|Home | About | Journals | Submit | Contact Us | Français|
Homeostatic plasticity may compensate for Hebbian forms of synaptic plasticity, such as long-term potentiation (LTP) and depression (LTD), by scaling neuronal output without changing the relative strength of individual synapses. This delicate balance between neuronal output and distributed synaptic weight may be necessary for maintaining efficient encoding of information across neuronal networks. Here, we demonstrate that Arc, an immediate early gene (IEG) that is rapidly induced by neuronal activity associated with information encoding in the brain, mediates homeostatic synaptic scaling of AMPA type glutamate receptors (AMPARs) via its ability to activate a novel and selective AMPAR endocytic pathway. High levels of Arc block the homeostatic increases in AMPAR function induced by chronic neuronal inactivity. Conversely, loss of Arc results in increased AMPAR function and abolishes homeostatic scaling of AMPARs. These observations, together with evidence that Arc is required for memory consolidation, reveal the importance of Arc’s dynamic expression as it exerts continuous and precise control over synaptic strength and cellular excitability.
Arc is an immediate early gene (IEG) that is dynamically regulated by neuronal activity and is tightly coupled to behavioral encoding of information in neuronal circuits (Guzowski et al., 2005). Arc mRNA traffics to distal dendrites where it accumulates at sites of synaptic activity and is locally translated (Steward et al., 1998). In vivo, Arc is coordinately induced in populations of neurons that mediate learning such as place cells of the hippocampus (Ramirez-Amaya et al., 2005), and behavior-specific neural networks in parietal (Burke et al., 2005), visual (Tagawa et al., 2005; Wang et al., 2006) and olfactory (Zou and Buck, 2006) cortices. For example, 5 minutes of spatial exploration elicits transcriptional induction of Arc in ~40% of CA1 neurons (Guzowski et al., 2005). Moreover, Arc is repeatedly induced in the same network during exploration of the same space (Guzowski et al., 2006) and during memory consolidation (Ramirez-Amaya et al., 2005), indicating that in vivo expression can be maintained at elevated steady-state levels in specific neuronal networks. Consistent with the notion that Arc protein is required for learning and memory (Guzowski et al., 2000), Arc KO mice demonstrate impaired maintenance of LTP and consolidation of long-term memory, but exhibit normal short-term memory (Plath et al., 2006). Molecular studies indicate that Arc modulates AMPAR trafficking by interacting with two proteins intricately involved in endocytosis, endophilin 2/3 and dynamin (Chowdhury et al., 2006).
Changes in synaptic strength are proposed to underlie memory storage in neuronal circuits (Malenka and Nicoll, 1999; Martin et al., 2000). Hebbian forms of plasticity such as LTP and LTD can modify the strength of individual synapses. However, unrestrained potentiation or depression can result in saturation of a neuron’s ability to encode information (Moser et al., 1998). Homeostatic compensation for these acute changes in synaptic strength is required to maintain neuronal output in the normal range, but must be accomplished without erasing information encoded by the distributed synaptic strengths (Davis and Bezprozvanny, 2001; Turrigiano and Nelson, 2004). For example, chronic blockade of network activity for several days results in an increase in surface and synaptic AMPARs, while a chronic increase in activity reduces surface and synaptic AMPARs (O’Brien et al., 1998; Turrigiano et al., 1998).
Here, we present evidence that the Arc-endocytic pathway mediates homeostatic scaling of AMPARs. Arc protein is dynamically regulated by chronic changes in neuronal activity that normally evokes synaptic scaling. Overexpression of Arc blocks the upregulation of surface AMPARs and mEPSCs induced by chronic neuronal inactivity. Conversely, Arc KO neurons exhibit a scaled increase in surface AMPARs and AMPAR-mediated mEPSCs that mimics the upregulation of synaptic function induced by chronic inactivity. Strikingly, Arc KO neurons exhibit gross deficits in homeostatic scaling of surface AMPARs and AMPAR-mediated mEPSCs. These findings support a simple model in which Arc regulates an endocytic pathway whose activity is continuously coupled to neuronal excitability by the transcription of Arc. This process regulates steady state AMPARs and permits the homeostatic scaling of synaptic strength.
Arc mRNA induction and protein synthesis are regulated by neuronal activity in vivo (Steward et al., 1998; Steward and Worley, 2001). To elucidate the synaptic function of Arc, we examined Arc expression in primary neuronal cultures. Many features of Arc that are observed in vivo are also recapitulated in culture (Rao et al., 2006). Arc expression is relatively low in preadolescent brain but reaches peak levels by postnatal day 28 (Lyford et al., 1995). Similarly, in primary hippocampal or cortical cultures, Arc expression is relatively low prior to 12 DIV, but increases markedly in older neurons (data not shown) and correlates with the appearance of mature synapses. In these older cultures, Arc protein is present throughout distal dendrites and spines and many of these puncta co-localize with the synaptic marker bassoon (Figure 1A). Arc expression in primary culture is also dependent on spontaneous neuronal activity, which is consistent with its role as an IEG (Figures 1B and 1C). Long-term TTX treatment of high-density cortical neurons, which blocks all evoked neuronal activity, significantly reduces Arc expression. Bicuculline, which blocks inhibitory neurotransmission mediated by GABAA receptors and increases neuronal firing, significantly increases Arc expression (Figures 1B and 1C). Similar results are seen in low-density hippocampal neurons (data not shown).
Since Arc expression is reciprocally regulated by activity and facilitates AMPAR endocytosis (Chowdhury et al., 2006), we hypothesized that Arc may be involved in mechanisms that permit homeostatic AMPAR scaling. Consistent with this idea, over-expression of Arc results in a cell-wide decrease in the surface expression of GluR1-containing AMPARs, mimicking the effect observed with chronic bicuculline treatment (Figures 2A and B). Arc over-expression also blocked the homeostatic upregulation of GluR1 induced by chronic TTX treatment (Figures 2A and B). This indicates that low levels of Arc protein are required for optimal homeostatic upregulation of AMPARs induced by chronic inactivity. Indeed, since TTX treatment also reduces endogenous Arc levels, we observed greater relative downregulation of AMPARs when Arc is over-expressed in the presence of TTX. We next examined AMPAR expression in high-density cortical neurons, a preparation optimized for biochemical assays of surface proteins and for electrophysiological assays of synaptic events (Rumbaugh et al., 2006; Rumbaugh et al., 2003). In contrast to hippocampal cultures, Arc expression driven by a sindbis viral vector had little effect on AMPAR surface expression as detected by surface biotinylation (Figure 2C). We asked whether this difference might be due to high levels of neuronal network activity and endogenous Arc expression common to high-density cortical cultures. Consistent with this hypothesis, treatment with TTX, which reduces the endogenous level of Arc, revealed a striking effect of Arc transgene expression on surface AMPARs. Arc-expressing neurons treated with TTX exhibited a ~50% reduction of surface AMPARs compared to control neurons expressing only GFP. Similarly, electrophysiological recordings from these neurons demonstrated that, in the absence of TTX, Arc transgene expression had no effect on the amplitude of AMPAR-mediated mEPSCs (Figure 2D). By contrast, in the presence of TTX, Arc transgene reduced the amplitude of AMPAR-mediated mEPSCs (Figure 2E) and blocked the TTX-induced scaling effect. Importantly, cumulative probability distributions of mEPSC amplitudes were uniformly reduced by Arc transgene expression (Figure 2F). This suggests that Arc scales AMPAR responses in a manner similar to activity-dependent homeostatic synaptic scaling. Strikingly, we found that a deletion mutant of Arc (aaΔ91-100) that does not interact with endophilin (Chowdhury et al., 2006) had no effect on mEPSCs and did not block the TTX-induced increase in mEPSCs (Figures 2D and E). Since this region is also required for downregulation of AMPARs (Chowdhury et al., 2006), these data provide compelling evidence that Arc regulates homeostatic scaling of AMPARs though a molecular interaction with the endocytic machinery.
To assess the function of endogenous Arc, we performed various experiments on Arc knockout (KO) mice. These mice are viable and do not exhibit any gross changes in neural architecture (Plath et al., 2006). We isolated primary neurons from these mice and compared them with neurons derived from wildtype (Wt) animals. To further explore the contribution of Arc in homeostasis, we examined surface AMPAR levels in Arc KO neurons using DIV 21-28 low-density hippocampal neurons. If endogenous Arc is involved in promoting the removal of AMPARs from synapses, then neurons lacking Arc might exhibit an increase in steady state levels of these receptors. In support of this hypothesis, we found that Arc KO neurons exhibit a dramatic increase in surface GluR1 when compared with Wt neurons (Figure 3A). Puncta from Arc KO neurons exhibit increased size, number and intensity compared with Wt neurons (Supplemental Figure 1), suggesting a global up-regulation of surface GluR1 levels. Importantly, cultured Arc KO neurons form similar numbers of synapses and have a normal compliment of synaptic proteins when compared to Wt neurons (Figure 3C and Supplemental Figure 2). In contrast to GluR1 surface levels, surface GluR2 levels were unchanged in Arc KO neurons (Figure 3B), suggesting that the removal of endogenous Arc from synapses changes the subunit composition of AMPARs to favor GluR1-containing receptors.
The cell biological observations made in Arc KO neurons cannot definitively show that increases in surface AMPARs lead to changes in synaptic strength. To directly examine this issue, AMPAR-mediated mEPSCs were recorded from Wt and Arc KO neurons. Arc KO neurons possessed larger AMPAR-mediated mEPSC amplitudes relative to Wt neurons (Figure 3D). Notably, the increase of mEPSC amplitudes was distributed over the entire range of recorded events as shown by cumulative probability distributions (Figure 3D). This is similar to the homeostatic synaptic scaling that occurs in Wt neurons treated with TTX. Thus, reduction of Arc, by either genetic deletion or TTX, results in a similar state of enhanced surface AMPAR expression and function.
To directly assess the effect of Arc in homeostatic scaling of AMPARs, we treated both Wt and Arc KO low-density hippocampal neurons with TTX or bicuculline for 48 hrs and examined surface AMPAR levels. Wt mouse neurons exhibited a robust increase in surface GluR1 after TTX treatment and significantly lower surface levels after bicuculline treatment (Figure 4A). Strikingly, Arc KO neurons exhibited no changes in surface GluR1 levels after either treatment indicating a complete absence of activity-dependent homeostatic scaling of AMPARs in hippocampal neurons (Figure 4A). In electrophysiological assays, TTX treatment of Wt neurons resulted in a significant increase in mEPSC amplitudes and a multiplicative shift in the cumulative probability distribution, indicative of synaptic scaling of AMPARs (Figure 4B). By contrast, Arc KO neurons treated with TTX failed to exhibit an enhancement of mEPSC amplitudes (Figure 4C), possibly because KO neurons are already homeostatically upregulated. In response to bicuculline, both Wt and Arc KO neurons exhibited a modest down regulation of AMPAR (Figures 4B and C). This result contrasts with histochemical assays in hippocampal neurons and suggests that mechanisms in addition to Arc may contribute to AMPAR downregulation in highly active cortical neurons.
We next asked whether the absence of homeostatic increases of AMPARs in Arc KO neurons following TTX treatment might be due to saturation of mechanisms that can deliver AMPARs to the cell surface. To address this issue, we briefly stimulated Wt and Arc KO neurons with glycine, a treatment that results in a rapid and persistent change in synaptic function though the insertion of AMPARs. This process is believed to mimic LTP (Liao et al., 2001; Lu et al., 2001). Glycine treatment produced an equivalent increase of surface AMPAR in Wt and Arc KO neurons (Figure 5). Thus, AMPAR insertion in response to transient synaptic activation is intact in Arc KO neurons.
The present study provides evidence that Arc mediates synaptic scaling of AMPARs over a broad range of synaptic activity in mature neurons. Arc directly interacts with components of the endocytic pathway, including dynamin and endophilin, and selectively increases the rate of AMPAR endocytosis (Chowdhury et al., 2006). These data suggest the following model: In the absence of Arc, or in conditions of persistent low activity where Arc expression is dramatically reduced, Arc-dependent endocytosis is minimized, causing a shift in the steady state AMPAR distribution towards membrane insertion. In conditions of persistent high activity, high levels of Arc are available to facilitate endocytosis of AMPARs, with consequent down regulation of synaptic AMPARs. Thus, Arc acts to titrate surface AMPARs and this allows optimal synaptic plasticity by maintaining distributed synaptic weights. Periods of sustained Arc expression act to regulate synaptic scaling and thus counteract saturation of synaptic strength inherent in Hebbian positive-feedback loops.
Homeostatic synaptic scaling of AMPARs has been elegantly described in physiological terms in both cortical and hippocampal neurons (O’Brien et al., 1998; Thiagarajan et al., 2005; Turrigiano et al., 1998). Cortical neurons consistently exhibit changes in mEPSC amplitude without any changes in frequency, suggesting a postsynaptic mechanism. In contrast, hippocampal neurons have been reported to exhibit increases in mEPSC amplitude and frequency (Thiagarajan et al., 2002). Presynaptic changes such as enlargement of presynaptic terminals and their vesicle pools have also been observed in hippocampal neurons (Murthy et al., 2001). Further complicating these results, a recent study showed that the expression locus of homeostasis is governed by how long the cultures are incubated in vitro and not by cell type. In both cortical and hippocampal neurons, two days of TTX treatment induced an increase in mEPSC amplitude in cells that were less than 14 DIV, without affecting mEPSC frequency. However, in cultures older than 18 DIV, the same treatment induced a large increase in mEPSC frequency and a reduced effect on amplitude (Wierenga et al., 2006). The molecular mechanisms underlying these changes are unclear. Our results implicate Arc as one member of a molecular pathway that permits AMPAR scaling but these data do not address other homeostatic mechanisms of regulating neuronal output. We find differences in the action of Arc in hippocampal versus cortical neurons. Arc is more effective in down regulating AMPARs in low density hippocampal cultures than in high density cortical neurons. This is probably due to lower basal Arc in low density neurons but we cannot rule-out region specific differences in co-functional molecules.
In hippocampal neurons homeostatic increases in surface AMPARs induced by prolonged inactivity have been shown to be mainly due to increases in GluR2-lacking receptors (Thiagarajan et al., 2005). In our studies of Arc KO neurons, Arc effects appear to be preferential for GluR1, as KO neurons show no overt changes in surface GluR2, suggesting an increase in GluR2 lacking receptors at the plasma membrane. Arc expression in pyramidal neurons of hippocampal slice cultures produces a selective down-regulation of AMPARs that requires Arc’s ability to bind endophilin, and is blocked by agents that inhibit NMDA receptor dependent LTD, including the calcineurin inhibitor FK506 and peptides that mimic the C-terminus of GluR2 (Verde et al, 2006). However, when Arc is expressed in Wt neurons, we see down regulation of both GluR1 and GluR2 (Figure 2C). These results confirm that Arc can endocytose GluR2 in an acute manner. It is possible that the apparent GluR1 selectivity in the Arc KO is a consequence of long term depletion of the protein. This may promote a shift in the subunit composition of AMPAR pools towards more GluR2-lacking AMPARs, similar to what is seen with prolonged inactivity. Further work is needed to clarify Arc’s precise role in determining AMPAR subunit specificity.
A recent report indicates that TNF-α, which is derived homeostatic up regulation of AMPARs in neurons induced by chronic activity blockade, but not down regulation due to increased activity (Stellwagen and Malenka, 2006). The molecular mechanism of TNF-αregulation remains to be determined in both homeostasis and synaptic maturation. In addition, it is unclear if TNF-αplays a direct role in information storage or is just a permissive factor needed for correct neuronal network development.
AMPARs are highly dynamic and undergo rapid shuttling between the plasma membrane and internal recycling pools (Luscher et al., 1999; Park et al., 2004). Both LTD and LTP involve modulation of endocytosis and exocytosis of AMPA receptors and numerous AMPA interacting molecules such as PICK1 and NSF have been shown to be important for regulating AMPAR trafficking and synaptic plasticity (Song and Huganir, 2002). Long-term maintenance of LTP and LTD requires new protein synthesis (Huber et al., 2000; Nguyen and Kandel, 1996; Otani and Abraham, 1989) and both LTP and LTD-inducing stimuli enhance the production of an overlapping set of proteins through shared biochemical pathways such as the MAPK cascade and the mTOR pathway. It has been hypothesized that synapses are “tagged” by plasticity inducing stimuli and capture proteins that subserve LTP and LTD (Frey and Morris, 1997). The molecular nature of the tag or the precise proteins that are captured are unknown. Both LTP and LTD are disrupted in Arc KO mice (Plath et al, 2006), and Arc-induced synaptic depression mimics LTD in slices (Verde et al, 2006). This evidence suggests that Arc may be a critical component for bidirectional plasticity. The direction may be governed by the nature of the tag or the context/prior history of the neuron (i.e Arc could critically regulate the metaplasticity of the cell). Basal synaptic transmission and mEPSCs are normal in slices from Arc KO mice (Plath et al., 2006). However, slices have much lower spontaneous network activity than primary culture (see Plath et al., 2006), and thus Arc-dependent homeostasis is more apparent in our culture model. In addition, there are numerous homeostatic mechanisms that control basal transmission including intrinsic excitability and modulation of ion channel properties (Marder and Goaillard, 2006). Since Arc is deleted from birth, other homeostatic processes could compensate and keep basal transmission normal. Further experiments that manipulate Arc for acute periods in vivo are needed to directly look at whether Arc is required for the maintenance of normal basal transmission in a large population of cells.
It is intriguing that Arc is induced in the genomic response to both homeostatic synaptic plasticity and protein synthesis-dependent Hebbian plasticity, as it has been proposed that they share underlying molecular mechanisms (Yeung et al., 2004). In vivo evidence for the importance of Arc’s role in homeostatic plasticity comes from recent work showing that Arc KO mice have deficits in orientation tuning in the visual cortex (Wang et al., 2006). Arc KO mice exhibit activation of a larger neuronal ensemble with reduced orientation specificity. As a consequence, the orientation selectivity of neuronal spiking activities is significantly reduced. Suppression of responses at non-preferred orientations was critically dependent on Arc protein, whereas responses at preferred orientations were not significantly affected, suggesting Arc plays a more critical role in less active neurons and thus may increase the signal-to-noise of the system.
Exactly how homeostasis affects Hebbian plasticity remains unclear. It is possible that Arc’s effects on LTP/LTD are secondary to poor homeostatic scaling in neurons or these two processes may be independent of each other. Our manipulations of Arc are cell wide, whereas in vivo regulation may only affect specific neuronal circuits. Thus, our data do not rule out a role of Arc in metaplasticity of specific synapses.
Arc KO mice exhibit deficits in long-term consolidation of memory (Plath et al., 2006) that could be linked to changes in AMPAR homeostasis. Arc is normally expressed at high levels in brain of awake, behaving animals (Lyford et al., 1995). The time course of the mRNA and protein expression is rapid and transient, and returns to baseline levels within 2 hrs (Ramirez-Amaya et al., 2005). However, Arc can be repeatedly induced in the same network with repetition of the same behavioral paradigm (Guzowski et al., 2006) and is reactivated during periods when animals are resting or sleeping as the hippocampus is “off line” (Ramirez-Amaya et al., 2005). Thus, Arc protein can be persistently upregulated in specific neurons as they engage in network-specific learning and consolidation. Our studies predict that homeostatic-scaling is an ongoing process in these neurons and is essential for consolidation.
In summary, our findings suggest that Arc controls surface AMPAR levels in a homeostatic manner, and acts to keep surface levels and subunit composition optimal for Hebbian plasticity. These processes are required for consolidation of information. Arc may also play a key role in cognitive disorders, as disruption of Arc expression has been observed in Fragile X mental retardation (Zalfa et al., 2003) and Alzheimer’s disease (Dickey et al., 2003; Palop et al., 2005). Further understanding of Arc’s function and regulated expression should help elucidate the molecular mechanisms of memory storage by linking protein-synthesis dependent synaptic plasticity, AMPAR trafficking and homeostatic synaptic scaling.
All the expression constructs were made by PCR. Internal deletion and point mutant were made either using QuikChange Site-Directed Mutagenesis Kit (Stratagene) or by megaprimer method (Barik, T. 2002.PCR Cloning Protocols, Vol 192, 2nd edn, Humana Press). The sequence of the primers used to generate each mutant will be supplied upon request. PCR products were cloned into expression vectorpRK5 (Genentech). All constructs were verified by sequencing.
All antibodies were previously described or were acquired commercially: Bassoon (Stressgen), GluR1-N (pAb, JH1816(Rumbaugh et al., 2003)), GluR1-C (JH1710;(Ye et al., 2000)), GluR2-C (pAb, JH1707 (Blackstone et al., 1992)) pAb, Arc (pAb (Lyford et al., 1995)), Arc (mAb, Santa Cruz),β-actin (mAb, Sigma),α-CaMKII (mAb, Boehringer Mannheim), PSD-95 (mAb, Affinity Bioreagents).
Arc ORF was first subcloned into pIRES2-EGFP and transferred into pSinRep5 (Invitrogen). At 14-21 days in vitro (DIV), cultured neurons were infected with virus. Experiments were usually performed 12-16 hrs after infection.
The subcellular fractionation procedure was performed according to the technique of Huttner et al.(Huttner et al., 1983). All procedures were performed at 4°C. In brief, rat brains were homogenized in 10 volumes of buffered sucrose (0.32 M sucrose, 4 mM HEPES/NaOH, pH 7.4, 1mM EDTA, 1mM EGTA and protease inhibitors cocktail) with a glass-Teflon homogenizer. The homogenate was centrifuged at 800Xg for 15 min and the supernatant was collected. The supernatant was again centrifuged at 9000Xg for 15 min and pellet was collected as crude synaptosomal fraction, P2.
Six to eight weeks old Arc Wt and KO mice were sacrificed by decapitation and forebrain regions dissected. Protein concentration of an aliquot of total homogenate was measured. Rest of the homogenate used for P2 fractionation as described above and protein concentration of an aliquot of P2 measured. Equal amount of total protein from wt and ko was loaded. For coimmunoprecipitation, crude synaptosomal fraction (P2) was sonicated in PBS with 1% Triton X-100, 1mM EDTA, 1mM EGTA and protease inhibitors cocktail (Roche). The homogenate was centrifuged at 100,000Xg for 20 min at 4°C and supernatants with equal amount of protein were incubated with 2 ug of Rabbit polyclonal antibodies for GluR1 and GluR2. After 1.5 hr of mixing at 4°C Protein A agarose slurry was added and incubated for another hour. The beads were washed with PBS+1%Triton X-100 three times and eluted with SDS loading buffer. The samples were than analyzed by SDS-PAGE and western blotting.
For surface biotinylation, infected or drug-treated cortical neurons were cooled on ice, washed twice with ice-cold PBS containing 1mM CaCl2 and 0.5 mM MgCl2, and then incubated with PBS containing 1mM CaCl2, 0.5 mM MgCl2, and 1 mg/ml Sulfo-NHS-SS-Biotin (Pierce) for 30 min at 4°C. Unreacted biotin was quenched by washing cells three times with ice-cold 100mM Glycine (pH7.4). Cultures were harvested in RIPA buffer. Homogenates were centrifuged at 132,000 rpm for 20 min at 4°C. The resulting supernatant volume was measured and 15% of it separated as the total protein. The remaining 85% of the homogenate was rotated overnight at 4°C with Streptavidin beads (Pierce). Precipitates were washed with RIPA buffer and analyzed by immunoblotting with each antibody.
Low-density hippocampal neurons were prepared as described previously (Banker and Cowan, 1977). High-density cortical cultures from embryonic day 18 (E18) rat pups were prepared as reported previously (ADD Refernce). Mouse cultures were prepared in a similar manner from E16.5-E17.5 mouse pups. Neuronal transfections were performed with Lipofectamine 2000 (Invitrogen, Carlsbad, CA) in DIV 12-14 neurons and were analyzed 16-24 hr after initial incubation.
Cells were fixed in 4% paraformaldehyde, 4% sucrose containing PBS solution for 20 min at 4°C and were subsequently permeabilized with 0.2% Triton X-100 in PBS for 10 min. Cells were then blocked for 1 hr in 10% normal donkey/goat serum (NGS). Primary antibodies were diluted in 10% NGS and incubated with neurons for 1 hr at room temperature or overnight at 4°C. Alexa488, Alexa555, or Alexa647-conjugated secondary antibodies (1:500; Molecular Probes, Eugene, OR) to the appropriate species were diluted in 10% NDS and incubated at room temperature for 1 hr. Coverslips were mounted on precleaned slides with PermaFluor and DABCO.
To label surface GluR1-containing AMPA receptors, 2.5 μg of GluR1-N JH1816 pAb was added to neuronal growth media and incubated at 10°C for 20 min. The unbound excess antibody was quickly washed with fresh warmed growth medium and then fixed and mounted according to the methods described above.
Glycine stimulation was carried out as described before (Watt et al., 2004). Neurons were incubated at 37 °C and 5% CO2 for 15 min in ACSF containing the following: NaCl, 126 mM; KCl, 5.5 mM; MgSO4, 0.4 mM; NaH2PO4, 1 mM; NaHCO3, 25 mM; CaCl2, 2 mM; dextrose, 14 mM; glycine, 0.2 mM; bicuculline, 0.01 mM. Then, we replaced the medium in the culture dishes and returned them to the tissue culture incubator until immunocytochemistry was performed. Immunofluorescence was viewed and captured using a Zeiss LSM 510 confocal laser scanning microscope. Quantification of surface GluR1 puncta were carried out essentially as described (Rumbaugh et al., 2003), using Metamorph imaging software (Universal Imaging, Downingtown, PA). Images were acquired and saved as multi-channel TIFF files with a dynamic range of 65536 gray levels (16-bit binary; MultiTrack acquisition for confocal). To measure punctate structures, neurons were thresholded by gray value at a level close to 50% of the dynamic range. Background noise from these images was negligible. After a dendrite segment was selected, all puncta were treated as individual objects and the characteristics of each, such as pixel area, average fluorescence intensity, and total fluorescence intensity, were logged to a spreadsheet. In addition, each dendrite length was logged in order to calculate puncta density and total intensity per dendritic length (all values shown are per 10 μm of dendrite). untransfected cells in individual coverslips. The average single pixel intensity from each region was calculated and averages from all regions were derived. Significance was determined by a paired Student’s T test.
Whole-cell patch-clamp recordings were performed from forebrain cultures at the day in vitro (DIV) indicated. To isolate AMPAR-mediated mEPSCs, neurons were continuously perfused with artificial cerebral-spinal fluid (aCSF) at a flow rate of 1 ml/min. The composition of aCSF was as follows (in mM): 150 NaCl, 3.1 KCl, 2 CaCl2,1 MgCl2, 10 HEPES, 0.1 DL-APV, 0.005 strychnine, 0.1 picrotoxin, and 0.001 tetrodotoxin. The osmolarity of aCSF was adjusted to 305-310 and pH was 7.3-7.4. Intracellular saline consisted of (in mM): 135 Cs-MeSO4, 10 CsCl, 10 HEPES, 5 EGTA, 2 MgCl2, 4 Na-ATP, and 0.1 Na-GTP. This saline was adjusted to 290-295 mOsm, and pH was 7.2.
Transfected neurons were selected based on fluorescent (eGFP) signal. Once the whole-cell recording configuration was achieved, neurons were voltage-clamped and passive properties were monitored throughout. In the event of a change in series resistance (Rs) or input resistance (Ri) >15% during the course of a recording, the data were excluded from the set. mEPSCs were acquired through a MultiClamp 700A amplifier (Axon Instruments, Union City, CA), filtered at 2 kHz, and digitized at 5 kHz. Sweeps of 20 seconds with zero latency (essentially “gap free”) were acquired until a sufficient number of events were recorded (a minimum of 5 and no longer than 30 minutes). Data were recorded continuously only after a period of two minutes, during which the cell was allowed to stabilize. mEPSCs were detected manually with MiniAnalysis software (Synaptosoft Inc, Decatur, GA) by setting the amplitude threshold to √ RMS x 3 (usually 4 pA). Once a minimum of 100 events had been collected from a neuron, the amplitude, frequency, rise time (time to peak), decay time (10-90%), and passive properties were measured. In all electrophysiological experiments, a similar amount of data was acquired from both transfected and untransfected neurons on the same day. We have found that recording transfected neurons followed by recording an untransfected neuron in the immediate vicinity yields remarkably consistent results. This is likely a result of reducing errors that arise from slight changes in neuronal density between preps or changes in the density of neurons from different areas of a coverslip. These parameters were crucial for obtaining reliable, low variability data between experimental populations. Data from each group were then averaged, and statistical significance was determined by Student’s T test (unless noted otherwise). All electrophysiological experiments were performed from at least two individual platings of neurons from three different transfections.
This work was supported by grants from NIMH (P.F.W), NIMH Conte Center (R.L.H. P.I.), and the Howard Hughes Medical Institute (R.L.H).