70, No. 5, November 1993. Printed in U.S.A.. NMDA Receptor-Mediated Components of Miniature Excitatory. Synaptic Currents in Developing Rat Neocortex.
JOURNALOF NEUROPHYSIOLOGY Vol. 70, No. 5, November 1993. Printed in U.S.A.
NMDA Receptor-Mediated Components of Miniature Excitatory Synaptic Currents in Developing Rat Neocortex EDWARD C. BURGARD AND JOHN J. HABLITZ Neurobiology Research Center and Department of Physiology and Biophysics, University of Alabama at Birmingham, Birmingham, Alabama 35294
SUMMARY
AND
INTRODUCTION
CONCLUSIONS
1. In vitro slices of frontal neocortex were prepared from rat pups at various agesafter birth: postnatal days (PN) 3-56-8, and 9- 14. Using whole-cell patch-clamp techniques, both spontaneous and evoked excitatory postsynaptic currents (EPSCs)were recorded from voltage-clamped layer II-III pyramidal neurons. Developmental changes in EPSCswere examined. 2. Four properties of miniature EPSCs(mEPSCs)were studied: rise time, amplitude, decay time constant (7), and frequency. These parameters were not tetrodotoxin sensitive and did not exhibit significant developmental changes during the first two postnatal weeks. 3. mEPSCs occurred approximately every 2-3 s and had peak amplitudes of 25-30 pA. Within each age group, certain parameters of mEPSCs were voltage dependent. mEPSC rise time and decay 7 were significantly increased at depolarized potentials (-30 to -45 mV) relative to hyperpolarized (-75 to -90 mV) or resting membrane potential (RMP) (-60 to -70 mV). 4. At threshold stimulation intensity, EPSCswere evoked in an “all-or-none” manner. The amplitude and decay 7 of evoked unitary EPSCsand mEPSCswere not significantly different. As stimulation intensity was increased, a late EPSC component appeared that was not seen in mEPSCs. At suprathreshold stimulus intensities, EPSCduration was significantly longer in PN 3-5 than in PN 9- 14 neurons. 5. The N-methyl-D-aspartate (NMDA) receptor antagonist D( -)2-amino-5-phosphonovaleric acid (APV, 10 PM) significantly decreased mEPSC decay 7 and frequency only at depolarized membrane potentials. Likewise, EPSCs were depressed by APV to a greater extent at depolarized potentials, and the depression was mainly of the late component. mEPSCsrecorded at RMP were blocked by the non-NMDA receptor antagonist 6-cyano-7nitroquinoxaline-2,3-dione (5 PM). 6. Removal of extracellular Mg2+ reversibly increased the decay 7 of mEPSCs at RMP but not at depolarized membrane potentials. The decay 7 and duration of evoked EPSCs were also increased in zero Mg2+. These effects were reversible with application of APV. All effects of zero Mg2+ on mEPSCsand EPSCswere observed as early as PN 3-5. 7. These results indicate that the basic kinetic properties of mEPSCs are present by PN 3-5 and do not change significantly over the first two postnatal weeks. NMDA receptor activation contributes to mEPSCs and sensitivity to Mg2+ appears as early as PN 3-5. Unitary EPSCs resemble mEPSCs, but a late NMDA receptor-mediated component appears in EPSCs as stimulus intensity is increased. The developmental decrease in the late EPSC component as well as its apparent increase in voltage sensitivity cannot be attributed to changes in characteristics of miniature synaptic currents.
0022-3077/93
Numerous studies have shown that functional synaptic activity is essential for development of the nervous system. Original studies on the neuromuscular junction described a trophic influence of presynaptic acetylcholine release on postsynaptic receptor clustering and strengthening of synaptic contacts (see Schuetze and Role 1987 for review). Various neurotransmitter systems have been implicated in functional development of the CNS (Mattson 1988). In particular, activation of the N-methyl-D-aspartate (NMDA) subtype of glutamate receptor has been shown to have a trophic effect on developing neurons (Balazs et al. 1988; Pearce et al. 1987), whereas blockade of NMDA receptors disrupts certain forms of developmental visual plasticity (Constantine-Paton et al. 1990). These studies suggest that NMDA receptor activation regulates the development of both neurons and synaptic connections. NMDA receptors are present and activated in embryonic neocortex (Lo Turco et al. 199 1) and contribute to evoked synaptic activity in embryonic spinal cord (Ziskind-Conhaim 1990). However, a number of developmental changes in receptor properties have been described. Postnatally, NMDA receptor density peaks during the first few weeks in hippocampus (Tremblay et al. 1988), neocortex (McDonald and Johnston 1990), and spinal cord (Kalb et al. 1992), and then declines. The voltage dependence (Ben-Ari et al. 1988), sensitivity to Mg2+ (Bowe and Nadler 1990; Kleckner and Dingledine 199 1; Morrisett et al. 1990), and affinity for various ligands (Kalb et al. 1992) of the NMDA receptor have all been reported to change during the first few postnatal weeks, suggesting a change in receptor structure during development. It is also likely that the expression of molecularly distinct subtypes of NMDA receptors (Kutsuwada et al. 1992; Monyer et al. 1992) will be developmentally regulated, as is the case for glutamate receptors (Bettler et al. 1990; Gallo et al. 1992; Monyer et al. 199 1). If NMDA receptor activation is necessary for neuronal and synaptic development, it would be expected that NMDA receptors contribute substantially to synaptic activity in the neonate. Likewise, if properties of NMDA receptors change during development, then NMDA receptormediated synaptic activity might also be expected to change. We have previously described developmental changes in NMDA receptor-mediated components of evoked excitatory postsynaptic potentials (EPSPs) and excitatory postsynaptic currents (EPSCs) during the first two
$2.00 Copyright 0 1993 The American Physiological Society
1841
1842
E. C. BURGARD
postnatal weeks in rat neocortex (Burgard and Hablitz 1993). Both EPSPs and EPSCs are long in duration and have substantial late components mediated by NMDA receptors during this period. However, the NMDA component decreases significantly between postnatal days (PN) 3-5 and 6-8. By PN 9- 14, expression of the NMDA component can be largely masked by the onset of evoked GABAergic inhibition. The mechanisms underlying developmental changes in NMDA receptor-mediated components remain unclear. One possibility considered is that EPSPs reflect the summation of activity at unitary synapses. This predicts that spontaneous miniature EPSCs (mEPSCs) would also display a substantial, long-duration NMDA component that decreases with development over the first two postnatal weeks. Alternatively, the changes associated with EPSCs could be due to processes other than changes in elementary components of synaptic transmission. In the present study, we have compared the properties of mEPSCs and evoked EPSCs with attention to the contribution of NMDA receptors. This study was designed to measure synaptic responses under relatively physiological recording conditions [ 1.3 mM extracellular Mg*+, membrane potentials -30 to -70 mV, and no y-aminobutyric acid (GABA) receptor antagonists] to define the contribution of NMDA receptors to normal synaptic activity. We report that, unlike EPSCs, mEPSCs do not exhibit appreciable developmental changes during the first two postnatal weeks. Preliminary reports of these studies have appeared in abstract form (Burgard and Hablitz 1992). METHODS
Slice preparation Rat pups were decapitated between PN 3 and 14 (day of birth = PN 0). The brains were removed by dissection and immersed in physiological saline at 4°C. Coronal slices of frontal neocortex (400 pm thick) were prepared using a Vibroslice (Camden Instruments) and maintained in an incubation chamber at room temperature for 2 1 h before recording. Sliceswere then transferred to an interface recording chamber where they were slowly warmed to 35 & l°C. Slices were continuously perfused with oxygenated (95% O,-5% CO& physiological saline (pH 7.4) at a rate of - 1 ml/min. The saline consisted of (in mM): 125 NaCl, 3.5 KCl, 26 NaHCO,, 1.25 NaH,PO,, 2.5 CaCl,, 1.3 MgSO,, and 10 dextrose. The osmolarity of this saline was 300-3 10 mOsm. A thin nylon net was placed on top of each slice in the recording chamber to aid mechanical stability.
Electrophysiology Whole-cell patch-clamp recording techniques were used as described previously (Blanton et al. 1989; Burgard and Hablitz 1993; Otis et al. 199 1). Briefly, patch pipettes (tip resistance 3-4 Mfi) were filled with a solution consisting of (in mM): 125 Cs-acetate, 10 CsCl, 0.5 ethylene glycol-bis(P-aminoethyl ether)-N, N, N’,N’tetraacetic acid, 10 N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid, and 2 MgATP. The osmolarity of this solution was 260-270 mOsm, and the pH was adjusted to 7.2. The recording pipette was positioned in upper cortical layers, corresponding to layer II-III in PN 6- 14 pups, or the cortical plate/presumptive layer II-III in PN 3-5 pups (Miller 198 1). Resting membrane potential (RMP) and input resistance (R,) were recorded in bridge mode immediately after obtaining a whole-cell recording. R, was
AND
J. J. HABLITZ
measured from the membrane voltage change in response to a hyperpolarizing current pulse (0.1 nA, 500 ms). Voltage-clamp recordings were obtained using a discontinuous single-electrode voltage-clamp amplifier (NPI SEC1L, Tamm, FRG) at a switching frequency of 27 kHz and a 25% duty cycle. Responses were filtered at 3 kHz and stored on videotape (Neurocorder DR484) for later analysis. A bipolar stimulating electrode (platinum-iridium wire, 30 pm diam) was placed intracortically below the recording pipette approximately in layers IV-V. Stimuli consisted of step current pulses between 5 and 100 PA in amplitude and 20-200 ps duration. Stimulation of fibers projecting to layers II-III was delivered at a frequency of 0.033 Hz to prevent the frequency-dependent depression of synaptic responses seen in immature neurons (Burgard and Hablitz 1993). Because of the low stimulus frequency used, evoked traces shown in the figures are averages of two to three consecutive responses. To study the developmental morphology of pyramidal neurons, the fluorescent dye Lucifer yellow CH (0.5-2% dipotassium salt, Sigma) was added to the recording pipette. Neurons were filled for l- 10 min by hyperpolarizing current injection. Sliceswere then fixed overnight in a 4% paraformaldehyde-0.1 M phosphate buffer, cleared in dimethyl sulfoxide, photographed, and drawn.
Data analysis Responses were filtered at l-3 kHz, digitized at 1O-20 kHz, and analyzed using SCAN software (courtesy J. Dempster, StrathClyde). EPSC parameters analyzed were lo-90% rise time, peak amplitude, frequency of events, and decay time constant (7). Decay 7s were calculated from single or double exponential fits of EPSC decays using a modified Levenberg-Marquardt leastsquares algorithm. Correlation coefficients (r>were calculated using a least-squares linear regression analysis. All values in the text are expressed as means t SD. “Control” responseswere recorded from an untreated sample, whereas “baseline” responseswere recorded in the same neurons before treatment. Statistical analysis consisted of unpaired t test or one-way analysis of variance (ANOVA) plus post hoc t tests with Bonferroni correction. A significance level of 0.05 was used.
Compounds The following compounds were bath applied to the slices: D( -)2-amino-5-phosphonovaleric acid (APV), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, Cambridge Research Biochemicals), bicuculline methiodide, and tetrodotoxin (TTX, Sigma). RESULTS
Pyramidal cells in layers II-III undergo rapid morphological development in the first two postnatal weeks. This is demonstrated in Fig. 1, which shows the appearance of two Lucifer yellow-injected neurons from slices of neocortex prepared at different postnatal ages. The soma of the PN 4 neuron in Fig. 1A is located in the cortical plate (presumptive layer II-III), and a single apical dendrite extends toward the pial surface. This was typical of PN 3-5 neurons, which were characterized by a relatively unbranched apical dendrite and a few small basal dendrites projecting from the soma. By PN 10 (Fig. 1B) the neurons had increased in both size and complexity. The apical dendrite was now typically bifurcated, with smaller dendrites also projecting from the major branches. Basal dendrites extending from the soma were numerous and exhibited a complex branching pattern approaching that of adult neurons (Miller 198 1). The increase in dendritic number and branches developed gradu-
MINIATURE
A
PN4 ----------me---
B
EPSCs DURING
PN10 -------------es
FIG. 1. Morphological development of layer II-III pyramidal neurons. Camera lucida drawings of Lucifer yellow-filled neurons obtained from postnatal day (PN) 4 (A) and PN 10 (B) neocortex. The basic development of apical and basal dendritic branching is represented, although the axons and ends of dendrites extend beyond the drawings. Dashed lines: pial surface. Calibration bar: 50 pm.
ally from PN 3 and continued through PN 14 (the extent of this study). Basic neuronal properties Recordings from 97 layer II-III pyramidal neurons, in pups ranging in age from PN 3 to 14, were included in this study. The neurons were divided into three groups on the basis of age (PN 3-5, PN 6-8, and PN 9-14). This follows the protocol established previously (Burgard and Hablitz 1993) where significant differences were observed between age groups with respect to the synaptic contribution of NMDA receptors and development of GABAergic inhibition. Membrane properties of neurons in each age group are shown in Table 1. RMP averaged between -60 and -65 mV and R, averaged between 200 and 300 MQ. No significant developmental changes were observed in these parameters. We have previously shown that when a potassium isethionate-based intracellular pipette solution is used RMP becomes more negative and R, decreases with age (Burgard and Hablitz 1993). In the present study, using cesium as the principal intracellular cation, no significant developmental differences were observed in either RMP or R, between groups. Developmental changes in potassium conductances (Spigelman et al. 1992), which may be masked when cesium is present intracellularly, may contribute to these differences. Spontaneous mEPSCs Spontaneous EPSCs were present in recordings from all age groups. Bath application of 6 PM TTX did not significantly change the frequency or amplitude of spontaneous
CORTICAL
DEVELOPMENT
1843
EPSCs (n = 7), suggesting that they were generated independently of action potentials. It is assumed that they represent quanta1 synaptic events, and the term mEPSC is therefore used to describe these events. Under the present recording conditions, spontaneous activity was relatively infrequent, with mEPSCs occurring approximately every 2-3 s in all age groups. Although mEPSCs were recorded in most neurons, the probability of encountering neurons without mEPSCs was higher in the PN 3-5 group. Although mEPSC frequency indicated an increasing trend over the first 2 wk, the differences were not significant. GABAergic inhibitory postsynaptic currents were not observed in PN 3-8 animals. Two factors account for this: first, the expected chloride reversal potential (-69 mV) was near RMP, minimizing the driving force for chloride, and second, evoked GABAergic inhibitory postsynaptic potentials are not evident in neocortex until the second postnatal week, suggesting that functional GABA synapses are late developing (Burgard and Hablitz 1993; Luhmann and Prince 199 1). Pharmacological blockade of inhibition was not employed because spontaneous chloride-mediated outward currents were only observed at depolarized membrane potentials in PN 13- 14 neurons, and analysis of evoked currents was restricted mainly to PN 3-8 neurons. Representative mEPSCs from a PN 8 neuron are shown in Fig. 2. At a holding potential of -80 mV, the mEPSC consisted of an inward current that decayed with a single exponential time course. When the holding potential was changed to -40 mV, the peak amplitude of the response was reduced and the decay prolonged. When the records were scaled to the same peak amplitude and superimposed, the increase in decay was apparent (Fig. 2C). To quantify developmental changes in mEPSCs, four basic properties of mEPSCs were measured. The results for each age group are summarized in Table 1. This table indicates that all parameters exhibited small nonsignificant increases with age. Because no significant developmental differences were observed, we have concluded that these four properties exhibit no general developmental pattern between PN 3 and 14. The basic kinetic properties of spontaneous mEPSCs appear to be determined by PN 3. As indicated in Fig. 2, mEPSCs exhibited voltage-depen1. Properties of cell membranes and mEPSCs in immature neurons
TABLE
Parameter Neurons RMP, mV R,, MQ mEPSCs Rise time, msec Amplitude, pA Decay T, ms Frequency, Hz
PN 3-5
n
PN 6-8
n
-62 + 8 278+91
43 43
-64 fi 6 227 I!I 104
40 39
-60 k 6 206 I!Z 104
14 14
1.5 27.3 3.6 0.29
20 22 22 20
1.8 ? 0.6 30.9 2 7.2 4.5k1.6 0.41 zk 0.26
23 23 23 23
1.8 30.9 4.7 0.56
13 13 13 12
I?I 0.5 + 7.6 + 1.3 + 0.24
PN9-14
f 0.7 + 7.0 k 2.0 AZ0.43
n
Values are means f SD. RMP was measured immediately on rupture of the cell membrane in whole-cell patch configuration. The membrane potential was subsequently voltage clamped between -60 and -70 mV and all other parameters were measured in this range. Amplitude was measured as the peak current amplitude at 2to 3-ms-latency. All mEPSC decays were best fit by a single exponential function. mEPSCs, minature excitatory postsynaptic currents; PN, postnatal day; RMP, resting membrane potential; R,, input resistance; 7, time constant.
1844
E. C. BURGARD
AND J. J. HABLITZ
A -40 mV
c
-80 mV
c 25
C
pA
20 msec
FIG. 2. Spontaneous miniature excitatory postsynaptic currents (mEPSCs) recorded in a PN 8 neuron. Traces shown are averaged currents recorded at depolarized (A, -40 mV, n = 8) and hyperpolarized (B, -80 mY, n = 14) membrane potentials. Single exponential decay time constant (7) values are given for each trace. C: the 2 traces are scaled and superimposed to demonstrate the prolonged decay at depolarized potentials.
dent changes in amplitude and time course. For analysis purposes, the membrane potentials were divided into three ranges: depolarized (-30 to -45 mV), RMP (-60 to -70 mV), and hyperpolarized (-75 to -90 mV). An ANOVA indicated that no significant differences were observed between any parameters measured at RMP or hyperpolarized potentials. Consequently, we have focused our analysis on differences between recordings obtained at RMP and depolarized potentials. As shown in Table 2, mEPSC decay 7s were significantly longer at depolarized membrane potentials than at RMP (or hyperpolarized potentials, not shown) in all age groups. In addition, mEPSC rise times were prolonged at depolarized potentials. These findings are consistent with increased contribution from NMDA receptors at depolarized potentials (Forsythe and Westbrook 1988), which effectively prolongs the currents. These parameters showed no significant developmental changes in voltage dependence. A histogram of mEPSC amplitudes from a PN 4 neuron is shown in Fig. 3A. It can be seen that the amplitudes apTABLE
2.
pear to be normally distributed around the mean amplitude. A single peak in the amplitude histograms was observed in each cell. Although the relatively small number of events recorded per neuron precluded a quantitative analysis of amplitude distributions, it is apparent that the noise level (~6 PA) was clearly below the mean mEPSC amplitude (35 PA). The range of mEPSC amplitudes was large and varied between neurons. A typical amplitude range for a single neuron was 13-47 pA (n = 30 events), whereas the largest range was 8-l 35 pA (n = 235 events). Examination of individual records indicated that large events were not due to superimposition of multiple small events. However, mEPSCs may result from activity at independent spatially distributed release sites. The anatomic and electrotonic location of the synapses responsible for generation of the mEPSCs recorded here is unknown. Synaptic signals from distant dendritic locations would be expected to be distorted (Rall 1977). However, the results of experiments with conventional intracellular microelectrodes have indicated that neocortical pyramidal cells are electrotonically compact (Larkman et al. 1992; Stafstrom et al. 1984). The higher R, and smaller leak conductances associated with patch-clamp recordings in immature neurons would be expected to make neurons more electrotonically compact, although not isopotential (Spruston and Johnston 1992). To assess the influence of dendritic filtering on mEPSCs, we examined the relationship between the rise time and decay 7 of mEPSCs. Figure 3B shows a plot of rise time versus decay 7 for an individual PN 4 neuron. Regression analysis indicated no significant correlation between these two parameters. This was true in 2 1 of 23 neurons (r < 0.40) from all age groups. The lack of correlation suggests that within individual neurons mEPSCs may be generated within a relatively confined region of the dendritic tree. However, when mean rise time and decay 7 were compared between neurons, a significant correlation was observed (Fig. 3C) regardless of age group. Thus, although mEPSCs may be generated within a distinct region on an individual neuron, the location appears to vary among neurons. Evoked unitary EPSCs To determine the relationship between mEPSCs and evoked EPSCs, we evoked unitary EPSCs by activating a minimal number of fibers using intracortical stimulation in layers IV/V. EPSCs in response to stimuli of increasing intensity are shown in Fig. 4A. At threshold, stimulation at RMP in all age groups (n = 6) evoked EPSCs with approxi-
Voltage dependence of mEPSC properties PN 3-5
Rise time, ms Decay 7, ms Peak/late ratio
PN 6-8
PN 9-14
RMP
Depolarized
RMP
Depolarized
RMP
Depolarized
1.5 k 0.5 (20) 3.6 -t 1.3 (22) 7.8 I!Z5.0 (21)
2.3 + 1.5 (7) 8.4 AI 7.3” (7) 3.7 f 1.9” (7)
1.8 t 0.6 (23) 14.5f 1.6 (23) 6.3 + 4.3 (22)
3.4 + 3.0* (18) 10.7 + 6.2* (17) 2.9 + 1.9* (17)
1.8 & 0.7 (13) 4.7 k 2.0 (13) 8.0 + 5.7 (11)
2.8 + 1.1 (8) 13.7 + 7.5* (8) 2.2 AI 0.6* (8)
Values are means + SD. Basic properties of mEPSCs are voltage dependent. Membrane potential was held at either RMP or depolarized potentials in each age group and mEPSC parameters were measured. Peak/late refers to the ratio of mEPSC amplitudes measured at 2- to 3-ms/ 10 ms onset latency. For abbreviations see Table 1. *Significant difference compared with respective RMP group (P < 0.05, t test) indicated in parentheses.
MINIATURE
EPSCs DURING
t = 2.8 msec --I 5 msec
10
25 40 55 Amplitude (PA)
20
pA
70
B 0 a,
g e
u2
10 8642
l l
0a
a
0 l
-
l
E 0
I
I
1
0.5 1.0 1.5 Rise time (msec)
I
2.0
C : g
8 6
s
2
Fi
0
1
1
I
1.0 2.0 Rise time (msec)
I
3.0
FIG. 3. Basic properties of mEPSCs. A: histogram of peak amplitudes measured at -65 mV in a PN 4 neuron. Events (n = 85) were recorded over a 2-min period and are binned in 1S-pA increments. Inset: digitized average of the 85 mEPSCs. B: plot of rise time (abscissa) vs. single exponential decay 7 (ordinate) for the responses shown in A. Regression analysis revealed no significant correlation between these two parameters in this neuron (r = 0.25). C: plot of rise time vs decay 7 for 20 PN 3-5 neurons recorded at resting membrane potential (RMP). Each point represents mean values for a single neuron. A significant correlation was observed (r = 0.76, P < 0.05) between neurons.
mately the same peak amplitude (38.9 t 2 1.2 PA) and decay 7 (5.4 t 2.1 ms) as mEPSCs. The input-output (I-O) curve for the neuron in Fig. 4A is shown in Fig. 4B. At threshold stimulation, EPSCs measured at 5 ms responded in an “all-or-none” manner, initially peaking near 30 pA. Because threshold stimulation produced EPSCs with amplitudes and decay 7 similar to those of mEPSCs, we consider threshold EPSCs to be unitary events due to activation of a single fiber. As stimulus intensity was increased, the amplitude of the early EPSC also increased and reached a level approximately twice the amplitude of the initial unitary event. In addition, a late component of the EPSC (measured at 25 ms latency) appeared. We have previously shown that this late component is NMDA receptor-mediated (Burgard and Hablitz 1993). The I-O curve for the late component showed a concomitant increase in amplitude with increased stimulus intensity. In six neurons, a late component was not clearly present at RMP until the stimulus intensity was increased to produce an early EPSC ampli-
CORTICAL
DEVELOPMENT
1845
tude of 54.8 t 20.7 pA, a value approximately twice the average mEPSC amplitude. This suggests that activation of more than one fiber may be needed to produce a late EPSC and cooperativity among afferents may be required. In contrast to mEPSCs, whose basic properties did not appear to change during development, EPSCs demonstrated distinct developmental changes in some parameters, as reported previously (Burgard and Hablitz 1993). As early as PN 3-5, threshold stimulation produced EPSCs resembling mEPSCs. However, at stimulus intensities suprathreshold for evoking both early and late EPSC components, developmental differences became apparent. EPSC durations were significantly longer in PN 3-5 than in PN 9-14 neurons (P < 0.05, ANOVA). This was true at both depolarized potentials (239 t 88 ms, n = 9 vs. 80 t 42 ms, n = 2) and RMP (207 t 95 ms, n = 16 vs. 66 t 42 ms, n = 4) Decay 7s of suprathreshold EPSCs exhibited developmental patterns that indicated age-dependent changes in their voltage sensitivity. For example, in PN 3-5 neurons, 25% of decay curves could not be fit with an exponential function, regardless of membrane potential, because of their long time course. Another 25% were best fit by a single long exponential decay 7 (50 t 77 ms) at both RMP and depolarized potentials. The remaining 50% were fit by a double exponential function at RMP (7i 3.4 t 1.9 ms, 72 139 t 69 ms; n = 9). In contrast, 80- 100% of PN 6-8 EPSC decays at both RMP and hyperpolarized potentials were best fit by a double exponential function (Fig. 4C, n = 17). The early decay 7 (TV, 4.9 t 2.4 ms) did not significantly differ from the decay 7 of mEPSCs. The late decay 7 (72, 87 t 39 ms) was absent in mEPSCs. Over 50% of PN 6-8 EPSCs measured at depolarized potentials were best fit by a single decay7(46& 17ms,n = 6) whose value was intermediate between those for 71 and 72 at more negative potentials. Only 11% of PN 6-8 EPSCs could not be fit with any exponential function because of their prolonged time course. In PN 9- 14 neurons (n = 5) all EPSC decays consisted of either a single early decay 7 or a double exponential decay similar to those from PN 6-8 neurons. Overall, PN 3-5 EPSCs were more likely to exhibit long, voltage-independent decays than PN 6-8 or PN 9- 14 EPSCs. The appearance during postnatal development of two distinct decay 7s at more negative membrane’ potentials implies that a change in the voltage dependence of evoked responses occurs with development. NMDA receptor-mediated
component of EPSCs
Activation of NMDA receptors appeared to be involved in generation of both mEPSCs and evoked EPSCs. This was borne out by experiments with bath application of the NMDA receptor antagonist APV. Blockade of NMDA receptors by 10 PM APV had voltage-dependent effects on EPSCs. As shown in Fig. 5A, APV shortened the decay of mEPSCs recorded at a holding potential of -40 mV whereas responses at -70 mV were relatively unaffected. Evoked EPSCs were similarly affected by application of NMDA receptor antagonists (Fig. 5B). APV significantly reduced the decay 7 of mEPSCs in PN 6-8 cells while producing a similar but nonsignificant depression of decay 7 in
1846
E. C. BURGARD
A
AND J. J. HABLITZ
C
-65 mV
40mV
+ 't = 31 msec
100 pA
c 25 msec
B
-80 mV *
5 = 58 msec 60 5
50 pA
-l 20 msec
Stimulus intensity
(nC)
FIG. 4. Evoked excitatory postsynaptic current (EPSC) components are dependent on stimulus intensity and membrane potential. A: 3 EPSCs recorded in response to increasing stimulus intensity (top to bottom) in a PN 6 neuron voltage clamped at -65 mV. Threshold stimulation (top trace) produced an EPSC similar to an mEPSC. Increasing stimulation produced an increase in the early component amplitude and the appearance of a late component. B: complete input-output curve from the neuron shown in A. Notice that the early component (5 ms) increased only after a threshold stimulation intensity (2.2 nC) was reached. The late component (25 ms) gradually increased without a threshold increase. C: EPSCs recorded at depolarized (-40 mV) and hyperpolarized (-80 mV) membrane potentials in a PN 8 neuron. At -40 mV, the decay 7 was best fit by a single exponential function. At -80 mV, the decay was best fit by 2 exponential functions, an early 71 and a late TV.Dashed line is a 0 reference line.
PN 3-5 neurons. In all age groups, the voltage-dependent increased rise time and decay 7 observed at depolarized potentials in controls was not seen in the presence of APV.
A
APV depressed mEPSCs at all times during development. However, the decrease in frequency was more dramatic in PN 3-5 neurons (n = 8), where APV completely blocked
B
Spontaneous
Evoked APV 2: = 8.4 msec
APV 2: = 6.3 msec -40 mV
-40 mV 716
msec
25 msec
APV 't = 6.2 msec
APV I: = 8.6 msec -70 mV
-70 mV k7.5
msec
I
20 pA
13.1 msec
I 50 pA
20 msec 20 msec 5. N-methyl-D-aspartate (NMDA) receptors contribute to activation of EPSCs. A: mEPSCs recorded at depolarized (-40 mV) and resting (-70 mV) membrane potentials in a PN 7 neuron. Responses recorded before (base) and during (APV) application of 10 PM D(-)2-amino-5-phosphonovaleric acid (APV) are superimposed. Each trace is the average of 23-66 (base) or 7-19 (APV) consecutive events. B: EPSCs recorded from the same neuron as in A. For all pairs of traces in this figure, responses to APV are top traces. The effects of APV were greater at depolarized membrane potentials in both mEPSCs and EPSCs. FIG.
MINIATURE
A
__--
IT-+peak
CORTICAL
B
Spontaneous -40 mV
EPSCs DURING
- ---__
late
2 _a 2 -2 #$
30 25 20 15 10
2 z
5 0
DEVELOPMENT
PN 7
25 pA I
20 msec
10
20
30
40
Peak Amplitude
50
(PA)
RMP
Depolarized 0
l 0 0 0
0
0 0 X
l
0
0
.
0
0
$0
0 0
,e”
l
I
I
10 20 30 Peak Amplitude
1
40 (PA)
1
50
2
0’
I
l
O
x
0
FIG. 6. Correlation between NMDA and non-NMDA components of spontaneous mEPSCs. A: averaged mEPSCs recorded in a single PN 7 neuron at -40 mV, indicating the location of peak (2-3 ms) and late ( 10 ms) amplitude measurements. Dashed line: 0 reference line. B: plot of peak (non-NMDA) vs. late (NMDA) amplitudes for 23 mEPSCs from the neuron presented in A. A significant correlation (Y = 0.88, P < 0.05) between amplitudes was observed within this neuron. C: plot of peak vs. late amplitudes of mEPSCs recorded at depolarized membrane potentials in 32 neurons. Each point represents mean values for a single neuron. A significant correlation (r = 0.66, P < 0.05) was observed between neurons at depolarized potentials. D: same as C, but mEPSCs from 54 ne.urons were recorded at RMP. No correlation was present at RMP. For both C and D, filled circles: PN 3-5 neurons; open circles: PN 6-8 neurons; crosses: PN 9- 14 neurons.
0
l
0
I
.oP
1
10 20 30 Peak Amplitude
mEPSCs in 50% of depolarized neurons and significantly decreased mEPSC frequency in the remaining 50%. In PN 6-8 neurons (n = 5), APV significantly decreased the frequency of mEPSCs only at depolarized membrane potentials (P < 0.05, t test vs. baseline). The mechanism underlying this effect remains unclear, but it could suggest the presence of presynaptic NMDA receptors that modulate glutamate release from neonatal nerve terminals. To determine the relative contribution of NMDA and non-NMDA receptors to mEPSCs, we measured mEPSC amplitudes at different latencies (Fig. 6A). Non-NMDA receptor activation primarily underlies the peak mEPSC amplitude, whereas NMDA receptors contribute to the amplitude at increased latencies (Bekkers and Stevens 1989). Peak (non-NMDA) amplitudes were measured at 2-3 ms, whereas late (NMDA) amplitudes were measured at 10 ms after onset. The latter latency was chosen because this time point was significantly affected by changes in membrane potential and addition of APV. Attempts to measure mEPSC amplitudes at longer latencies (25 ms) were unsuccessful because responses had returned to baseline noise levels. Amplitude ratios of peak (non-NMDA) to late (NMDA) components were voltage dependent in neurons from all age groups (Table 2). The mean ratio at depolarized potentials was significantly smaller than the mean ratio at RMP, indicating a larger NMDA component in depolarized mEPSCs. No significant developmental changes in this ratio were observed. Because NMDA receptors appeared to contribute to mEPSCs at depolarized potentials, we examined the relationship between NMDA and non-NMDA receptor activation in individual mEPSCs by comparing the amplitudes of each component. In a PN 7 neuron voltage clamped at -40 mV (Fig. 6, A and B), a significant correlation existed between the peak and late amplitudes of mEPSCs. When
0
QI
l o x.e:
/‘/./ I
00 0 OXOo~Ox 06 *
*ox
0 X
1847
!
i
40 50 (PA)
mean amplitudes of neurons from all age groups were plotted, peak and late amplitudes were significantly correlated only at depolarized potentials (Fig. 6c). At RMP no correlation was seen (Fig. 60) because NMDA receptors contribute negligibly to mEPSCs at RMP (Fig. 5A). Within each age group there was also a significant correlation at depolarized potentials, but not at RMP. These results suggest that NMDA and non-NMDA receptors are both colocalized and coactivated at a constant ratio at depolarized membrane potentials, regardless of mEPSC amplitude. No developmental change in this correlation was observed in mEPSCs. To isolate NMDA receptor-mediated mEPSCs in PN 9-14 neurons (n = 3) we recorded spontaneous currents in the presence of CNQX (3 PM), bicuculline ( 10 PM), and 1.3 mM extracellular Mg2+. No mEPSCs were observed at RMP. However, at depolarized potentials mEPSCs were recorded that had mean rise times, decay TS,and amplitude ratios not significantly different from those recorded at depolarized potentials under physiological conditions. The peak amplitudes were significantly smaller under these conditions (13.8 t 0.9 pA, P < 0.05, t test). The similarities between these NMDA mEPSCs and mEPSCs recorded at depolarized potentials under physiological conditions suggest that NMDA receptors contribute predominantly to the mEPSC waveform at depolarized potentials. To quantify the contribution of NMDA receptors to evoked EPSCs, we calculated a ratio of amplitudes measured at peak (2-5 ms) and late (25 ms) latencies, representing contributions from non-NMDA and NMDA receptors, respectively. This ratio was significantly smaller at depolarized potentials (1.9 t 1.0, yt = 9) than RMP (2.8 t 1.O, y2= 14) in PN 6-8 neurons (P < 0.05, ANOVA). This difference was absent in the PN 3-5 age group, where ratios were voltage independent. This indicates that the NMDA compo-
1848
E. C. BURGARD
AND J. J. HABLITZ
C Depolarized
100
RMP
75 I 50
0 @
25
e /
a’.
, ’
20
J,
40
I
60
I
80
Peak Amplitude (PA) FIG. 7. Correlation between NMDA and non-NMDA components of evoked EPSCs. A: plot of peak (non-NMDA, 2-5 ms) vs. late (NMDA, 25 ms) amplitudes for the evoked responses presented in Fig. 4, A and B. Stimuli of increasing intensity produced increasing EPSC amplitudes at both latencies. Peak and late components were significantly correlated (r = 0.85, P < 0.05) in this neuron. Subthreshold responses were plotted but not included in the regression analysis. B: plot of peak vs. late amplitudes for evoked EPSCs recorded at depolarized membrane potentials in 2 1 neurons. Each point represents mean values for a single neuron. A significant correlation (r = 0.63, P < 0.05) was observed between neurons at depolarized potentials. C: same as B, but EPSCs from 33 neurons were recorded at RMP. A significant correlation (r = 0.54, P < 0.05) was observed between neurons at RMP. For both B and C, filled circles: PN 3-5 neurons; open circles: PN 6-8 neurons.
nent observed at RMP and hyperpolarized membrane potentials decreases between PN 3 and PN 8, presumably because of development of a more pronounced voltage-dependent block by Mg2+. As mentioned above, the amplitudes of NMDA and nonNMDA components of spontaneous mEPSCs were correlated, suggesting activation of both receptor subtypes at a constant ratio during synaptic events. To determine whether evoked EPSCs followed the same pattern, we examined the relationship between peak and late EPSC amplitudes. In a single PN 6 neuron (Fig. 7A), increasing the stimulus intensity produced EPSC components of increased amplitude (see also Fig. 4B). The amplitudes of peak and late EPSCs were correlated in this neuron. When mean amplitudes from all age groups were plotted, a significant correlation was observed at both depolarized potentials (Fig. 7B) and RMP (Fig. 7C). The NMDA receptor contribution to EPSCs at RMP is typically greater than that to mEPSCs (ratios of -3 and 7, respectively), which may account for the amplitude correlation at RMP. Although a significant correlation existed when all age groups were combined, the correlation was much stronger for PN 6-8 EPSCs (depolarized, Y = 0.79; RMP, r = 0.77, P < 0.05) than for PN 3-5 EPSCs (depolarized, r = 0.46; RMP, r = 0.43, not significant). Like mEPSCs, evoked EPSCs exhibit overall activation of NMDA and non-NMDA receptors at a relatively constant ratio, and this correlation appeared to increase between the first and second postnatal week. In evoked EPSCs from all age groups recorded at depolarized potentials, the ratios of non-NMDA to NMDA EPSC components were significantly increased in APV (6.2 t 0.2) compared with baseline (1.8 t 0.1) responses (n = 3, P < 0.05, t test). Ratios of components recorded at RMP in APV (4.3 t 3.4) were also increased compared with baseline (2.2 t 0.7), but not significantly. EPSCs, like mEPSCs, revealed a greater NMDA component at depolarized membrane potentials.
NMDA components of suprathreshold EPSCs show significant changes during postnatal development (Burgard and Hablitz 1993). NMDA components of EPSCs are depressed by APV at RMP significantly more in PN 3-5 than in PN 6-8 neurons. We have confirmed these findings in the present study where APV increased the non-NMDA to NMDA EPSC ratio by 220% in PN 3-5 neurons (n = 3) and by 176% in PN 6-8 neurons (n = 3) at RMP. To further examine the contribution of NMDA receptors to EPSCs, extracellular Mg2+ was omitted from the perfusion saline. As shown in Fig. 8A, the decay 7 of mEPSCs recorded in PN 6-8 neurons at RMP was significantly increased in zero-Mg2+ solution (P < 0.05, t test vs. controls, n = 7). The effects of zero Mg2+ were reversible with washin of normal 1.3.mM Mg2+ saline (n = 3). Removal of extracellular Mg2+ produced effects as early as PN 3-5 (n = 7), where the mEPSC decay 7 was increased in PN 3-5 neurons (4.8 t 1.4 ms). There were no significant differences between zero-Mg2+ versus control mEPSCs parameters at depolarized membrane potentials in either age group. In addition, the voltage dependence of rise time and decay 7 was absent in zero-Mg2+ medium, indicating that mEPSCs display a Mg2+ sensitivity early in development. Evoked EPSCs recorded in zero-Mg2+ medium were increased in duration and decay 7 compared with controls as early as PN 3-5. This is demonstrated in Fig. 8B, where EPSCs in response to stimuli of increasing intensity in the absence of Mg2+ are shown. Under these conditions, a late component was recorded at stimulus intensities subthreshold for evoking an early EPSC. At suprathreshold intensities (bottom trace), the early component was followed by a large, prolonged late component. If responses were first obtained in normal Mg2+ medium (as in Fig. 8C) and then Mg2+ was washed out of the preparation, the late component was enhanced relative to the early component. As shown in Fig. SC, subsequent addition of APV selectively blocked the late component of the EPSC, indicating that
MINIATURE
EPSCs DURING
CORTICAL
DEVELOPMENT
1849
Evoked in 0 Mg"
base z: = 6.0 msec -65 mV
65 mV 0 Mg" 'I = 12 msec 25
100 pA
pA
-l
50 msec
10 msec
C base
0 Mg"
0 Mg" plus APV
-65 mV
50 msec 8. Removal of extracellular Mg2+ prolongs EPSCs. A: mEPSCs recorded in a PN 7 neuron voltage clamped at -65 mV. Traces are averaged events recorded before (base, n = 24) and after (0 Mg2+, n = 48) washout of Mg2+. B: series of EPSCs recorded at -65 mV in the absence of extracellular Mg2+. The EPSCs were evoked in a PN 4 neuron in response to increasing stimulation intensities of (top to bottom): subthreshold, threshold for late component, and threshold for early component. Notice that the late component appeared before threshold for the early component was reached. C: time series showing EPSCs recorded at -65 mV before washout (base), after washout of Mg2+ (0 Mg2+), and after subsequent addition of 10 PM APV in a PN 3 neuron. Dashed line: 0 reference line. FIG.
activation of NMDA receptors is involved in the prolongation of EPSCs in zero Mg*+. DISCUSSION
NMDA receptors contribute to mEPSCs We have shown that mEPSCs are present in neonatal neocortex as early as PN 3 and that their basic biophysical properties do not significantly change during the first two postnatal weeks. The basic biophysical properties of mEPSCs measured under control conditions in the present study are similar to those reported using whole-cell patchclamp techniques in early postnatal hippocampus (Livsey and Vicini 1992; McBain and Dingledine 1992), spinal cord (Hori and Endo 1992), and embryonic turtle cerebral cortex (Blanton and Kriegstein 199 1). Our results also demonstrate that mEPSC rise time and decay 7 are dependent on membrane potential. The voltage-dependent increase in these parameters observed under control conditions was absent at all ages studied in the presence of APV, indicating that NMDA receptor activation contributed to mEPSCs. This was not unexpected, because NMDA receptors are present in neocortex at this age (McDonald and Johnston 1990) and activation of NMDA receptors by endogenous transmitter occurs prenatally during embryogenesis (Lo Turco et al. 199 1). It appears that by PN 3 mEPSCs are generated by activation of both NMDA and non-NMDA receptors and these receptors are colocalized at individual synapses. This is in agreement with previous reports of dual NMDA-non-NMDA component mEPSCs in hippocampal (Bekkers and Stevens 1989; McBain and Dingledine 1992) and spinal cord neurons (Hori and Endo 1992). A significant correlation existed between the amplitudes of NMDA and non-NMDA components of mEPSCs. This suggests that activation of these glutamate receptor subtypes occurs at an almost constant ratio during synaptic activity of varying intensity. Similar findings have been de-
scribed for synaptic activity on stellate cells of visual cortex (Stern et al. 1992). We observed no developmental changes in coactivation of receptor subtypes in mEPSCs. This is in agreement with the overall absence of changes in other mEPSC parameters. Our previous results with evoked EPSCs suggested that an unconventional voltage dependence, characteristic of Mg*+ block of the NMDA receptor (Mayer et al. 1984), was present in neurons as early as PN 3. In the present study we found that the decay 7 of mEPSCs was reversibly increased at all ages after removal of extracellular Mg*+. This effect was not observed at depolarized membrane potentials and the usual voltage dependence of rise time and decay 7 were absent in zero-Mg*+ medium. These findings are consistent with the presence of a functional voltage-dependent Mg*+ block of the NMDA receptor in early postnatal development. Similar Mg*+ sensitivity of NMDA receptors has been reported in immature spinal cord (Hori and Endo 1992), cerebellum (Rosenmund et al. 1992), and embryonic cortex (Lo Turco et al. 199 1). In contrast to the absence of postnatal changes reported here, distinct developmental increases in mEPSC rise time, frequency, and decay 7 were observed during embryonic development in turtle cortex (Blanton and Kriegstein 199 1). This may represent a species difference, or possibly significant maturation occurs during the prenatal period in the turtle with little further development of mEPSC basic properties during the first two postnatal weeks. In both superior colliculus (Hestrin 1992) and visual cortex (Carmignoto and Vicini 1992) a progressive decrease in the duration of NMDA EPSCs begins at the end of the second postnatal week and continues into adulthood. These two reports focused on isolating the NMDA components of EPSCs by blocking non-NMDA and GABA, receptors and either routinely omitting extracellular Mg*+ or recording EPSCs at positive membrane potentials. Compared with the results obtained here, decays were significantly longer at
1850
E. C. BURGARD
all ages and showed a decline with maturation. In the present study, under relatively physiological conditions, no developmental change in mEPSC properties was seen. Longduration mEPSCs were also reported in visual cortical interneurons (Stern et al. 1992) but not hippocampal interneurons (McBain and Dingledine 1993). The visual cortex may display different patterns of activity during development. NMDA receptors contribute to evoked EPSCs Some properties of EPSCs were different from mEPSCs and these differences were dependent on the intensity of stimulation. The amplitudes and decay 7s of EPSCs evoked at threshold stimulation intensity were similar to those of mEPSCs. We consider these all-or-none responses to be unitary events produced by stimulation of a single axon, resembling the unitary responses described by Stern et al. (1992) in stellate cells of visual cortex. At all ages studied, unitary EPSCs had properties similar to mEPSCs and no developmental changes were observed. At suprathreshold stimulation intensities, a long-duration late NMDA component of the EPSC appeared. Long-duration EPSPs have been recorded in immature cortical neurons (Connors et al. 1983; Kriegstein et al. 1987; Purpura et al. 1965; Schwartzkroin 1982) that result, in part, from prolonged NMDA components of EPSCs (Burgard and Hablitz 1993). Suprathreshold stimulation has also been shown to evoke EPSCs with a prominent NMDA component in neonatal somatosensory neurons (Agmon and O’Dowd 1992). Although mEPSCs and unitary EPSCs displayed no developmental changes, suprathreshold EPSCs showed three main developmental changes. First, they become progressively shorter with age. The longer duration of PN 3-5 EPSCs corresponds with a period of greater NMDA receptor contribution to EPSCs. The NMDA component was primarily the late component seen with increased stimulation intensity. A late EPSC component has previously been shown to be mediated by NMDA receptors (Dale and Roberts 1985; Forsythe and Westbrook 1988) and to be voltage sensitive (Konnerth et al. 1990). Second, the voltage dependence of the NMDA component changed with age. The ratios of NMDA to non-NMDA components were significantly greater at depolarized potentials than at RMP or hyperpolarized potentials in PN 6-8 neurons. The voltage dependence of this ratio was absent in PN 3-5 neurons, indicating a significant NMDA contribution at all membrane potentials in PN 3-5 neurons. Third, decay properties became more voltage dependent with development. These EPSC changes indicate that some voltage and/or Mg2+ dependence of suprathreshold EPSC NMDA components develops postnatally. The developmental increase in voltage sensitivity of EPSCs represents a quantitative change because the NMDA components were prolonged in the absence of extracellular Mg2+ as early as PN 3. The apparent increase in voltage sensitivity of the NMDA component in the evoked EPSC is intriguing because no developmental differences in voltage, APV, or Mg2+ sensitivity were observed in mEPSCs. Other studies demonstrate a significant voltage-dependent Mg2+ sensitivitv of embrvonic NMDA receptors (Lo Turco et al. 199 1)
AND
J. J. HABLITZ
and NMDA components of early postnatal evoked responses (Stern et al. 1992). However, other investigators have reported developmental increases in both voltage dependence (Ben-Ari et al. 1988) and Mg2+ sensitivity (Bowe and Nadler 1990; Kleckner and Dingledine 199 1; Morrisett et al. 1990) of the NMDA receptor. The possibility exists that a population of NMDA receptors undergoes a developmental change in voltage or Mg2+ sensitivity. This population does not underlie the NMDA component of mEPSCs but may contribute to the NMDA component of suprathreshold EPSCs. Possible mechanisms underlying developmental changes in NMDA components The mechanisms underlying the prolongation of neonatal EPSCs remain unknown. Kraszewski and Grantyn (1992) have shown that evoked synaptic transmission at developing GABAergic synapses in culture also changes without accompanying changes in miniature synaptic currents. They report an increase in the evoked current with development that is attributed to an increase in presynaptic sprouting. Our results show a decrease in EPSC duration during development without a corresponding change in the mEPSCs. A number of groups have shown that postsynaptic responses evoked by orthodromic stimulation in adult cortex have both NMDA and non-NMDA components. The NMDA component may be monosynaptic (Thomson 1986), polysynaptic (Hablitz and Sutor 1990; Sutor and Hablitz 1989a,b), or both (Shirokawa et al. 1989). The long-duration NMDA component in PN 3-5 neurons could be explained by activation of polysynaptic NMDA receptor-mediated circuits, which prolong the EPSC. This would also require that either the number or the efficacy of these circuits decrease over the first postnatal week. Although cortical innervation (Wise and Jones 1978) and synaptic density (Blue and Parnavelas 1983; Kristt and Molliver 1976) are increasing during this time period, several regressive phenomena are also underway. These include retraction of inappropriate axons (Ivy and Killackey 1982) and dendritic sculpting (Koester and O’Leary 1992; Murphy and Magness 1984). These anatomic changes in circuitry could be associated with alterations in physiological responsiveness. Alternately, NMDA receptors may still be distributed widely or unevenly (Onodera and Takeuchi 199 1) over the surface of neonatal neurons. Glutamate released in response to presynaptic stimulation may spill over and activate extrasynaptic NMDA receptors with a temporal lag that extends the EPSC. As synapses develop and NMDA receptors cluster in postsynaptic regions, the EPSC duration would decrease. However, studies in more mature neurons argue against this, because the EPSC decay properties in these studies are mediated by channel kinetics rather than rebinding (Lester et al. 1990) or diffusion (Hestrin et al. 1990) of glutamate. In turn, evoked synaptic activity in neonates may be similar to that described at Mauthner cell synapses (Faber and Korn 1988), where lateral diffusion of neurotransmitter from adjacent activated terminals synergistically potentiates evoked responses. In summarv. these results indicate that NMDA receptors
MINIATURE
EPSCs DURING
contribute to both mEPSCs and evoked EPSCs in the developing neocortex. Our results indicate that only a few fibers need to be simultaneously active to produce a long-lasting NMDA component in EPSCs. Such events would be expetted to be associated with Ca*+ influx and subsequent activation of intracell .ular signaling mechanisms. These processes could serve as stimuli for process extension (Pearce et al. 1987), synapse stabilization (Constantine-Paton et al. 1990), and perhaps elimination (Rabacchi et al. 1992). Decreases in the duration and or alterations in threshold for evoking NMDA EPSCs could be a mechanism underlying reduction of plasticity with maturation. This work was supported by National Institute of Neurological Disorders and Stroke Javits Neuroscience Investigator Award NS-22373. Address reprint requests to J. J. Hablitz. Received 5 March 1993; accepted in final form 26 June 1993. REFERENCES A. AND O’DOWD, D. K. NMDA receptor-mediated currents are prominent in the thalamocortical synaptic response before maturation of inhibition. J. Neurophysiol. 68: 345-349, 1992. BALAZS, R., HACK, N., AND JORGENSEN, 0. S. Stimulation of the Nmethyl-D-aspartate receptor has a trophic effect on differentiating cerebellar granule cells. Neurosci. Lett. 87: 80-86, 1988. BEKKERS, J. M. AND STEVENS, C. F. NMDA and non-NMDA receptors are co-localized at individual excitatory synapses in cultured rat hippocampus. Nature Lond. 341: 230-233, 1989. BEN-ARI, Y., CHERUBINI, E., AND KRNJEVIC, K. Changes in voltage dependence of NMDA currents during development. Neurosci. Lett. 94: 88-92, 1988. BETTLER, B., BOULTER, J., HERMANS-BORGMEYER, I., O’SHEA-GREENFIELD, A., DENERIS, E. S., MOLL, C., BORGMEYER, U., HOLLOMAN, M., AND HEINEMANN, S. Cloning of a novel glutamate receptor subunit, GluR5: expression in the nervous system during development. Neuron 5: 583-595, 1990. BLANTON, M. G. AND KRIEGSTEIN, A. R. Spontaneous action potential activity and synaptic currents in the embryonic turtle cerebral cortex. J. Neurosci. 11: 3907-3923, 199 1. BLANTON, M. G., Lo TURCO, J. J., AND KRIEGSTEIN, A. R. Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex. J. Neurosci. Methods 30: 203-2 10, 1989. BLUE, M. E. AND PARNAVELAS, J. G. The formation and maturation of synapses in the visual cortex of the rat. I. Qualitative analysis. J. Neurocytol. 12: 599-6 16, 1983. BOWE, M. A. AND NADLER, J. V. Developmental increase in the sensitivity to magnesium of NMDA receptors on CA1 hippocampal pyramidal cells. Dev. Brain Res. 56: 55-6 1, 1990. BURGARD, E. C. AND HABLITZ, J. J. Voltage dependence of spontaneous excitatory currents during postnatal development. Sot. Neurosci. Abstr. 18: 804, 1992. BURGARD, E. C. AND HABLITZ, J. J. Developmental changes in NMDA and non-NMDA receptor-mediated synaptic potentials in rat neocortex. J. Neurophysiol. 69: 230-240, 1993. CARMIGNOTO, G. AND VICINI, S. Activity-dependent decrease in NMDA receptor responses during development of the visual cortex. Science Wash. DC258: 1007-1011, 1992. CONNORS, B. W., BENARDO, L. S., AND PRINCE, D. A. Coupling between neurons of the developing rat neocortex. J. Neurosci. 3: 773-782, 1983. CONSTANTINE-PATON, M., CLINE, H. T., AND DEBSKI, E. Patterned activity, synaptic convergence, and the NMDA receptor in developing visual pathways. Annu. Rev. Neurosci. 13: 129-154, 1990. DALE, N. AND ROBERTS, A. Dual-component amino-acid-mediated synaptic potentials: excitatory drive for swimming in Xenopus embryos. J. Physiol. Lond. 363: 35-59, 1985. FABER, D. S. AND KORN, H. Synergism at central synapses due to lateral diffusion of transmitter. Proc. Natl. Acad. Sci. USA 85: 8708-87 12, 1988. FORSYTHE, I. D. AND WESTBROOK, G. L. Slow excitatory postsynaptic AGMON,
CORTICAL
DEVELOPMENT
1851
currents mediated by N-methyl-D-aspartate receptors on cultured mouse central neurones. J. Physiol. Loid. 396: 5 151533, 1988. GALLO, V., UPSON, L. M., HAYES, W. P., VYKLICKY, L., JR., WINTERS, C. A., AND BUONANNO, A. Molecular cloning and developmental analysis of a new glutamate receptor subunit isoform in cerebellum. J. Neurosci. 12: 1010-1023, 1992. HABLITZ, J. J. AND SUTOR, B. Excitatory postsynaptic potentials in rat neocortical neurons in vitro. III. Effects of a quinoxalinedione nonNMDA receptor antagonist. J. Neurophysiol. 64: 1282- 1290, 1990. HESTRIN, S. Developmental regulation of NMDA receptor-mediated synaptic currents at a central synapse. Nature Land. 357: 686-689, 1992. HESTRIN, S., SAH, P., AND NICOLL, R. A. Mechanisms generating the time course of dual component excitatory synaptic currents recorded in hippocampal slices. Neuron 5: 247-253, 1990. HORI, Y. AND ENDO, IS. Miniature postsynaptic currents recorded from identified rat spinal dorsal horn projection neurons in thin slice preparations. Neurosci. Lett. 142: 19 I- 195, 1992. IVY, G. 0. AND KILLACKEY, H. P. Ontogenetic changes in the projections of neocortical neurons. J. Neurosci. 2: 736-743, 1982. KALB, R. G., LIDOW, M. S., HALSTED, M. J., AND HOCKFIELD, S. Nmethyl-D-aspartate receptors are transiently expressed in the developing spinal cord ventral horn. Proc. Natl. Acad. Sci. USA 89: 8502-8506, 1992. KLECKNER, N. W. AND DINGLEDINE, R. Regulation of hippocampal NMDA receptors by magnesium and glycine during development. 1MoZ. Brain Res. 11: 151-159, 1991. KOESTER, S. E. AND O’LEARY, D. D. M. Functional classes of cortical projection neurons develop dendritic distinctions by class-specific sculpting of an early common pattern. J. Neurosci. 12: 1382-l 393, 1992. KONNERTH, A., KELLER, B. U., BALLANYI, K., AND YARRI, Y. Voltage sensitivity of NMDA-receptor mediated postsynaptic currents. Exp. Brain Res. 8 1: 209-2 12, 1990. KRASZEWSKI, K. AND GRANTYN, R. Unitary, quantal, and miniature GABA-activated synaptic chloride currents in cultured neurons from the rat superior colliculus. Neuroscience 47: 555-570, 1992. KRIEGSTEIN, A. R., SUPPES, T., AND PRINCE, D. A. Cellular and synaptic physiology and epileptogenesis of developing rat neocortical neurons in vitro. Dev. Brain Res. 34: 161-171, 1987. KRISTT, D. A. AND MOLLIVER, M. E. Synapses in newborn rat cerebral cortex: a quantitative ultrastructural study. Brain Res. 108: 180- 186, 1976. KUTSUWADA, T., KASHIWABUCHI, N., MORI, H., SAIUMURA, K., KUSHIYA, E., ARAKI, K., MEGURO, H., MASAKI, H., KUMANISHI, T., ARAKAWA, M., AND MISHINA, M. Molecular diversity of the NMDA receptor channel. Nature Lond. 358: 36-4 1, 1992. LARKMAN, A. U., MAJOR, G., STRATFORD, K. J., AND JACK, J. J. B. Dendritic morphology of pyramidal neurones of the visual cortex of the rat. IV. Electrical geometry. J. Camp. Neural. 323: 137-152, 1992. LESTER, R. A. J., CLEMENTS, J. D., WESTBROOK, G. L., AND JAHR, C. E. Channel kinetics determine the time course of NMDA receptor-mediated synaptic currents. Nature Lond. 346: 565-567, 1990. LIVSEY, C. T. AND VICINI, S. Slower spontaneous excitatory postsynaptic currents in spiny versus aspiny hilar neurons. Neuron 8: 745-755, 1992. Lo TURCO, J. J., BLANTON, M. G., AND KRIEGSTEIN, A. R. Initial expression and endogenous activation of NMDA channels in early neocortical development. J. Neurosci. 11: 792-799, 199 1. LUHMANN, H. J. AND PRINCE D. A. Postnatal maturation of the GABAergic system in rat neocortex. J. Neurophysiol. 65: 247-263, 199 1. MATTSON, M. P. Neurotransmitters in the regulation of neuronal cytoarchitecture. Brain Res. Rev. 13: 179-2 12, 198. MAYER, M. L., WESTBROOK, G. L., AND GUTHRIE, P. B. Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurons. Nature Land. 309: 26 l-263, 1984. MCBAIN, C. J. AND DINGLEDINE, R. Dual-component miniature excitatory synaptic currents in rat hippocampal CA3 pyramidal neurons. J. Neurophysiol. 68: 16-27, 1992. MCBAIN, C. J. AND DINGLEDINE, R. Heterogeneity of synaptic glutamate receptors on CA3 stratum radiatum interneurones of rat hippocampus. J. Physiol. Land. 462: 373-392, 1993. MCDONALD, J. W. AND JOHNSTON, M. V. Physiological and pathophysiological roles of excitatory amino acids during central nervous system development. Brain Res. Rev. 15: 4 l-70, 1990. MILLER, M. Maturation of rat visual cortex. I. A quantitative study of
1852
E. C. BURGARD
Go&impregnated pyramidal neurons. J. Neurocytol. 10: 859-878, 1981. MONYER, H., SEEBURG, P. H., AND WISDEN, W. Glutamate-operated channels: developmentally early and mature forms arise by alternative splicing. Neuron 6: 799-8 10, 199 1. MONYER, H., SPRINKLE, R., HERB, A., HIGUCHI, M., LCIMELI, H., BURNASHEV, N., SAKMANN, B. AND SEEBURG, P. H. Heteromeric NMDA receptors: molecular and functional distinction of subtypes. Science Wash. DC256: 1217-1221, 1992. MORRISETT, R. A., MOTT, D. D., LEWIS, D. V., WILSON, W. A., AND SWARTZWELDER, H. S. Reduced sensitivity of the N-methyl-D-aspartate component of synaptic transmission to magnesium in hippocampal slices from immature rats. Dev. Brain Res. 56: 257-262, 1990. MURPHY, E. H. AND MAGNESS, R. Development of the rabbit visual cortex: a quantitative Golgi analysis. Exp. Brain Res. 53: 304-3 14, 1984. ONODERA, K. AND TAKEUCHI, A. Uneven distribution of excitatory amino acid receptors on ventral horn neurones of newborn rat spinal cord. J. Physiol. Lond. 439: 257-276, 199 1. OTIS, T. S., STALEY, K. J., AND MODY, I. Perpetual inhibitory activity in mammalian brain slices generated by spontaneous GABA release. Brain Res. 545: 142-150, 1991. PEARCE, I. A., CAMBRAY-DEAKIN, M. A., AND BURGOYNE, R. D. Glutamate acting on NMDA receptors stimulates neurite outgrowth from cerebellar granule cells. FEBS Lett. 223: 143- 147, 1987. PURPURA, D. P., SHOFER, R. J., AND SCARFF, T. Properties of synaptic activities and spike potentials of neurons in immature neocortex. J. Neurophysiol. 28: 925-942, 1965. RABACCHI, S., BAILLY, Y., DELHAYE-BOUCHRD, N., AND MARIANI, J. Involvement of the N-methyl-D-aspartate (NMDA) receptor in synapse elimination during cerebellar development. Science Wash. DC 256: 1823-1825, 1992. RALL, W. Core conductor theory and cable properties of neurons. In: Handbook of Physiology. The Nervous System. Cellular Biology rons. Bethesda, MD: Am. Physiol. Sot., 1977, sect. 1, vol. I,
of Neu-
part 1, p.
39-97. C., LEGENDRE, P., AND WESTBROOK, G. L. Expression of NMDA channels on cerebellar Purkinje cells acutely dissociated from newborn rats. J. Neurophysiol. 68: 190 1- 1905, 1992.
ROSENMUND,
AND J. J. HABLITZ S. M. AND ROLE, L. W. Developmental regulation of nicotinic acetylcholine receptors. Annu. Rev. Neurosci. 10: 403-457, 1987. SCHWARTZKROIN, P. A. Development of rabbit hippocampus: physiology. Dev. Brain Res. 2: 469-486, 1982. SHIROKAWA, T., NISHIGORI, A., KIMURA, F., AND TSUMOTO, T. Actions of excitatory amino acid antagonists on synaptic potentials of layer II/III neurons of the cat’s visual cortex. Exp. Brain Res. 78: 489-500, 1989. SPIGELMAN, I., ZHANG, L., AND CARLEN, P. L. Patch-clamp study of postnatal development of CA1 neurons in rat hippocampal slices: membrane excitability and K+ currents. J. Neurophysiol. 68: 55-69, 1992. SPRUSTON, N. AND JOHNSTON, D. Perforated patch-clamp analysis of the passive membrane properties of three classes of hippocampal neurons. J. Neurophysiol. 67: 508-529, 1992. STAFSTROM, C. E., SCHWINDT, P. C., AND CRILL, W. E. Cable properties of layer V neurons from cat sensorimotor cortex in vitro. J. Neurophysiol. 52: 278-289, 1984. STERN, P., EDWARDS, F. A., AND SAKMANN, B. Fast and slow components of unitary EPSCs on stellate cells elicited by focal stimulation in slices of rat visual cortex. J. Physiol. Lond. 449: 247-278, 1992. SUTOR, B. AND HABLITZ, J. J. EPSPs in rat neocortical neurons in vitro. I. Electrophysiological evidence for two distinct EPSPs. J. Neurophysiol. 6 1: 607-620, 1989a. SUTOR, B. AND HABLITZ, J. J. EPSPs in rat neocortical neurons in vitro. II. Involvement of N-Methyl-D-aspartate receptors in the generation of EPSPs. J. Neurophysiol. 6 1: 62 l-634, 1989b. THOMSON, A. M. A magnesium-sensitive post-synaptic potential in rat cerebral cortex resembles neuronal responses to N-methylaspartate. J. Physiol. Land. 370: 531-549, 1986. TREMBLAY, E., ROISIN, M. P., REPRESA, A., CHARRIAUT-MARLANGUE, C., AND BEN-ARI, Y. Transient increased density of NMDA binding sites in the developing rat hippocampus. Brain Res. 46 1: 393-396, 1988. WISE, S. P. AND JONES, E. G. Developmental studies of thalamocortical and commissural connections in the rat somatic sensory cortex. J. Comp. Neural. 178: 187-208, 1978. ZISKIND-CONHAIM, L. NMDA receptors mediate poly- and monosynaptic potentials in motoneurons of rat embryos. J. Neurosci. 10: 125-l 35, 1990. SCHUETZE,
