression
(LTP/LTD)

Christian Lüscher1 and Robert C. Malenka2

1
Department of Basic Neurosciences and Clinic of Neurology, University of Geneva and Geneva University
Hospital, 1211 Geneva, Switzerland
2
Nancy Pritzker Laboratory, Department of Psychiatry and Behavioral Sciences, Stanford University School of
Medicine, Palo Alto, California 94305-5453
Correspondence: [email protected] and [email protected]

Long-term potentiation and long-term depression (LTP/LTD) can be elicited by activating

N-methyl-D-aspartate (NMDA)-type glutamate receptors, typically by the coincident activity
of pre- and postsynaptic neurons. The early phases of expression are mediated by a redistri-
bution of AMPA-type glutamate receptors: More receptors are added to potentiate the
synapse or receptors are removed to weaken synapses. With time, structural changes
become apparent, which in general require the synthesis of new proteins. The investigation
of the molecular and cellular mechanisms underlying these forms of synaptic plasticity has
received much attention, because NMDA receptor–dependent LTP and LTD may constitute
cellular substrates of learning and memory.

ong-term synaptic plasticity is a generic term duction), followed by an investigation of the

L that applies to a long-lasting experience-de-
pendent change in the efficacy of synaptic trans-
mechanism of expression (hours) and mainte-
nance (days). The best-characterized form of
mission. Here we will focus on N-methyl-D-as- NMDA receptor (NMDAR)-dependent LTP oc-
partate (NMDA) receptor– dependent synaptic curs between CA3 and CA1 pyramidal neurons
potentiation (LTP) and depression (LTD), two of the hippocampus (Fig. 1). Throughout the
forms of activity-dependent long-term changes chapter we will mostly refer to this specific form
in synaptic efficacy that have been extensively of LTP. At these CA3-CA1 Schaffer collateral
studied. Because both LTP and LTD are believed synapses, the loci of both induction and expres-
to represent cellular correlates of learning and sion are situated in the postsynaptic neuron.
memory, they have attracted considerable inter-
est. In this article we will focus on the molecular
AMPA-TYPE AND NMDA-TYPE
and cellular mechanisms associated with LTP
GLUTAMATE RECEPTORS
and LTD. As for other forms of long-term syn-
aptic plasticity, a characterization of LTP and After exocytotic release of the content of the
LTD involves describing the molecular mecha- synaptic vesicle from the presynaptic specializa-
nisms that are required to elicit the change (in- tion (see Castillo 2012), the neurotransmitter

Editors: Morgan Sheng, Bernardo Sabatini, and Thomas C. Südhof

Additional Perspectives on The Synapse available at www.cshperspectives.org
Copyright # 2012 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a005710
Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710

1
Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

C. Lüscher and R.C. Malenka

A B
CA1
2
2

Dentate 1 1
CA3

EPSP slope (% baseline) 200 1 2 3

150 3
100
1
50 2

0 1 Hz for 10 min 2 × 100 Hz

0 20 40 60 80 100
min

Figure 1. NMDAR-dependent LTD and LTP in the hippocampus. (A) Historical drawing by Ramon y Cajal
(1909) of the trisynaptic pathway in the hippocampus. LTP and LTD are induced by activation of NMDARs at
synapses between CA3 and CA1 pyramidal neurons (blue and red). In contrast, LTP at mossy fiber synapses onto
CA3 neurons (green on blue) is NMDAR-independent. (B) This electron microscopy image shows the densely
packed neuropil in the CA1 region of the hippocampus and highlights two asymmetric CA3-CA1 synapses. Note
the typical “bouton en passant” configuration of synapse 1 and the prominent spine in synapse 2. The post-
synaptic densities (PSDs) are visible. Scale bar, 200 nm. (Image kindly provided by Rafael Luján, Universitad de
Castilla-La Mancha.) (C) Bidirectional change in CA3-CA1 synaptic efficacy by LTD and LTP in the same
synapses monitored by extracellular field recordings in an acute slice preparation of the hippocampus. Note
the contrasting induction protocols (Data from C Lüscher, unpubl.).

rapidly diffuses across the synaptic cleft and Kþ. Activation of ionotropic glutamate recep-
reaches the postsynaptic membrane, where it tors leads to strong influx of Naþ and only a
binds to its respective receptors. The neuro- small efflux of Kþ, such that the net effect is
transmitter concentration in the synaptic cleft the depolarization of the postsynaptic neuron.
remains high for only a very brief period. Iono- The workhorse of glutamatergic transmis-
tropic receptors are direct ligand-gated ion sion is the AMPAR, which drives large and rapid
channels that respond rapidly to the brief pulse synaptic signaling (Fig. 2). There are four genes
of neurotransmitter released from the synaptic encoding AMPARs (GRIA1 to -4). Each AM-
vesicle. The evoked synaptic currents are typi- PAR is composed of four subunits, which can
cally believed to last only a few milliseconds. be a homomeric or heteromeric mixture of
The most common excitatory transmitter is glu- GluA1 to -4. Most AMPARs contain at least
tamate, whereas many inhibitory synapses use one subunit of GluA2. Following transcription
g-aminobutyric acid (GABA) for transmission. this subunit undergoes RNA editing, whereby
Here we will first discuss the ionotropic the RNA coding for a glutamine residue in the
members that are directly involved in synaptic ion channel pore-forming region is exchanged
plasticity of excitatory transmission, NMDA for the RNA codon for arginine. This process is
and AMPA-type glutamate receptors (NMDARs, an essential step in the quality control. Noned-
AMPARs) (Dingledine et al. 1999). Both are ion- ited GluA2 will be retained in the endoplasmic
otropic receptors that are permeable to Naþ and reticulum. The large arginine residue located in

2 Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent LTP/LTD

A AMPA receptor NMDA receptor

Na+ Ca2+ and Na+

0 mV
Mg2+
Glu Glu

–60 mV K+ K+

B
AMPA receptor NMDA receptor
I +GluA2 I

–60 mV –GluA2 –60 mV

V V

Figure 2. Major ionotropic glutamate receptors involved in LTD and LTP. (A) When glutamate binds to AMPA
receptors, many sodium ions flow into the cell while only some potassium ions leave the neuron, causing a net
depolarization of the membrane. NMDA receptors are also permeable for calcium but only if the magnesium is
expelled by a slight depolarization of the neuron. (B) The current –voltage (I –V ) relationship provides a
biophysical signature for the different receptors. AMPA receptors have a linear I –V relationship when they
contain the subunit GluA2, but are inward-rectifying (see text for definition) without GluA2. NMDA receptors
have a complex I –V curve because Mg blocks the pore at negative potentials.

the pore region of the channel limits the flow of GluA2. Such channels have a high conductance
Naþ and Kþ ions and prevents divalent ions for sodium and are even permeable for calcium.
from entering the cell. This design renders AM- Because endogenous polyamines, which are
PARs that contain GluA2 calcium-impermeable negatively charged, can also access a site close
(Liu and Zukin 2007). to the cytoplasmic mouth of the pore, the chan-
AMPARs carry inward currents at negative nels are inhibited at positive potentials. As a
potentials and outward currents at positive po- consequence, GluA2-lacking AMPARs have an
tentials, and the reversal potential is 0 mV. For inward-rectifying current – voltage relationship
receptors that contain GluA2, this relationship (i.e., they conduct current more easily into
is symmetrical, i.e., the current –voltage rela- the cell than out of the cell; Fig. 2). Such calci-
tionship is linear. In contrast, for receptors um-permeable AMPARs are expressed on many
that lack GluA2 (e.g., GluA1 homomeric or GABAergic neurons.
GluA1/3 heteromeric channels), the current – The current – voltage relationship of NM-
voltage relationship shows an interesting non- DARs is even more complex. At negative mem-
linear voltage-dependent interaction. GluA2- brane potentials close to the resting membrane
lacking AMPARs have a glutamine residue in potential, magnesium ions enter the pore of
the pore region instead of the arginine in the the NMDAR, blocking the passage for all other

Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710 3

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

C. Lüscher and R.C. Malenka

ions. Upon depolarization the magnesium is tens of minutes using mechanisms that allow
expelled from the pore, allowing sodium, potas- rapid removal from and insertion into the PSD.
sium, and, importantly, calcium ions to pass. At Although under baseline conditions the num-
positive potentials NMDARs then show maxi- ber of receptors remains stable, rapid redistri-
mal permeability (i.e., large outward currents bution leading to a change in the number of
can be observed under these circumstances; synaptic receptors is a widely used mechanism
Fig. 2). NMDARs also have much slower kinet- for synaptic plasticity (see below).
ics than AMPARs. Following release of gluta- Although the major role of ionotropic glu-
mate, NMDARs activate more slowly, having a tamate receptors is postsynaptic, they have also
peak conductance long after the AMPAR peak been found on presynaptic boutons, where they
conductance. NMDARs can also remain open may participate in the regulation of transmitter
for hundreds of milliseconds after presynaptic release. Such presynaptic receptors can be ac-
release of glutamate, much longer than the AM- tivated by strong release from the bouton on
PARs, which typically open for only a few mil- which they are located (homosynaptic modula-
liseconds. It is important to note that NMDARs tion) or by glutamate released by neighboring
conduct currents only when glutamate is bound synapses (heterosynaptic modulation). Wheth-
and the postsynaptic neuron is depolarized. In er this activation enhances or inhibits further
other words, both the pre- and postsynaptic release varies from synapse to synapse.
neurons need to be active to open NMDARs.
Through this mechanism NMDARs play the
INDUCTION OF LTP AND LTD
role of molecular coincidence detectors, which,
as we will see below, is essential for several forms LTP and LTD are induced by specific patterns
of synaptic plasticity. of activity (Malenka 1994). For LTP induction
Ionotropic glutamate receptors do not work both pre- and postsynaptic neurons need to
in isolation. They interact with many proteins be active at the same time because the postsyn-
of the postsynaptic density (PSD) (Elias and aptic neuron must be depolarized when gluta-
Nicoll 2007). Some of these proteins modulate mate is released from the presynaptic bouton to
glutamate receptor function, whereas others fully relieve the Mg2þ block of NMDARs. As a
control their membrane insertion and removal. consequence of coincident depolarization and
The number of glutamate receptors at a synapse glutamate binding, calcium influx through
(normally only a few dozens of receptors) can NMDARs is maximal, which activates intracel-
therefore be regulated through these interac- lular signaling cascades that ultimately are re-
tions. A particularly important interacting pro- sponsible for the altered synaptic efficacy.
tein is stargazin, as well as other members of NMDAR-dependent LTP is therefore an associ-
the transmembrane AMPAR regulatory protein ative form of plasticity and fulfills the criteria for
(TARP) family. Stargazin is in fact an auxiliary correlated activity as the origin of the strength-
subunit of most AMPARs that affects con- ening of the connection between two neurons
ductance, kinetics, and rectification. In other proposed by Donald Hebb more than 60 years
words, in the physiological situation in which ago (Hebb 1949). Conversely, LTD can be in-
TARPs are present, the kinetics and pharmacol- duced by repeated activation of the presynap-
ogy of neuronal AMPARs differ from those pre- tic neuron at low frequencies without postsyn-
dicted by classical in vitro expression studies of aptic activity (see below). Because the driving
AMPARs. PICK1 is another example of a pro- force for calcium entry is very large for a neuron
tein that regulates the number of receptors at the at rest and the block of NMDARs by Mg2þ is
synapse. Unlike the TARPs, which have little incomplete even at resting potentials (Jahr and
subunit specificity, PICK1 is believed to interact Stevens 1993), significant Ca enters the cell in
primarily with GluA2. Interfering with PICK1 response to synaptic stimulation (Sabatini et al.
and other interacting proteins has shown that 2002; Bloodgood and Sabatini 2007; Blood-
synaptic AMPARs dynamically recycle within good et al. 2009) during low-frequency synaptic

4 Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent LTP/LTD

stimulation. Presumably it is repeated occur- However, this makes sense because high-fre-
rence of this smaller NMDAR-dependent Ca in- quency stimulation protocols cause a strong
flux that triggers LTD induction. temporal summation of the excitatory postsyn-
Because NMDAR-dependent calcium in- aptic potentials (EPSPs), and the resultant large
flux induces both LTP and LTD, the cell must depolarization of the postsynaptic cell is suffi-
have a way to decide whether to potentiate or cient to relieve the Mg block of the NMDAR
depress a synaptic connection. An early model and allow a large amount of calcium to enter
invoked the level of activity or depolarization in the postsynaptic cells during the induction pro-
the postsynaptic cell as the critical variable dic- tocol. In contrast, low-frequency stimulation of
tating whether LTP or LTD was evoked (or no presynaptic axons is commonly used to induce
change in synaptic strength). Indeed, it is now LTD. Typically such protocols involve stimula-
fairly well accepted that modest activation of tion at 1– 3 Hz for 5– 15 min. This causes only a
NMDARs leading to modest increases in post- modest amount of postsynaptic depolarization,
synaptic calcium are optimal for triggering LTD, resulting in a modest but prolonged increase in
whereas much stronger activation of NMDARs postsynaptic calcium due to modest and repet-
leading to much larger increases in postsynaptic itive activation of NMDARs. A variant of these
calcium are required to trigger LTP (Malenka induction protocols is the so-called pairing pro-
1994). tocol, whereby during low-frequency activation
An important property of these NMDAR- of axons (0.1 – 1 Hz), individual cells are held
dependent forms of LTP and LTD is that they are at depolarized membrane potentials by passing
input- or synapse-specific. That is, only syn- current through the recording electrode. This
apses that actively contribute to the induction provides relief of the Mg-dependent block of
process will undergo the plasticity. This is be- NMDARs, allowing calcium to enter the post-
cause NMDARs on the synapses must be acti- synaptic cell when the NMDARs are activated
vated by synaptically released glutamate during by the presynaptically released glutamate. Con-
the induction protocol. The synapse specificity sistent with original ideas (Bienenstock et al.
of LTP has been explicitly tested using photolyt- 1982), modest depolarization can elicit LTD,
ic release of glutamate to bypass the presynaptic whereas stronger depolarization leads to LTP.
terminal and directly stimulate a visualized den- More recently studies have shown that LTP
dritic spine with glutamate. Experiments using and LTD can be elicited if action potentials in
this approach to induce NMDAR-dependent the pre- and postsynaptic neurons coincide with
LTP have shown that LTP can be induced at the appropriate timing. From the perspective
a single synapse without causing postsynap- of the postsynaptic neuron (which is where
tic potentiation or depression of neighboring the recordings are made), these manipulations
synapses (Matsuzaki et al. 2004; Harvey and lead to both a synaptic potential and an action
Svoboda 2007; but see Engert and Bonhoeffer potential that backpropagates into the den-
1997). However, interactions between neigh- drites, which presumably provides the addi-
boring synapses do occur, such that the intra- tional depolarization required to facilitate cal-
dendritic spread of the activated Ras from an cium influx through the activated NMDARs
active synapse briefly lowers the threshold for (Stuart et al. 1997; Waters et al. 2005). If a pre-
subsequent LTP induction at neighboring syn- synaptic spike is repetitively elicited slightly be-
apses (Harvey et al. 2008). fore (e.g., 5 msec) the postsynaptic neuron is
In acute slices of the hippocampus, LTP is fired, the EPSP precedes the backpropagating
often induced by tetanic stimulation of Schaffer action potential and such repetitive “pre-post”
collaterals. High-frequency stimulation proto- action potential firing can generate LTP. Con-
cols typically comprise delivery of one or several versely, when the backprogating action poten-
trains of pulses at 50 – 100 Hz for 1 sec. It may tial is repetitively elicited before the presynaptic
be surprising that stimulation of the presynap- spike, “post-pre” firing, LTD is often observed.
tic axons alone is sufficient for LTP induction. The time window for such pairings is critical,

Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710 5

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

C. Lüscher and R.C. Malenka

such that the closer in time presynaptic and quired because it is well established that
postsynaptic spikes are fired, the larger the mag- NMDAR-dependent LTP and LTD are triggered
nitude of the plasticity induced. If both stimu- in the postsynaptic cell. Indeed, during the de-
lations, however, occur at once, an asymptote bate about the locus of expression for LTP (and
is reached, in which the resultant long-term LTD), there was much discussion about the pos-
change in synaptic strength can no longer be sible identity of so-called retrograde messen-
reliably predicted. As the direction of synaptic gers, the substances that might be released
plasticity depends on the timing between the from postsynaptic cells following appropriate
presynaptic and postsynaptic spikes, this phe- activation of NMDARs and modify presynaptic
nomenon has been named spike-timing-depen- function (Arancio et al. 1996; but see Williams
dent plasticity (Dan and Poo 2006; Caporale et al. 1993).
and Dan 2008). It is often a considered a more To solve the pre- versus post- debate, which
physiological method of inducing LTP or LTD primarily focused on the locus of expression of
because it is imagined that these patterns of LTP, a large number of experiments were per-
spiking may occur during real behavior. formed. To assess whether presynaptic function
changed, the assays included:
EXPRESSION MECHANISM
† Using use-dependent open channel blockers
During the last two decades of the last century,
of postsynaptic receptors to estimate the re-
the locus of expression of NMDAR-dependent
lease probability. Most importantly, dizocil-
LTP was the focus of an intense debate (Ma-
pine (MK-801) irreversibly blocks NMDARs
lenka and Nicoll 1999; Nicoll 2003). Although
once they are activated by glutamate. Thus,
in hindsight it seems difficult to understand
once glutamate is released from a single pre-
how an apparently simple question could keep
synaptic terminal and activates its correspond-
many researchers in the field busy, the confu-
ing postsynaptic NMDARs, those NMDARs
sion could largely be attributed to the lack of
are irreversibly blocked. This means that
understanding of the basic physiology of excit-
application of MK-801 while activating a
atory synapses in the mammalian central ner-
population of synapses leads to a gradual de-
vous system. Although a minority of researchers
crease of the postsynaptic currents generated
in the field would argue that the question is still
by NMDARs, and the rate of this decrease
not fully answered, there is general agreement
directly correlates with the overall probability
that the experiments that were performed to
of release of the synapses being activated.
address this topic significantly furthered our
Importantly, it was found that the rate of
understanding of the basic properties of synap-
this decrease remains unaltered by LTP but
tic transmission.
is clearly affected by other manipulations
In theory, how LTP and LTD are expressed is
that are known to influence the probability
a simple question. For LTP the increase in syn-
of transmitter release (Manabe and Nicoll
aptic strength could be due either to more trans-
1994).
mitter being released from the presynaptic ax-
ons being activated during the course of the † Monitoring short-term plasticity before and
experiment or to the same amount of transmit- after the induction of LTP. Paired-pulse ratios
ter being released but having a greater effect and short-term facilitation or depression of
because the postsynaptic cell was more sensitive synaptic responses during short, high-fre-
to the same amount of released neurotransmit- quency bursts of stimulation reflect short-
ter. It is important to note that if LTP (or LTD) is term changes in release probability. These
caused by enhanced (or decreased) transmitter ubiquitous forms of short-term synaptic
release, it requires that the postsynaptic cell plasticity are greatly influenced by the base-
somehow communicate to the presynaptic ter- line release probability yet were unaffected by
minals and modify their function. This is re- the generation of LTP (McNaughton 1982;

6 Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent LTP/LTD

Manabe et al. 1993; but see Schulz et al. synapse, whereas conversely, LTD may involve
1995). the removal or endocytosis of synaptic AMPARs
(Fig. 3) (Lledo et al. 1998; Lüscher et al. 1999).
† Monitoring glial glutamate transporter cur-
A large body of experimental evidence now
rents before and after LTP. The reuptake of
supports this hypothesis (Lüscher et al. 2000;
glutamate by astrocytes is electrogenic, and
Lüscher and Frerking 2001; Malinow and Mal-
thus currents can be measured from astro-
enka 2002; Nicoll 2003; Collingridge et al. 2004;
cytes that reflect the amount of transmitter
Malenka and Bear 2004). In fact, AMPARs can
released. There is, however, no change in the
be quite mobile and recycle between the cyto-
size of these currents after induction of LTP
plasm and the cell membrane even under base-
even though other manipulations that affect
line conditions within tens of minutes. This can
transmitter release have clear effects. These
be shown, for example, by interfering with en-
experiments are particularly convincing, be-
docytosis, which leads to a run-up of synaptic
cause this approach relies on a readout that
responses. It is presumably this mobile pool of
is independent of glutamate receptors and
AMPARs that allows for rapid but sustained
therefore unlikely to participate in the ex-
changes in synaptic efficacy. The insertion
pression of LTP (Diamond et al. 1998; Lü-
and removal of AMPARs during LTP and LTD,
scher et al. 1998).
respectively, is believed to involve classical me-
† Visualization of presynaptic exocytosis with chanisms of SNARE protein – mediated exocy-
styryl dyes such as FM1-43. These dyes stain tosis and dynamin-dependent endocytosis via
presynaptic vesicles and are washed out once clathrin-coated vesicles (Lüscher et al. 1999;
the vesicles fuse with the membrane in an Carroll et al. 2001; Kennedy and Ehlers 2011).
activity-dependent fashion. Just as with MK- Current evidence favors the idea that the endo-
801, the destaining curves were superimpos- cytosis and exocytosis of AMPARs during LTD
able before and after tetanic induction of and LTP happens not directly at the synapse but
NMDAR-dependent LTP in CA1 neurons at at slightly perisynaptic locations, from where
50 – 100 Hz (Zakharenko et al. 2001). the receptors reach the postsynaptic density by
lateral diffusion.
Taken together, these findings (and addi- A large family of proteins associates with
tional experiments that are not covered because the AMPARs to regulate their mobility and bio-
of space limitations) make it very unlikely that physical properties as well as their stabilization
LTP is associated with an increase in release within the PSD (see Sheng and Kim 2012). Dif-
probability or the amount of glutamate released ferent AMPAR subunits may play distinct roles
from a presynaptic vesicle. But they say nothing in this redistribution process. Heteromeric
about what postsynaptic mechanisms contrib- GluA1/GluA2 receptors seem to be the primary
ute to LTP (and LTD). An important clue to this subtype of AMPAR that is inserted into syn-
question came from the observation that in the apses during LTP (Adesnik and Nicoll 2007).
hippocampi of very young animals, some syn- In addition, there are also forms of synaptic
apses contain only NMDARs and no, or very potentiation that are expressed by an exchange
few, AMPARs. Thus, these synapses are func- of GluA2-containing AMPARs in the synapse
tionally silent under baseline conditions. After for GluA2-lacking AMPARs (Liu and Zukin
the application of an LTP induction protocol, 2007). Because the latter have a higher conduc-
these synapses “wake up” and become function- tance, this will potentiate the synapse even if the
al because of the insertion of AMPARs into their total number remains the same. It has also been
postsynaptic membrane (Isaac et al. 1995; Liao suggested that during the very early phase of
et al. 1995). This result immediately raised the LTP (its first 10 – 20 min), there is a transient
possibility that at both silent synapses and syn- appearance of GluA2-lacking AMPARs at the
apses that already contain AMPARs, LTP in- synapse and this is required for the maintenance
volves the insertion of more AMPARs into the of LTP (Plant et al. 2006). However, this finding

Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710 7

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

C. Lüscher and R.C. Malenka

A
AMPA

NMDA Weak Strong

P
P
Ca2+
Endocytosis Ca2+
PP2B Exocytosis
PP1 CaMKll

Weak depolarization Strong depolarization

LTD LTP

B
Blocking LTD by disruption of dynamin– Blocking LTP with tetanus toxin
amphiphysin interaction (D15 peptide )

1 2
EPSC (% baseline)

EPSC (% baseline)

1+2 1+2
120 300
2
80
200
40
1 2 2
100 1
0
5 Hz, 3 min, –40 mV 2 × 100 Hz, 0 mV
0 10 20 30 0 10 20 30 40
min min

Figure 3. Postsynaptic expression mechanisms of LTP and LTD. (A) Weak activity of the presynaptic neuron leads
to modest depolarization and calcium influx through NMDA receptors. This preferentially activates phospha-
tases that dephosphorylate AMPA receptors, thus promoting receptor endocytosis. Strong activity paired with
strong depolarization triggers LTP in part via CaMKII, receptor phosphorylation, and exocytosis. (B) When
endocytosis is blocked with a dynamin dominant-negative peptide, LTD is inhibited. Conversely, blocking
exocytosis abolishes LTP. (Left panel: adapted and reprinted, with permission, from Lüscher et al. 1999; right
panel: W Morishita and RC Malenka, unpubl.)

is controversial (Adesnik and Nicoll 2007). Of et al. 1999; Lüthi et al. 1999; Collingridge and
particular interest are members of the TARP Isaac 2003).
family, which interact with all AMPA receptor
subunits and control not only membrane inser-
SIGNALING CASCADES FOR TRIGGERING
tion but also lateral redistribution (Chen et al.
LTP AND LTD
2000; Tomita et al. 2005; see Blakely and Ed-
wards 2012). If interactions with scaffolding We have briefly reviewed the mechanisms un-
proteins are manipulated through genetic inter- derlying the induction of LTP and LTD, as well
ventions or the perfusion of dominant-negative as their expression mechanisms. Appropriate
proteins, LTP and LTD can be blocked (Lüscher coincident activity of the pre- and postsynaptic

8 Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent LTP/LTD

neurons causes an influx of calcium through If LTP involves the activation of CaMKII
NMDARs, and depending on the quantitative (and other kinases) and LTD represents the in-
characteristics of this calcium signal, AMPARs verse of LTP, then a logical hypothesis is that
are either inserted into or removed from the LTD involves the preferential activation of pro-
synapses, resulting in LTP or LTD, respectively. tein phosphastases. Indeed, a very influential
But how are the triggering of LTP and LTD and model proposed that NMDAR-dependent LTD
their expression linked? What are the interme- depends on the calcium/calmodulin-depen-
diate signaling events that translate the increase dent protein phosphatase calcineurin as well as
in postsynaptic calcium into receptor redistri- on protein phosphatase 1 (PP1) (Lisman 1989).
bution? For LTP there is strong evidence that the This is a very attractive model because calci-
opening of NMDARs increases calcium concen- neurin has a much higher affinity for calci-
tration sufficiently in the dendritic spine to ac- um/calmodulin than does CaMKII and thus
tivate calcium/calmodulin-dependent kinase II will be preferentially activated by a modest in-
(CaMKII), which is found at very high concen- crease in calcium, the exact trigger for LTD.
trations in spines and which is clearly required There is now strong evidence that these two
for LTP (Lisman et al. 2002). This leads to the phosphatases do indeed play a role in LTD
phosphorylation of a number of proteins in- (Mulkey et al. 1993, 1994: Carroll et al. 2001),
cluding AMPARs themselves (Derkach et al. perhaps in part by influencing the phosphory-
1999). The phosphorylation of AMPAR sub- lation state of AMPARs. As was the case for LTP,
units can cause an increase in the conductance the intracellular signaling cascades underlying
of the AMPAR channel (Benke et al. 1998), an- LTD are certainly more complex than simply the
other postsynaptic mechanism that contributes activation of two phosphatases. For example, it
to at least the early phase of LTP. In addition, in has been suggested that apoptotic mechanisms
ways that remain to be determined, the increase including activation of caspase 3 via mitochon-
in CaMKII activity contributes to the insertion dria are critical for LTD (Li et al. 2010). Further-
of AMPARs (Ehlers 2000). This can be shown by more, the AMPARs that have internalized may
perfusing the postsynaptic neurons with acti- need to be degraded via lysosomal or proteaso-
vated CaMKII, which not only leads to an in- mal pathways so that they are not returned to
crease of synaptically evoked currents but also the plasma membrane.
enhances responses to iontophoretically applied
AMPA. Importantly, this manipulation oc-
MAINTENANCE OF LTP AND LTD
cludes further LTP, suggesting a shared mech-
anism between the ways of increasing synaptic AMPARs are tightly anchored in the PSD by
strength (Malenka and Nicoll 1999; Lisman a large number of scaffolding proteins linking
et al. 2002). them to cytoskeletal elements including actin.
Although CaMKII is well accepted to be one Insertion of additional receptors therefore is
major requisite trigger for LTP, like many other likely to affect the ultrastructure of the synapse,
cell biological phenomena, the signaling cas- and in fact, spines associated with synapses
cades underlying the induction and mainte- that underwent LTP enlarge (Matsuzaki et al.
nance (see below) of LTP are extremely complex. 2004; Harvey and Svoboda 2007; Holtmaat
A host of additional protein kinases, such as and Svoboda 2009; Kasai et al. 2010). Further-
cAMP-dependent protein kinase (PKA), pro- more, at the ultrastructural level in electron
tein kinase C (PKC), mitogen-activated protein micrographs, synapses that underwent LTP had
kinases, and tyrosine kinases, have all been sug- enlarged PSDs that were often discontinuous
gested to contribute to LTP in various ways (Toni et al. 2001). Such perforated synapses
(Bliss and Collingridge 1993; Malenka and Nic- also contain a higher proportion of smooth
oll 1999; Salter and Kalia 2004; Sweatt 2004). endoplasmic reticulum and a spine apparatus
However, many of the details of their precise (Lüscher et al. 2000). A second morphological
function in LTP remain to be worked out. correlate of LTP besides the physical enlargement

Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710 9

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

C. Lüscher and R.C. Malenka

of spines is the appearance of new spines from for LTP, as synapses made directly on the den-
the shaft within minutes of the induction pro- drites (i.e., shaft synapses) are capable of ex-
tocol ( perhaps also by division of an existing pressing and maintaining LTP. Moreover, selec-
spine). Consistent with this idea, the spine den- tive interference with the structural changes that
sity increases following LTP induction, and a accompany LTD do not seem to affect the in-
significantly higher frequency of multiple-spine crease in synaptic strength, at least not in early
synapses (Toni et al. 1999) is observed, as well as a stages, up to 1 – 2 hr following induction (Fig. 4)
reduction of their turnover (De Roo et al. 2008). (Wang et al. 2007; Redondo and Morris 2011).
Conversely, the triggering of LTD is associated
with the shrinkage of dendritic spines and per-
NMDAR-DEPENDENT LTP OF INHIBITORY
haps even their disappearance (Nägerl et al. TRANSMISSION
2004; Zhou et al. 2004; Wang et al. 2007; Kasai
et al. 2010). These observations raise the ques- Although the vast majority of work on the
tion of the interdependence of the mechanisms mechanisms and functions of LTP and LTD
underlying structural and functional plasticity have focused on excitatory synapses, potentia-
during LTPand LTD, a topic that is being actively tion and depression of inhibitory transmission
investigated. (I-LTP/I-LTD) can also be observed in the brain
Not only does long-term plasticity cause (Castillo et al. 2011). The underlying molecular
structural changes in synapses, but the mainte- mechanisms of I-LTP and I-LTD are variable,
nance of the change in synaptic strength dur- but one common feature is that they are often
ing LTP is protein synthesis – dependent. PKA, heterosynaptic. Synaptic plasticity of GABAergic
CaMKIV, protein kinase M-z, and extracellular
signal – regulated kinase (ERK), as well as other
A LTP
signaling molecules, initiate protein synthesis
either locally in the dendrites from prefabricat-
ed mRNA or by nuclear transcription (Sacktor LTD
2008). The latter involves interactions with tran-
scription factors, including cAMP response el-
ement – binding protein (CREB). Both local B
12 h
dendritic and nuclear transcription and somatic
translation together are believed to synthesize
the proteins required for the maintenance of
functional and structural plasticity following
the triggering of LTP. In reality, the events of
LTP expression and maintenance do not occur 5 mV
sequentially, and the first structural changes,
such as the increase of the size of the PSD and
spine growth, can be observed rapidly after in- 10 min

duction. Through these mechanisms LTP may Figure 4. Structural changes associated with LTP and
guide the selective stabilization of synaptic in- LTD. (A) Synaptic strength correlates with spine vol-
puts that show coincident activity, whereas non- ume and the area of the postsynaptic density (or-
activated inputs may be removed and replaced ange). Note that the PSD in potentiated synapses is
by new spines. Arc is one immediate early gene often perforated. (B) LTP can also lead to the appear-
that may orchestrate the translation of dendritic ance of new spines. Within 30 min of triggering LTP
mRNA required for actin polymerization and (30 stimuli applied to the presynaptic axon at 10-sec
intervals paired with depolarizing current injection
stable expansion of dendritic spines during LTP into the postsynaptic neuron; black bar) of a synaptic
(Bramham et al. 2010). Although such struc- connection in the hippocampus, new spines appear.
tural changes in spines have been studied in (Panel created from data adapted from Engert and
some detail, they are not absolutely required Bonhoeffer 1999.)

10 Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent LTP/LTD

transmission needs the activation of glutama- ders, depression, addiction, posttraumatic stress
tergic synapses and a specific messenger that syndrome, neuropathic pain, and anxiety disor-
passes the signal from one type of synapses to ders. To illustrate this emerging field and to em-
the other. Here we will focus on the forms of phasize the point that there are multiple ways by
I-LTP/I-LTD that are induced by NMDARs, which LTP and LTD can be involved in disease
which, as for the plasticity of excitatory trans- pathophysiology, we will focus on two contrast-
mission, are classified according to the locus of ing conditions: the loss of synaptic plasticity
induction and expression. associated with Alzheimer disease’s (AD) and
In the hippocampus, visual cortex, and op- excessive plasticity observed after exposure to
tic tectum, a form of I-LTP dependent on brain- addictive drugs.
derived neurotrophic factor (BDNF) has been A definitive diagnosis of AD requires the
described (Inagaki et al. 2008). It is induced by visualization of amyloid plaques and neurofi-
activation of NMDARs (and sometimes voltage- brillary tangles in histological sections of the
gated calcium channels) and modulated by brain. The cognitive decline, however, starts
GABAB receptors, which together drive an in- well before this stage, and there is growing evi-
crease in cytoplasmic calcium, in part by trig- dence that one of the toxic protein species be-
gering the release of calcium from intracellular lieved to be etiologically related to AD, soluble
stores. In this model NMDAR activation leads Ab oligomers, causes early memory problems
to postsynaptic release of BDNF, which func- by disrupting LTP and LTD mechanisms (Walsh
tions as a retrograde messenger and causes an et al. 2002; Tanzi 2005; Shankar et al. 2008; Cisse
increase in GABA release through activation of et al. 2011). Direct application or overproduc-
presynaptic TrkB receptors. tion of Ab oligomers both inhibits LTP and
In the ventral tegmental area, a similar form triggers LTD-like changes (Fig. 5). The net result
of plasticity called LTPGABA is induced by strong is weaker synapses that have difficulty generat-
activation of NMDARs on dopamine neurons ing LTP. Furthermore, the toxic Ab also de-
(Nugent et al. 2007). This leads to activation of a creases synaptic NMDARs, a change that con-
Ca2þ-dependent nitric oxide synthase, which tributes to the impaired LTP (Kamenetz et al.
generates nitric oxide, which acts as a retrograde 2003). Based on these findings, there is great
messenger by diffusing back to presynaptic neu- interest in finding compounds that prevent the
rons. Nitric oxide in turn causes activation of synaptic effects of Ab oligomers with the hope
guanylate cyclase and synthesis of cGMP in the that such compounds will be therapeutically
synaptic terminals of inhibitory afferents onto beneficial if given to patients early enough dur-
dopamine neurons. The release probability for ing disease progression. See Sheng et al. (2012)
GABA then increases through a still unknown for a comprehensive discussion of synaptic
mechanism involving activation of the cGMP- changes associated with AD.
dependent protein kinase, PKG. Addictive drugs also have profound effects
The slow GABAB receptor– mediated inhib- on synaptic transmission that may influence
itory postsynaptic potential (IPSP) can also LTP and LTD (Wolf 2003; Kauer and Malenka
be potentiated if the postsynaptic neuron is 2007; Lüscher and Malenka 2011). For example,
strongly depolarized (Huang et al. 2005). The a single dose of a drug of abuse such as cocaine
increase of this IPSP also depends on NMDAR enhances excitatory transmission onto dopa-
activation and CaMKII, thus sharing two key mine neurons of the ventral tegmental area
properties with hippocampal LTP of AMPARs. (Ungless et al. 2001). This drug-evoked plastic-
ity requires activation of type 1 dopamine re-
ceptors along with NMDAR activation and is
SYNAPTIC PLASTICITY AND DISEASE
expressed by a redistribution of AMPARs and
Altered LTP and LTD has been implicated as a NMDARs. Calcium-permeable, GluA2-lacking
mechanism that may contribute to brain dis- AMPARs appear (Bellone and Lüscher 2006),
eases as diverse as dementia, movement disor- while NMDAR function is decreased. As a result,

Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710 11

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

C. Lüscher and R.C. Malenka

AMPAR
mGluRs
NMDAR
Thinning and loss
of synapses

Healthy synapse
Late stage AD

Aβ oligomers

Early stage AD

Figure 5. Block of LTP and triggering of LTD by Ab oligomers in Alzheimer’s disease (AD). Soluble Ab oligomers
strongly activate metabotropic glutamate receptors (mGluRs), which leads to the internalization of AMPARs
and LTD. As a consequence, normal NMDAR-dependent LTD is occluded. At the same time they inhibit
NMDARs, precluding the induction of LTP. In a second step synapses become thinner and some disappear. It
is believed that these changes underlie the early cognitive decline observed in AD.

the primary source for synaptic calcium entry dent activity in pre- and postsynaptic neurons
shifts from NMDARs in naı̈ve animals to resulting in calcium influx through synaptic
GluA2-lacking AMPARs after a dose of an ad- NMDARs is well established to be necessary
dictive drug. As a consequence, the rules for for the triggering of both LTP and LTD. This
subsequent activity-dependent plasticity are causes the activity-dependent redistribution of
inverted (Mameli and Lüscher 2011). These AMPARs, a general mechanism for modifying
early drug-evoked changes in the properties synaptic strength in many neuronal cell types.
of excitatory synapses on dopamine cells can Elucidating the detailed molecular mechanisms
be very long-lasting and are followed by many by which AMPARs traffic to and away from syn-
additional synaptic modifications in related apses is an active topic of current investigations
structures. For example, in the nucleus accum- (see West and Greenberg 2012). An equally im-
bens chronic (5-d) administration of cocaine portant topic is the identification of the pro-
triggers an internalization of AMPARs in medi- teins that need to be synthesized to maintain
um spiny neurons, a change that occludes the LTP and LTD. Through coordination of func-
subsequent synaptic induction of NMDAR- tional changes in synaptic efficacy to structural
dependent LTD (Thomas and Malenka 2003; changes in synapses, the reorganization of neu-
Kauer and Malenka 2007; Wolf and Ferrario ral networks can be structurally maintained for
2010). long periods of time. Certainly, more complete
understanding of the mechanisms underlying
LTP and LTD will continue to contribute to
CONCLUSIONS
our understanding of the mechanisms of learn-
Because of a major effort by a large number ing and memory (see Mayford et al. 2012), as
of investigators, the mechanisms underlying well as provide insight into the pathophysiology
NMDAR-dependent LTP and LTD are under- of a broad spectrum of brain diseases (see Sheng
stood in reasonable molecular detail. Coinci- et al. 2012).

12 Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent LTP/LTD

REFERENCES Collingridge GL, Isaac JT, Wang YT. 2004. Receptor traffick-
 Reference is also in this collection. ing and synaptic plasticity. Nat Rev Neurosci 5: 952– 962.
Dan Y, Poo MM. 2006. Spike timing-dependent plasticity:
From synapse to perception. Physiol Rev 86: 1033–1048.
Adesnik H, Nicoll RA. 2007. Conservation of glutamate
receptor 2-containing AMPA receptors during long- De Roo M, Klauser P, Muller D. 2008. LTP promotes a se-
lective long-term stabilization and clustering of dendritic
term potentiation. J Neurosci 27: 4598– 4602.
spines. PLoS Biol 6: e219.
Arancio O, Kiebler M, Lee CJ, Lev-Ram V, Tsien RY, Kandel
Derkach V, Barria A, Soderling TR. 1999. Ca2þ/calmodulin-
ER, Hawkins RD. 1996. Nitric oxide acts directly in the
kinase II enhances channel conductance of a-amino-
presynaptic neuron to produce long-term potentiation in
3-hydroxy-5-methyl-4-isoxazolepropionate type gluta-
cultured hippocampal neurons. Cell 87: 1025–1035.
mate receptors. Proc Natl Acad Sci 96: 3269–3274.
Bellone C, Lüscher C. 2006. Cocaine triggered AMPA recep-
Diamond JS, Bergles DE, Jahr CE. 1998. Glutamate release
tor redistribution is reversed in vivo by mGluR-depen-
monitored with astrocyte transporter currents during
dent long-term depression. Nat Neurosci 9: 636– 641.
LTP. Neuron 21: 425–433.
Benke TA, Lüthi A, Isaac JT, Collingridge GL. 1998. Modu- Dingledine R, Borges K, Bowie D, Traynelis SF. 1999. The
lation of AMPA receptor unitary conductance by synaptic glutamate receptor ion channels. Pharmacol Rev 51:
activity. Nature 393: 793–797. 7 –61.
Bienenstock EL, Cooper LN, Munro PW. 1982. Theory for Ehlers MD. 2000. Reinsertion or degradation of AMPA re-
the development of neuron selectivity: Orientation spe- ceptors determined by activity-dependent endocytic
cificity and binocular interaction in visual cortex. J Neu- sorting. Neuron 28: 511–525.
rosci 2: 32– 48.
Elias GM, Nicoll RA. 2007. Synaptic trafficking of glutamate
 Blakely RD, Edwards RH. 2012. Vesicular and plasma mem- receptors by MAGUK scaffolding proteins. Trends Cell
brane transporters for neurotransmitters. Cold Spring Biol 17: 343 –352.
Harb Perspect Biol doi: 10.1101/cshperspect.a005595.
Engert F, Bonhoeffer T. 1997. Synapse specificity of long-
Bliss TV, Collingridge GL. 1993. A synaptic model of mem- term potentiation breaks down at short distances. Nature
ory: Long-term potentiation in the hippocampus. Nature 388: 279–284.
361: 31–39.
Harvey CD, Svoboda K. 2007. Locally dynamic synaptic
Bloodgood BL, Sabatini BL. 2007. Nonlinear regulation of learning rules in pyramidal neuron dendrites. Nature
unitary synaptic signals by CaV2.3 voltage-sensitive calci- 450: 1195– 1200.
um channels located in dendritic spines. Neuron 53:
Harvey CD, Yasuda R, Zhong H, Svoboda K. 2008. The
249–260. spread of Ras activity triggered by activation of a single
Bloodgood BL, Giessel AJ, Sabatini BL. 2009. Biphasic syn- dendritic spine. Science 321: 136 –140.
aptic Ca influx arising from compartmentalized electrical Hebb DO. 1949. The organization of behavior: A neuropsy-
signals in dendritic spines. PLoS Biol 7: e1000190. chological theory. John Wiley & Sons, New York.
Bramham CR, Alme MN, Bittins M, Kuipers SD, Nair RR, Holtmaat A, Svoboda K. 2009. Experience-dependent struc-
Pai B, Panja D, Schubert M, Soule J, Tiron A, et al. 2010. tural synaptic plasticity in the mammalian brain. Nat Rev
The Arc of synaptic memory. Exp Brain Res 200: 125 – Neurosci 10: 647 –658.
140.
Huang CS, Shi SH, Ule J, Ruggiu M, Barker LA, Darnell RB,
Caporale N, Dan Y. 2008. Spike timing-dependent plasticity: Jan YN, Jan LY. 2005. Common molecular pathways me-
A Hebbian learning rule. Annu Rev Neurosci 31: 25– 46. diate long-term potentiation of synaptic excitation and
Carroll RC, Beattie EC, von Zastrow M, Malenka RC. 2001. slow synaptic inhibition. Cell 123: 105– 118.
Role of AMPA receptor endocytosis in synaptic plasticity. Inagaki T, Begum T, Reza F, Horibe S, Inaba M, Yoshimura Y,
Nat Rev Neurosci 2: 315– 324. Komatsu Y. 2008. Brain-derived neurotrophic factor-me-
 Castillo PE. 2012. Presynaptic LTP and LTD of excitatory and diated retrograde signaling required for the induction of
inhibitory synapses. Cold Spring Harb Perspect Biol doi: long-term potentiation at inhibitory synapses of visual
10.1101/cshperspect.a005728. cortical pyramidal neurons. Neurosci Res 61: 192– 200.
Castillo PE, Chiu CQ, Carroll RC. 2011. Long-term plastic- Isaac JTR, Nicoll RA, Malenka RC. 1995. Evidence for silent
ity at inhibitory synapses. Curr Opin Neurobiol 21: synapses. Implications for the expression of LTP. Neuron
328–338. 15: 427– 434.
Chen L, Chetkovich DM, Petralia RS, Sweeney NT, Kawasaki Jahr CE, Stevens CF. 1993. Calcium permeability of the N-
Y, Wenthold RJ, Bredt DS, Nicoll RA. 2000. Stargazin methyl-D-aspartate receptor channel in hippocampal
regulates synaptic targeting of AMPA receptors by two neurons in culture. Proc Natl Acad Sci 90: 11573– 11577.
distinct mechanisms. Nature 408: 936 –943. Kamenetz F, Tomita T, Hsieh H, Seabrook G, Borchelt D,
Cisse M, Halabisky B, Harris J, Devidze N, Dubal DB, Sun B, Iwatsubo T, Sisodia S, Malinow R. 2003. APP processing
Orr A, Lotz G, Kim DH, Hamto P, et al. 2011. Reversing and synaptic function. Neuron 37: 925– 937.
EphB2 depletion rescues cognitive functions in Alz- Kasai H, Fukuda M, Watanabe S, Hayashi-Takagi A, Nogu-
heimer model. Nature 469: 47– 52. chi J. 2010. Structural dynamics of dendritic spines in
Collingridge GL, Isaac JT. 2003. Functional roles of protein memory and cognition. Trends Neurosci 33: 121– 129.
interactions with AMPA and kainate receptors. Neurosci Kauer JA, Malenka RC. 2007. Synaptic plasticity and addic-
Res 47: 3 –15. tion. Nat Rev Neurosci 8: 844–858.

Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710 13

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

C. Lüscher and R.C. Malenka

Kennedy MJ, Ehlers MD. 2011. Mechanisms and function of tiation: Effects on paired pulse facilitation and EPSC
dendritic exocytosis. Neuron 69: 856–875. variance in the CA1 region of the hippocampus. J Neuro-
Li Z, Jo J, Jia JM, Lo SC, Whitcomb DJ, Jiao S, Cho K, Sheng physiol 70: 1451– 1459.
M. 2010. Caspase-3 activation via mitochondria is re- Matsuzaki M, Honkura N, Ellis-Davies GC, Kasai H. 2004.
quired for long-term depression and AMPA receptor in- Structural basis of long-term potentiation in single den-
ternalization. Cell 141: 859 –871. dritic spines. Nature 429: 761 –766.
Liao DZ, Hessler NA, Malinow R. 1995. Activation of post-  Mayford M, Siegelbaum SA, Kandel ER. 2012. Synapses and
synaptically silent synapses during pairing induced LTP memory storage. Cold Spring Harb Perspect Biol doi:
in CA1 region of hippocampal slice. Nature 375: 400 – 10.1101/cshperspect.a005751.
404. McNaughton BL. 1982. Long-term synaptic enhancement
Lisman J. 1989. A mechanism for the Hebb and the anti- and short-term potentiation in rat fascia dentata act
Hebb processes underlying learning and memory. Proc through different mechanisms. J Physiol 324: 249 –262.
Natl Acad Sci 86: 9574– 9578. Mulkey RM, Herron CE, Malenka RC. 1993. An essential
Lisman J, Schulman H, Cline H. 2002. The molecular basis role for protein phosphatases in hippocampal long-term
of CaMKII function in synaptic and behavioural memo- depression. Science 261: 1051– 1055.
ry. Nat Rev Neurosci 3: 175 –190. Mulkey RM, Endo S, Shenolikar S, Malenka RC. 1994. In-
Liu SJ, Zukin RS. 2007. Ca2þ-permeable AMPA receptors in volvement of a calcineurin/inhibitor-1 phosphatase cas-
synaptic plasticity and neuronal death. Trends Neurosci cade in hippocampal long-term depression. Nature 369:
30: 126 –134. 486 –488.
Lledo PM, Zhang X, Sudhof TC, Malenka RC, Nicoll RA. Nägerl UV, Eberhorn N, Cambridge SB, Bonhoeffer T. 2004.
1998. Postsynaptic membrane fusion and long-term po- Bidirectional activity-dependent morphological plastici-
tentiation. Science 279: 399 –403. ty in hippocampal neurons. Neuron 44: 759– 767.
Lüscher C, Frerking M. 2001. Restless AMPA receptors: Im- Nicoll RA. 2003. Expression mechanisms underlying long-
plications for synaptic transmission and plasticity. Trends term potentiation: A postsynaptic view. Philos Trans R Soc
Neurosci 24: 665–670. Lond B Biol Sci 358: 721– 726.
Lüscher C, Malenka RC. 2011. Drug-evoked synaptic plas- Nugent FS, Penick EC, Kauer JA. 2007. Opioids block long-
ticity in addiction: From molecular changes to circuit term potentiation of inhibitory synapses. Nature 446:
remodeling. Neuron 69: 650 –663. 1086–1090.
Lüscher C, Malenka RC, Nicoll RA. 1998. Monitoring glu- Plant K, Pelkey KA, Bortolotto ZA, Morita D, Terashima A,
tamate release during LTP with glial transporter currents. McBain CJ, Collingridge GL, Isaac JT. 2006. Transient
Neuron 21: 435–441. incorporation of native GluR2-lacking AMPA receptors
Lüscher C, Xia H, Beattie EC, Carroll RC, von Zastrow M, during hippocampal long-term potentiation. Nat Neuro-
Malenka RC, Nicoll RA. 1999. Role of AMPA receptor sci 9: 602– 604.
cycling in synaptic transmission and plasticity. Neuron Redondo RL, Morris RG. 2011. Making memories last: The
24: 649 –658. synaptic tagging and capture hypothesis. Nat Rev Neuro-
Lüscher C, Nicoll RA, Malenka RC, Muller D. 2000. Synaptic sci 12: 17–30.
plasticity and dynamic modulation of the postsynaptic Sabatini BL, Oertner TG, Svoboda K. 2002. The life cycle of
membrane. Nat Neurosci 3: 545– 550. Ca2þ ions in dendritic spines. Neuron 33: 439–452.
Lüthi A, Chittajallu R, Duprat F, Palmer MJ, Benke TA, Kidd Sacktor TC. 2008. PKMz, LTP maintenance, and the dynam-
FL, Henley JM, Isaac JT, Collingridge GL. 1999. Hippo- ic molecular biology of memory storage. Prog Brain Res
campal LTD expression involves a pool of AMPARs reg- 169: 27–40.
ulated by the NSF–GluR2 interaction. Neuron 24: Salter MW, Kalia LV. 2004. Src kinases: A hub for NMDA
389–399. receptor regulation. Nat Rev Neurosci 5: 317–328.
Malenka RC. 1994. Synaptic plasticity in the hippocampus: Schulz PE, Cook EP, Johnston D. 1995. Using paired-pulse
LTP and LTD. Cell 78: 535– 538. facilitation to probe the mechanisms for long-term po-
Malenka RC, Bear MF. 2004. LTP and LTD: An embarrass- tentiation (LTP). J Physiol Paris 89: 3 –9.
ment of riches. Neuron 44: 5 –21. Shankar GM, Li S, Mehta TH, Garcia-Munoz A, Shepardson
Malenka RC, Nicoll RA. 1999. Long-term potentiation—a NE, Smith I, Brett FM, Farrell MA, Rowan MJ, Lemere
decade of progress? Science 285: 1870–1874. CA, et al. 2008. Amyloid-b protein dimers isolated di-
Malinow R, Malenka RC. 2002. AMPA receptor trafficking rectly from Alzheimer’s brains impair synaptic plasticity
and synaptic plasticity. Annu Rev Neurosci 25: 103–126. and memory. Nat Med 14: 837– 842.
Mameli M, Lüscher C. 2011. Synaptic plasticity and addic-  Sheng M, Kim E. 2012. The postsynaptic organization of
tion: Learning mechanisms gone awry. Neuropharmacol- synapses. Cold Spring Harb Perspect Biol doi: 10.1101/
ogy 61: 1052– 1059. cshperspect.a005678.
Manabe T, Nicoll RA. 1994. Long-term potentiation: Evi-  Sheng M, Sabatini B, Südhof TC. 2012. Synapses and Alz-
dence against an increase in transmitter release probabil- heimer’s disease. Cold Spring Harb Perspect Biol doi:
ity in the CA1 region of the hippocampus. Science 265: 10.1101/cshperspect.a005777.
1888– 1892. Stuart G, Spruston N, Sakmann B, Hausser M. 1997. Action
Manabe T, Wyllie DJ, Perkel DJ, Nicoll RA. 1993. Modula- potential initiation and backpropagation in neurons of
tion of synaptic transmission and long-term poten- the mammalian CNS. Trends Neurosci 20: 125–131.

14 Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710

Downloaded from http://cshperspectives.cshlp.org/ on September 30, 2014 - Published by Cold Spring Harbor Laboratory Press

NMDA Receptor-Dependent LTP/LTD

Sweatt JD. 2004. Mitogen-activated protein kinases in syn- Wang XB, Yang Y, Zhou Q. 2007. Independent expression of
aptic plasticity and memory. Curr Opin Neurobiol 14: synaptic and morphological plasticity associated with
311–317. long-term depression. J Neurosci 27: 12419– 12429.
Tanzi RE. 2005. The synaptic Ab hypothesis of Alzheimer Waters J, Schaefer A, Sakmann B. 2005. Backpropagating
disease. Nat Neurosci 8: 977– 979. action potentials in neurones: Measurement, mecha-
Thomas MJ, Malenka RC. 2003. Synaptic plasticity in the nisms and potential functions. Prog Biophys Mol Biol
mesolimbic dopamine system. Philos Trans R Soc Lond B 87: 145– 170.
Biol Sci 358: 815–819.  West AE, Greenberg ME. 2012. Neuronal activity-regulated
Tomita S, Stein V, Stocker TJ, Nicoll RA, Bredt DS. 2005. gene transcription in synapse development and cognitive
Bidirectional synaptic plasticity regulated by phosphor- function. Cold Spring Harb Perspect Biol doi: 10.1101/
ylation of stargazin-like TARPs. Neuron 45: 269–277. cshperspect.a005744.
Toni N, Buchs PA, Nikonenko I, Bron CR, Muller D. 1999. Williams JH, Li YG, Nayak A, Errington ML, Murphy KP,
LTP promotes formation of multiple spine synapses be- Bliss TV. 1993. The suppression of long-term potentia-
tween a single axon terminal and a dendrite. Nature 402: tion in rat hippocampus by inhibitors of nitric oxide
421–425. synthase is temperature and age dependent. Neuron 11:
877 –884.
Toni N, Buchs PA, Nikonenko I, Povilaitite P, Parisi L, Muller
D. 2001. Remodeling of synaptic membranes after induc- Wolf ME. 2003. LTP may trigger addiction. Mol Interv 3:
tion of long-term potentiation. J Neurosci 21: 6245– 248 –252.
6251. Wolf ME, Ferrario CR. 2010. AMPA receptor plasticity in the
Ungless MA, Whistler JL, Malenka RC, Bonci A. 2001. Sin- nucleus accumbens after repeated exposure to cocaine.
gle cocaine exposure in vivo induces long-term potenti- Neurosci Biobehav Rev 35: 185–211.
ation in dopamine neurons. Nature 411: 583–587. Zakharenko SS, Zablow L, Siegelbaum SA. 2001. Visualiza-
Walsh DM, Klyubin I, Fadeeva JV, Cullen WK, Anwyl R, tion of changes in presynaptic function during long-term
Wolfe MS, Rowan MJ, Selkoe DJ. 2002. Naturally secreted synaptic plasticity. Nat Neurosci 4: 711– 717.
oligomers of amyloid b protein potently inhibit hip- Zhou Q, Homma KJ, Poo MM. 2004. Shrinkage of dendritic
pocampal long-term potentiation in vivo. Nature 416: spines associated with long-term depression of hippo-
535–539. campal synapses. Neuron 44: 749 –757.

Cite this article as Cold Spring Harb Perspect Biol 2012;4:a005710 15