Driving AMPA receptors into synapses by synaptic activity (LTP) or spontaneous activity during early postnatal development

• - Excitatory synaptic transmission in mature hippocampal neurons is mediated by the actions of glutamate, on two different ion-permeable receptors, NMDA-Rs and AMPA-Rs.

• -NMDA-Rs require membrane depolarization to transmit and are largely inactive at resting membrane potentials.

• Excitatory synaptic contacts initially contain primarily NMDA-Rs, and are electrophysiologically “silent” at normal resting potentials. Such synapses must acquire AMPA-Rs to form connections that can function.

• AMPA-Rs delivered to synapses during early postnatal development of the hippocampus or,

• in mature neurons, long-term potentiation (LTP) induced delivery AMPA-Rs from non-synaptic to synaptic regions.

• Escitatory actions at chemical synapses result from the opening of chennels permeable to both Na+ and K+.

• The magnitude and sign of the synaptic current is determined by the sum of the fluxes of K+ and Na+ through the synaptic conductance.

• The reversal potential for the synaptic current can be determined using a voltage clamp.

• At the resting membrane potential and at more negative clamped potentials(-70 and -80mV) the synaptic current is large and inward because the electrochemical driving force is inward.

• When the membrane potential is made less negative(-20mV), the magnitude of the inward synaptic current decreases.

• When the membrane potential is made more positive, the synaptic current is outward.

Electrophysiological tagging to monitor synaptic delivery of recombinant AMPA receptors

• AMPA Rs are oligomers formed by a combination of four different subunits, GluR1 to 4.

• Most endogenous AMPA-Rs contain the GluR2 subunit and can pass current equally well in both inward and outward directions.

• In contrast, AMPA-Rs lacking GluR2(or containing GluR2 that is genetically modified) exhibit profound inward rectification such that they can pass minimal current in the outward direction when the cell is depolarized to +40 mV.

• Rectification is an intrinsic biophysical property of a receptor to conduct ions more readily in one direction than in the other and can be detected as the ratio of the response observed at -60mV to that at +40mV.

• Thus, incorporation of recombinant AMPA-Rs into synapses and their contribution to real synaptic transmission can be monitored functionally.

• Transmission mediated by activity of AMPA-Rs on neurons expressing GluR1-GFP or GluR4-GFP shows significantly increased rectification.

Tetanic stimulation induces spine delivery and clustering of GluR1-GFPS-H Shi, at el, Science, 1999

• GluR1-GFP was introduced into neurons of organotypic hippocampal slice cultures with Sindbis virus expression system

• To test the effect of synaptic activity on receptor distribution, small glass-stimulating electrode was placed near a group of dendrites labeled with GluR1-GFP.

• In the absence of evoked activity, the GluR1-FP distribution pattern was stable for hours.

• Delivery of a brief tetanic stimulus, which was sufficient to induce LTP, produced a rapid redistribution of GluR1-GFP

•(A) Column 1, GluR1-GFP expression in apical dendritic region. Stimulation electrode was placed in nearby region (~5 to 10 µm from top left corner, outside imaged region). Column 2, region near stimulation electrode (top and middle: two different magnifications of same region) and another region (bottom) imaged before tetanus. a and b denote locations of interest. Column 3, same regions imaged 30 min after tetanic stimulation. Arrows mark regions a and b in column 2. Column 4, surface GluR1-GFP assessed with antibody to GFP immunostaining in nonpermeabilized fixation conditions. At region showing redistribution (top and middle), immunostaining detected increased GluR1-GFP on dendrite and spine-like structures. Scale bars, 2 µm.

NMDA receptor antagonist reversibly blocks tetanus-induced redistribution of GluR1-GFP.

• To determine whether the redistribution of GluR1-GFP by tetanic stimulation requires synaptic activation of NMDA receptors, the authors have tested the effect of (D,L)-2-amino-5-phosphono valeric acid(APV), a reversible NMDA receptor antagonist.

• (A) Images of apical dendritic segments obtained at different times during experimental period. (Top) In the presence of 100 µM D,L-APV, tetanus produces little change in fluorescence pattern. (Bottom) After 45 min of drug wash, tetanus at the same site produces both delivery to spines and clustering of GluR1-GFP. In this example, spine delivery persisted at 50 min after tetanus in only two of five spines and clustering reverted to smooth distribution. Time stamps are in minutes relative to tetanus. Scale bar, 5 µm.

- The CaMKII holoenzyme is a dodecamer of subunits held together through subunit association domains. In the absence of Ca/CaM, each subunit is inactive because the regulatory domain binds to the catalytic domain.- When Ca2+ enters the cell, it binds to CaM, which in turn binds to the regulatory subunit. The catalytic subunit is then released and can phosphorylate substrates.-As Ca2+ levels decline, CaM is released and the subunits return to their inactive state. The autoregulatory properties of CaMKII subunits within the holoenzyme make CaMKII an excellent sensor of intracellular Ca2+ concentration and oscillations.

• If two adjacent subunit should become bound to Ca/CaM simultaneously, one subunit can phosphorylate the T286 autophosphorylation site of its adjacent partner. Phosphorylation at T286 increases the subunit’s affinity for CaM and keeps the subunit in a constitutively active state.

• Because CaMKII autophosphorylation could result in Ca/CaM-independent kinase activity long after the initial Ca signaling, it was suggested the CaMKII autophosphorylation might represent a type of molecular memory.

• Long-term potentiation(LTP) of synaptic transmission is a form of activity-dependent plasticity likely to play important roles in learning and memory. CaMKII is thought to be a key mediator of this plasticity, and it strongly expressed at excitatory synapses.

Enhanced synaptic transmission in neurons expressing catalytic domain of CaMKII fused with GFP.

A) Hippocampal CA1 pyramidal neuron infected with Sindbis virus and expressing GFP. Fluorescent (top) and differential interference contrast (bottom) images of the same field during electrophysiological recording. Bar: 30 µm. Schematic diagram on right: Whole-cell recordings were obtained from a fluorescent (infected) and an adjacent nonfluorescent (uninfected) neuron with identical stimulation position and intensity.B)CaMKII-independent kinase activity of tCaMKII constructs measured in BHK(baby hamster cells).

(C) Synaptic responses from neurons expressing GFP or from nearby nonexpressing cells. (Left panel) For each pair of cells, the amplitude of response from infected cell is plotted against amplitude of response in uninfected cell (n = 27). (Right panel) Summary results of measured rectification for uninfected (2.4 ± 0.1, n = 15) and infected (2.4 ± 0.1, n = 17) cells. Sample responses from nearby uninfected and infected cells are overlaid and shown on right side of each panel. (D) tCaMKII-GFP produces enhancement of synaptic transmission in expressing neurons (left panel: uninfected: 15.4 ± 2.4; infected: 25.3 ± 2.5, n = 35) with no effect on rectification

To examine the effect of elevated CaMKII activity on neuronal function, construct encoding the catalytic domain of this enzyme fused with GFP was infected into neurons in hippocampal slices. Synaptic responses in two nearby neurons, one infected and one uninfected, were measured:

Increased CaMKII activity delivers GluR1-GFP into synapses.

• (A)Fluorescence image of a hippocampal slice expressing GluR1-GFP. Sindbis virus particles expressing GluR1-GFP were injected into several sites of CA1 pyramidal cell layer. DG, dentate gyrus. Bar: 300 µm. (B) Immunoprecipitation study indicates that GluR1-GFP forms largely homomers. (C) Synaptic responses from neurons expressing GluR1-GFP and nearby nonexpressing cells. Expression of GluR1-GFP did not affect the amplitude (uninfected: 32.7 ± 4.1; infected: 31.0 ± 3.3, n = 13) or rectification (2.2 ± 0.1, n = 42 for uninfected; 2.4 ± 0.1, n = 41 for infected cells). (D) Coexpression of GluR1-GFP with tCaMKII increased amplitude (uninfected: 25.7 ± 5.3; infected: 41.3 ± 5.1, n = 13) and rectification (2.2 ± 0.4, n = 75 for uninfected;4.3 ± 0.7, n = 29 for infected cells)

To examine if this increase in AMPA-R-mediated transmission was due to a delivery of receptors to synapses, the authors overexpressed GluR1GFP and tCaMKII by using an internal ribosomal entry site (IRES) construct.

• GluR1 is phosphorylated by CaMKII at Ser831during LTP. Substitution of Ser with Ala did not block delivery.

• Mutation in PDZ –binding region of GluR1:• COOH-terminus of the cytosolic tail of many membrane proteins has a consensus

(S/T)X(V/L) responsible for the interaction with PDZ domain containing proteins.• The last three acids of the COOH-terminus of GluR1 are TGL. • COOH-terminus of GluR1 was converted from TGL to AGL. • When expressed in hippocampal neurons, this protein was detected in dendrites. This

construct showed no effect on transmission when expressed alone. However, when GluR1(T887A)-GFP and tCaMKII were coexpressed in hippocampal slice neurons, the effects of tCaMKII on synaptic response amplitude and rectification were completely blocked.

Conclusion:• An interaction between GluR1 and a PDZ-domain protein is necessary for LTP or

CaMKII to drive synaptic delivery of GluR1. Neither the identity of the GluR1-interacting PDZ-domain protein(s) responsible for LTP, nor the subcellular site where these interactions occurs is known.

Postnatal synaptic potentiation: Delivery of GluR4-containing AMPA receptors by

spontaneous activity

J. Julius Zhu1, 3, José A. Esteban1, 3, Yasunori Hayashi2 & Roberto Malinow1

• Developmental profile of AMPA receptor subunits in hippocampus was assayed by western blots.

• GluR4 expression was largely restricted to the first postnatal week.

• According to immunohistochemistry analysis, pyramidal cells expressed GluR4 (Fig. 1b) at this early stage. Staining at the soma and apical dendrite of CA1 pyramidal neuron indicated by arrows. So, stratum oriens; sp, stratum pyramidale; sr, stratum radiatum

Dendritic clustering and surface delivery of GluR4-GFP

• To examine the potential role of GluR4 in synaptic plasticity during this early period, GluR4 tagged with GFP(GluR4-GFP) was expressed in hipocampal slices obtained from P5-7 animals.

• Approximately 36 hours after expression, GluR4-GFP appeared clustered in dendrites, with some delivery to the surface (Fig. 1c) Organotypic hippocampal slices expressing GluR4-GFP were processed for anti-GFP immunohistochemistry in non-permeabilized conditions (Methods). Two-photon images of GFP fluorescence and surface anti-GFP (Texas Red) are shown, together with their overlay.

• Electron microscopy indicated that GluR4-GFP was delivered to dendrites, and some reached synaptic sites (Fig. 1D)

Synaptic delivery of recombinant GluR4

• To study the regulation of GluR4-GFP delivery to synapses, electrophysiological tagging was used. GluR4-GFP expressed in CA1 cells mainly formed receptors lacking GluR2 subunits (Fig. 2a); these receptors show inward rectification (Fig. 2b). Because AMPA-receptor-mediated transmission in pyramidal neurons is largely non-rectifying(Fig. 2e, left; and f, right), delivery of recombinant receptors can be detected by an increased rectification of synaptic responses.

• By comparing transmission simultaneously evoked onto pairs of nearby infected and non-infected neurons (Fig. 2c–f), we found that responses in cells expressing GluR4-GFP had both larger amplitudes at -60 mV and greater rectification, indicating synaptic delivery of homomeric GluR4-GFP receptors.

Spontaneous activity delivers GluR4-GFP into synapses

• Recordings from CA1 neurons demonstrated intermittent synaptic activity (Fig. 3a) that could be blocked by addition of high Mg2+ (n = 16; Fig. 3a) or TTX (n = 8; data not shown) in the bath. To determine if this activity is required to deliver receptors to synapses, the authors infected slices with GluR4-GFP and immediately exposed them to high Mg2+ or TTX.

• Thirty-six hours after expression in slices whose activity was blocked, GluR4-GFP distribution in dendritic regions was largely unchanged. However, synaptic delivery was blocked.(Fig.3b-e)

• To test whether this activity-dependent delivery of GluR4-GFP required NMDA receptor activation, the authors incubated slices in APV(NMDA inhibitor) during the expression of GluR4-GFP. Although spontaneous activity was not blocked (n = 6; Fig. 3a), synaptic delivery of the receptor was largely attenuated (Fig. 3d and e).

• KN-93, a membrane-permeable CaMKII inhibitor that blocked LTP was used to test if the delivery requires activation of CaMKII. In slices maintained in the presence of KN-93 (20 M), transmission onto cells expressing GluR4-GFP was still enhanced and was more rectified (Fig. 3d and e). This indicates that, in contrast to LTP in mature animals, CaMKII activation is not required for synaptic delivery of this receptor.

Delivery of GluR4-GFP to silent synapses

-Early in postnatal development, a large fraction of excitatory synapses produce only NMDA-receptor-mediated responses . Such synapses produce no response at resting membrane potentials, and therefore have been termed 'silent' synapses

-To determine whether GluR4-GFP is delivered to such synapses, the authors measured the response success rate (fraction of successful responses) on nearby infected and non-infected cells. Evoked transmission onto cells expressing GluR4-GFP showed a greater success rate when cells were clamped at the normal resting potential (infected, 54.4 4.6%; control, 38.9 3.2%; n = 32; p < 0.01), but not when clamped at depolarized potential (infected, 60.3 4.5%; ctrl, 58.0 5.6%; n = 23; p = 0.68; Fig. 4a and b, left). These data indicated that these cells had fewer silent synapses (Fig. 4b, right).

-In cells expressing GluR4-GFP, there were significantly more successes at hyperpolarized potentials compared to depolarized potentials ( Fig. 4c and d). These data indicated that cells expressing GluR4-GFP, have a significant fraction of synapses containing only rectifying AMPA receptors, which suggested that GluR4-GFP is delivered to synapses that do not have endogenous AMPA receptors (silent synapses).

-GluR1-GF, which can be delivered to synapses by LTP13, was not delivered to synapses by spontaneous activity (Fig. 4e). Neither success rate at hyperpolarized potentials nor fraction of silent synapses changed in infected cells ( Fig. 4e and f). Similar results were found for cells infected with GluR2-GFP and GluR3-GFP (Fig. 4e and f). These results indicated that spontaneous activity specifically promotes the synaptic delivery of GluR4-GFP, converting silent synapses into functional ones at early stages of circuit formation.

Delivery of endogenous GluR4 into synapses

• Because synaptic trafficking of glutamate receptors seems to be controlled mainly by interactions mediated by their cytoplasmic tail, overexpression of the cytoplasmic tail would be expected to block such interactions and therefore interfere with synaptic delivery. Cells expressing the GluR4 cytoplasmic tail (GluR4ct-GFP) for 36 hours showed markedly decreased transmission relative to nearby uninfected cells (Fig. 5a and b). This decrease was accompanied by a decrease in success rate (infected, 72.6 4.5; ctrl, 97.6 1.0; n = 20, p < 0.001), suggesting that delivery of endogenous GluR4-containing receptors to silent synapses was blocked by GluR4ct-GFP.

• If synaptic delivery of endogenous GluR4-containing receptors is mediated by spontaneous activity, then overexpression of GluR4ct-GFP should have no effect on slices maintained with activity blocked. Indeed, in slices that had been incubated in high Mg2+, cells expressing GluR4ct-GFP for 36 hours had transmission indistinguishable from nearby uninfected cells (Fig. 5c and d).

Exchange of synaptic GluR4-GFP with endogenous GluR2-containing receptor maintains potentiated transmission

• To determine the long-term fate of GluR4-containing AMPA receptors after delivery to synapses by spontaneous activity slices were prepared, and expression of GluR4-GFP was allowed to proceed with normal spontaneous activity to drive the synaptic incorporation of the recombinant protein. Thirty-six hours later, high Mg2+ was added, which should have blocked any subsequent synaptic delivery of GluR4. Transmission was then examined for the next several days. Transmission onto cells expressing GluR4-GFP was still enhanced for several days after activity was blocked (Fig. 6). This enhanced transmission no longer showed the increased rectification characteristic of GluR4-GFP-containing receptors. This indicates that endogenous AMPA receptors containing GluR2 (non-rectifying receptors) have replaced the rectifying recombinant synaptic receptors. Indeed, the existence of endogenous GluR2-lacking AMPA-Rs was supported by co-immunoprecipitation.

What is the mechanism underlying this replacement? Studies suggest that a pool of AMPA receptors can cycle between synaptic and non-synaptic sites.

  NSF binding to GluR2 regulates synaptic transmission.Nishimune A, Isaac JT, Molnar E, Noel J, Nash SR, Tagaya M, Collingridge GL, Nakanishi S, Henley JM. Neuron 1998 Jul;21(1):87-97

• NSF: N-ethylmaleimide-sensitive fusion protein is a homohexameric ATPase that is an essential component of the protein machinery responsible for various membrane fusion events, including intercisternal Golgi protein transport and the exocytosis of synaptic vesicles.

• It has been shown show that N-ethylmaleimide-sensitive fusion protein (NSF) interacts directly and selectively with the intracellular C-terminal domain of the GluR2 subunit of AMPA receptors.

• Loading of decapeptides corresponding to the NSF-binding domain of GluR2 into rat hippocampal CA1 pyramidal neurons results in a marked, progressive decrement of AMPA receptor-mediated synaptic transmission. This reduction in synaptic transmission was also observed when an anti-NSF monoclonal antibody (mAb) was loaded into CA1 neurons.

• It has been suggested that cycling of AMPA receptors depends on interactions between GluR2 and NSF and can be prevented by a peptide (pep2m/G10) containing the NSF-interacting region of GluR2.

Maintanance of enhanced transmission depends on interactions between GluR2 and NSF

• To test whether such interactions are involved in the plasticity characterized in this study, slices were infected with GluR4-GFP and allowed 36 hours of spontaneous activity to deliver the recombinant receptor to synapses. At this point, intracellular loading of pep2m/G10 (peptide, containing the NSF-interacting region of GluR2) decreased transmission markedly in control cells but had diminished effects on GluR4-GFP-expressing cells (Fig. 7a–c). Furthermore, the decrease of transmission in infected cells was accompanied by an additional increase in rectification (4.66 0.97, n = 10 versus 2.91 0.26, n = 34; t-test, p < 0.05; comparison with infected cells not infused with pep2m/G10, Fig. 7f).

• In case of suppression of transmission by exposure of slices to high Mg concentration for 48-72 hours, peptide had similar effect on infected and control cells.

• Thus, an interaction between GluR2 and NSF seems to be involved in the maintenance of the enhanced transmission, but not in the initial delivery.

Experience strengthening transmission by driving AMPA receptors into synapses.

Takahashi T, Svoboda K, Malinow R. Science 2003 Mar 7;299(5612):1585-8