| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Molecular Biosciences, University of Kansas, Lawrence, Kansas; and 2Department of Physiology, The University of Wisconsin, Madison, Wisconsin
Submitted 28 June 2004; accepted in final form 15 August 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Heynen et al. 2000; Lin et al. 2000; Luscher et al. 1999; Shi et al. 2001). Many examples of synaptic plasticity rely on postsynaptic Ca2+ signaling pathways that increase or decrease synaptic transmission and are thought to result from changes in the number of postsynaptic AMPARs (Beattie et al. 2000; Linden 2001; Luthi et al. 1999; Man et al. 2000). The C-termini of AMPAR GluR2/3 subunits interact with the Discs large/zona occludens-1 (PDZ) domains of glutamate receptor interacting protein (GRIP1/2) or AMPA receptor-binding protein (ABP) (two homologous proteins referred to as GRIP/ABP), and protein interacting with C-kinase-1 (PICK1) (Chung et al. 2000; Dong et al. 1997, 1999; Sheng 1996; Srivastava et al. 1998; Wyszynski et al. 1999; Xia et al. 1999). Interactions between PDZ domains and AMPARs are thought to regulate the anchoring, insertion and internalization of synaptic receptors (Daw et al. 2000; Kim et al. 2001; Li et al. 1999). The C-terminal amino acids SVKI of GluR2/3 bind specifically to PDZ domains, and these interactions regulate AMPAR trafficking during long-term potentiation/depression (LTP/LTD; Daw et al. 2000; Kim et al. 2001; Li et al. 1999; Shi et al. 2001). Soluble n-ethylmaleimide-sensitive factor attachment protein (SNAP) receptor SNARE complexes may also regulate the trafficking of synaptic AMPARs via vesicular exocytosis (Beattie et al. 2000; Ehlers 2000; Lin et al. 2000; Lu et al. 2001; Luscher et al. 1999; Man et al. 2000).
The exocytosis of vesicles at the cell surface involves interactions between SNARE proteins that mediate vesicle fusion with target membranes. Synaptotagmin I is a major Ca2+ sensor that triggers SNARE proteins to fuse vesicle lipids with target membrane lipids for secretion (Chapman 2002). Synaptotagmin 1 contains two regulatory C2 domains (Perin et al. 1990), and its C2A domain penetrates lipid bilayers in response to Ca2+ (Davis et al. 1999). C2A domains also interact with syntaxin and SNAP-25 during SNARE complex assembly (Augustine 2001). Earles et al. (2001) developed a recombinant protein with two C2A domains (C2A–C2A) that binds to SNARE complexes and rapidly inhibits exocytosis. Other results indicate that dendritic synaptotagmin 1 undergoes Ca2+-dependent translocation, suggesting a postsynaptic role for synaptotagmin 1 in Ca2+-regulated, SNARE-catalyzed vesicle fusion (Schwab et al. 2001).
We examined the effects of postsynaptic Ca2+ release on synaptic transmission in CA1 pyramidal neurons. We show that postsynaptic perfusion of adenophostin A [AdA, potent inositol 1,4,5-tris-phosphate receptor (IP3R) agonist] significantly increases excitatory postsynaptic current (EPSC) amplitudes, which are blocked by the Ca2+ chelator bis-(o-aminophenoxy)-N,N,N',N'-tetraacetic acid (BAPTA). These increases are also dependent on Ca2+/calmodulin-dependent protein kinase II (CaM-KII) and/or PKC activities (J.-H. Wang, R. L. MacKinnon II, B. J. Maher and P. Kelly, unpublished data), two kinases implicated in regulating AMPAR trafficking (Daw et al. 2000; Shi et al. 2001). The studies herein examined if intracellular Ca2+ release plays a role in regulating AMPAR insertion and/or anchoring in postsynaptic neurons. We investigated the role of AMPAR-PDZ domain interactions in postsynaptic neurons by co-perfusing AdA with peptides corresponding to the C-termini of GluR2/3. These manipulations significantly attenuated increases in EPSC amplitudes observed with AdA alone, suggesting that PDZ domain interactions between GluR2/3 and GRIP/ABP/PICK1 can modulate synaptic plasticity. We also observed that co-perfusion of C2A–C2A, an inhibitor that disrupts synaptotagmin-SNARE-dependent processes, significantly attenuated AdA-enhanced increases in EPSC amplitudes. Last, we saw that postsynaptic perfusion of peptides that disrupt PDZ domain interactions, or synaptotagmin-SNARE-mediated vesicular exocytosis, altered basal synaptic transmission but in opposite directions. Perfusion of peptides that disrupt GluR2/3 binding to GRIP/ABP increased basal synaptic transmission, whereas C2A–C2A caused a decrease in transmission. These results suggest that AMPAR trafficking and/or stabilization are important for maintaining basal synaptic transmission.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Transverse hippocampal slices (400 µM) were prepared from Harlan Sprague-Dawley rats [postnatal days (PND) 14–21]. Animals were anesthetized with isoflurane and decapitated, and brains were quickly removed. Brains were dissected while immersed in ice-cold (4°C) artificial cerebral spinal fluid (ACSF) containing (in mM) 124 NaCl, 3 KCl, 1.3 NaH2PO4, 26 NaHCO3, 4 MgCl2, 0.5 CaCl2, 10 dextrose, and 10 HEPES (310–315 mOsmol), and gassed with 95% O2-5% CO2 (pH 7.35). Tissue blocks containing the hippocampus and overlying cerebral cortex were sliced using a Vibratome. Slices were then transferred to oxygenated "standard" ACSF (1.3 mM MgCl2, 2.4 mM CaCl2 plus other salts/HEPES at the preceding concentrations) at 25°C for 1–2 h before being transferred to a submersion chamber (300 µl volume) for recordings. All whole cell recordings were done at 31°C at a perfusion rate of 1–2 ml/min. CA1 pyramidal neurons were selected on the basis of morphological appearance using an infrared DIC microscope (x40 water-immersion objective). Somas of CA1 pyramidal neurons in stratum pyramidale/oriens were triangular or olive in shape with one major apical dendrite and a few basal dendrites. All experiments using animals were conducted according to an IACUC-approved protocol (University of Kansas).
Electrophysiology
EPSCs were recorded under whole cell patch-clamp conditions. Bipolar tungsten electrodes (12 M
) were used to stimulate Schaffer collateral/commissural (S/C) axons and evoke population EPSCs in CA1 neurons at a stimulus frequency of 0.1 Hz. Patch pipettes (4–7 M) were used to perfuse modulators of intracellular signaling pathways during whole cell recordings. Standard pipette solution contained (in mM) 155 K-gluconate, 4 NaCl, 10 HEPES, 0.5 EGTA, 4 Mg-ATP, and 0.5 tris-GTP (310–315 mOsmol), and was filtered with a 0.1-µm centrifuge filter (Millipore) before use. Where specified, pipette solutions also contained 1 µM AdA (IP3R agonist), 100 µM AMPAR C-terminal peptides (NVYGIE-SVKI, -SVKE, -SVAI, or -EVKI, see RESULTS), or 20 µM synaptotagmin I reagents (C2A–C2A, MC2A–MC2A, or C2A). Recombinant synaptotagmin I reagents were expressed and purified as described in Earles et al. (2001) and then extensively dialyzed at 4°C in dialysis cassettes (Pierce) against standard pipette solution (minus ATP and GTP). To facilitate obtaining high-resistance seals, patch pipette tips were filled with standard solution and backfilled with AMPAR C-terminal peptides or synaptotagmin reagents in standard solution. EPSCs were recorded with an Axopatch-1D amplifier in voltage-clamp mode at –75 mV. Output bandwidth was filtered at 1 kHz. Series resistances were monitored throughout all voltage-clamp recordings by measuring responses to 5 mV (50 ms) pulses.
Data analysis
Initial baseline values of synaptic transmission (defined as 100%) were obtained by averaging the peak amplitudes of the first five EPSCs after stable whole cell configuration was obtained and correct stimulus intensity (i.e., 75–150 pA EPSCs) was achieved. Synaptic responses are represented as means ± SE. The peak amplitude of Iin was measured throughout experiments and only neurons whose amplitudes remained within 20% of initial baseline levels were considered acceptable for analysis. Data were obtained with pClamp 7.0 or 8.0 and analyzed using Clampfit. t-tests were used for comparisons between experimental groups from stable recordings obtained 25 min after whole cell access was achieved. Waveforms displayed in all figures are averaged from five consecutive synaptic responses selected from a representative experiment.
Materials
AdA was purchased from Calbiochem; Dr. Morgan Sheng gratefully provided all C-terminal peptides; all other chemicals were reagent grade.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
One pathway that increases intracellular [Ca2+] is the depletion of IP3R-mediated Ca2+ stores (Supattapone et al. 1988). We have shown that perfusion of AdA, a potent IP3R agonist (Takahashi et al. 1994), into postsynaptic CA1 neurons greatly increased AMPAR-mediated EPSC amplitudes. Postsynaptic perfusion of AdA (1 µM) dramatically increased EPSC amplitudes in a representative CA1 neuron (Fig. 1A, 
). When all neurons were averaged, AdA perfusion substantially increased EPSC amplitudes [373 ± 46% at 25 min (n = 27); P < 0.0001, Fig. 1B, ] compared with control neurons [100 ± 6% at 25 min (n = 39), Fig. 1B, 
]. To examine the role of intracellular Ca2+ release on increases in AMPAR-mediated EPSC amplitudes, we co-perfused AdA with the fast-acting Ca2+ chelator BAPTA (5 mM). Co-perfusion of BAPTA with AdA virtually eliminated increases in EPSC amplitudes in a representative neuron (Fig. 1A, 
). When all neurons co-perfused with AdA plus BAPTA were averaged, BAPTA strongly inhibited AdA-enhanced increases in EPSC amplitudes [97 ± 14% at 25 min (n = 7); P < 0.01 Fig. 1B, ] compared with interleaved neurons perfused with AdA alone [406 ± 92% at 25 min (n = 8); data not shown]. No significant differences were observed between group data for co-perfusion of BAPTA with AdA compared with control intracellular solution (P = 0.85) or intracellular solution containing BAPTA alone [Fig. 2 (n = 7), P = 0.42]. These data suggest that activation of postsynaptic IP3Rs with AdA and subsequent intracellular Ca2+ release increases AMPAR-mediated EPSCs.
|
|
The delivery of AMPARs to the postsynaptic membrane during basal transmission is believed to occur by vesicular fusion that requires specific interactions of v-SNAREs with cognate t-SNAREs (Luscher et al. 1999). Luscher et al. (1999) used botulinum toxin type B to inhibit exocytosis in CA1 neurons. This toxin inhibits exocytosis by proteolysis of the v-SNARE synaptobrevin (VAMP) (Schiavo et al. 1992) prior to the formation in v/t-SNARE fusion complexes. We examined the involvement of postsynaptic AMPAR insertion during basal synaptic transmission by perfusing a protein reagent called C2A–C2A. C2A–C2A (Mr = 35 K) is derived from synaptotagmin 1, a vesicle associated protein that interacts with SNAREs, and has tandem Ca2+ binding domains that regulate exocytosis (Chapman 2002). C2A–C2A binds strongly to SNAREs, displaces native synaptotagmin from preformed SNARE complexes, and rapidly inhibits exocytosis (Earles et al. 2001). Thus C2A–C2A acts at a different step and interacts with different components of the exocytotic machinery compared with type B botulinum toxin.
Postsynaptic perfusion of C2A–C2A into a representative CA1 neuron caused a gradual decrease of EPSC amplitudes relative to baseline values of 100% [58 ± 4% at 25 min; Fig. 3A, ]. When all neurons perfused with C2A–C2A were averaged, a consistent decrease of EPSC amplitudes was observed relative to baseline [40 ± 7% at 25 min (n = 7); P < 0.01, Fig. 4A, ] compared with interleaved controls [103 ± 19% at 25 min (n = 7); Fig. 4A, ]. To examine C2A–C2A's specificity, interleaved experiments using two different synaptotagmin 1 control reagents were performed. The first control reagent is a mutated version of C2A–C2A (MC2A–MC2A) that has two aspartate residues substituted for asparagines (D230N, D232N) in the Ca2+ binding region of each C2A domain. MC2A–MC2A has <30% of the inhibitory effect on exocytosis compared with C2A–C2A (Earles et al. 2001). The second control reagent is a single C2A domain (C2A) that still binds CA2+ but has little effect on vesicular exocytosis (Earles et al. 2001). Perfusion of MC2A–MC2A did not significantly alter basal EPSC amplitudes [100 ± 13% at 25 min (n = 6); P = 0.89, Fig. 4B]. Likewise, postsynaptic perfusion of C2A had no significant effect on basal EPSC amplitudes relative to baseline [90 ± 19% at 25 min (n = 8); P = 0.64, Fig. 4C]. Our results using C2A indicate that C2A-C2A’s ability to inhibit basal transmission is not simply due to its Ca2+ binding property. Together, these results suggest the presence of a pool of constitutively cycling AMPARs that require synaptotagmin-SNARE-mediated exocytosis to maintain stable basal synaptic transmission.
|
|
; Daw et al. 2000; Kim et al. 2001; Li et al. 1999; Shi et al. 2001). We tested the role of AMPAR anchoring to PDZ domain-containing proteins during basal synaptic transmission by perfusing a 10 amino acid peptide containing SVKI into CA1 neurons while monitoring evoked EPSCs. Perfusion of SVKI produced small but not highly significant increases in EPSC amplitudes [138 ± 25% at 25 min (n = 12); P = 0.156, Fig. 5A, ] compared with interleaved controls [104 ± 8% at 25 min (n = 16); Fig. 5A, ]. Changing the C-terminal isoleucine of SVKI to glutamate (i.e., SVKE) completely abolishes the peptide's ability to interact with PDZ domains in GRIP/ABP/PICK1. SVKE therefore has been used as a control peptide (Daw et al. 2000; Li et al. 1999). Postsynaptic perfusion of SVKE produced no significant change in basal EPSC amplitudes relative to baseline [103 ± 16% at 25 min (n = 7); P = 0.91, Fig. 5B].
|
) and PICK1 (R. Huganir, personal communication) but cannot be phosphorylated by PKC because changing Lys882 to Ala disrupts its PKC consensus phosphorylation sequence. Postsynaptic perfusion of SVAI resulted in a large increase in basal synaptic transmission in a representative individual neuron (Fig. 3B). When all neurons perfused with SVAI (Fig. 5C, ) were compared with interleaved control neurons (Fig. 5C, ), SVAI significantly increased basal EPSC amplitudes [165 ± 21% at 25 min, (n = 12); P < 0.01]. These results suggest that disrupting interactions between GluR2/3 and GRIP/ABP and/or PICK1 can enhance basal synaptic transmission.
The specificity of SVKI is broad because it can bind the PDZ domains of GRIP1/ABP as well as PICK1 (Daw et al. 2000; Li et al. 1999) and possibly other unknown PDZ domain-containing proteins. The phosphorylation state of GluR2/3 subunits appears to be important for PDZ domain interactions. PKC phosphorylation of Ser880 within the GluR2/3 PDZ-ligand prevents GRIP/ABP interactions but does not effect PICK1 interactions (Chung et al. 2000). This phosphorylation effect can be mimicked by changing Ser880 to glutamic acid (i.e., SVKI to EVKI) (Daw et al. 2000; Kim et al. 2001; Li et al. 1999). EVKI is therefore useful for determining which specific PDZ domain proteins are important for various neuronal signaling pathways because it specifically disrupts PICK1 interactions. Postsynaptic perfusion of EVKI did not influence basal EPSC amplitudes in a representative neuron (Fig. 3C). Likewise, when all neurons perfused with EVKI were averaged, there was no significant change in basal EPSC amplitudes [109 ± 19% at 25 min (n = 11); P = 0.81, Fig. 5D, ] compared with interleaved controls (Fig. 5D, ). Together, these results indicate that disrupting interactions between GluR2/3 and GRIP/ABP are primarily responsible for increasing basal synaptic transmission.
Postsynaptic Ca2+ release modulates synaptotagmin-SNARE-dependent exocytosis of AMPARs
Fusion of receptor-containing vesicles with membranes requires the interaction of v-SNAREs with cognate t-SNAREs via Ca2+-regulated synaptotagmin-mediated exocytosis (Chapman 2002). To test the hypothesis that intracellular Ca2+ release may enhance AMPAR trafficking to postsynaptic membranes, we tried to inhibit synaptotagmin-SNARE-dependent exocytosis with C2A–C2A. Co-perfusion of C2A-C2A with AdA strongly attenuated increases in EPSC amplitudes in a representative neuron [Fig. 6A, ] compared with an interleaved neuron co-perfused with AdA plus the control reagent C2A (Fig. 6A, ). Comparison of all neurons co-perfused with AdA plus C2A–C2A showed significant attenuation of increases in EPSC amplitudes [153 ± 53% at 25 min (n = 7); Fig. 7A, ] compared with neurons co-perfused with AdA plus the control reagent C2A [341 ± 67% (n = 7); P = 0.047, Fig. 7A, ]. To further examine C2A–C2A's specificity, the control reagent MC2A–MC2A (a version of C2A–C2A with its Ca2+-binding domains mutated) was co-perfused with AdA. When all neurons co-perfused with AdA plus MC2A–MC2A were averaged, MC2A–MC2A perfusion had no significant effect on AdA-induced increases in EPSC amplitudes [350 ± 85% (n = 9); P = 0.81, Fig. 7B]). These results suggest that increases in AMPAR-mediated EPSCs resulting from AdA-enhanced intracellular Ca2+ release are dependent on synaptotagmin-SNARE-mediated insertion of AMPARs into postsynaptic membranes.
|
|
Previous results suggest that postsynaptic perfusion of AdA can convert silent synapses to active synapses (J.-H. Wang, R. L. MacKinnon II, B. J. Maher and P. Kelly, unpublished data). Thus we hypothesized that intracellular Ca2+ release may strengthen the anchoring of existing AMPARs and/or anchor new AMPARs in the postsynaptic membrane. We tested this hypothesis by co-perfusing AdA with the SVKI peptide that disrupts the binding of AMPARs to the PDZ domain containing proteins GRIP1/ABP and PICK1 (Daw et al. 2000; Li et al. 1999). Co-perfusion of AdA with SVKI (100 µM) strongly attenuated increases in AMPAR-mediated EPSCs [167 ± 40% at 25 min, (n = 7); P = 0.017, Fig. 8A, ] compared with interleaved neurons perfused with AdA alone [336 ± 39% at 25 min, (n = 16), Fig. 8A, ]. Co-perfusion of AdA plus SVKE, a control peptide that doesn't disrupt PDZ domain interactions, had no significant effect on AdA-induced increases in EPSC amplitudes [378 ± 79% at 25 min, (n = 8); P = 0.59, Fig. 8B, ]. These results suggest that intracellular Ca2+ release is a trigger for PDZ domain-dependent trafficking and/or anchoring of AMPARs in the postsynaptic membrane.
|
] compared with an interleaved neuron perfused with just AdA [Fig. 6B, ]. When comparing all neurons co-perfused with AdA plus SVAI (Fig. 8C, ) to interleaved neurons perfused with AdA alone (Fig. 8C, ), SVAI perfusion significantly attenuated increases in EPSC amplitude [119 ± 20% at 25 min, P < 0.001 (n = 10)]. These results suggest that SVKI is not acting as a competitive substrate for PKC but is disrupting PDZ domain interactions between GRIP/ABP/PICK1 and AMPARs that are important for AdA-enhanced increases in EPSC amplitudes.
To examine PICK1's role in AdA's ability to increase AMPAR EPSC amplitudes, we tested a peptide with the motif EVKI that interacts with PICK1 but not GRIP/ABP. Co-perfusion of AdA plus EVKI (100 µM) strongly attenuated increases in EPSC amplitudes in a representative neuron (Fig. 6C, ) compared with an interleaved neuron perfused with AdA alone (Fig. 6C, ). Comparing all neurons co-perfused with AdA plus EVKI (Fig. 8D, ) to interleaved neurons perfused with AdA alone (Fig. 8D, ), EVKI perfusion significantly attenuated EPSC amplitudes [131 ± 16% at 25 min, (n = 9); P < 0.001]. These results suggest that interactions between AMPARs and the PDZ domain of PICK1 are important for AdA-induced increases in EPSC amplitudes.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
) greatly increased EPSC amplitudes in CA1 neurons (Fig. 1). This increase is blocked by a calmodulin-binding peptide, and pseudo-substrate inhibitors of PKC or CaM-KII (J.-H. Wang, R. L. MacKinnon II, B. J. Maher and P. Kelly, unpublished data). Postsynaptic perfusion of AdA also converts sub-threshold stimulation that initially failed to evoke EPSCs into supra-threshold stimulation that produced failure-free EPSCs (J.-H. Wang, R. L. MacKinnon II, B. J. Maher and P. Kelly, unpublished data). Previous studies have used similar protocols to activate silent synapses (Isaac et al. 1995; Li et al. 1999; Liao et al. 1995). One mechanism to activate silent synapses is the insertion of functional AMPARs into the postsynaptic membrane. Previous results indicate that AMPAR trafficking is important for synaptic plasticity and maintaining basal synaptic transmission (Beattie et al. 2000; Linden 2001; Luscher et al. 1999; Luthi et al. 1999; Man et al. 2000).
We observed that stimulating postsynaptic IP3Rs with AdA increased EPSC amplitudes and that co-perfusion of BAPTA virtually eliminated this increase (Fig. 1). These results suggest that postsynaptic Ca2+ release from IP3R-regulated stores is responsible for increasing EPSC amplitudes. It is possible that AdA could stimulate Ca2+ influx through plasmalemmal stores-operated Ca2+ channels (SOCs) (Minke and Cook 2002). Although it was suggested that AdA may directly stimulate SOCs (DeLisle et al. 1997), Ca2+-imaging experiments by Machaca and Hartzell (1999) clearly showed that AdA's ability to increase Ca2+ in Xenopus oocytes depends on IP3R-mediated Ca2+ release. Machaca (2003) has shown that CaM-KII activity potentiates Ca2+ currents through SOCs in Xenopus oocytes. This finding is interesting because we have shown that AdA-stimulated increases in EPSC amplitudes is strongly blocked by inhibitors of Ca2+/calmodulin or CaM-KII (J.-H. Wang, R. L. MacKinnon II, B. J. Maher, and P. Kelly, unpublished data).
SOCs are related to the Drosophila transient receptor potential (TRP) protein and many have been identified in brain (Clapham 2003; Moran et al. 2004). TRP1 and TRP5 are enriched in CA1 pyramidal neurons and dendrites and appear to be primarily extrasynaptic; they form nonselective cation channels that display voltage dependence similar to NMDARs (Strubing et al. 2001). TRP1/5 heteromers are activated by Gq-coupled receptors and not by depletion of intracellular Ca2+ stores (Strubing et al. 2001). Philipp et al. (1998) showed that TRP5 and TRP4 are co-expressed in CA1 neurons and form Ca2+ channels that are activated by depleting intracellular Ca2+ stores. Baba et al. (2003) have shown that cultured hippocampal neurons contain SOC-like channels that modulate NMDA-induced Ca2+ transients. They also reported that SOC inhibitors attenuated tetanus-induced increases in dendritic Ca2+ and LTP induction at CA1 synapses. Thus it is possible that AdA-stimulated increases in postsynaptic Ca2+ could stimulate additional Ca2+ influx via SOCs that together enhance AMPAR trafficking and synaptic transmission.
Ca2+ stores in dendrites/spines
Studies at light and EM levels showed that the type-1 IP3R is a major IP3R isoform in CA1 neurons and is predominantly enriched in postsynaptic structures including dendritic shafts and spines (Sharp et al. 1993). Ryanodine receptors are also enriched in CA1 dendrites and spines (Sharp et al. 1993) where they could contribute to IP3-mediated Ca2+ release through Ca2+-induced Ca2+ release (Friel and Tsien 1992). Ca2+ imaging results by Ross and co-workers (Nakamura et al. 2000) indicate that large-amplitude regenerative Ca2+ release occurs primarily in apical dendrites in CA1 neurons with much less release in oblique distal dendrites. These large Ca2+ release events in apical dendrites are elicited by high-frequency afferent stimulation in the presence of the mGluR agonist t-ACPD. Electrophysiological studies have shown that mGluR1 and mGluR5 receptors are present on CA1 pyramidal neurons where they regulate neuronal excitability and intracellular Ca2+ release (Mannaioni et al. 2001; Mutel et al. 2000; Rae and Irving 2004). In addition, mGluR5 knockout mice show impaired learning and reduced LTP in CA1 (Lu et al. 1997). Together, these findings indicate that mGluR signaling pathways underlying IP3R-mediated Ca2+ release are present in dendrites and can regulate neuronal excitability, local Ca2+ release, and synaptic plasticity.
Synaptotagmin-SNARE-dependent insertion of AMPARs during intracellular Ca2+ release and basal synaptic transmission
The N-ethylmaleimide-sensitive factor (NSF) is a critical component of the membrane fusion machinery (Rothman 1994) and is involved in GluR2 insertion/stabilization at synapses (Braithwaite et al. 2002; Luscher et al. 1999; Nishimune et al. 1998; Song et al. 1998). NSF is required for the recycling of SNARE complexes (Chapman 2002) and is believed to be important for the insertion of synaptic AMPARs (Luscher et al. 1999). Synaptotagmin 1, a critical component of SNARE-dependent exocytosis, has been visualized in dendrites where it could modulate postsynaptic functions (Schwab et al. 2001). We examined the role of SNARE proteins in AdA-enhanced increases in synaptic transmission by co-perfusing C2A–C2A, a synaptotagmin-1-derived inhibitor of SNARE-mediated membrane fusion. C2A–C2A binds strongly to SNAREs, displaces native synaptotagmin 1 from SNARE complexes, and rapidly inhibits exocytosis (Earles et al. 2001). C2A–C2A strongly attenuated AdA-enhanced EPSC amplitudes (Fig. 7A), suggesting that intracellular Ca2+ release increases AMPAR insertion into postsynaptic membranes via synaptotagmin-SNARE-mediated exocytosis. Two control reagents (C2A and MC2A–MC2A) failed to inhibit AdA-enhanced increases in EPSC amplitudes (Fig. 7, A and B). Our results provide the first evidence suggesting a postsynaptic function for synaptotagmin 1–SNARE interactions in AMPAR trafficking.
Proteins that regulate exocytosis/endocytosis are thought to modulate basal synaptic transmission (Luscher et al. 1999; Luthi et al. 1999; Song et al. 1998). The balance between exocytosis and endocytosis may be regulated by AMPARs that constitutively cycle using SNARE-dependent processes to deliver receptors to synapses. C2A–C2A decreased basal EPSC amplitudes (Fig. 4A), suggesting that C2A–C2A inhibits the delivery of AMPARs to synapses, leaving endocytosis unchecked and thus decreases the number of synaptic AMPARs.
Postsynaptic perfusion of C2A–C2A depressed basal transmission (Fig. 4A), whereas C2A–C2A attenuated AdA-enhanced increases in EPSC amplitudes (Fig. 7A). These results suggest that C2A–C2A may not directly inhibit AdA's action in increasing EPSC amplitudes but simply depresses transmission by the same mechanism that it depresses basal synaptic transmission. This seems unlikely because all known examples of vesicular exocytosis require SNARE-dependent mechanisms (McNew et al. 2000) and therefore should be inhibited by C2A–C2A. Therefore we believe that both basal and AdA-enhanced synaptic transmission require SNARE-dependent exocytosis of AMPARs.1
AMPAR interactions with PDZ domains during intracellular Ca2+ release and basal synaptic transmission
PDZ domains are important protein–protein interacting motifs involved in regulating the trafficking/anchoring of AMPARs (Dong et al. 1997) and NMDARs (Kornau et al. 1995). The SVKI sequence in GluR2/3 preferentially binds class II PDZ domains in GRIP/ABP and PICK1. Because the two major types of heteromeric AMPARs in CA1 neurons are GluR1/2 and GluR2/3 (Wenthold et al. 1996), the anchoring/trafficking of CA1 AMPARs could be modulated by interactions between GluR2 and GluR3 and the PDZ domains in GRIP/ABP and PICK1. Postsynaptic perfusion of peptides containing the PDZ-binding sequence SVKI in GluR2/3 can alter synaptic transmission and plasticity (Daw et al. 2000; Kim et al. 2001; Li et al. 1999; Matsuda et al. 1999). The overexpression of GRIP1 in neurons reduces the number of cell-surface AMPARs (Dong et al. 1997), suggesting that GRIP1 regulates AMPAR trafficking. GRIP/ABP has been localized near the postsynaptic density (PSD) and in membrane vesicles in dendritic shafts (Dong et al. 1999; Wyszynski et al. 1999). The palmitoylation of ABP is thought to target ABP to spine plasma membranes where it anchors GluR2s, whereas nonpalmitoylated ABP localizes GluR2 to intracellular storage sites (DeSouza et al. 2002). Thus GRIP/ABP is believed to anchor GluR2/3 in both synaptic and dendritic compartments. PICK1 is also found in spines and dendrites (Xia et al. 1999), and phorbol esters induce PICK1, GluR2 and PKC to become preferentially co-localized in spines (Chung et al. 2000; Perez et al. 2001).
We tested if PDZ domains were involved in AdA-enhanced increases in EPSC amplitudes. Co-perfusion of SVKI, which disrupts interactions between GluR2/3 and GRIP/ABP/PICK1, strongly attenuated AdA-enhanced increases in EPSC amplitudes (Fig. 8A), suggesting that interactions between GluR2/3 and class II PDZ domains of GRIP/ABP/PICK1 play a role in AdA-enhanced increases in EPSC amplitudes. In addition, SVAI which disrupts GluR2/3 binding to GRIP/ABP and PICK1 and does not contain a PKC phosphorylation motif, strongly attenuated AdA's ability to increase EPSC amplitudes (Fig. 8C). This suggests that SVKI acts by disrupting interactions between GluR2/3 and GRIP/ABP/PICK1 rather than acting as a competitive PKC inhibitor.
Are both GRIP/ABP and PICK1 involved in AdA-enhanced increases in EPSC amplitudes? Changing the serine in SVKI to glutamic acid (i.e., EVKI) mimics PKC phosphorylation. The resulting EVKI peptide selectively interacts with PICK1 but not GRIP/ABP. EVKI strongly attenuated AdA-enhanced increases in EPSC amplitudes (Fig. 8D), suggesting that interactions between GluR2/3 and PICK1 are mainly responsible for AdA's enhancement of synaptic transmission.1
Role of PICK1 and PKC in AdA-induced increases in synaptic transmission
AdA-enhanced increases in EPSC amplitudes were significantly attenuated by SVKI (Fig. 8A). SVKI inhibits interactions between GluR2/3 and GRIP/ABP/PICK1 and could inhibit PKC by acting as a competitive substrate. In addition, we have shown that inhibition of postsynaptic PKC activity strongly attenuated AdA-induced increases in synaptic transmission (J.-H. Wang, R. L. MacKinnon II, B. J. Maher and P. Kelly, unpublished data). Because the PICK1 selective peptide EVKI was effective in attenuating AdA-enhanced increases in EPSC amplitudes (Fig. 8D), it appears that signaling between PKC, PICK1, and GluR2/3 may be critical in this type of synaptic plasticity. Although the exact role of PKC is unclear, we suggest that AdA-enhanced Ca2+ release stimulates PKC phosphorylation of Ser880, which releases GluR2/3 tethered to intracellular storage sites by GRIP/ABP. This facilitates GluR2/3's binding to PICK1, promoting receptor trafficking to the postsynaptic membrane and increasing synaptic transmission. A model sharing some of these properties1 has been proposed by DeSouza et al. (2002) and Daw et al. (2000).
The mechanism proposed in the preceding text shares some similarities with recent results describing the effects of overexpression of PICK1 in CA1 neurons. PICK1 overexpression significantly increased AMPAR EPSC amplitudes, decreasing synaptic/cell-surface GluR2s without changing surface GluR1 levels or the total number of surface AMPARs (Terashima et al. 2004). AMPARs lacking GluR2 display greater single-channel conductance than those that contain GluR2 (Swanson et al. 1997), and this may be the basis for increased synaptic transmission during PICK1 overexpression. PICK1-mediated increases in synaptic transmission observed by Terashima et al. were blocked by inhibitors of PKC or CaM-KII. This dependence on PKC and CaM-KII activities is similar to our results on AdA-stimulated increases in synaptic transmission (J.-H. Wang, R. L. MacKinnon II, B. J. Maher and P. Kelly, unpublished data). Terashima et al. (2004) also observed that viral-mediated expression of SVKI/EVKI peptides in CA1 neurons decreased AMPAR EPSC amplitudes, which is different from the actions of these peptides on basal synaptic transmission that we (Fig. 5) and others have observed (Daw et al. 2000; Kim et al. 2001). The differences between our results and those of Terashima et al. (2004) may be due to experimental conditions. We have studied short-term effects (30 min) of SVKI/EVKI peptides and IP3R agonists during whole cell perfusion in acute hippocampal slices whereas they examined viral-mediated expression of PICK1 and SVKI/EVKI peptides over longer times (20–48 h) using organotypic slices or cultured neurons. Despite these differing results, it is possible that AdA perfusion activates PICK1 via PKC and CaM-KII signaling pathways, decreases the proportion of GluR2s in synaptic AMPARs and thereby enhances EPSC amplitudes.
Although there are several targets of IP3R-mediated Ca2+ release that could modulate synaptic transmission, two noteworthy candidates are calpain and the clathrin-mediated endocytotic complex. The Ca2+-activated protease calpain is present in dendrites (Perlmutter et al. 1990) and calpain inhibitors attenuate LTP (Denny et al. 1990; Oliver et al. 1989; Vanderklish et al. 1996). NMDAR stimulation activates calpain in dendritic spines (Vanderklish et al. 2000). Calpain degrades GRIP in vitro and in hippocampal slices stimulated with N-methyl-D-aspartate (NMDA) show enhanced GRIP proteolysis (Lu et al. 2001). AdA-stimulated Ca2+ release could activate postsynaptic calpain which degrades GRIP, disrupts its binding to GluR2/3 and thereby modulates AMPAR trafficking.
Cell-surface receptors undergo endocytosis by the action of a protein complex containing clathrin, clathrin adapters, dynamin, amphiphysin, and synaptojanin (Slepnev and De Camilli 2000). Lai et al. (1999) reported that calcineurin (CaN), a Ca2+/calmodulin-dependent phosphatase, is physically associated with dynamin, and this association is Ca2+- dependent. They showed that dynamin physically links CaN to the endocytic complex and most importantly, that inhibition of CaN in vivo attenuates transferrin endocytosis. These results suggest that the CaN-dynamin complex is a Ca2+ sensor for vesicular endocytosis. Consistent with this role for CaN, we have shown that inhibiting postsynaptic CaN activity increased basal synaptic transmission (Wang and Kelly 1997). Beattie et al. (2000) have shown that activation of NMDA receptors triggers AMPAR endocytosis by a mechanism that involves Ca2+ influx and CaN activity. The dephosphorylation of dynamin by CaN is an important step in the functional assembly of the endocytosis complex (Slepnev et al. 1998). It is possible that Ca2+-mediated endocytosis may be activated following AdA-stimulated intracellular Ca2+ release; however, our results indicate that the balance between AMPAR insertion versus internalization greatly favors insertion. Clearly, more information is needed to understand how these complicated but opposing pathways regulate AMPAR trafficking.
What role do PDZ domains play in regulating basal synaptic transmission? We observed that SVAI significantly increased basal transmission (Fig. 5C), while SVKI perfusion produced small but not significant increases in basal EPSC amplitudes (Fig. 5A). One explanation for the different actions of SVAI versus SVKI may be that SVKI is phosphorylated by PKC under basal conditions and becomes a selective PICK1 inhibitor (Kim et al. 2001) and therefore doesn't modulate basal synaptic transmission. Consistent with this notion, we observed that EVKI (a selective PICK1 inhibitor) did not alter basal synaptic transmission (Fig. 5D); this is in agreement with Daw et al. (2000; however, see Kim et al. 2001). Because SVAI significantly increased basal synaptic transmission and isn't phosphorylated by PKC, we believe SVAI is acting primarily by inhibiting interactions between GluR2/3 and GRIP/ABP. Two potential mechanisms could account for SVAI's ability to increase basal transmission. First, SVAI may disrupt GluR2/3s that are anchored to GRIP/ABP at intracellular storage sites and allow these AMPARs to traffic to the postsynaptic membrane. Second, SVAI may inhibit the intracellular storage of GluR2/3s that are being constitutively internalized and thus allow GluR2/3s to return directly to the postsynaptic membrane. Both of these potential mechanisms could increase the number of synaptic AMPARs being inserted versus the number being constitutively internalized and therefore contribute to SVAI's ability to increase basal EPSC amplitudes.1
Our findings suggest that stimulating postsynaptic Ca2+ release with AdA increases synaptic transmission by enhancing the insertion and anchoring of AMPARs. AdA-enhanced postsynaptic Ca2+ release increases AMPAR function, possibly via synaptotagmin-SNARE-dependent exocytosis that controls receptor insertion and interactions with PICK1 PDZ domains that mediate receptor anchoring. In addition, the regulation of basal synaptic transmission involves synaptotagmin-SNARE-dependent exocytosis and PDZ domain interactions of GRIP/ABP that regulate the constitutive cycling of AMPARs.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present address of B. Maher: Vollum Institute, Oregon Health Sciences University, Portland, OR 97239.
|
FOOTNOTES |
|---|
1 The Supplementary Material for this article (a figure) illustrates the roles of synaptotagmin-SNARE-mediated exocytosis, GRIP/ABP, and PICK1 in AMPAR cycling during basal (A) and AdA-enhanced synaptic transmission (B). This figure is available online at http://jn.physiology.org/cgi/content/full/00651.2004/DC1. 
Address for reprint requests and other correspondence: P. T. Kelly, Dept. of Molecular Biosciences, 4006 Haworth Hall, The University of Kansas, Lawrence, KS 66045-2106 (E-mail: ptkelly{at}ku.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Baba A, Yasui T, Fujisawa S, Yamada RX, Yamada MK, Nishiyama N, Matsuki N, and Ikegaya Y. Activity-evoked capacitative Ca2+ entry: implications in synaptic plasticity. J Neurosci 23: 7737–7741, 2003.
Beattie EC, Carroll RC, Yu X, Morishita W, Yasuda H, von Zastrow M, and Malenka RC. Regulation of AMPA receptor endocytosis by a signaling mechanism shared with LTD. Nat Neurosci 3: 1291–1300, 2000.[CrossRef][ISI][Medline]
Braithwaite SP, Xia H, and Malenka RC. Differential roles for NSF and GRIP/ABP in AMPA receptor cycling. Proc Natl Acad Sci USA 99: 7096–7101, 2002.
Chapman ER. Synaptotagmin: a Ca(2+) sensor that triggers exocytosis? Nat Rev Mol Cell Biol 3: 498–508, 2002.[CrossRef][ISI][Medline]
Chung HJ, Xia J, Scannevin RH, Zhang X, and Huganir RL. Phosphorylation of the AMPA receptor subunit GluR2 differentially regulates its interaction with PDZ domain-containing proteins. J Neurosci 20: 7258–7267, 2000.
Clapham DE. TRP channels as cellular sensors. Nature 426: 517–524, 2003.[CrossRef][Medline]
Davis AF, Bai J, Fasshauer D, Wolowick MJ, Lewis JL, and Chapman ER. Kinetics of synaptotagmin responses to Ca2+ and assembly with the core SNARE complex onto membranes. Neuron 24: 363–376, 1999.[CrossRef][ISI][Medline]
Daw MI, Chittajallu R, Bortolotto ZA, Dev KK, Duprat F, Henley JM, Collingridge GL, and Isaac JT. PDZ proteins interacting with C-terminal GluR2/3 are involved in a PKC-dependent regulation of AMPA receptors at hippocampal synapses. Neuron 28: 873–886, 2000.[CrossRef][ISI][Medline]
DeLisle S, Marksberry EW, Bonnett C, Jenkins DJ, Potter BV, Takahashi M, and Tanzawa K. Adenophostin A can stimulate Ca2+ influx without depleting the inositol 1,4,5-trisphosphate-sensitive Ca2+ stores in the Xenopus oocyte. J Biol Chem 272: 9956–9961, 1997.
Denny JB, Polan-Curtain J, Ghuman A, Wayner MJ, and Armstrong DL. Calpain inhibitors block long-term potentiation. Brain Res 534: 317–320, 1990.[CrossRef][ISI][Medline]
DeSouza S, Fu J, States BA, and Ziff EB. Differential palmitoylation directs the AMPA receptor-binding protein ABP to spines or to intracellular clusters. J Neurosci 22: 3493–3503, 2002.
Dong H, O'Brien RJ, Fung ET, Lanahan AA, Worley PF, and Huganir RL. GRIP: a synaptic PDZ domain-containing protein that interacts with AMPA receptors. Nature 386: 279–284, 1997.[CrossRef][Medline]
Dong H, Zhang P, Song I, Petralia RS, Liao D, and Huganir RL. Characterization of the glutamate receptor-interacting proteins GRIP1 and GRIP2. J Neurosci 19: 6930–6941, 1999.
Earles CA, Bai J, Wang P, and Chapman ER. The tandem C2 domains of synaptotagmin contain redundant Ca2+ binding sites that cooperate to engage t-SNAREs and trigger exocytosis. J Cell Biol 154: 1117–1123, 2001.
Ehlers MD. Reinsertion or degradation of AMPA receptors determined by activity-dependent endocytic sorting. Neuron 28: 511–525, 2000.[CrossRef][ISI][Medline]
Friel DD and Tsien RW. A caffeine- and ryanodine-sensitive Ca2+ store in bullfrog sympathetic neurones modulates effects of Ca2+ entry on [Ca2+]i. J Physiol 450: 217–246, 1992.
Heynen AJ, Quinlan EM, Bae DC, and Bear MF. Bidirectional, activity-dependent regulation of glutamate receptors in the adult hippocampus in vivo. Neuron 28: 527–536, 2000.[CrossRef][ISI][Medline]
Isaac JT, Nicoll RA, and Malenka RC. Evidence for silent synapses: implications for the expression of LTP. Neuron 15: 427–434, 1995.[CrossRef][ISI][Medline]
Kim CH, Chung HJ, Lee HK, and Huganir RL. Interaction of the AMPA receptor subunit GluR2/3 with PDZ domains regulates hippocampal long-term depression. Proc Natl Acad Sci USA 98: 11725–11730, 2001.
Kornau HC, Schenker LT, Kennedy MB, and Seeburg PH. Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95. Science 269: 1737–1740, 1995.
Lai MM, Hong JJ, Ruggiero AM, Burnett PE, Slepnev VI, De Camilli P, and Snyder SH. The calcineurin-dynamin 1 complex as a calcium sensor for synaptic vesicle endocytosis. J Biol Chem 274: 25963–25966, 1999.
Li P, Kerchner GA, Sala C, Wei F, Huettner JE, Sheng M, and Zhuo M. AMPA receptor-PDZ interactions in facilitation of spinal sensory synapses. Nat Neurosci 2: 972–977, 1999.[CrossRef][ISI][Medline]
Liao D, Hessler NA, and Malinow R. Activation of postsynaptically silent synapses during pairing-induced LTP in CA1 region of hippocampal slice. Nature 375: 400–404, 1995.[CrossRef][Medline]
Lin JW, Ju W, Foster K, Lee SH, Ahmadian G, Wyszynski M, Wang YT, and Sheng M. Distinct molecular mechanisms and divergent endocytotic pathways of AMPA receptor internalization. Nat Neurosci 3: 1282–1290, 2000.[CrossRef][ISI][Medline]
Linden DJ. The expression of cerebellar LTD in culture is not associated with changes in AMPA-receptor kinetics, agonist affinity, or unitary conductance. Proc Natl Acad Sci USA 98: 14066–14071, 2001.
Lu X, Wyszynski M, Sheng M, and Baudry M. Proteolysis of glutamate receptor-interacting protein by calpain in rat brain: implications for synaptic plasticity. J Neurochem 77: 1553–1560, 2001.[CrossRef][ISI][Medline]
Lu YM, Jia ZP, Janus C, Henderson JT, Gerlai R, Wojtowicz JM, and Roder JC. Mice lacking metabotropic glutamate receptor 5 show impaired learning and reduced CA1 long-term potentiation (LTP) but normal CA3 LTP. J Neurosci 17: 5196–5205, 1997.
Luscher C, Xia H, Beattie EC, Carroll RC, von Zastrow M, Malenka RC, and Nicoll RA. Role of AMPA receptor cycling in synaptic transmission and plasticity. Neuron 24: 649–658, 1999.[CrossRef][ISI][Medline]
Luthi A, Chittajallu R, Duprat F, Palmer MJ, Benke TA, Kidd FL, Henley JM, Isaac JT, and Collingridge GL. Hippocampal LTD expression involves a pool of AMPARs regulated by the NSF-GluR2 interaction. Neuron 24: 389–399, 1999.[CrossRef][ISI][Medline]
Machaca K. Ca2+-calmodulin-dependent protein kinase II potentiates store-operated Ca2+ current. J Biol Chem 278: 33730–33737, 2003.
Machaca K and Hartzell HC. Adenophostin A and inositol 1,4,5-trisphosphate differentially activate Cl- currents in Xenopus oocytes because of disparate Ca2+ release kinetics. J Biol Chem 274: 4824–4831, 1999.
Man HY, Lin JW, Ju WH, Ahmadian G, Liu L, Becker LE, Sheng M, and Wang YT. Regulation of AMPA receptor-mediated synaptic transmission by clathrin-dependent receptor internalization. Neuron 25: 649–662, 2000.[CrossRef][ISI][Medline]
Mannaioni G, Marino MJ, Valenti O, Traynelis SF, and Conn PJ. Metabotropic glutamate receptors 1 and 5 differentially regulate CA1 pyramidal cell function. J Neurosci 21: 5925–5934, 2001.
Matsuda S, Mikawa S, and Hirai H. Phosphorylation of serine-880 in GluR2 by protein kinase C prevents its C terminus from binding with glutamate receptor-interacting protein. J Neurochem 73: 1765
