A developmental switch of AMPA receptor subunits in neocortical pyramidal neurons

Abstract

AMPA receptors mediate most of the fast excitatory neurotransmission in the brain, and those lacking the glutamate receptor 2 (GluR2) subunit are Ca(2+)-permeable and expressed in cortical structures primarily by inhibitory interneurons. Here we report that synaptic AMPA receptors of excitatory layer 5 pyramidal neurons in the rat neocortex are deficient in GluR2 in early development, approximately before postnatal day 16, as evidenced by their inwardly rectifying current-voltage relationship, blockade of AMPA receptor-mediated EPSCs by external and internal polyamines, permeability to Ca(2+), and GluR2 immunoreactivity. Overall, these results indicate that neocortical pyramidal neurons undergo a developmental switch in the Ca(2+) permeability of their AMPA receptors through an alteration of their subunit composition. This has important implications for plasticity and neurotoxicity.

Fig. 1.

Synaptically activated AMPA currents in pyramidal neurons. A–E, Schematic of the model for studying intracortical excitability and properties of pharmacologically isolated monosynaptic EPSCs evoked in pyramidal cells during callosal stimulation. A, Cortical slice preparation used in the study showing placement of stimulating (S₁, intracortical; S₂, callosal) and recording (R) electrodes; Fr1,Fr2, frontal cortex areas 1 and 2; CA1, region of hippocampus; cg, cingulum; cc, corpus callosum (principal commissural pathway linking both cerebral hemispheres). Scale bar, 1 mm. B, High-magnification view (Scale bar, 40 μm) depicting the typical morphology of the cells used in electrophysiological recordings.C, Photomicrograph of one such cell (P15) filled with biocytin and visualized after fixation with diaminobenzidine (Scale bar, 100 μm) showing extensive arborization of apical and basal dendrites typical of a pyramidal neuron. C, Left panel, Portion of the adjacent cortex stained with thionin to reveal cortical lamination. D, E, Examples of averaged EPSCs from the same neuron evoked by callosal stimulation (S₂) under the indicated conditions. PTX, Picrotoxin. Note that responses were evoked with a fixed latency from stimulus onset (filled triangle) and the late, slowly decaying component at +30 mV inD was modified by d-APV (E) indicating activation of NMDA receptors. Responses in E were mediated entirely by AMPA receptors, had identical rise-times at ±30 mV, and were best fit with single exponentials with the indicated time constants. The above approach (D, E) was used to isolate the pure AMPAR-mediated component in all subsequent experiments. F, G, Rectification of AMPAR-mediated EPSCs is dependent on age but is a property common to synaptic receptors on pyramidal neurons.F₁, F₂, Averages of EPSCs evoked at various holding potentials in the same neurons during concomitant alternate stimulation of the callosum (filled arrowheads) and local intracortical excitatory afferents (open arrowheads). Peak synaptic currents in P13-P15 neurons (F₁) were generally smaller at positive holding potentials than those at corresponding negative levels, in contrast to P16–P21 neurons (F₂), whose EPSCs had similar amplitudes. Callosally evoked EPSCs showed similar age-dependent rectification as those evoked via intracortical stimulation (n = 11). G, Rectification indices for the two stimulation paradigms were similar within the respective age groups but differed significantly between age groups. *p < 0.05; **p < 0.01.

Fig. 2.

GluR2 expression in pyramidal neurons at different ages. A, B, Dual-channel images of recorded biocytin-filled layer 5 pyramidal neurons from P14 (A₁–A₃) and P18 (B₁–B₃) animals immunostained with the GluR2 antibody (A₁,B₁, biocytin-Texas Red;A₂, B₂, GluR2-FITC;A₃, B₃, color overlap). Note that pyramidal neurons, including biocytin-filled cells (A₂, B₂,arrows), in the older age group show higher levels of GluR2 immunofluorescence compared with younger under similar conditions of illumination. Scale bars: A₁,A₂, B₁,B₂, 50 μm;A_3, B₃, 20 μm.C, D, High-magnification images of young (C₁–C₃) and old (D₁–D₃) biocytin-filled neurons showing that although the intracellular level of GluR2-FITC fluorescence in the young animal (C₂) is comparable with adjacent neuropil and with the intranuclear staining, it is markedly more intense in the corresponding regions of the older neuron (C₂ vs D₂). Note the absence of cross-talk between Texas Red (C₁,D₁) and FITC (C₂,D₂) channels. Immunofluorescence was measured at four locations (C₂ andD₂, red circles) within the perikaryon (outlined with dashed yellow lines), and the average of these values was used to represent cell brightness (see Materials and Methods). Scale bars, 10 μm. E, GluR2 levels in P13-P15 (n = 6) and P16-P21 (n = 9) biocytin-filled pyramidal neurons based on perikaryon immunofluorescence measurements compared with background as outlined in C, D. F, Changes in GluR2 immunoreactivity relative to coexpressed GluR1 (or 4) measured separately in a population of layer 5 pyramidal cells from corresponding cortical regions in intact brain sections of animals at the indicated ages (P12, n = 142 cells, 11 sections; P21, n = 83 cells, 10 sections). *p < 0.05; ***p < 0.0001.

Fig. 3.

Inclusion of NHPP-spermine (50 μm) in the pipette does not alter age-dependent differences inI–V characteristics of AMPAR-mediated EPSCs.A, Superimposed averaged records of synaptic currents evoked in a P13 neuron at various holding potentials (−70 to +40 mV; step size, 10 mV; isolated in cocktail solution and 50 μmd-APV) with the polyamine included in the patch pipette. The corresponding I–V curve, normalized to the EPSC amplitude at −70 mV, is shown in the inset.B, Normalized I–V relationship of the pooled data taken from P13–P15 neurons showing inward rectification of the EPSCs. Each point on the plot (open circle) represents an ensemble average of 14 experiments, and error bars indicate SEM where this is greater than the size of the symbol. The dashed line is an extension of the linear regression fit of data points at negative holding potentials. C, D, In contrast with A, B, EPSCs recorded from neurons in an older animal were nonrectifying (C), and the composite I–Vprofile, averaged from eight P16–P21 neurons, was linear throughout the entire voltage-range (D). E, F, RIs for the two age groups and averaged holding currents recorded at various membrane potentials in these experiments, respectively. ***p < 0.0001.

Fig. 4.

Effects of extracellular (A) and intracellular (B) polyamine on EPSC amplitude. A₁, A₂, Pooled data (n = 7) showing the time course of changes in response amplitudes (percentage of control, −70 mV) of neurons in different age groups in the presence of 5 μmNHPP-spermine applied externally at the times indicated (black bar; 100% = average of all control values). EPSCs inA₁ were blocked by 37.1 ± 8.1% as opposed to 1.8 ± 4.6% in A₂ (values correspond to the second designated time point in A₁,A₂, respectively). Tracesrecorded at the times indicated are shown below their respective time plots. EPSCs in all these experiments were isolated in a cocktail andd-APV. B₁,B₂, Inclusion of NHPP-spermine (50 μm) in the patch solution had differential effects on EPSCs depending on the developmental age of the animal. Changes in EPSC amplitude after break-in (t = 0 min; holding potential, −70 mV, ACSF) from neurons in the younger (B₁; n = 6) and older (B₂; n = 7) age groups under different experimental conditions are shown. NHPP-spermine reduced the amplitude of evoked responses during both callosal (●) and intracortical (▴) stimulation in the younger (B₁) but not in the older (B₂) age group. Callosal EPSCs recorded with polyamine-containing patch electrodes were significantly smaller compared with those with control solution (○). Eachpoint on the plot represents an ensemble average of the EPSC amplitudes, and error bars indicate SEM. Responses are normalized to the averaged EPSC at t = 0 min (100%). Series resistance (■) in these experiments was constant, as shown inB₁. Traces at thetop of the plots are averaged sample records obtained at the times indicated.

Fig. 5.

Experimental design and validation of the procedure used to determine Ca²⁺ permeability of synaptic AMPARs. A, Schematic of the experimental configuration showing relative placements of the intracortical stimulating electrode and local perfusion pipette with respect to the patch recording electrode. B, Representative traces of AMPAR-mediated responses (−70 mV) during perfusion of solutions with different Ca²⁺ concentrations. Tracesshown are from a P18 animal. Note that it required higher stimulus intensities to evoke EPSCs in high Ca²⁺. The averaged (percent) increments in charge required to elicit responses when [Ca²⁺]_o was sequentially elevated from 1.8 to 10 and from 10 to 30 mm were 60.9 and 20.8%, respectively, for P13–P15 neurons and 70.9 and 41.9%, respectively, for P16–P21 neurons. Inclusion of 10 μm NBQX in the local perfusate blocked all EPSCs evoked in high Ca²⁺. The numbered responses inB are overlaid such that the second response is scaled to match the amplitude of the first response. Note the fixed latency from stimulus onset for the two responses and the similarity in the shape of the waveforms. C, D, mEPSC frequency is increased after local elevation of [Ca²⁺]_o via the local perfusion system. C, Representative traces of mEPSCs recorded from a P19 layer 5 pyramidal neuron in the presence of 1 μmTTX, 50 μm picrotoxin, and 100 μm APV under the indicated conditions. D, Time course of changes taken from a different neuron showing the fast onset and offset of the altered mEPSC frequency. E, Bar plot of the averaged mEPSC frequency under different [Ca²⁺]_o conditions (n = 8 and 5 cells for 10 mm Ca²⁺ andwash, respectively) normalized to the mean value incontrol. *p < 0.05.

Fig. 6.

Differential Ca²⁺ permeability of synaptic AMPARs within the identified age groups.A–C, Families of representative traces (A, B, ±40 mV; step size, 10 mV) and their correspondingI–V relationships (C₁,C₂) recorded from animals in the two age groups under various [Ca²⁺]_o conditions. Theasterisks in A indicate the response at a holding potential of 0 mV. D, Plot of theE_rev values, measured fromI–V relationships, as a function of [Ca²⁺]_o. Each pointrepresents the mean of the indicated number of experiments in the respective age groups. Note the leftward shift inE_rev for the older animals in contrast with the opposite trend (dashed lines) observed in the younger animals. Statistical comparisons are between the reversal potentials for animals in the two age groups. *p < 0.02; **p < 0.005.