camkii in learning and memory

8/3/2019 CamkII in Learning and Memory

1/14

Long-term potentiation (LTP) is a process wherebybrief periods of synaptic activity can produce a long-lasting increase in the strength of a synapse, as shown byan increase in the size of the excitatory postsynaptic cur-rent (EPSC). LTP, in vivo, can last for at least a month1,and similar changes in synaptic strength occur duringlearning2,3. When LTP is prevented by various muta-tions, memory is impaired4. Taken together, these find-ings indicate that LTP has an important role in learningand memory, and a major effort has therefore been madeto understand the mechanisms underlying this process.

The CA1 region of the hippocampus, a brain regionthat is vital for long-term memory formation, has beenused as the primary model system for understandingLTP5. The process of LTP at hippocampal CA1 syn-apses is initiated by brief periods of high-frequencypresynaptic stimulation. Glutamate, the neurotrans-mitter that is released at these synapses, activates post-synaptic AMPA- and NMDA-type glutamate receptors(AMPARs and NMDARs, respectively). The opening

of NMDARs during the induction of LTP leads to cal-cium entry that triggers a biochemical cascade, the endproduct of which is the long-lasting potentiation of theAMPAR-mediated EPSC.

Strong evidence now shows that calcium elevationinitiates this biochemical cascade through activationof calcium/calmodulin-dependent protein kinase II(CaMKII), which is a major synaptic protein6. Thiskinase is a large holoenzyme consisting of 12 subunits.There are two principal types of CaMKII subunits CaMKII and CaMKII both of which are capableof catalysis. In the rodent forebrain, CaMKII holoen-zymes mainly comprise CaMKII homomers and

CaMKIICaMKII heteromers. Recent work has ledto the crystallization of the holoenzyme and the elucida-tion of how the enzymes structure relates to its function7(BOX 1). The kinase can act as a protein switch8; onceactivated by calcium/calmodulin, the enzyme can beautophosphorylated at T286 (on CaMKII subunits;T287 on CaMKII subunits), an event that makesCaMKII activity persist even after the calcium concen-tration falls to baseline levels9 (this autophosphorylatedstate of CaMKII is termed the autonomous state). Ifactive CaMKII subunits or holoenzymes are introducedinto CA1 neurons, the EPSC becomes large and LTP canno longer be induced by synaptic stimulation10,11. If per-sistent activation of CaMKII is prevented by a knock-inmutation that encodes T286A in CaMKII, LTP induc-tion is prevented and mice show profound memoryimpairments, notably after one-trial learning12,13. Inmany mutant mice that show enhanced learning,CaMKII activation is increased14. Taken together, theseresults indicate that activated CaMKII is sufficient to

trigger LTP and that CaMKII activation is necessary forLTP induction and certain forms of learning.

In this Review, we describe recent progress in under-standing the role of CaMKII in LTP. It has been shownthat LTP occurs specifically at synapses that are stimu-lated (inactive synapses that are only micrometres awayfrom a stimulated synapse remain unpotentiated). Wereview recent data showing that activation of CaMKIIfollowing synaptic stimulation is synapse-specific andelucidating how this specificity is achieved. We alsodescribe experiments that have identified the down-stream process by which activated CaMKII causes EPSCenhancement during the early stages of LTP. Finally,

1Department of Biology,

Brandeis University,

Waltham, Massachusetts

02454, USA.2Department of

Neurobiology, Duke

University, Durham, North

Carolina 27710, USA.

Correspondence to J.L.

email: [email protected]

doi:10.1038/nrn3192

Published online

15 February 2012

Mechanisms of CaMKII actionin long-term potentiation

John Lisman1, Ryohei Yasuda2 and Sridhar Raghavachari2

Abstract | Long-term potentiation (LTP) of synaptic strength occurs during learning and can

last for long periods, making it a probable mechanism for memory storage. LTP induction

results in calcium entry, which activates calcium/calmodulin-dependent protein kinase II

(CaMKII). CaMKII subsequently translocates to the synapse, where it binds to NMDA-type

glutamate receptors and produces potentiation by phosphorylating principal and auxiliarysubunits of AMPA-type glutamate receptors. These processes are all localized to stimulated

spines and account for the synapse-specificity of LTP. In the later stages of LTP, CaMKII has a

structural role in enlarging and strengthening the synapse.

R E V I E W S

NATURE REVIEWS |NEUROSCIENCE VOLUME 13 | MARCH 2012 |169

2012 Macmillan Publishers Limited. All rights reserved
mailto:[email protected]:[email protected]


2/14

a

b

c

Kinase domain Hub domain

90

36

1 274 314 475

Linker region

Kinase domain N-lobe

Kinase domain C-lobe

Regulatory segment

Central hub

Kinase domain N-lobe

Kinase domain C-lobe

Regulatory segment

Phosphorylation sites

Central domain

Substratebinding region

T-site

T286

T305306

Autoinhibited compact(calmodulin-inaccessible)

Autoinhibited extended(calmodulin-accessible)

Activated(calmodulin-accessible)

Ca2+/calmodulin

Regulatorysegment

Hub domain

Kinasedomain

Ca2+/calmodulinbound (active)

Box 1 | Structural and functional properties of CaMKII

The architecture of the calcium/calmodulin-dependent protein kinase II (CaMKII) holoenzyme has been determined and is

based on its crystal structure (see the figure, part a)7. The holoenzyme has 12 subunits. Each of these subunits has a

carboxy-terminal association (hub) domain and, together, these domains form a central organizing hub (grey) onto which

the subunits catalytic kinase domains (blue) fold into two rings of six domains. The catalytic domains have a bilobed

structure with an ATP binding site located between the two lobes. In each subunit, a regulatory segment (yellow) lies

adjacent to the kinase domain (which contains the substrate-binding site, or S-site), and a linker region connects the

regulatory domain to the hub domain, which has a central pore of unknown function146

.In the autoinhibited state, the regulatory segment forms an -helix that blocks the catalytic S-site in a pseudosubstrate

manner (see the figure, part b). This inactive state is further stabilized by sequestering the T286 phosphorylation site

(amino acid numbering relates to CaMKII subunits) into a hydrophobic groove (the T-site)147. The crystal structure reveals acompact state in which the portion of the regulatory domain that is recognized by calmodulin is sequestered into the

association domain, making it impossible for calmodulin to bind CaMKII.

Activation of this enzyme is only possible because the compact state isin equilibrium with an extended state to which

calmodulin can bind (see the

figure, partc)7,148,149. The linker

region varies in length

between different CaMKII

isoforms and determines both

the set-point of this

equilibrium and the calcium

level required for subunit

activation. Upon calmodulinbinding, the regulatory

segment is stripped from the

S-site7,150, allowing substrate

phosphorylation and initiating

the slower process of T286

intersubunit autophosphoryl-

ation. If T286 phosphorylation

has not occurred, lowering the

calcium concentration rapidly turns off

activity. If T286 phosphorylation has

occurred, however, activity remains despite the low calcium

concentration (this is termed the autonomous state) and the

T-site is open for binding (for example, to the NR2B subunit of

NMDA-type glutamate receptors). T286 phosphorylation

dramatically strengthens the binding of calmodulin, a processcalled trapping. The autonomous activity of CaMKII for

phosphorylating various substrates is up to five times higher

with calmodulin bound than it is when calmodulin is not

bound126. Formation of autonomous CaMKII has a Hill

coefficient of approximately eight with respect to calcium,

indicating a virtual threshold151. Various cooperative reactions

underlie this ultrasensitivity. Four calcium binding events are

required to fully activate calmodulin, which displays

cooperativity when binding to the holoenzyme152. Calmodulin

can activate a CaMKII subunit, but the subunit can only

phosphorylate a neighbouring subunit if the adjacent subunit is

bound to a calmodulin molecule (as such binding is required to

expose T286). The top right part of panel a, panelb, and panel c

are reproduced, with permission, from REF. 7 (2011) Elsevier.

R E V I E W S

170 | MARCH 2012 | VOLUME 13 www.nature.com/reviews/neuro



3/14

4 pA

10 ms

Stimulated spine

Ca2+

LTP (EPSCs)

CaMKII activation

CaMKII translocation

4 12 204 s

1 m

1 m

32 ms 32 96 160 352

36 8460 108

Adjacent spine

a

b

c

d

7 min

3 m

0 min 15 25

Spine

Dendrite

28 min

7 min

28 min

Two-photon glutamate

uncaging

This technique involves

focusing a pulsed laser beam

into a solution that includes

caged glutamate. The

glutamate is released into the

diffraction-limited volume

when photolysis of the caged

molecules by a two-photon

excitation process occurs at

the focus.

Two-photon fluorescence

lifetime imaging

(2pFLIM). The fluorescence

lifetime can be determined by

measuring fluorescence decay

(which occurs over a period of

nanoseconds) after a short

pulse of fluorescence

excitation. Probes can be

made in which the lifetime

changes with the conformation

of the protein, thereby

providing a measure of itsactivation state.

we discuss experiments that suggest that CaMKII cancontribute to the later stages of LTP, when structuralchanges in the synapse become important for long-termmemory. Owing to space limitations, we do not discussthe presynaptic functions of CaMKII in LTP15, the roleof other kinases and second messengers in LTP16 or thephosphatase-dependent processes that limit or reverseCaMKII-dependent activation17.

Synapse-specific activation

Early experimental data suggested that LTP is a syn-apse-specific event18,19, and optical methods have nowmade it possible to directly demonstrate that this is cor-rect. An identified spine can be stimulated using two-photon glutamate uncaging. Such uncaging gives rise toa glutamate pulse with submicrometre dimensions thatdirectly targets the identified spine and induces LTP atthe synapse on that spine. Nearby synapses that are onlymicrometres away are not potentiated20(FIG. 1). Given thecomplex biochemical cascade that is involved in LTP andthe small distances that separate different synapses, it is

remarkable that synapse-specific LTP can be achieved.High-resolution imaging of CaMKII activation

during LTP induction has been made possible by thedevelopment of the Camui activity sensor21,22. This sen-sor comprises two different f luorophores monomericenhanced green fluorescent protein (mEGFP) and reso-nance energy-accepting chromoprotein (REAch) thatare attached to different ends of CaMKII subunits. Thelevel of fluorescence resonance energy transfer (FRET)that occurs between the two fluorophores is an indica-tor of the distance between them: when the enzyme isactivated, conformational changes increase the dis-tance between the fluorophores and thus decrease theamount of FRET. Through use of this sensor and

two-photon fluorescence lifetime imaging (2pFLIM), whichallows for the quantification of FRET, the conforma-tional change in CaMKII that is associated with activa-tion can be monitored at the level of a single spine inreal time. Moreover, such monitoring can be conductedwhile LTP is induced at the same spine using a secondtwo-photon laser to uncage glutamate, which activatesNMDARs on that spine. Using this approach, it can beseen that CaMKII is activated within seconds after thestart of synaptic stimulation and that the activation isrestricted to the stimulated spine22(FIG. 1). At the end ofstimulation, CaMKII activity decays to baseline levelswithin ~1 min. By contrast, in cells expressing a mutant

variant of CaMKII (T286A) that cannot be autophos-phorylated, the activity of the enzyme decays within sec-onds. Taken together, these results show that CaMKIIactivation has the localization that is required to accountfor the synapse-specificity of LTP. The results also showthat autophosphorylation makes CaMKII activity persistfor much longer than the duration of calcium elevation.The slow rate at which active CaMKII turns off (~1 min)

means that during this period, the enzyme acts as a bio-chemical integrator of the multiple calcium pulses thatgenerally occur during LTP induction. This integrationis important for the full activation of CaMKII that isrequired to produce LTP.

Investigations have examined the mechanisms ofCaMKII activation, and the results of these studiesprovide a framework for understanding why activationof the kinase is localized to the stimulated spine. Theprimary source of calciumentry into spines is throughNMDARs. The opening of these channels requires boththe presynaptic release of glutamate and strong post-synaptic depolarization, which relieves the magnesiumblockade of these receptors23. These channel properties

Figure 1 | Synapse-specificity of the processes

involved in long-term potentiation induction.

a | A transient elevation of calcium (detected by the

calcium-sensitive dye Fluo4-FF; green signal) in a spine can

be observed in response to glutamate uncaging at that

spine (arrow). The calcium-insensitive dye Alexa-594

outlines the spine volume (red signal)35. b | Local glutamate

uncaging can potentiate AMPA receptor-mediated

excitatory postsynaptic currents (EPSCs) in a stimulated

spine but not in an adjacent spine154. c | After long-term

potentiation (LTP) induction by local glutamate uncaging

(arrow), transient activation of calcium/calmodulin-

dependent protein kinase II (CaMKII) can be observed (red

denotes high activation; level of activation measured byfluorescence resonance energy transfer (Camui sensor))22.

d | Stimulation of a spine (arrow) in this fashion can cause a

persistent increase in spine volume and translocation of

CaMKII to spines (here shown by the translocation of green

fluorescent protein-labelled CaMKII; red denotes highlevels of this kinase)41. Panel a is reproduced from REF. 35

(2009) Bentham Open. Panel b is reproduced, withpermission, from REF. 154(2008) American Association

for the Advancement of Science. Panelc is modified, with

permission, fromREF. 22(2009) Macmillan Publishers

Ltd. All rights reserved. Paneld is reproduced, with

permission, fromREF. 41(2008) National Academy

of Sciences.

R E V I E W S




4/14

Hebbian condition

The Hebb rule indicates that

for long-term synaptic

strengthening of excitatory

synapses to occur, two

conditions must be met: the

presynaptic input at that

synapse must be active and

the postsynaptic neuron must

be strongly depolarized by

multiple excitatory inputs.

Calcium nanodomain

A region that extends a few

tens of nanometres from a

calcium-permeable channel

and where calcium ions that

have come through the channel

are at a high concentration.

Intracellular signalling is

affected by the positioning of

calcium sensors within the

nanodomain.

allow NMDARs to compute the Hebbian condition that iscrucial for associative memory storage. Calcium indi-cators have been used to study calcium entry throughNMDARs into spines, and such studies have demon-strated that calcium elevation is localized to stimulatedspines22,2427(FIG. 1).

Spines largely contain two types of NMDARs: thosecontaining the NMDAR subtype 2A (NR2A) subunit andthose containing the NR2B subunit28. NR2A-containingNMDARs open briefly but reliably following glutamaterelease, whereas NR2B-containing receptors have alower probability of opening but continue to have manybrief openings for several hundred milliseconds afterglutamate release29,30. When calcium enters through anNMDAR, the concentration of calcium becomes highnear the inner mouth of the channel, a region called acalcium nanodomain31. Calcium then diffuses into the bulkof the spine head within microseconds.

Experiments have been conducted to determinewhether CaMKII activation depends on events thatoccur in the nanodomain or in the bulk of the spine

head; these were based on the calcium-buffering capabil-ities of BAPTA, a fast calcium buffer, and EGTA, a slowcalcium buffer (the steady-state dissociation constant issimilar for both of these buffers). As calcium traversesthe nanodomain quickly, only a fast buffer can blockcalcium elevation in the nanodomain, whereas either afast or a slow buffer can block calcium elevation in thebulk of the spine head. The aforementioned experiments

showed that CaMKII activation can be inhibited byeither BAPTA or EGTA, implying that activation ofthis enzyme is dependent on an elevation of calcium inthe bulk of the spine head22. However, a nanodomainprocess may also be required for such activation, as abulk calcium signal, when produced by depolarizationof voltage-dependent calcium channels, is not sufficientto activate CaMKII.

The actual calcium sensor for CaMKII activation isnot the enzyme itself but the small diffusible moleculecalmodulin. A new view of calmodulin activation hasbeen recently presented32. This view is based on meas-urements that show that the on-rate constant for calciumbinding to calmodulin is significantly higher than it isfor calcium binding to calbindin, the other major cal-cium buffer in pyramidal cells. Reconstitution experi-ments and computer simulations indicate that becauseof the high intracellular concentration of calmodulin(~100 M) and its high on-rate constant for calciumbinding, this protein can account for much of what hasbeen termed the fast calcium buffer27,33. From a signal-

ling perspective, these conditions make sense: a substan-tial fraction of the calcium that enters the spine becomesbound to calmodulin, making it an excellent detectorof calcium and an eff icient activator of CaMKII subu-nits, which are present in spines at a concentration of~100 M34.

The ability of local activation of NMDARs to lead tolocal activation of CaMKII is not trivial because calcium,calmodulin and CaMKII can all diffuse. For a smalldiffusible molecule, the diffusion length scale is of theorder of 1 m per ms. A spine can be as close as 1 mto another, yet somehow LTP remains localized to thestimulated spine. A rather satisfactory biophysical expla-nation can now be given for how the synapse-specificityof CaMKII activation is achieved (BOX 2). In brief, thenarrow spine neck slows diffusion out of the spine, andthe lifetime of the activated state of these molecules isshort compared with the time that would be required fordiffusion to carry the signal to other spines35,36.

Translocation and binding to the NMDA receptor

Once activated by calcium entry, CaMKII moves fromthe cytoplasm to the synapse, a phenomenon calledtranslocation37. The drivers of this translocation aresimple diffusion and the binding of activated CaMKIIto synaptic proteins, most importantly NMDARs (seebelow). However, the translocation of this enzyme is also

driven by spine enlargement: the resulting increase inF-actin in spines leads to an increase in the number ofCaMKIICaMKII heteromers that are bound to thiscytoskeletal protein3840.

An important issue is whether translocation is syn-apse-specific. This was demonstrated to be the case inexperiments using two-photon glutamate uncaging toinduce LTP at an identified spine (FIG. 1). Experimentsinvolving a second method (synaptically evoked LTP)showed similar results: in this case, active spines wereidentified by synaptically induced calcium elevation41.

Electron microscopy immunolabelling42 has beenused to determine whether some of the CaMKII that

Box 2 | Biophysical basis of the synapse-specificity of long-term potentiation

In neurons, calcium, calmodulin and calcium/calmodulin-dependent protein kinase II

(CaMKII) are all diffusible (albeit not freely), but the narrow spine neck that connects

the spine head to the dendrite helps the compartmentalization of these molecules35,36.

In general, any signal in the spine will be localized to the spine if the decay rate of its

activity is faster than the spineneck diffusion coupling. Early experiments using

two-photon imaging of caged fluorescein showed that the spineneck diffusion

coupling time can be estimated as:

t=Vl/(Ds)where Vis the spine volume, l is the spine neck length, s is the cross sectional area of the

spine neck and D is the diffusion constant153. Assuming V =~0.2m3, l =~1m ands= ~0.052, tcan be calculated as ~0.1 s for small molecules, with D of the order of100m/s2. Thus, the active state of a small signalling molecule will be localized to thespine if the lifetime of this state is


5/14

Postsynaptic density

(PSD). A structure, rich in

scaffolding proteins and

enzymes, that is attached to

the postsynaptic membrane.

translocates into the spine binds to the postsynapticdensity (PSD). In this study, LTP induction, which waschemically induced, produced a twofold increase inthe amount of CaMKII per unit length of the PSD, asmeasured one hour after LTP induction (FIG. 2). Otherwork has revealed the molecular processes that con-tribute to CaMKII binding to the PSD. Of note, acti-

vated CaMKII has been shown to bind NMDARs,most strongly to NR2B43,44. Several binding sitesinvolved in the formation of CaMKIINMDAR com-plexes have been identified. The region near S1303 inNR2B binds activated CaMKII in a way that does notrequire T286 phosphorylation; a second binding site inNR2B between amino acids 839 and 1120 does requireT286 phosphorylation45. An I205K-encoding muta-tion in CaMKII, which affects the so-called T-siteof the enzyme (BOX 1), blocks complex formation andstrongly attenuates CaMKII translocation in HEK293cells that express CaMKII and NMDARs46. Activationof NMDARs in hippocampal slices leads to a twofoldincrease in the amount of the CaMKIINMDAR

complex, as determined by co-immunoprecipitationexperiments43(FIG. 2). In experiments in which CaMKIItranslocation was caused by ionophore-mediated cal-cium entry, once CaMKII became bound to an NMDAR

via the T-site, the complex persisted for more than30 min after the removal of calcium47. Importantly, thebinding of CaMKII to NR2B locks the enzyme in anactive state45 (for conflicting results, see REF. 48).

Formation of the CaMKIINMDAR complex seemsto have a key role in LTP induction and learning. LTPis greatly reduced by the overexpression of mutant vari-ants of NR2B (variants with mutations affecting resi-dues near S1303; AS/QD) that block CaMKII bindingto the NMDAR49(FIG. 2). The results of this study arein agreement with other experiments showing that totaldeletion of the cytoplasmic tail of NR2B can block LTPdespite having no effect on NMDAR-mediated cur-rents50. Transgenic mice have been generated that over-express the cytoplasmic tail of NR2B, which interfereswith the formation of the CaMKIINMDAR complexand reduces the level of this complex by 2050%51. Theseanimals have a reduction in LTP of ~50% and exhibitlearning deficits.

Although the NMDAR is probably of primary impor-tance to the initiation of CaMKII binding to the PSD,other CaMKII-binding proteins may also have a role inthis process. PSDs in the average spine contain approxi-

mately 80 CaMKII holoenzymes and the molar ratioof PSD95 (a synaptic scaffolding protein also knownas disks large homologue 4 (DLG4)) to the NMDARNR1 subunit is approximately 7:1 (REFS 52,53). Thus,assuming that each NMDAR has two NR1 subunits,the PSD95:NMDAR molar ratio is 14:1. The absolutenumber of PSD95 molecules is ~300 (REFS 52,54), so arough estimate of NMDARs per synapse is 2025. Thus,many of the CaMKII holoenzymes may be held in thePSD by other binding partners, including leucine-richrepeat-containing protein 7 (LRRC7 also known as den-sin 180)55, -actinin55, DLG1 (also known as SAP97)56and multiple PDZ domain protein (mPDZ)57. Also,

CaMKII is known to associate with other channels andreceptors, including L-, T- and P/Q-type voltage-gatedcalcium channels5860 and dopamine receptors61. Manyof these channels and receptors are associated with thePSD and could therefore contribute to CaMKII bind-ing. CaMKII holoenzymes can also bind to other holo-enzymes in a self-association process62. Such bindingserves to amplify the amount of CaMKII that is boundin the PSD and probably gives rise to the CaMKII towersthat are observed in the PSD63. CaMKII in the PSD hasspecial importance in the control of synaptic strength(see below), even though it is only a small fraction of theCaMKII in spines34.

In summary, the available data point to the followingmodel: activated CaMKII diffuses to the synapse andbinds to NR2B, an event that is necessary for LTP (FIG. 3).Other binding partners may act to further enhancethe amount of CaMKII in the PSD and spine head.Interestingly, the number of NR2B-containing NMDARsis independent of synapse size53. Thus, one possibility isthat the amount of the CaMKIINMDAR complex

is graded; with increasing rounds of stimulation, NR2Bsubunits eventually become saturated with CaMKII.This scenario could potentially explain why the strongera synapse becomes, the less it can be potentiated64.Similarly, this scenario would explain why the higher thefraction of CaMKII that is bound in the spine, the fewerthe number of CaMKII molecules that can be addedto the spine by further LTP42.

The enzymatic and structural processes by whichCaMKII enhances AMPAR-mediated transmission arediscussed below.

Potentiation of AMPA receptor function

The enhancement of AMPAR-mediated transmissionduring LTP occurs through two processes (FIG. 3). Oneprocess increases the average single channel conduct-ance of AMPARs, which can be measured by non-sta-tionary noise analysis65. A similar increase occurs afteroverexpression of an active truncated form of CaMKII(REF. 66). The second process involves an increase in thenumber of AMPARs at the synapse67,68. One of the stud-ies that revealed this process showed that LTP producedan increase in the rectification of the EPSC of the typeexpected from insertion into the synapse of AMPARglutamate receptor 1 (GluR1) subunit homomers(GluR1 was overexpressed in this study). Further analy-sis revealed that extrasynaptic AMPARs that diffused

into the synapse could be phosphorylated by CaMKII(REF. 68). This phosphorylation occurs on the AMPARauxiliary protein stargazin and allows the receptorsto bind to PSD95 (REFS 69,70). In this way, AMPARsbecome captured by the synapse. Below we review theevidence for these processes.

GluR1 contains several important phosphorylationsites: S831, which is phosphorylated by protein kinase C(PKC) or CaMKII (REFS 71,72); S567, which is phospho-rylated by CaMKII (REF. 73); S845, which is phosphoryl-ated by protein kinase A (PKA)74; and S818, which isa substrate for PKC75. Evidence now suggests that theincrease in AMPAR conductance that is observed during

R E V I E W S




6/14

a

e f

b c d

* ***

*

*

**

0.0

0 10 20 30 40 50

0.2

0.4

0.6

0.8

1.0

Cumulativeprobability

CaMKII particles/m of PSD

NR2B

NR1

1 2 3 4

196 kDa

116 kDa

CaMKII

Vehi

cle

NMDA

MK8

01/NMDA

KN62

/NMDA

Time (min) Time (min) Time (min)

0

0

0 20 40 60 80 0 20 40 60 80100 120

0.0

0.5

1.0

1.5

2.0

1

2

3

4

10 20

ControlNR1 + NR2B RS/QD

NormalizedEP

SCamplitude

Normalize

dEPSPslope

4T 1T 4T

tatCN21

PSD

Before LTP

Afer LTP

100 nm

the early stages of LTP is due to S831 phosphorylation.Of note, the phosphorylation of this site increases during

LTP76,77. Moreover, S831 pseudophosphorylation blocksthe increase in AMPAR conductance that is observedwhen active CaMKII is expressed78(FIG. 4). Finally, aknock-in mutation that encodes S831 pseudophospho-rylated GluR1 increases channel conductance. Single-channel recordings show that S831 phosphorylationdoes not change the magnitude of any of the conduct-ance states of AMPARs but, rather, enhances the prob-ability that high conductance states will be activated byintermediate glutamate concentrations78. Interestingly,for AMPARs containing both GluR1 and GluR2, theability of S831 phosphorylation to increase channelconductance only occurs if stargazin is present 78,79.

Thus, an understanding of the interactions between thesubunits within the AMPAR is likely to provide mecha-

nistic insight into channel regulation. Given that S831phosphorylation occurs during LTP and that such phos-phorylation increases the average channel conductance,one would expect that a knock-in mutation encodingGluR1 S831A by itself would decrease the magnitudeof LTP. Such an effect, however, is not observed unlessGluR1 S845 phosphorylation is also prevented80. Thus,the increase in AMPAR conductance during LTP maydepend on interactions between these two sites.

The CaMKII-dependent phosphorylation of GluR1probably occurs at the synapse itself. This scenario issuggested by the fact that extrasynaptic membranepatches that are excised after LTP induction have a

Figure 2 | Role of CaMKIINMDAR complex in long-term potentiation maintenance. a,b | Electron microscopy

immunolabelling has been used to measure calcium/calmodulin-dependent protein kinase II (CaMKII) levels in the

postsynaptic density (PSD) before (a) and after (b) long-term potentiation (LTP) (CaMKII immunolabelling is indicated by

asterisks). c | The histogram shows that there is more CaMKII per unit length of the PSD after LTP induction than before

induction of LTP42. d | The CaMKII NMDA-type glutamate receptor (NMDAR) complex can be immunoprecipitated with

an antibody that detects CaMKII (lane 1). Synaptic stimulation with NMDA increases the amount of NMDAR subunits

NR2B and NR1 that are found in complexes with CaMKII (lane 2). These increases can be blocked by the NMDAR

antagonist MK801 (lane 3) or by the CaMK inhibitor KN62 (lane 4)43

. e | The overexpression of a mutant form of NR2B (RS/QD), which has a low binding affinity for CaMKII, blocks LTP (timing of LTP induction is denoted by the arrow)49. f| LTP can

be reversed by the CN compound tatCN21, a peptide that interferes with CaMKII binding to NR2B. In this example, LTP

was induced by 4 tetani (T), and an additional tetanus had no effect on the excitatory postsynaptic potential (EPSP),

indicating that LTP saturation had been reached (left-hand side). Subsequent application of tatCN21 reversed LTP. After

30 min of tatCN21 application, the peptide was removed and only partial recovery of LTP was observed (note that there

was an acute inhibitory effect that was reversed when the peptide was removed). Additional LTP could then be induced

using a 4T protocol (right-hand side)108. Panels a, b and c are reproduced, with permission, from REF. 42 (2004) Society

for Neuroscience. Panel d is reproduced, with permission, from REF. 43 (1999) National Academy of Sciences. Panel e is

modified, with permission, from REF. 49 (2005) Elsevier. Panel fis reproduced, with permission, from REF. 108 (2011)

Society for Neuroscience. EPSC, excitatory postsynaptic current.

R E V I E W S




7/14

Dendrite Spine head

CaMKII

AMPAR

NMDAR

P

P

Ca2+

Ca2+ bulk

Ca2+/calmodulin

?

RASERKsignalling

Translocationof CaMKIIto synapse

Activationof CaMKII

Ca2+

nanodomain

P

Exoctyosisof AMPAR-containingvesicles

Stargazin on mobileAMPARs is phosphorylated,allowing binding ofstargazin to PSD95 andthereby immobilizingAMPARs at the synapse

Phosphorylation of GluR1S831 increases the averagesingle channel conductanceof AMPARs

PSD

Diffusion of CaMKII

from dendrite tospine head

P

Stargazin

PPSD95

higher number of AMPARs than such patches havebefore LTP induction (see below) but show no changein single-channel conductance81. The possibility thatphosphorylation occurs at the synapse itself is furthersupported by the observation that activation of CaMKIIin PSD preparations leads to strong phosphorylation ofGluR1 S831 (REF. 82) (for conflicting results, see REF. 83).

As stated, the second mechanism that leads to theenhancement of AMPAR-mediated currents duringthe early phases of LTP is the insertion of AMPARsinto the synapse. One process driving this mechanism

could involve the exocytosis of AMPAR-containingvesicles into the plasma membrane, thereby supplyinga pool of receptors that could be inserted into the syn-apse8486. Such exocytosis has been shown to occur byimaging of pHluorin-tagged AMPARs that only exhibitfluorescence when they are exposed to the high pHextracellular environment. After induction of LTP atsingle spines, a rapid increase occurs in the rate of exo-cytotic events, a rise that lasts for about 1 minute. Theseevents occur in the stimulated spine and also in den-drites within 3 m of the stimulated spine. Exocytosisdepends on the RASextracellular signal-regulatedkinase (ERK) pathway86, RAB-GTPase proteins, soluble

N-ethylmaleimide-sensitive-factor attachment proteinreceptors (SNARE proteins), syntaxin 4 and 13, and themotor protein myosin V. However, inhibition of anyof these proteins does not affect the first 20 minutes ofLTP, although it does strongly reduce potentiation atlater times8794(FIG. 4). The fact that inhibition of exocy-tosis does not affect the first 20 minutes of potentiationsuggests that this early phase is due to the capture ofAMPARs that are present extrasynaptically before LTPinduction85.

The identity of the calcium sensor that triggers the

exocytosis pathway is unknown. The RASERK pathwaycan be stimulated by CaMKII (REF. 94), but exocytosisis only partially blocked by the CaMK inhibitor KN62(REF. 86). This finding raises the possibility that there areother calcium sensors for exocytosis, perhaps one ofthe calcium-sensitive RAS activators (guanine nucleo-tide exchange factors (GEFs)) or inactivators (GTPase-activating proteins (GAPs))95.

Insertion of AMPARs into the extrasynaptic mem-brane is not sufficient to enhance transmission; trans-location of these channels into the synapse requiresthe phosphorylation of stargazin, probably by CaMKII(REFS 69,83). This requirement was demonstrated by

Figure 3 | Role of CaMKII in AMPAR-mediated transmission during early long-term potentiation. Calcium entry

through NMDA-type glutamate receptors (NMDARs) leads to the activation of calmodulin both in the bulk cytoplasm and

in the channel nanodomain. Calmodulin activates calcium/calmodulin-dependent protein kinase II (CaMKII), which then

translocates to the postsynaptic density (PSD), where it enhances AMPA-type glutamate receptor (AMPAR)-mediated

transmission in two ways. First, CaMKII phosphorylates AMPAR glutamate receptor 1 (GluR1) subunits at S831, which

leads to an increase in the average conductance of such channels. Second, CaMKII phosphorylates the AMPAR-binding

protein stargazin, which causes stargazin to bind PSD95, thereby increasing the number of AMPARs at the synapse.

CaMKII activity, perhaps with other calcium-dependent processes, stimulates the fusion of vesicles containing AMPARs

with the plasma membrane, increasing the extrasynaptic concentration of such channels. This increase is not necessary for

the earliest stage of long-term potentiation, but is necessary for subsequent stages. Activated CaMKII also has effects on

protein synthesis, spine and synapse growth, proteolysis, microtubule motors and enhancement of inhibition. Other

binding partners of this kinase include multiple PDZ domain protein (mPDZ), actin, -actinin, leucine-rich

repeat-containing protein 7 (LRRCT) and disks large homologue 1 (DLG1). ERK, extracellular signal-regulated kinase.

R E V I E W S




8/14

a

c

e

d

b10

8

6

4

2

0

C

onductance(pS)

Control

CaMKII

Wild-type

C57BI/6GluR1-

S831A

GluR1-

S831D,S845D

GluR1-

S831A,S845A

Knock-in mice

GluR2 GluR2 + tCaMKII

%F

reque

ncy

%F

reque

ncy

Diffusion (m2/s) Diffusion (m2/s1)

Immobile Mobile MobileImmobile

10

0

105 103 101

20

10

0

105 103 101

20

0

100

200

300

400

500

600

5 0 0 10 20 30

0

100

300

500

60 20 20 60

E

PSC

(%o

fbaseline)

EPSC

slope(%)

EPSC

slope(%)

Time (min)

Time (min)

BTX Inactivated BTX

0

100

300

500

60 20 20 60Time (min)

0

1

2

3

0 10 20 30

Time (min)

Wild type

Control

S9A

S9D

RelativeAMPARcurrent

WT CaMKII

CaMKII T286A

Figure 4 | Mechanisms of expression processes by which CaMKII enhances AMPAR-mediated transmission.

a | Calcium/calmodulin-dependent protein kinase II (CaMKII) enhances AMPAR conductance in isolated CA1 pyramidal

cells of wild-type mice but not in knock-in mice in which phosphorylation of the AMPA-type glutamate receptor (AMPAR)

subunit glutamate receptor 1 (GluR1) on S831 or both S831 and S845 is prevented by mutation of the serine residues to

alanine residues. Pseudophosphorylation of these sites (that is, expression of S831D or S845D GluR1) enhances AMPAR

conductance and prevents further enhancement by CaMKII (REF. 78). b | In addition to affecting AMPAR conductance,

CaMKII affects the mobility of these receptors. Single-particle tracking experiments have shown that in neurons under

control conditions (left-hand panel), GluR2 subunits exhibit rapid diffusion that follows a broad path (three examples of

particle paths are shown at the top and a histogram of diffusion constants for GluR2-labelled particles is shown at the

bottom). In neurons that overexpress an active (truncated) form of CaMKII (tCaMKII; right-hand panel), GluR2 molecules

show limited diffusion paths (top) and a lower average diffusion constant (bottom), as AMPARs are captured at the

synapse70. c | The influence of CaMKII on AMPAR-mediated transmission can be observed in long-term potentiation (LTP)

experiments. LTP can be induced by a pairing protocol (arrow denotes the point of protocol initiation) in wild-type CA1

pyramidal cells but is prevented in cells expressing a mutant variant of CaMKII (T286A) that cannot become persistentlyactive12. d | Stargazin has an important role in CaMKII-enhanced AMPAR-mediated transmission. In control CA1 pyramidal

cells or cells expressing wild-type stargazin, LTP can be induced (arrow denotes start of LTP induction protocol). However,

LTPis blocked in CA1 pyramidal cells expressing a mutant variant of stargazin in which the nine potential CaMKII or

protein kinase C serine phosphorylation sites are mutated to alanine residues (S9A). Pseudophosphorylation of stargazin

at these sites (S9D) led to enhanced AMPAR-mediated transmission that occluded LTP69. e | CaMKII activity can stimulate

the exocytosis of AMPAR-containing vesicles; however, this does not seem to be important for the earliest phase of LTP.

Postsynaptic injection of botulinum toxin (BTX), which blocks exocytosis, before induction of LTP (denoted by arrow) does

not affect the earliest phase of LTP but does block later phases (left-hand side); in the control condition, inactivated BTX is

injected (right-hand side)91. Panel a is reproduced, with permission, from REF. 78 (2011) Macmillan Publishers Ltd. All

rights reserved. Panel b is reproduced, with permission, from REF. 70 (2010) Elsevier. Panel c is modified, with permission,

from REF. 12 (1998) American Association for the Advancement of Science. Panel d is modified, with permission, from

REF. 91 (1998) American Association for the Advancement of Science. Panel e is reproduced, with permission, from REF.

69 (2005) Elsevier. EPSC, excitatory postsynaptic current.

R E V I E W S




9/14

mutating the putative CaMKII phosphorylation sitesof stargazin to preclude phosphorylation. Althoughthese mutations did not affect the surface expressionof AMPARs, they did prevent these channels fromenhancing synaptic transmission after LTP induction(FIG. 4). Conversely, pseudophosphorylation of stargazinled to enhanced transmission that occluded LTP. Thesefindings suggest that LTP-induced phosphorylation ofstargazin is required for the capture of extrasynapticAMPARs by the synapse.

This capturing process has now been directlyobserved using quantum-dot labelling of AMPARsand single-particle tracking70. Under basal conditions,AMPARs diffused freely in the extrasynaptic space,pausing only briefly at synaptic locations. Upon activa-tion of CaMKII by NMDA application or synaptic input,diffusing AMPARs were captured at synapses (FIG. 4).The arrest of AMPARs at stimulated synapses isdependent on the phosphorylation of stargazin, asreceptors with non-phosphorylatable stargazin did notexhibit immobilization at synapses. Immobilization

of AMPARs could also be blocked by inhibition ofCaMKII. Thus, although stargazin can be phosphoryl-ated by either CaMKII or PKC, CaMKII seems to havethe dominant role in stimulating the capture process.Capture was also prevented by expressing a mutant vari-ant of CaMKII that is unable to bind NMDARs but stillallows CaMKII activation. Given that NR2B is at thesynapse, these results suggest that the stargazin phos-phorylation that triggers AMPAR capture occurs at thesynapse rather than extrasynaptically.

It was initially thought that AMPARs were trappedin the synapse through the association of GluR1 subu-nits with PDZ-containing PSD proteins, but this nowseems not to be the case96. Rather, the synaptic localiza-tion of AMPARs depends on the binding of stargazinto PSD95 (REFS 97,98). Synaptic clustering of AMPARscan be inhibited by mutating the binding sites on star-gazin that bind to PSD95 or by mutating the bindingsite on PSD95 that binds stargazin98. As noted earlier,the early phase of LTP can be blocked by deleting theCaMKII phosphorylation sites on stargazin. If the roleof this phosphorylation is to allow stargazin to bind toPSD95, then early LTP should also be blocked by delet-ing the PDZ1 and PDZ2 domains in PSD95 to whichstargazin binds. Consistent with this prediction, overex-pression of PSD95 lacking these domains prevents LTPinduction99.

Additional evidence for the importance of the star-gazinPSD95 interaction comes from the study of basaltransmission in CA1 (REF. 100). Application of a ligandthat disrupts both PDZ1 and PDZ2 interactions pro-duced a rapid reduction in EPSC and miniature EPSC(mEPSC) amplitude. The fact that the saturating con-centration of the ligand produced only a 45% reduc-tion in these currents indicates that only a componentof AMPAR-mediated transmission is dependent on thestargazinPSD95 interaction. This notion is consistentwith previous work showing that multiple mechanismsunderlie basal AMPAR-mediated transmission at CA1synapses101,102.

A structural picture for how phosphorylation ofstargazin leads to the anchoring of receptors at the syn-apse is beginning to emerge. The PDZ-binding site onstargazin is normally situated close to the membrane,where it can be held by electrostatic interactions withpositively charged lipids103. Synaptic PSD95 has been

visualized by electron microscope tomography and isoriented perpendicular to the membrane with the PDZ1and PDZ2 domains that bind stargazin situated quitefar (15 nm) from the membrane104. CaMKII phospho-rylation of stargazin may interfere with the electrostaticinteractions at the membrane, allowing the PDZ ligandsites on stargazin to extend from the membrane andbind to PSD95.

Thus, there is now strong support for a relatively sim-ple model of early LTP (FIG. 3). The process begins withCaMKII activation. CaMKII then binds to the NMDAR,putting the kinase in close juxtaposition to GluR1 and tothe carboxy-terminal region of stargazin83. The resultingphosphorylation of GluR1 enhances AMPAR conduct-ance, while the resulting phosphorylation of stargazin

leads to the anchoring of additional AMPARs at thesynapse. Together, these processes enhance synaptictransmission by an enzymatic process. The duration ofthis enzymatic action may be only several minutes105, butthe consequences of CaMKII enzymatic activity may lastlonger because it takes time for the substrates to becomedephosphorylated.

Maintenance of long-term potentiation

We now turn our attention to the later phases of LTPand the processes that maintain synaptic strength. Lessis known about these processes and the specific role ofCaMKII. Most of the CaMKII in spines is active onlyfor about a minute after synaptic stimulation22, sug-gesting that CaMKII may exert its enzymatic effects foronly this brief period. Thus, one view is that the involve-ment of CaMKII in LTP is only transient and that dif-ferent downstream processes are required to maintainLTP106. Support exists, however, for an alternative sce-nario in which a small subset of CaMKII molecules hasa persistent role in LTP, a role that may be structuralrather than catalytic. Several lines of evidence pointto a long-lasting role. As mentioned above, formationof CaMKIINMDAR complexes is required for LTPinduction49(FIG. 2). In a neuronal cell-culture model ofLTP, these complexes, once formed, can be persistent47.Furthermore, the LTP-induced association of CaMKII

with the PSD can persist for at least 60 min after induc-tion (FIG. 2). Finally, large pools of CaMKIINMDARcomplexes exist in the hippocampus under basal condi-tions43, perhaps resulting from previous learning (FIG. 2).Thus, one hypothesis is that CaMKII has a role in basaltransmission, early LTP and late LTP, and that synap-tic strength is controlled by the amount of CaMKIINMDAR complex at the synapse.

A key test of any model of synaptic memory mech-anisms is to inhibit or disrupt the putative memorymolecule after LTP induction and show that saturatedLTP can be reversed. However, this result might sim-ply indicate interference of LTP expression; that is, the

R E V I E W S




10/14

CN compounds

Peptides that are derived from

an endogenous protein that

inhibits calcium/calmodulin-

dependent protein kinase II.

processes that couple the memory mechanism to thechanges in AMPAR properties. To show that a main-tenance mechanism has been affected, two fur thertests are required. First, the attacking molecule mustbe subsequently removed; if an expression process wasaffected, LTP will recover, whereas if a maintenanceprocess was affected, LTP will not recover. Second, toexclude the possibility that the manipulations havedamaged a cell, it must be shown that LTP can bere-induced.

Tests of this kind have been conducted to examinethe role of the CaMKIINMDAR complex in the main-tenance of LTP. These tests used CN compounds, whichtogether comprise a class of CaMKII inhibitor that isknown to interfere with the binding of CaMKII to NR2Bin vitro107. Transient treatment of hippocampal slicecultures with a CN compound reversed LTP, allowingadditional LTP to be induced108(FIG. 2). In parallel exper-iments, it was shown that a CN inhibitor reduced theamount of the CaMKIINMDAR complex in the slices.

One conclusion from these experiments is that CN

inhibitors reverse LTP by interfering with the bindingreactions of CaMKII rather than the catalytic actions ofthe kinase. CN compounds bind strongly to the T-siteof CaMKII, to which the NMDAR also binds107. OtherCaMKII inhibitors bind mainly to the catalytic site ofthe enzyme (the S-site) and only weakly to the T-site,and do not reverse LTP109,110. These findings also explainwhy low concentrations of CN compounds that are suffi-cient to block CaMKII activity111, but are not sufficient todisrupt the CaMKIINMDAR complex, do not reverseLTP108. Taken together, these results suggest that in addi-tion to the catalytic action of CaMKII, which is impor-tant for synaptic strengthening during early LTP, thisenzyme is involved in a structural process that strength-ens transmission during late LTP.

Not much can be said with certainty about whatthis structural process might be, but there are severallines of relevant evidence. LTP is accompanied by arapid and persistent increase in spine volume (FIG. 1),an increase that can be mimicked by overexpres-sion of active CaMKII holoenzymes11. In addition tothe increase in spine volume, there seems to be a slowincrease in the size of the synapse itself (both pre- andpostsynaptically)112. The late phase of spine growth isdependent on protein synthesis113, which is stimulatedby brain-derived growth factor and dopamine andinvolves the selective targeting of newly synthesized

proteins to the synapses that have undergone LTP114116.Changes in protein synthesis are due to local controlmechanisms117 and to transcriptional changes118, andchanges in DNA methylation control some of theselong-term transcriptional changes119. Ultimately, thereare changes to the synthesis of many proteins, includ-ing CaMKII. A mutation in CaMKII that causes spe-cific interference with dendritic synthesis blocks lateLTP and greatly reduces the amount of CaMKII in thePSD120.

Another perspective on the structural changes thatoccur in late LTP focuses on changes in quantal trans-mission. According to one model155, the synapse is a

modular structure, with each module capable of anchor-ing ~20 AMPARs (the total number of AMPARs at thesynapse can be in the hundreds). During LTP, the addi-tion of AMPARs causes these modules to become func-tional while coordinated presynaptic processes increasethe size of the active zone and enhance the full-fusionmode of vesicle release. Thus, in the late phase of LTP,a synapse is stronger both because of enhanced gluta-mate release and because there are additional AMPAR-containing postsynaptic modules. The mechanisms bywhich these structural changes are triggered and therole of CaMKII in these processes are not known.

Conclusions and perspectives

Although there has been progress in understanding LTP,many important questions remain unanswered. Theability to optically measure LTP-induced calcium eleva-tion and CaMKII activation in single spines raises theprospect of understanding the initial processes of LTPat a quantitative level. Models of increasing sophistica-tion121,122 have been developed for how CaMKII is acti-

vated by calcium and calmodulin in vitro. Use of thesemodels to understand how CaMKII is activated in liv-ing cells requires knowledge of the state of calmodulin,about which less is known. The abundant protein neu-rogranin binds to calmodulin and releases it when cal-cium binds to calmodulin. Mutations in the neurograningene have strong effects on synaptic plasticity123, someof which can be accounted for by computational mod-els124. Direct measurements of the amounts of free andbound calmodulin will be important in providing theinformation that is necessary to quantitatively accountfor CaMKII activation.

The focus of this Review has been on T286 phospho-rylation, but there are other phosphorylation sites onCaMKII about which less is known. A short time afterT286 phosphorylation, T305 and T306 in the calmo-dulin-binding region become phosphorylated125. Thisdelayed phosphorylation strongly reduces autonomouskinase activity126,127. The phosphorylation of T305 andT306 has physiological consequences: overexpressionof a mutant variant of CaMKII that is pseudophos-phorylated at T286 actually leads to synaptic weaken-ing (and occlusion with long-term depression (LTD))unless T305 and T306 phosphorylation is prevented byadditional mutations128. Phosphorylation of T305 andT306 prevents CaMKII from binding to the PSDand causes learning deficits129. Understanding the con-

ditions that lead to T305 and T306 phosphorylationwill have medical implications because Angelmansdisease, a form of intellectual disability, is associatedwith enhanced phosphorylation at these sites that pre-

vents normal LTP130. CaMKII contains other phospho-rylation sites, for example T253, the functional role ofwhich is unclear131,132. Recent work shows that stimulithat induce moderate levels of calcium elevation causetranslocation of CaMKII to inhibitory synapses, leadingto enhancement of GABA

Areceptor insertion133. It will

be interesting to determine the properties of CaMKIIthat target this enzyme to inhibitory versus excitatorysynapses.

R E V I E W S




11/14

There has been substantial progress in understand-ing the role of CaMKII in LTP expression, but thereremain uncertainties. Experiments suggest that thedifferent expression processes (that is, enhancementof channel conductance versus an increase in channelnumber) may dominate at different synapses on thesame cell65. Why this should be the case is unclear. Therole of different GluR1 phosphorylation sites and theirinteractions remain unclear, particularly the role ofCaMKII-dependent phosphorylation of S567. Aspectsof the role of stargazin in LTP also remain uncertain. Itis likely that stargazin phosphorylation is a required stepfor the transition of AMPARs from the plasma mem-brane to the synapse, but the kinetics of this phospho-rylation step are not known. Furthermore, if the bindingof stargazin to PSD95 leads to AMPAR-mediated trans-mission, then the effects on LTP of deleting the PDZl-binding domain of stargazin should be equivalent to theexpression of a non-phosphorylatable mutant of thisprotein, but they are dramatically different, suggestingthat the stargazin-mediated mechanism involves addi-

tional complexity.The mechanisms of LTP maintenance remain contro-

versial. Evidence has been presented that protein kinase M(PKM) is responsible for the maintenance of LTP.Moreover, various forms of memory, as measured at thebehavioural level, can be erased by treatment with PKMinhibitors134,135. However, questions remain regardingthe specificity of zeta inhibitory protein (ZIP)115,136, themain pharmacological agent used to inhibit PKM, andit will be important to test the PKM hypothesis of LTPmaintenance by using knockout methods. A preliminaryreport indicates that late LTP is not affected by knock-ing out PKM expression and that the effects of ZIP stilloccur in knockout animals137.

An alternative to the PKM hypothesis is that LTPis maintained by the CaMKIINMDAR complex. Thismodel is supported by the ability of CN compounds toreverse LTP maintenance108(FIG. 2). Although it is knownthat formation of this complex is necessary for LTPand that the complex forms during chemical-inducedforms of LTP, methods need to be developed to moni-tor CaMKIINMDAR complexes during synapticallyinduced LTP. Such methods would make it possible todetermine whether the complexes that form have thestability to underlie long-term memory. The fractionof total spine CaMKII subunits that are bound to theNMDAR is probably too small (


12/14

1. Abraham, W. C. How long will long-term potentiation

last? Phil. Trans. R. Soc. B. 358, 735744 (2003).2. Whitlock, J. R., Heynen, A. J., Shuler, M. G. & Bear,

M. F. Learning induces long-term potentiation in the

hippocampus. Science313, 10931097 (2006).

3. Gruart, A., Munoz, M. D. & Delgado-Garcia, J. M.

Involvement of the CA3-CA1 synapse in the acquisition

of associative learning in behaving mice.J. Neurosci.

26, 10771087 (2006).4. Grant, S. G. & Silva, A. J. Targeting learning. Trends

Neurosci.17, 7175 (1994).

5. Kerchner, G. A. & Nicoll, R. A. Silent synapses and theemergence of a postsynaptic mechanism for LTP.

Nature Rev. Neurosci.9, 813825 (2008).

6. Lisman, J., Schulman, H. & Cline, H. The molecular

basis of CaMKII function in synaptic and behavioural

memory. Nature Rev. Neurosci.3, 175190 (2002).

7. Chao, L. H. et al.A mechanism for tunable

autoinhibition in the structure of a human Ca2+/

calmodulin- dependent kinase II holoenzyme. Cell

146, 732745 (2011).

A breakthrough paper that provides the first

crystal structure of the CaMKII holoenzyme. It

should be noted, however, that the crystallized

isoform is not the alpha or beta isoform, which are

the most prevalent forms in the brain.

8. Lisman, J. E. A mechanism for memory storage

insensitive to molecular turnover: a bistable

autophosphorylating kinase. Proc. Natl Acad. Sci. USA

82, 30553057 (1985).9. Miller, S. G. & Kennedy, M. B. Regulation of brain

type II Ca2+/calmodulin-dependent protein kinase by

autophosphorylation: a Ca2+-triggered molecular

switch. Cell44, 861870 (1986).

A study showing that purified CaMKII can become

persistently active following autophosphorylation.

10. Lledo, P. M. et al. Calcium/calmodulin-dependent

kinase II and long-term potentiation enhance synaptic

transmission by the same mechanism. Proc. Natl

Acad. Sci. USA 92, 1117511179 (1995).11. Pi, H. J. et al. CaMKII control of spine size and

synaptic strength: role of phosphorylation states and

nonenzymatic action. Proc. Natl Acad. Sci. USA107,

1443714442 (2010).

12. Giese, K. P., Fedorov, N. B., Filipkowski, R. K. & Silva,

A. J. Autophosphorylation at Thr286 of the alpha

calcium-calmodulin kinase II in LTP and learning.

Science279, 870873 (1998).

This paper demonstrates that mutation of a single

amino acid, T286, onCaMKII prevents LTP

induction and has profound effects on memory.13. Lucchesi, W., Mizuno, K. & Giese, K. P. Novel insights

into CaMKII function and regulation during memoryformation. Brain Res. Bull.85, 28 (2011).

14. Lee, Y. S. & Silva, A. J. The molecular and cellular

biology of enhanced cognition. Nature Rev. Neurosci.

10, 126140 (2009).

15. Hojjati, M. R. et al. Kinase activity is not required for

CaMKII-dependent presynaptic plasticity at CA3-CA1synapses. Nature Neurosci.10, 11251127 (2007).

16. Sanderson, J. L. & DellAcqua, M. L. AKAP signaling

complexes in regulation of excitatory synaptic

plasticity. Neuroscientist17, 321336 (2011).

17. Blitzer, R. D. et al. Gating of CaMKII by cAMP-

regulated protein phosphatase activity during LTP.

Science280, 19401942 (1998).

Paper discussing the control of CaMKII activation

by phosphatase and the role of cAMP in the control

of phosphatase.

18. Andersen, P., Sundberg, S. H., Sveen, O. &

Wigstrom, H. Specific long-lasting potentiation of

synaptic transmission in hippocampal slices. Nature

266, 736737 (1977).19. Lynch, G. S. , Dunwiddie, T. & Gribkoff, V.

Heterosynaptic depression: a postsynaptic correlate of

long-term potentiation. Nature266, 737739

(1977).

20. Matsuzaki, M., Honkura, N., Ellis-Davies, G. C. &

Kasai, H. Structural basis of long-term potentiation in

single dendritic spines. Nature429, 761766 (2004).

21. Takao, K. et al. Visualization of synaptic Ca2+ /

calmodulin-dependent protein kinase II activity in

living neurons.J. Neurosci.25, 31073112 (2005).

22. Lee, S. J., Escobedo-Lozoya, Y., Szatmari, E. M. &

Yasuda, R. Activation of CaMKII in single dendritic

spines during long-term potentiation. Nature458,

299304 (2009).

Study describing a breakthrough optical method

that allows for the determination of the spatial and

temporal properties of CaMKII activation during

LTP induction.

23. Ascher, P. & Nowak, L. Early biophysics of the NMDA

receptor channel.J. Physiol.587, 45634564

(2009).

24. Mainen, Z. F. et al. Two-photon imaging in living brain

slices. Methods18, 231239, (1999).

25. Emptage, N. J., Reid, C. A., Fine, A. & Bliss, T. V.

Optical quantal analysis reveals a presynaptic

component of LTP at hippocampal Schaffer-

associational synapses. Neuron38, 797804

(2003).

26. Kovalchuk, Y., Eilers, J., Lisman, J. & Konnerth, A.

NMDA receptor-mediated subthreshold Ca

2+

signalsin spines of hippocampal neurons.J. Neurosci. 20,

17911799 (2000).

27. Sabatini, B. L., Oertner, T. G. & Svoboda, K. The life

cycle of Ca2+ ions in dendritic spines. Neuron33,

439452 (2002).

28. Sobczyk, A., Scheuss, V. & Svoboda, K. NMDA

receptor subunit-dependent [Ca2+] signaling in

individual hippocampal dendritic spines.J. Neurosci.

25, 60376046 (2005).

29. Santucci, D. M. & Raghavachari, S. The effects of NR2

subunit-dependent NMDA receptor kinetics on

synaptic transmission and CaMKII activation. PLoS

Comput. Biol.4, e1000208 (2008).

30. Erreger, K., Dravid, S. M., Banke, T. G., Wyllie, D. J. &

Traynelis, S. F. Subunit-specific gat ing controls rat

NR1/NR2A and NR1/NR2B NMDA channel kinetics

and synaptic signalling profiles.J. Physiol.563,

345358 (2005).

31. Schneggenburger, R. & Neher, E. Presynaptic calcium

and control of vesicle fusion. Curr. Opin. Neurobiol.

15, 266274 (2005).32. Faas, G. C., Raghavachari, S., Lisman, J. E. & Mody, I.

Calmodulin as a direct detector of Ca2+ signals. Nature

Neurosci.14, 301304 (2011).

This paper provides evidence showing that

calmodulin binds calciumvery rapidly and so is

likely to be the first molecule to bind calcium after

it enters the cytoplasm. These properties make

calmodulin an ideal calcium signal detector.

33. Neher, E. & Augustine, G. J. Calcium gradients and

buffers in bovine chromaffin cells.J. Physiol.450,

273301 (1992).

34. Feng, B., Raghavachari, S. & Lisman, J. Quantitative

estimates of the cytoplasmic, PSD, and NMDAR-

bound pools of CaMKII in dendritic spines. Brain Res.

1419, 4652 (2011).

35. Lee, S. J. & Yasuda, R. Spatiotemporal regulation of

signaling in and out of dendritic spines: CaMKII and

Ras. Open Neurosci. J.3, 117127 (2009).

36. Yasuda, R. & Murakoshi, H. The mechanisms

underlying the spatial spreading of signalingactivity. Curr. Opin. Neurobiol. 21, 313321

(2011).

37. Shen, K. & Meyer, T. Dynamic control of CaMKII

translocation and localization in hippocampal

neurons by NMDA receptor stimulation. Science284,

162166 (1999).

A classic study that demonstrates that CaMKII

translocates towards the synapse after activation.38. Lin, B. et al. Theta stimulation polymerizes actin in

dendritic spines of hippocampus.J. Neurosci.25,

20622069 (2005).39. Ahmed, R., Zha, X. M., Green, S. H. & Dailey, M. E.

Synaptic activity and F-actin coordinately regulate

CaMKII localization to dendritic postsynaptic sites indeveloping hippocampal slices. Mol. Cell. Neurosci.

31, 3751 (2006).

40. Shen, K., Teruel, M. N., Subramanian, K. & Meyer, T.

CaMKII functions as an F-actin targeting module thatlocalizes CaMKII/ heterooligomers to dendritic

spines. Neuron21, 593606 (1998).41. Zhang, Y. P., Holbro, N. & Oertner, T. G. Optical

induction of plasticity at single synapses reveals input-

specific accumulation ofCaMKII. Proc. Natl Acad.Sci. USA105, 1203912044 (2008).

42. Otmakhov, N. et al. Persistent accumulation of

calcium/calmodulin-dependent protein kinase II in

dendritic spines after induction of NMDA receptor-

dependent chemical long-term potentiation.

J. Neurosci.24, 93249331 (2004).

Using electron microscopy, this study shows that

CaMKII translocation after LTP induction results in

a persistent increase in CaMKII in the postsynaptic

density.

43. Leonard, A. S., Lim, I. A., Hemsworth, D. E., Horne,

M. C. & Hell, J. W. Calcium/calmodulin-dependent

protein kinase II is associated with the N-methyl-

d-aspartate receptor. Proc. Natl Acad. Sci. USA96,

32393244 (1999).

44. Strack, S. & Colbran, R. J. Autophosphorylation-

dependent targeting of calcium/ calmodulin-dependent

protein kinase II by the NR2B subunit of the

N-methyl-d-aspartate receptor.J. Biol. Chem.273,

2068920692 (1998).

References 43 and 44 were the first studies to

demonstrate that activated CaMKII binds to the

NMDAR.

45. Bayer, K. U., De Koninck, P., Leonard, A. S., Hell, J. W.

& Schulman, H. Interaction with the NMDA receptor

locks CaMKII in an active conformation. Nature411,

801805 (2001).46. Strack, S., McNeill, R. B. & Colbran, R. J. Mechanism

and regulation of calcium/calmodulin-dependent

protein kinase II targeting to the NR2B subunit of the

N-methyl-d-aspartate receptor.J. Biol. Chem.275,

2379823806 (2000).

47. Bayer, K. U. et al. Transition from reversible to

persistent binding of CaMKII to postsynaptic sites and

NR2B.J. Neurosci.26, 11641174 (2006).

48. Robison, A. J., Bartlett, R. K., Bass, M. A. & Colbran,

R. J. Differential modulation of Ca2+/calmodulin-

dependent protein kinase II activity by regulated

interactions with N-methyl-d-aspartate receptor

NR2B subunits and -actinin.J. Biol. Chem. 280,3931639323 (2005).

49. Barria, A. & Malinow, R. NMDA receptor subunit

composition controls synaptic plasticity by regulating

binding to CaMKII. Neuron 48, 289301 (2005).

A key paper demonstrating that the binding of

CaMKII to the NMDAR is vital for LTP induction.

50. Foster, K. A. et al. Distinct roles of NR2A and NR2B

cytoplasmic tails in long-term potentiation.

J. Neurosci.30, 26762685 (2010).

51. Zhou, Y. et al. Interactions between the NR2B

receptor and CaMKII modulate synaptic plasticity and

spatial learning.J. Neurosci.27, 1384313853

(2007).

52. Chen, X. et al. Mass of the postsynaptic density and

enumeration of three key molecules. Proc. Natl Acad.

Sci. USA102, 1155111556 (2005).

53. Shinohara, Y. et al. Left-right asymmetry of the

hippocampal synapses with differential subunit

allocation of glutamate receptors. Proc. Natl Acad.

Sci. USA105, 1949819503 (2008).

54. Sugiyama, Y., Kawabata, I., Sobue, K. & Okabe, S.

Determination of absolute protein numbers in single

synapses by a GFP-based calibration technique.

Nature Methods2, 677684 (2005).

55. Walikonis, R. S. et al. Densin-180 forms a ternary

complex with the -subunit of Ca2+/calmodulin-dependent protein kinase II and -actinin. J. Neurosci.

21, 423433 (2001).56. Nikandrova, Y. A., Jiao, Y., Baucum, A. J., Tavalin, S. J.

& Colbran, R. J. Ca2+/calmodulin-dependent protein

kinase II binds to and phosphorylates a specific SAP97

splice variant to disrupt association with AKAP79/150

and modulate -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-type glutamate receptor

(AMPAR) activity.J. Biol. Chem.285, 923934

(2010).

57. Krapivinsky, G., Medina, I., Krapivinsky, L., Gapon, S.

& Clapham, D. E. SynGAP-MUPP1-CaMKII synaptic

complexes regulate p38 MAP kinase activity and

NMDA receptor-dependent synaptic AMPA receptor

potentiation. Neuron43, 563574 (2004).

58. Hudmon, A. et al. CaMKII tethers to L-type Ca2+

channels, establishing a local and dedicated integrator

of Ca2+ signals for facilitation.J. Cell Biol. 171,

537547 (2005).

59. Welsby, P. J. et al. A mechanism for the direct

regulation of T-type calcium channels by Ca2+/

calmodulin-dependent kinase II.J. Neurosci.23,1011610121 (2003).

60. Jiang, X. et al. Modulation of Cav2.1 channels by

Ca2+/calmodulin-dependent protein kinase II bound to

the C-terminal domain. Proc. Natl Acad. Sci. USA105,

341346 (2008).61. Liu, X. Y. et al. Activity-dependent modulation of limbic

dopamine D3 receptors by CaMKII. Neuron61,

425438 (2009).62. Hudmon, A. et al. A mechanism for Ca2+/calmodulin-

dependent protein kinase II clustering at synaptic and

nonsynaptic sites based on self-association.

J. Neurosci.25, 69716983 (2005).63. Petersen, J. D. et al. Distribution of postsynaptic

density (PSD)-95 and Ca2+/calmodulin-dependent

protein kinase II at the PSD.J. Neurosci.23,

1127011278 (2003) .

This paper provides the best visualization to date

of CaMKII within the PSD.

R E V I E W S




13/14

64. Liao, D., Jones, A. & Malinow, R. Direct measurement

of quantal changes underlying long-term potentiation

in CA1 hippocampus. Neuron9, 10891097

(1992).

65. Luthi, A. et al. Bi-directional modulation of AMPA

receptor unitary conductance by synaptic activity.

BMC Neurosci.5, 44 (2004).

66. Poncer, J. C., Esteban, J. A. & Malinow, R. Multiple

mechanisms for the potentiation of AMPA receptor-

mediated transmission by -Ca2+/calmodulin-dependent protein kinase II.J. Neurosci.22,

44064411 (2002).A key demonstration of the multiple ways in which

CaMKII can lead to synaptic strengthening.

67. Shi, S. H. et al. Rapid spine delivery and redistribution

of AMPA receptors after synaptic NMDA receptor

activation. Science284, 18111816 (1999).

68. Opazo, P. & Choquet, D. A three-step model for the

synaptic recruitment of AMPA receptors. Mol. Cell.

Neurosci.46, 18 (2011).

69. Tomita, S., Stein, V., Stocker, T. J., Nicoll, R. A. &

Bredt, D. S. Bidirectional synaptic plasticity regulated

by phosphorylation of stargazin-like TARPs. Neuron

45, 269277 (2005).

The strongest evidence to date for the key role of

CaMKII-dependent phosphorylation of stargazin

in LTP.70. Opazo, P. et al. CaMKII triggers the diffusional

trapping of surface AMPARs through phosphorylation

of stargazin. Neuron 67, 239252 (2010).

This paper tracks single AMPA channels and

provides the first visualization of the trapping of

the AMPAR at the synapse. The role of

CaMKII-dependent phosphorylation of stargazin in

this process is confirmed.

71. Barria, A., Derkach, V. & Soderling, T. Identification of

the Ca2+/calmodulin-dependent protein kinase II

regulatory phosphorylation site in the

-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate-type glutamate receptor.J. Biol. Chem.272,

3272732730 (1997).

72. Roche, K. W., OBrien, R. J., Mammen, A. L.,

Bernhardt, J. & Huganir, R. L. Characterization of

multiple phosphorylation sites on the AMPA receptor

GluR1 subunit. Neuron 16, 11791188 (1996).

73. Lu, W., Isozaki, K., Roche, K. W. & Nicoll, R. A.

Synaptic targeting of AMPA receptors is regulated by

a CaMKII site in the first intracellular loop of GluA1.

Proc. Natl Acad. Sci. USA107, 2226622271

(2010).

74. Banke, T. G. et al. Control of GluR1 AMPA receptor

function by cAMP-dependent protein kinase.

J. Neurosci.20, 89102 (2000).75. Boehm, J. et al. Synaptic incorporation of AMPA

receptors during LTP is controlled by a PKC

phosphorylation site on GluR1. Neuron51, 213225

(2006).

76. Lee, H. K., Barbarosie, M., Kameyama, K., Bear, M. F.

& Huganir, R. L. Regulation of distinct AMPA receptor

phosphorylation sites during bidirectional synaptic

plasticity. Nature405, 955959 (2000).77. Barria, A., Muller, D., Derkach, V., Griffith, L. C. &

Soderling, T. R. Regulatory phosphorylation of AMPA-

type glutamate receptors by CaM-KII during long-term

potentiation. Science276, 20422045 (1997).

78. Kristensen, A. S. et al. Mechanism of Ca2+/calmodulin-

dependent kinase II regulation of AMPA receptor

gating. Nature Neurosci.14, 727735 (2011).

This paper resolves a problem that the field had

been struggling with: under what conditions does

CaMKII-dependent phosphorylation of GluR1

enhance channel conductance? It is shown that this

form of regulation requires the presence ofstargazin.

79. Derkach, V. A., Oh, M. C., Guire, E. S. & Soderling,

T. R. Regulatory mechanisms of AMPA receptors in

synaptic plasticity. Nature Rev. Neurosci.8, 101113

(2007).

80. Lee, H. K., Takamiya, K., He, K., Song, L. & Huganir,

R. L. Specific roles of AMPA receptor subunit GluR1

(GluA1) phosphorylation sites in regulating synaptic

plasticity in the CA1 region of hippocampus.

J. Neurophysiol.103, 479489 (2010).

81. Andrasfalvy, B. K. & Magee, J. C. Changes in AMPA

receptor currents following LTP induction on rat CA1

pyramidal neurones.J. Physiol. 559, 543554

(2004).

82. Vinade, L. & Dosemeci, A. Regulation of the

phosphorylation state of the AMPA receptor GluR1

subunit in the postsynaptic density. Cell. Mol.

Neurobiol.20, 451463 (2000).

83. Tsui, J. & Malenka, R. C. Substrate localization creates

specificity in calcium/calmodulin-dependent protein

kinase II signaling at synapses.J. Biol. Chem.281,

1379413804 (2006).

84. Yudowski, G. A., Puthenveedu, M. A. & von Zastrow, M.

Distinct modes of regulated receptor insertion to the

somatodendritic plasma membrane. Nature Neurosci.

9, 622627 (2006).

85. Makino, H. & Malinow, R. AMPA receptor

incorporation into synapses during LTP: the role of

lateral movement and exocytosis. Neuron64,

381390 (2009).86. Patterson, M. A., Szatmari, E. M. & Yasuda, R. AMPA

receptors are exocytosed in stimulated spines and

adjacent dendrites in a Ras-ERK-dependent manner

during long-term potentiation. Proc. Natl Acad. Sci.

USA107, 1595115956 (2010).

87. Park, M., Penick, E. C., Edwards, J. G., Kauer, J. A. &

Ehlers, M. D. Recycling endosomes supply AMPA

receptors for LTP. Science305, 19721975 (2004).

88. Park, M. et al. Plasticity-induced growth of dendritic

spines by exocytic trafficking from recycling

endosomes. Neuron52, 817830 (2006).

89. Wang, Z. et al. Myosin Vb mobilizes recycling

endosomes and AMPA receptors for postsynaptic

plasticity. Cell135, 535548 (2008).90. Kennedy, M. J., Davison, I. G., Robinson, C. G. &

Ehlers, M. D. Syntaxin-4 defines a domain for activity-

dependent exocytosis in dendritic spines. Cell141,

524535 (2010).

91. Lledo, P. M., Zhang, X., Sudhof, T. C., Malenka, R. C. &

Nicoll, R. A. Postsynaptic membrane fusion and long-

term potentiation. Science279, 399403 (1998).92. Yang, Y., Wang, X. B., Frerking, M. & Zhou, Q. Spine

expansion and stabilization associated with long-term

potentiation.J. Neurosci.28, 57405751 (2008).

93. English, J. D. & Sweatt, J. D. A requirement for the

mitogen-activated protein kinase cascade in

hippocampal long term potentiation.J. Biol. Chem.

272, 1910319106 (1997).94. Zhu, J., Qin, Y., Zhao, M., Van Aelst, L. & Malinow, R.

Ras and Rap control AMPA receptor trafficking during

synaptic plasticity. Cell110, 443455 (2002).95. Cullen, P. J. & Lockyer, P. J. Integration of calcium

and Ras signalling. Nature Rev. Mol. Cell. Biol.3,

339348 (2002).

96. Kim, C. H. et al. Persistent hippocampal CA1 LTP in

mice lacking the C-terminal PDZ ligand of GluR1.

Nature Neurosci.8, 985987 (2005).

97. Schnell, E. et al. Direct interactions between PSD-95

and stargazin control synaptic AMPA receptor

number. Proc. Natl Acad. Sci. USA99, 1390213907

(2002).98. Bats, C., Groc, L. & Choquet, D. The interaction

between Stargazin and PSD-95 regulates AMPA

receptor surface trafficking. Neuron53, 719734

(2007).

References 97 and 98 provide the key evidence for

the role of stargazin interaction with PSD95 in

anchoring AMPARs at the synapse.

99. Ehrlich, I. & Malinow, R. Postsynaptic density 95

controls AMPA receptor incorporation during

long-term potentiation and experience-driven

synaptic plasticity.J. Neurosci.24, 916927

(2004).

100. Sainlos, M. et al. Biomimetic divalent ligands for the

acute disruption of synaptic AMPAR stabilization.

Nature Chem. Biol.7, 8191 (2011).

101. Kim, C. H. & Lisman, J. E. A labile component of

AMPA receptor-mediated synaptic transmission is

dependent on microtubule motors, actin, and

N-ethylmaleimide-sensitive factor.J. Neurosci.21,

41884194 (2001).102. Sanhueza, M., McIntyre, C. C. & Lisman, J. E. Reversal

of synaptic memory by Ca2+/calmodulin-dependent

protein kinase II inhibitor.J. Neurosci.27,

51905199 (2007).

103. Chen, X. et al. PSD-95 is required to sustain the

molecular organization of the postsynaptic density.

J. Neurosci.31, 63296338 (2011).

104. Dani, A., Huang, B., Bergan, J., Dulac, C. & Zhuang, X.

Superresolution imaging of chemical synapses in the

brain. Neuron68, 843856 (2010).

105. Lengyel, I. et al. Autonomous activity of CaMKII is only

transiently increased following the induction of long-

term potentiation in the rat hippocampus. Eur.

J. Neurosci.20, 30633072 (2004).

106. Murakoshi, H., Wang, H. & Yasuda, R.

Local, persistent activation of Rho GTPases during

plasticity of single dendritic spines. Nature472,

100104 (2011).

107. Vest, R. S., Davies, K. D., OLeary, H., Port, J. D. &

Bayer, K. U. Dual mechanism of a natural CaMKII

inhibitor. Mol. Biol. Cell18, 50245033 (2007).

108. Sanhueza, M. et al. Role of the CaMKII/NMDA

receptor complex in the maintenance of synaptic

strength.J. Neurosci.31, 91709178 (2011).

The elucidation of the molecular basis of memory

requires the demonstration that if a putative

memory mechanism is attacked, saturated LTP can

be reversed. Furthermore, it should then be

possible to re-induce LTP. This paper shows that

these criteria can be met using CN compounds,which reduce the amount of CaMKIINMDAR

complex.

109. Otmakhov, N., Griffith, L. C. & Lisman, J. E.

Postsynaptic inhibitors of calcium/calmodulin-

dependent protein kinase type II block induction but

not maintenance of pairing-induced long-term

potentiation.J. Neurosci.17, 53575365 (1997).

110. Tsien, R. W., Schulman, H. & Malinow, R. Peptide

inhibitors of PKC and CaMK block induction but not

expression of long-term potentiation.Adv. Second

Messenger Phosphoprotein Res. 24, 101107

(1990).

111. Buard, I. et al. CaMKII autonomy is required for

initiating but not for maintaining neuronal long-term

information storage.J. Neurosci.30, 82148220

(2010).

112. Ostroff, L. E., Fiala, J. C., Allwardt, B. & Harris, K. M.

Polyribosomes redistribute from dendritic shafts into

spines with enlarged synapses during LTP in

developing rat hippocampal slices. Neuron35,

535545 (2002).

113. Tanaka, J. I. et al. Protein synthesis and neurotrophin-

dependent structural plasticity of single dendritic

spines. Science319, 16831687 (2008).

114. Smith, W. B., Starck, S. R., Roberts, R. W. & Schuman,

E. M. Dopaminergic stimulation of local protein

synthesis enhances surface expression of GluR1 and

synaptic transmission in hippocampal neurons.

Neuron 45, 765779 (2005).

115. Lisman, J. Memory erasure by very high

concentrations of ZIP may not be due to PKM-zeta.

Hippocampus28 Sep 2011 (doi:10.1002/

hipo.20980).

116. Frey, U. & Morris, R. G. Synaptic tagging: implications

for late maintenance of hippocampal long-term

potentiation. Trends Neurosci.21, 181188 (1998).

117. Schuman, E. M., Dynes, J. L. & Steward, O. Synaptic

regulation of translation of dendritic mRNAs.

J. Neurosci.26, 71437146 (2006).

118. Bramham, C. R., Worley, P. F., Moore, M. J. &

Guzowski, J. F. The immediate early gene arc/arg3.1:regulation, mechanisms, and function. J. Neurosci.28,

1176011767 (2008).

119. Day, J. J. & Sweatt, J. D. Cognitive neuroepigenetics: a

role for epigenetic mechanisms in learning and

memory. Neurobiol. Learn. Mem.96, 212 (2011).

120. Miller, S. et al. Disruption of dendritic translation of

CaMKII impairs stabilization of synaptic plasticity andmemory consolidation. Neuron36, 507519 (2002).

121. Pepke, S., Kinzer-Ursem, T., Mihalas, S. & Kennedy,

M. B. A dynamic model of interactions of Ca2+,

calmodulin, and catalytic subunits of Ca2+/calmodulin-

dependent protein kinase II. PLoS Comput. Biol.6,

e1000675 (2010).

122. Byrne, M. J., Putkey, J. A., Waxham, M. N. &

Kubota, Y. Dissecting cooperative calmodulin binding

to CaM kinase II: a detailed stochastic model.

J. Comput. Neurosci.27, 621638 (2009).

123. Huang, K. P. et al. Neurogranin/RC3 enhances long-

term potentiation and learning by promoting calcium-

mediated signaling.J. Neurosci.24, 1066010669(2004).

124. Zhabotinsky, A. M., Camp, R. N., Epstein, I. R. &

Lisman, J. E. Role of the neurogranin concentrated in

spines in the induction of long-term potentiation.

J. Neurosci.26, 73377347 (2006).125. Hanson, P. I. & Schulman, H. Inhibitory

autophosphorylation of multifunctional Ca2+/

calmodulin-dependent protein kinase analyzed by site-

directed mutagenesis.J. Biol. Chem.267,

1721617224 (1992).

126. Coultrap, S. J., Buard, I., Kulbe, J. R., DellAcqua, M. L.

& Bayer, K. U. CaMKII autonomy is substrate-

dependent and further stimulated by Ca2+/calmodulin.

J. Biol. Chem.285, 1793017937 (2010).

127. Jama, A. M., Fenton, J., Robertson, S. D. & Torok, K.

Time-dependent autoinactivation of phospho-

Thr286-Ca2+/calmodulin-dependent protein kinase II.J. Biol. Chem.284, 2814628155 (2009).

R E V I E W S




14/14

128. Pi, H. J., Otmakhov, N., Lemelin, D., De Koninck, P. &

Lisman, J. Autonomous CaMKII can promote either

long-term potentiation or long-term depression,

depending on the state of T305/T306

phosphorylation.J. Neurosci.30, 87048709 (2010).

The surprising complexity of CaMKII regulation of

synaptic strength is demonstrated in this paper.

Notably, if T305 and T306 phosphorylation is

allowed to occur, activated CaMKII promotes LTD

rather than LTP.129. Elgersma, Y. et al. Inhibitory autophosphorylation of

CaMKII controls PSD association, plasticity, andlearning. Neuron36, 493505 (2002).

130. Weeber, E. J. et al. Derangements of hippocampal

calcium/calmodulin-dependent protein kinase II in a

mouse model for Angelman mental retardation

syndrome.J. Neurosci. 23, 26342644 (2003).

131. Skelding, K. A. et al. Regulation of CaMKII by

phospho-Thr253 or phospho-Thr286 sensitive

targeting alters cellular function. Cell. Signal.22,

759769 (2010).

132. Gurd, J. W. et al. Ischemia and status epilepitcus result

in enhanced phosphorylation of calcium and

calmodulin-stimulated protein kinase II on threonine

253. Brain Res.1218, 158165 (2008).

133. Marsden, K. C., Shemesh, A., Bayer, K. U. & Carroll,

R. C. Selective translocation of Ca2+/calmodulin

protein kinase II (CaMKII) to inhibitory synapses.Proc. Natl Acad. Sci. USA 107, 2055920564

(2010).134. Sacktor, T. C. How does PKM maintain long-term

memory? Nature Rev. Neurosci.12, 915 (2011).

135. Shema, R. et al. Enhancement of consolidated long-

term memory by overexpression of protein kinase Min the neocortex. Science331, 12071210 (2011).

136. Sacktor, T. C. & Fenton, A. A. Appropriate application

of ZIP for PKM inhibition, LTP reversal, and memoryerasure. Hippocampus4 Oct 2011 (doi:10.1002/

hipo.20978).

137.Volk, L. J., Bachman, J., Johnson, R. C., Yu, Y. &

Huganir, R. L. Insights into synaptic plasticity and

memory maintenance from the protein kinase C knockout mouse. Soc.Neurosci. Abstr. 238.06

(Washington DC, 13 Nov 2011).

138. Miller, P., Zhabotinsky, A. M., Lisman, J. E. & Wang,

X. J. The stability of a stochastic CaMKII switch:

dependence on the number of enzyme molecules and

protein turnover. PLoS Biol.3, e107 (2005).

139. Mullasseril, P., Dosemeci, A., Lisman, J. E. & Griffith,

L. C. A structural mechanism for maintaining the

on-state of the CaMKII memory switch in the post-

synaptic density.J. Neurochem.103, 357364

(2007).

140. Cheriyan, J., Kumar, P., Mayadevi, M., Surolia, A. &

Omkumar, R. V. Calcium/calmodulin dependent

protein kinase II bound to NMDA receptor 2B subunit

exhibits increased ATP affinity and attenuated

dephosphorylation. PLoS ONE6, e16495 (2011).

141. Irvine, E. E. et al. Properties of contex

camkii in learning and memory

Documents