The Institute for Neuroscience and the Division of Pharmacology and
Toxicology, College of Pharmacy, University of Texas, Austin, Texas
Synaptic mechanisms underlying hyperexcitability due to withdrawal from
chronic ethanol exposure were investigated in a hippocampal explant
model system using electrophysiological techniques. Whole-cell voltage
clamp recordings from CA1 pyramidal cells demonstrated that acute
ethanol exposure inhibited N-methyl-D-aspartate receptor (NMDAR)-mediated excitatory postsynaptic currents by over 40%. Chronic
ethanol exposure for 6 to 11 days at 35 or 75 mM induced no differences
from control explants in the fast component of the population synaptic
response (non-NMDAR-mediated). Prolonged field potential recordings (to
10 hr) were used to monitor the withdrawal process in
vitro. Ethanol-exposed explants from both 35 and 75 mM groups
displayed an increase (60% and 89%, respectively) in the
NMDAR-mediated component of synaptic transmission on withdrawal from
chronic exposure. Prolonged tonic-clonic electrographic seizure activity was consistently observed after ethanol withdrawal only after
the increase in NMDAR function. This hyperexcitability was inhibited by
the NMDAR antagonist D-2-amino-5-phosphonovaleric acid and
returned once the NMDAR component was reestablished after antagonist
washout. In situ hybridization studies suggest that expression of NR2B subunit mRNA may be enhanced in explants after chronic ethanol exposure. No lasting differences were observed in the
NMDAR component after acute in vitro ethanol exposure
and withdrawal. These data suggest that the occurance of ethanol
withdrawal hyperexcitability in this system may be directly dependent
on alterations in NMDAR function after chronic exposure. Since this region and others that contain ethanol sensitive NMDARs may serve as
epileptic foci, long term alterations in NMDAR function may be expected
to generate paroxysmal depolarizing shifts underlying ictal events
after withdrawal from ethanol exposure.
 |
Introduction |
Abstinence
after chronic exposure to ethanol often leads to a physical withdrawal
syndrome including symptoms in humans such as tremor, agitation,
delirium and in severe cases, convulsions and brain damage (Victor and
Brausch, 1967
; Tabakoff and Rothstein, 1983). NMDARs are a subclass of
excitatory neurotransmitter receptors that play important physiological
roles in vertebrate nervous systems (McBain and Mayer, 1994). Since
studies have demonstrated that intoxicating concentrations of ethanol
inhibit the NMDAR (Hoffman et al., 1989; Lovinger et
al., 1989), it is conceivable that a substantial component of
alcohol withdrawal hyperexcitability may be related to an adaptive
upregulation of NMDAR function due to chronic exposure. Evidence
supporting this hypothesis has accumulated over the past few years.
Grant et al. (1990) demonstrated an increase in the number
of binding sites for the NMDAR channel blocker, MK-801, in the
hippocampus after chronic ethanol exposure. These investigators and
others (Morrisett et al., 1990) showed that MK-801 protected
against withdrawal hyperexcitability in rodents. This effect has also
been demonstrated using a competitive NMDAR antagonist (CGP39551)
in vivo (Liljequist, 1991) and in hippocampal slices taken
from rats chronically exposed to ethanol (Ripley and Little, 1995).
Biochemical studies have demonstrated an increased sensitivity to NMDA
in cultured neurons after chronic ethanol exposure (Iorio et
al., 1992; Chandler et al., 1993; Blevins et al., 1995). Recently, electrophysiological evidence for a chronic ethanol-induced enhancement of NMDAR-mediated synaptic responses was
provided by Whittington et al. (1995) in hippocampus. Taken together these studies suggest a role for the NMDAR in the alterations in neural function due to chronic ethanol exposure.
In this study, we test the hypothesis that NMDARs play a critical role
in the expression of ethanol withdrawal hyperexcitability. Specifically, we have analyzed the temporal relationship between the
alteration in NMDAR function and the occurrence of neuronal hyperexcitability due to removal from chronic ethanol exposure. We
propose that several criteria must be satisfied to provide strong
support for this hypothesis. First, direct evidence must be provided of
enhanced NMDAR function on ethanol withdrawal in comparison with
nonethanol-exposed or nonwithdrawn tissue. Second, the tissue must
exhibit evidence of withdrawal hyperexcitability. Third, a temporal
relationship should exist such that the enhanced NMDAR function should
precede (or occur in parallel with) the withdrawal hyperexcitability.
Fourth, the withdrawal hyperexcitability should be prevented by
antagonists of NMDAR function. Finally, the altered NMDAR function and
withdrawal hyperexcitability should be dependent on chronic exposure
and not observed with acute exposure.
Several experimental systems have been utilized to study the cellular
and molecular mechanisms mediating ethanol withdrawal hyperexcitability
(Trevisan et al., 1994; Blevins et al., 1995; Whittington et al., 1995; Hu et al., 1996). While
the existing model systems have provided important evidence regarding
these mechanisms, an in vitro system that enables the
observation of synaptic responses during seizure expression is
preferable. Brain slice culture preparations may provide such a system
for the study of chronic ethanol effects on neural function. Recently,
a technique has been developed in which brain slices are maintained in
culture on a porous membrane at an air-medium interface (Stoppini
et al., 1991). The general morphology and synaptic
connections of neurons in organotypic hippocampal slices cultured using
this interface technique largely resemble that seen in native tissue
(Buchs et al., 1993), and analyses of several molecular
markers suggest many similarities to mature brain (Bahr, 1995). The
explants exhibit excitatory and inhibitory synaptic responses and
several forms of synaptic plasticity including NMDAR-dependent
long-term potentiation (Stoppini et al., 1991; Muller
et al., 1993). Of particular relevance for the study of
postwithdrawal hyperexcitability, hippocampal explants are also capable
of generating epileptiform burst discharges and tonic-clonic seizure
events under certain conditions (McBain et al., 1989).
We used electrophysiological techniques to assess the effects of acute
and chronic ethanol exposure and withdrawal on neural excitability
associated with NMDAR function in hippocampal explant cultures. These
techniques included whole-cell voltage clamp as well as field potential
recordings to maintain long-term recordings during the withdrawal
period. Additionally, recent evidence suggests that the upregulation of
NMDAR function after chronic ethanol exposure may involve the NR2B
subunit of the heteromeric receptor complex (Trevisan et
al., 1994; Hu et al., 1996). We thus also used in
situ hybridization techniques to address whether alterations in
NR2B subunit message occur in the explants after chronic ethanol exposure. Some of the data reported here have been presented previously in abstract form (Thomas et al., 1995, 1996).
| |
Methods |
Slice preparation and culture.
Slices used in this study
were prepared from 9- to 11-day-old Sprague-Dawley rat pups of both
sexes. The animals were anesthetized with halothane and decapitated,
and the brain was rapidly removed and placed in ice-cold oxygenated
ACSF for 3 to 4 min. The hippocampi were then removed bilaterally and
500 micron transverse sections were cut, transferred to a holding
chamber containing ACSF bubbled with 95% O2/5%
CO2 (carbogen) and maintained at 32° to 35°C.
ACSF consisted of (in mM): NaCl, 120; NaHCO3, 25;
KCl, 3.3; NaH2PO4, 1.2;
CaCl2, 1.8; MgSO4, 2.4;
dextrose, 10.
Slices were cultured using a modification of methods described by
Stoppini et al. (1991). After incubation in ACSF at 32° to
35°C for 1 to 2 hr, slices were transferred onto membrane inserts (Millicell-CM; Millipore Corp., Bedford, MA) in 6-well plates containing 1 ml of culture medium/well. The plates were placed in an
incubator at least 15 min before tissue transfer to allow the pH and
temperature of the medium to stabilize. The cultures were then
maintained in the incubator at 37°C in a 5%
CO2-enriched atmosphere. The culture medium
consisted of 75% minimum essential medium (GIBCO 12360; Gaithersburg,
MD), 25% heat-inactivated horse serum (GIBCO 26050), glutamine (3 mM),
glucose (5.5 mg/ml medium) and penicillin/streptomycin (100 U/ml and
100 µg/ml, respectively; GIBCO 15140). Medium was replaced on the day
after slice preparation and then on alternate days.
Prior to recording, slice cultures were transferred to 35-mm petri
dishes mounted on the stage of an inverted microscope. The slices were
submerged in and continuously perfused with, carbogen-saturated ACSF at
a flow rate of ~1.5 ml/min (Rainin Rabbit-Plus peristaltic pump,
Woburn, MA). The recording ACSF was the same as that used for tissue
preparation except that the concentrations of
CaCl2 and MgSO4 were 2.0 mM
and 0.9 mM, respectively (except where noted). The tissue was perfused
in the acute recording solution for 1 hr before beginning recordings to
insure washout of the culture medium. Drugs were dissolved in ACSF and
applied to the tissue by switching solution reservoirs. All drugs were
obtained from Sigma (St. Louis, MO) except CGS-19755, which was
obtained from Nova Pharmaceuticals (Baltimore, MD), and ethanol (95%),
from University of Nebraska Medical Center Hospital Supply.
Patch clamp recording of NMDAR-mediated synaptic currents.
Tight-seal whole-cell patch recordings were made at room temperature
from CA1 pyramidal neurons using techniques previously described
(Morrisett and Swartzwelder, 1993). Recording electrodes were made from
thin-walled borosilicate glass (TW150F-4, WPI, Sarasota, FL, 1.2-2.2
M
) and filled with (in mM): CsMeSO3, 135; NaCl, 8; EGTA, 0.5; HEPES, 10; MgCl2, 2;
Tris-ATP, 2; Tris-GTP, 0.3; 260-270 mOsm, pH 7.2 with CsOH. Access
resistance was partially compensated after cell break-in and monitored
throughout the experiment. Recordings where access resistance increased
more than 10% during the course of the experiment were not included in
data analyses. EPSCs were evoked by stimulation of Schaffer collateral
fibers in the stratum radiatum layer of area CA3 via monopolar tungsten electrodes. Constant-current pulses (100 µsec duration, 9-34 µA amplitude) were applied through a stimulus isolation unit driven by a
digital stimulator (Master-8, A.M.P.I., Israel). Recordings were made
using a Warner PC-501A amplifier (Warner Instrument Corp., Hamden, CT),
filtered at 1 kHz and digitized at 10-20 kHz using a Lab Master DMA
interface (Scientific Solutions, Solon, OH). Data acquisition and
device control were implemented using AxoBASIC software (Axon
Instruments Inc., Foster City, CA).
The NMDAR-mediated component of synaptic current was recorded after
pharmacological elimination of other major synaptic components. The
fast, non-NMDA EPSC was blocked by bath application of 10 µM DNQX.
The fast IPSC was eliminated by holding the membrane potential near
ECl (approximately -50 mV). The slow,
GABAB receptor-mediated K+
current was eliminated by including Cs+ in the
patch pipette solution. Finally, activation of NMDARs was facilitated
by lowering the extracellular Mg2+ concentration
to 0.1 mM during patch clamp recording.
Chronic ethanol exposure.
Slice explants were cultured for 6 to 7 days before exposure to ethanol, and explants prepared from the
same animals were then randomly distributed between the control and
ethanol groups. Slices were exposed to either 35 mM (7-11 days) or 75 mM (6 days) ethanol. Ethanol was added to the preequilibrated medium at
the desired concentration and the culture plates were placed in a sealed, humidified plastic container to prevent evaporation. No changes
in pH were noted using this exposure paradigm and no differences in
field potentials were noted between explants cultured in the containers
and normal explants (data not shown). Ethanol was added to the water in
the chamber floor at a slightly elevated concentration (empirically
determined to be 90 mM to give 75 mM and 42 mM to give 35 mM, Sigma kit
333-A) to prevent evaporation loss on establishing equilibrium. Control
slices for the two ethanol groups were maintained in standard culture
medium in separate sealed, humidified containers. Media ethanol
concentrations were monitored regularly using a standard diagnostic kit
(Sigma 333-A).
Extracellular recording.
Experiments on control and
ethanol-exposed slices were performed on alternate days. The
extracellular recordings used to monitor synaptic potentials before and
during ethanol withdrawal were performed in normal recording ACSF, that
is, with no receptor antagonists, 0.9 mM MgSO4
and 2.0 mM CaCl2. Population field potentials were recorded at 32 ± 1°C from the CA1 pyramidal cell layer
with glass microelectrodes filled with 150 mM NaCl (1-3 M).
Synaptic responses were evoked by stimulation of Schaffer collateral
fibers and the responses were amplified using a DC-coupled amplifier, with data acquisition and analysis implemented as described above for
patch clamp recordings.
Plots of the population synaptic potential (PSP) amplitude
vs. stimulus amplitude (input/output curves) were obtained
after an initial 1-hr period of equilibration in ACSF ± ethanol.
Averaged responses (n = 5) were then collected hourly
at a stimulus intensity that evoked a near maximal peak PSP. Recordings
were made for 2 hr before withdrawal of ethanol and for up to 7 hr
after withdrawal in ethanol-exposed slices. Input/output curves were
obtained near the end of the recording period, and in some experiments
were obtained hourly during the entire experiment. Spontaneous and evoked electrographic seizure events were recorded using a digital tape
recorder (Sony DAT Model 75ES).
In situ hybridization.
Standard techniques were
used to detect the presence of mRNA for the NR1, NR2A and NR2B receptor
subunits (see Monyer et al., 1992, for oligonucleotide
sequences). Antisense oligonucleotide probes were 3' end-labeled with
-35S-dATP using terminal deoxynucleotidyl
transferase (New England Nuclear, Boston, MA) and unincorporated
nucleotide was removed by Nensorb chromatography (New England Nuclear).
After a brief rinse in chilled phosphate-buffered saline (PBS, pH 7.3)
the tissue slices were fixed on the membrane insert for 5 min in 4%
paraformaldehyde in PBS, rinsed with chilled PBS, removed from the
inserts, freeze-mounted to slides (Superfrost Plus, Fisher Scientific,
Pittsburg, PA) and stored frozen overnight. The tissue was then
immersed in chilled 70% ethanol for 3 min and stored in 95% ethanol
at -20°C. 35S-Labeled oligonucleotide probe
was dissolved in hybridization buffer (2000 cpm/µl; New England
Nuclear) containing 0.2 M dithiothreitol and applied to the tissue
slices, which were then sealed in special slide chambers (Probe Clip,
RPI Corp., Mount Prospect, IL). The tissue was incubated overnight at
42°C (NR1 and NR2A) or 48°C (NR2B). After a high-stringency rinse
(1 hr at 60°C for NR1 and NR2A, 20 min at 65°C for NR2B) in 1× SSC
buffer (0.015 M sodium citrate, 0.15 M NaCl), the tissue was incubated
at room temperature for 5 min each in 1× SSC, 0.1× SSC, 70% ethanol
and 95% ethanol and then air-dried. Slides were then applied to film
(
max; Amersham, Arlington Heights, IL) and developed for 2 days
(NR1), 4 days (NR2A) or 4 to 6 days (NR2B).
Data analysis.
Measures were expressed as mean ± S.E.M. for all experiments. Drug effects were evaluated using
Student's t test, and differences between chronic
ethanol-treated and control groups for input-output curves and time
course experiments were evaluated using ANOVA followed by Bonferroni
corrected comparisons between the intervals indicated for each
experiment. Differences were considered significant at a confidence
interval of P < .05.
A quantitative analysis of NR2B subunit-specific oligonucleotide probe
density was performed on explant tissue subjected to in situ
hybridization after chronic ethanol exposure using image analysis
software (MCID, Imaging Research, St. Catherines, Ontario, Canada).
Probe density per slice was estimated by averaging values from 10 circular regions in the cell body layers of area CA1, CA3 and dentate
gyrus, calibrated to a density standard curve to give relative probe
density. Mean values for control and chronic ethanol-treated slices
were compared using Student's t test.
| |
Results |
Acute effects of ethanol on NMDAR-mediated synaptic responses.
NMDAR-mediated EPSCs were recorded from 20 CA1 pyramidal cells in
hippocampal explants that were cultured for 8 to 33 days. NMDA EPSCs
recorded in the explants showed kinetic and pharmacological properties
characteristic of currents mediated by this subtype of glutamate
receptor. Figure 1 shows the inward
current evoked in a CA1 pyramidal cell by stimulation of Schaffer
collaterals in an explant slice after 8 days in vitro. The
stimulus intensity for all patch clamp recordings was adjusted to evoke
the maximal inward current obtainable without apparent loss of voltage
clamp (activation of voltage-dependent conductances). The mean peak amplitude for the evoked EPSCs was 172 ± 13 pA and ranged from 65 to 265 pA (n = 20). The evoked currents exhibited the
relatively slow rise and decay times characteristic of NMDAR-mediated
synaptic currents. The mean rise time for the NMDA EPSCs was 28 ± 1.4 msec with a range from 18 to 41 msec (n = 20),
while the current decay was fitted to a single exponential curve with a
mean time constant of 140 ± 22 msec (r = 0.96 ± 0.01, n = 7). The evoked currents were reversibly blocked by
bath application of 2 µM CGS-19755 (97 ± 3% inhibition,
n = 4 cells), a specific NMDAR competitive antagonist
(Lehmann et al., 1988), confirming that they were mediated by activation of NMDA receptors (fig. 1).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1.
NMDAR-mediated EPSCs show characteristic time
course and pharmacology in hippocampal explants. A, Evoked EPSCs
recorded from a CA1 pyramidal cell in an explant after 8 days in
culture (stimulus amplitude = 20 µA; holding potential = -52 mV; all traces are average of 5 responses). Traces are shown for
the current evoked in 10 µM DNQX and 0.1 mM Mg2+, after
perfusion with 2 µM CGS-19755, and after drug washout. Trace at
bottom shows control waveform with mean values for key parameters. B,
Mean amplitude of NMDA EPSCs (n = 4 cells) in
control solution, during perfusion with 2 µM CGS-19755 and after
washout of drug. Inhibition by CGS-19755 was significant (Student's
t test, P < .05), and mean amplitude after drug
washout was not different than control value. The amplitude of the
responses was measured 100 msec after the shock artifact.
|
|
Bath application of ethanol reversibly inhibited NMDA EPSCs recorded
from CA1 pyramidal cells in the explants (fig.
2). In the recording shown, application
of 75 mM ethanol reduced the EPSC by 60% relative to the control
response measured at a latency of 100 msec from the stimulus artifact.
Exposure to 75 mM ethanol reduced the amplitude of NMDA EPSCs on
average by 42 ± 6%, with a range of inhibition from 18 to 60%
(n = 6 cells from explants 9-28 days in
vitro; the level of inhibition was not significantly correlated
with explant age). This inhibition was reversible after washout of
ethanol, returning to 95 ± 8% of control amplitude (n = 5 cells). Thus, the NMDAR-mediated component of
synaptic transmission in the hippocampal explants is inhibited by
ethanol to a similar extent as previously demonstrated for native
hippocampal tissue (Lovinger et al., 1989; Morrisett and
Swartzwelder, 1993).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 2.
Ethanol inhibits NMDAR-mediated EPSCs in
hippocampal explants. A, Evoked NMDAR-mediated EPSCs recorded from a
CA1 pyramidal cell in an explant after 14 days in culture (stimulus
amplitude = 12 µA; holding potential = -50 mV; all traces
are average of 5 responses). Ethanol was applied for 15 min, and
effects were seen within 3 to 5 min. Wash responses were obtained 15 to
30 min after switching back to control solution. Trace at bottom shows
control response overlapped with ethanol response scaled to the same
peak amplitude. B, Mean EPSC amplitude (n = 6 cells) in control solution, during exposure to 75 mM ethanol and after
washout of ethanol. Inhibition by 75 mM ethanol was significant
(Student's t test, P < .01), and mean amplitude
after drug washout was not different than control value. The amplitude
of the responses was measured 100 msec after the shock artifact.
|
|
Since we eliminated GABAA-mediated currents by
clamping near ECl, it is conceivable that an
ethanol-induced enhancement of GABAA channel
activity could lead to an apparent decrease in NMDAR current via a
membrane shunting effect. This seems unlikely, however, because the
resulting change in membrane time constant would likely alter the
kinetics of the EPSC decay, and scaling of EPSC waveforms evoked in the
presence of ethanol to control waveforms (fig. 2, bottom) indicate that
no change in decay kinetics occurs.
Chronic effects of ethanol on population field recordings.
Population recordings were used to assess the effects of chronic
ethanol exposure on excitatory synaptic function in the explants. A
highly reproducible population waveform was recorded with an electrode
placed in the cell body layer of area CA1 after stimulation of fibers
in stratum radiatum of area CA3 (presumed Schaffer collaterals). The
waveform evoked consisted of a slow positive potential on which, at
higher stimulus amplitudes, was superimposed a sharp, negative-going
spike at a variable latency from the shock artifact (fig.
3A). These potentials correspond to the
PSP and PS, respectively, recorded with this electrode configuration in
acute hippocampal slices. In this study we focused on the field PSP as
a general measure of synaptic function.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 3.
Population PSP recordings in explants show steep
input/output relationship. A, Representative population synaptic
potentials (PSPs) evoked in a hippocampal explant at several stimulus
intensities (shown to the left in µA). Note that at the two highest
stimulus amplitudes, a slowly relaxing component appears in the PSP
waveform (especially prominent at 20 µA). The PS indicated by the
arrow is just noticable at 20 µA. B, PSP peak amplitude plotted
vs. stimulus amplitude for the evoked responses shown in
A.
|
|
As seen in acute slices, increases in stimulation intensity resulted in
a graded increase in the peak amplitude of the PSP from a threshold
value until a maximal PSP was evoked (fig. 3A). Figure 3B shows a plot
of the PSP peak amplitude vs. stimulus current amplitude for
the waveforms shown in figure 3A. Note that the amplitude of the PSP
rises steeply as a function of stimulus intensity; the response grades
from threshold to maximal amplitude over a 5 µA range of stimulus
amplitude. Further increases in stimulus amplitude increased the PS
amplitude and tended to lead to a decrease in the observed peak of the
PSP. At higher stimulus amplitudes, a slowly decaying component of the
PSP was observed (fig. 3A, bottom waveform). This component appeared at
stimulus amplitudes that evoked near-maximal peak PSP responses. Test
responses, recorded hourly during the recording period, were evoked at
a stimulus amplitude that produced a near-maximal peak PSP and a measurable slow component.
The relationship between PSP peak amplitude and stimulus intensity was
not changed after chronic ethanol exposure, neither before nor after
the withdrawal period (fig. 4). Figure 4A
shows cumulative input/output plots obtained before and after the
withdrawal period comparing data from 6 control slices and 8 slices
exposed to 75 mM ethanol for 6 days. No difference was observed between chronic ethanol-treated (CE) and control groups for the threshold stimulus intensity (8.3 ± 1.1 µA for ethanol vs.
8.3 ± 1.2 µA for controls, fig. 4B). The maximal PSP peak
amplitude (measured at twice the threshold stimulus) was 4.0 ± 1.0 mV for CE slices before the withdrawal period compared to 4.0 ± 0.7 mV for control slices. No significant difference was seen
between input/output curves obtained from these same sets of slices 7 hr into the withdrawal period. Thus, the threshold stimulus intensity
was 8.7 ± 1.5 µA for CE slices vs. 8.0 ± 1.1 µA for control slices (fig. 4B) and the maximal PSP peak amplitude
was 4.3 ± 0.8 mV for the CE group compared to 3.9 ± 0.8 mV
for controls.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 4.
PSP input/output curves are unchanged after chronic
ethanol exposure and withdrawal. A, Plot of PSP peak amplitude
vs. stimulus amplitude for control and 75 mM chronic
ethanol (CE) explant slices before ethanol withdrawal (solid lines) and
at the end of the recording period (dotted lines). The stimulus
amplitude was normalized to the threshold intensity for each experiment
for comparison between slices. Error bars (±S.E.M.) are shown only for
the postwithdrawal control slices for clarity and were similar for the
other groups. No significant differences in the curves were observed,
either before or after withdrawal, between 75 mM CE and control groups
(ANOVA, comparing 1×, 1.5×, and 2× threshold values for 75 mM CE,
n = 8 and control groups, n = 6). B, Mean stimulus amplitude for evoking a measurable PSP for 75 mM
CE and control groups before and after withdrawal.
|
|
Differences in the slow synaptic component were apparent in chronic
ethanol-treated explants. Figure 5 shows
evoked waveforms recorded from an explant slice after chronic exposure
to 35 mM ethanol compared to corresponding waveforms recorded from a
control explant. Maximal evoked PSPs are shown for three recording
conditions (and the corresponding periods for the control slice):
before withdrawal of ethanol from the recording media, several hours after the withdrawal of ethanol and after application of the
competitive NMDAR antagonist, D-APV. A comparison of the prewithdrawal
waveforms indicates that, while the peak amplitudes of the PSPs were
similar, the slow component was substantially larger in the waveform
from the CE explant (38% of peak for the ethanol slice vs.
10% of peak for the control slice). After the removal of ethanol from
the recording solution, the slow component was further enhanced (to 92% of peak compared to 23% for the control slice at the
corresponding time period), greatly prolonging the duration of the PSP
(fig. 5, post-W/D). This enhancement occurred without a significant change in the peak of the PSP over the duration of the recording period.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 5.
Increase in NMDAR-dependent slow synaptic component
after withdrawal from chronic ethanol exposure. Evoked PSPs recorded in
a control explant (A) and from an explant exposed to 35 mM ethanol for
6 days (B). Traces are averages of 5 responses. Representative
waveforms are shown for the prewithdrawal period, at 5 hr after
withdrawal (Post-W/D), and after application of 25 µM D-APV (bottom
trace shows D-APV and postwithdrawal waveforms superimposed). At right,
arrows indicate how the peak PSP amplitude and the PSP slow component
amplitude (at 50 msec after the stimulus artifact) were measured.
|
|
The slowly decaying component of the PSP was strongly inhibited by the
application of D-APV in control and CE slices (fig. 5), indicating that
this component is dependent on the activation of NMDA receptors. PSPs
recorded in the presence of 25 µM D-APV (i.e., the
non-NMDAR dependent component) were characterized by a rapidly decaying
waveform that, in control explants from the 35 mM CE experiment,
returned to near baseline by 50 msec from stimulus onset (fig. 5,
control, D-APV). Note that for the 35 mM CE explant the slow component
was not entirely eliminated in the presence of 25 µM D-APV. Thus, for
the control slice the PSP amplitude at 50 msec was reduced from 1.0 mV
to 0 mV after application of 25 µM D-APV, whereas in the 35 mM CE
slice, D-APV reduced the PSP amplitude at 50 msec from 3.3 mV to 0.9 mV. For the 75 mM CE experiments, we used 50 µM D-APV which virtually
eliminated the slow component measured at 50 msec latency (see below).
Thus, we used the amplitude of the PSP at 50 msec (fig. 5B) as an
indirect measure of the NMDAR-dependent component because the waveform at this latency is relatively free from contamination by the non-NMDAR dependent component.
Figure 6 shows the time course of the
slow, NMDAR-dependent synaptic component for all experiments in
explants chronically exposed to 35 mM ethanol (fig. 6A) and 75 mM
ethanol (fig. 6B). The graphs compare mean values at each time point
for all experiments in the two CE groups with the values at
corresponding time points for their respective control groups. Note
that the slow component was larger in the 35 mM CE group before the
removal of ethanol from the recording solution when compared with
control explants (mean for 2-hr period was 35 ± 8% of peak PSP
for 35 mM ethanol, n = 7 and 17 ± 6% of peak PSP
for controls, n = 4, fig. 6A), although the difference
was not significant. In contrast, no difference was observed before
ethanol withdrawal in the 75 mM CE group relative to controls (2-hr
mean was 27 ± 4% of peak PSP for 75 mM ethanol, n = 8 and 24 ± 4% of peak PSP for controls,
n = 5, fig. 6B).
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 6.
Increase in PSP slow component after ethanol
exposure occurs without a significant change in PSP peak amplitude.
Plots of slow PSP component vs. time for all slices
exposed to 35 mM (A) and 75 mM B, chronic ethanol. Responses were
measured as shown in figure 5 and expressed as percent of peak PSP.
Mean values ± S.E.M. are shown for each time point (35 mM
ethanol, n = 7; 35 mM control,
n = 4; 75 mM ethanol, n = 8; 75 mM control, n = 6). Asterisks indicate time points
where values for CE explants are different from corresponding control
values, and plus signs indicate time points where postwithdrawal values
are different from the 2-hr averaged prewithdrawal values (* = P < .05, ** = P < .01, + = P < .01, ++ = P < .001 or smaller; ANOVA followed by Bonferroni corrected comparisons at each
time point). Final points show mean slow component in the presence of
D-APV (25 µM for 35 mM CE experiment, 50 µM for 75 mM CE
experiment). All measures in D-APV were significantly reduced compared
with final predrug values (35 mM CE, P < .001, n = 7; 35 mM controls, P < .05, n = 3; 75 mM CE, P < 1E-6,
n = 8; 75 mM controls, P < .01, n = 5; Student's t test). Insets
show plots of PSP peak amplitude for the same experiments. Mean
values ± S.E.M. are shown for each time point. Values for the
peak PSP were unchanged across time for both 35 mM (A) and 75 mM B,
experiments, and values for CE explants were not different from
corresponding control values (ANOVA). Final points in all plots show
mean PSP peak amplitude in the presence of D-APV for both CE and
control explants. Measures in D-APV were not significantly different
compared with final predrug values for any group (Student's
t test).
|
|
For both groups, a marked and significant increase in the slow
component was observed within 1 hr from removal of ethanol from the
perfusate. This increase occured without a significant change in the
peak of the PSP. In the 35 mM group, the NMDAR-dependent component
increased on average from 35 ± 8% of peak PSP before withdrawal
to 56 ± 12% of peak PSP 1 hr after withdrawal (n = 7), an increase of 60%. In the 75 mM group a larger relative
increase of 89% was observed, from 27 ± 4% to 51 ± 7% of
peak PSP (n = 8). Comparing hour by hour, the
difference in the slow component between the 75 mM CE group and
controls was significant during the entire withdrawal period (reaching
P < .01 by 4 hr after withdrawal), where no significant change in
the slow component was observed in control slices (fig. 6B,
n = 5). Control slices interleaved with the 35 mM CE
group did display a slow, modest increase in the slow component on
average during the course of the experiment (fig. 6A, n = 4). This experiment was performed first, before the 75 mM experiment,
and because no increase in the slow component was observed in the 75 mM
controls, this slow increase may represent an artifact of experimental
procedure. The slow component was significantly larger during the
entire withdrawal period relative to the prewithdrawal period in the 35 mM CE explants; however, there was no statistically significant
difference in the slow component between the CE and control groups
during the withdrawal period.
The final points for each plot in figure 6 show the mean amplitude of
the PSP slow component in the presence of D-APV (25-50 µM). Compared
to values measured just before drug exposure, 25 µM D-APV
significantly reduced the slow component from 63 ± 9% to 21 ± 5% of peak PSP in the 35 mM CE group (n = 7) and
from 39 ± 3% to 7 ± 4% of peak PSP in the control group
(n = 3, fig. 6A). Relative to predrug values, 50 µM
D-APV significantly reduced the slow component from 71 ± 4% to
4 ± 2% of peak PSP in the 75 mM CE group (n = 8)
and from 22 ± 11% to 3 ± 2% of peak PSP in the control
group (n = 5, fig. 6B).
The increase in the slow NMDAR-dependent PSP component that was
observed after chronic ethanol exposure occurred without a significant
change in the peak of the PSP. The insets in figure 6 show the time
course of the peak PSP amplitude for all slices in the two ethanol
groups and their corresponding controls. No significant change in peak
PSP amplitude occurred in either experimental group at any time point.
The final points for these plots show the mean peak amplitude of the
PSP in the presence of D-APV (25-50 µM). These values were not
significantly different from the mean peak PSP amplitude before drug
application (5.8 ± 1.2 vs. 6.3 ± 1.0 mV predrug
for 35 mM CE, n = 7 and 3.0 ± 0.6 vs.
3.9 ± 0.5 mV predrug for controls, n = 3;
4.1 ± 1.0 vs. 4.5 ± 1.1 mV predrug for 75 mM CE,
n = 8 and 3.9 ± 0.8 vs. 4.8 ± 0.9 mV predrug for controls, n = 5) further
demonstrating that the peak PSP can be used as a measure of the
non-NMDAR mediated synaptic component.
Acute ethanol exposure and withdrawal.
The enhancement of the
slow synaptic component after withdrawal from chronic ethanol exposure
could conceivably be due, at least in part, to a rebound of NMDAR
function from the acute inhibitory effects of ethanol. To test for this
possibility, we studied the effects of acute ethanol exposure and
withdrawal on the slow synaptic component in recordings from
ethanol-naive explants. Figure 7 shows
the results of these experiments. Evoked PSPs were recorded before,
during and for 3 hr after a 1-hr exposure to 75 mM ethanol. In these
experiments, a stimulus intensity was used that produced a large slow
component in the field PSP so that an acute inhibitory effect of
ethanol could be readily observed. Application of ethanol to the
bathing solution significantly decreased the slow PSP component to
70 ± 10% of the control value within 15 min (n = 4 slices). After washout of ethanol, the slow component returned toward
control levels within 15 min, with a transient enhancement observed in 3 of 4 slices (mean of 126 ± 12% of control for all 4 slices at 1 hr after washout); however, the response returned to control levels
by 2 hr after washout.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 7.
Acute ethanol exposure does not result in lasting
enhancement of NMDAR-dependent PSP component in ethanol-naive explants.
Plot shows the slow PSP component before, during and for 3 hr after
exposure to 75 mM ethanol for 1 hr. Explants (n = 4) were treated as described for previous control experiments,
including placement in a humidified chamber for 6 days. Values are
expressed as a percentage of the mean value calculated over the 3 intervals before ethanol exposure. The mean value for the 1-hr interval
in ethanol was significantly reduced compared to the mean for the
control period (P < .05), whereas mean values for each 1-hr
interval after ethanol removal were not significantly different from
the control period (ANOVA followed by Bonferroni corrected comparisons
for each hour interval).
|
|
Ethanol withdrawal and ictal events in explants.
Explant
slices that were exposed to chronic ethanol displayed a marked
hyperexcitability during the period after withdrawal from ethanol. This
electrical hyperactivity was manifested as spontaneous epileptiform
burst discharges, large abrupt depolarizing shifts and long-lasting
phasic or phasic/tonic electrographic seizure (EGS) events. The
long-lasting EGSs often followed single electrical shocks applied to
evoke PSPs but they occasionally occurred spontaneously, after large
amplitude burst discharges. Figure 8
shows an EGS induced after a single stimulus in an explant slice after
withdrawal from chronic exposure to 35 mM ethanol. This event lasted
approximately 90 sec and showed a morphology that was typical of the
phasic/tonic type of ictal events seen during withdrawal. After the
initial evoked response (from less than a second to several seconds),
ictal events typically were initiated by spontaneous population spikes
(fig. 8a) which increased in frequency, producing a slow depolarization
envelope (fig. 8b). This was followed by high-amplitude, phasic bursts
that often led to a period of tonic firing (fig. 8c). Seizure events
that progressed from phasic to tonic bursting eventually returned to a
phasic bursting phase. Finally, the hyperpolarizations that followed
phasic bursts (fig. 8d) gradually increased in amplitude, terminating
the spontaneous bursting activity (fig. 8e) and resulting in a
quiescent period where excitability (evoked and spontaneous) was
greatly reduced. During this quiescent period, spontaneous epileptiform
burst discharges and depolarizing shifts were much less frequent or
entirely absent, and evoked PSPs were greatly reduced and required 30 min or more to return to their previous amplitude.
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 8.
Ethanol withdrawal induces ictal events in
chronically exposed explants. Top trace shows an electrographic seizure
recorded in area CA1 evoked by a single shock (30 µA, arrow) in an
explant slice after withdrawal from 35 mM ethanol. Traces labeled a
through e show expanded waveforms at points corresponding to letters in
top trace. The duration of EGS events were measured from the onset of
the first spontaneous burst to the last phasic burst before the
quiescent period.
|
|
The ictal events were all-or-none in nature, and their duration tended
to be similar in a particular explant. Thus, once initiated they
exhibited a full phasic/tonic discharge sequence with no events of
intermediate amplitude or duration observed. For the explant shown in
figure 8, four events were measured during the withdrawal period with a
mean duration of 93 sec and a range from 85 to 105 sec. In other
explants, long-lasting seizures occurred that had a similar morphology
except that there was little or no tonic, high-frequency firing phase,
and the event consisted of phasic bursts separated by hyperpolarizing
shifts. The duration of this type of electrographic seizure was more
variable from slice to slice, ranging from 30 sec to several minutes in
duration, but again tended to be similar for a particular explant.
The expression of ictal events during ethanol withdrawal was dependent
on the activation of NMDA receptors (fig.
9A). Application of D-APV (25 µM for 35 mM CE slices, n = 5; 50 µM for 75 mM CE slices,
n = 6) resulted in the complete loss of seizure events in all slices that had previously displayed ictal activity. The amplitude of spontaneous burst events and abrupt depolarizing shifts
was also reduced after drug exposure. In several experiments where
input/output curves were recorded hourly during the entire recording
period, higher stimulus intensities typically evoked a larger slow
component. Analysis of the relationship between stimulus intensity,
slow PSP amplitude and EGS expression within single explants revealed
that EGSs occurred after evoked PSPs that exhibited larger slow
components whereas they were never observed after PSPs with smaller
components evoked by lower stimulus amplitudes. For the 35 mM CE group,
the slow component was 84 ± 4% of peak for PSPs that evoked EGSs
and 19 ± 8% of peak for PSPs evoked at lower stimulus amplitude
where no EGS occurred (P < .01, n = 4). For the
75 mM CE group, these values were 78 ± 6% of peak and 31 ± 10% of peak, respectively (P < .01, n = 6). In 2 explants where recordings were continued after D-APV washout, EGS
events were observed only after the slow component of the PSP had
reappeared. Therefore, the induction of ictal events was time-locked
with the expression of the NMDAR-dependent synaptic component.
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 9.
Ictal events seen during ethanol withdrawal are
NMDAR-dependent. A, Histograms show percentage of explant slices in
each group that showed ictal activity at some point in the recording
period (left), and frequency of ictal events in slices displaying ictal
activity (right). Seizure frequency was calculated as the average
number of hours in the 7-hr withdrawal period where ictal events were
observed for each slice. B, Histogram shows the number of slices in
each CE group exhibiting at least one ictal event during each 1-hr
period during withdrawal.
|
|
Ictal events were observed at nearly equal frequency in slices exposed
to 35 mM and 75 mM ethanol and tended to occur more frequently later in
the recording period after ethanol withdrawal (fig. 9B). In the 35 mM
CE group, EGSs were exhibited in 5 of 7 slices during the 7-hr
withdrawal period. On average, these events occurred at least once
during 4 of the 7 hr of the withdrawal period. For the 75 mM group,
where 6 of 8 slices exhibited EGSs, the events occurred in 5 of the 7 hr on average. No ictal events were observed in ethanol-exposed slices
from either group during the prewithdrawal period where ethanol was
present in the recording solution. In the control groups, only one
slice from the 35 mM group displayed an ictal event, and this event
occurred only once, late in the recording period (i.e., one
event detected in over 100 hr of recordings from 10 slices).
NMDAR subunit composition in hippocampal explants.
In
situ hybridization techniques were used to determine whether mRNA
coding for NMDAR subtypes known to be present in native hippocampus
were present in the explants after 2-3 weeks in culture. Figure
10A shows images of representative
autoradiograms from slices incubated with oligonucleotide probes for
the NR1, NR2A and NR2B subunit message. The oligonucleotide sequence
for the NR1 subunit (pan-R1) was complementary to a sequence encoding
amino-acid residues 566-580, common to all splice variants of the NR1
polypeptide (Monyer et al., 1992). Binding was observed for
all three probes, primarily in the cell body layer of areas CA1 and CA3
and the granule cell layer of dentate gyrus.
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 10.
In situ hybridization reveals
presence of mRNA for NMDAR subunits. A, Representative autoradiograms
showing NR1, NR2A, and NR2B subunit mRNA distribution in explants
cultured for 2-3 weeks. Slices incubated with excess cold
oligonucleotide probe for the 2B subunit displayed no nonspecific
binding (not shown). B, Relative NR2B subunit probe density in CE
vs. control explants. Values are shown for each
concentration of ethanol (35 mM and 75 mM) pooled from two experiments
for each group relative to control values. Pooled values for the CE
explants were not different from control values for either group
(Student's t test). Note: To control for
variations in slice thickness, we established a thickness calibration
assay using the coomassie blue protein stain. Native brain tissue was
sliced from 10 to 150 (thick using a microtome and thaw-mounted to
microscope slides. The tissue slices were treated in the same fashion
as the explant slices (ie. fixation, in
situ rinsing, and dehydration conditions) and then incubated in
a 0.001% coomassie blue solution for 1.5 hr, followed by overnight
incubation in a destaining solution (45% methanol, 10% glacial acetic
acid). Explant tissue from 35 mM study #2 was subjected to the same
staining protocol after the in situ protocol, and the
density of each explant slice was compared to a calibration curve
established using the native tissue slices. This procedure did not
significantly affect the difference between control and ethanol-exposed
tissue.
|
|
Chronic effects of ethanol on NR2B subunit expression.
Alterations in NR2B mRNA and protein expression after chronic ethanol
exposure have been reported (Follesa and Ticku, 1996; Hu et
al., 1996). Expression of mRNA encoding the NR2B subunit was
therefore examined in explants after chronic exposure to either 35 mM
or 75 mM ethanol. Two independent sets of experiments were performed
for each ethanol concentration. These data are summarized in figure
10B. Exposure to 35 mM ethanol resulted in an average increase in NR2B
subunit message to 120 ± 9% of control (n = 21 control, n = 21 ethanol slices, both experiments
pooled). Exposure to 75 mM ethanol resulted in an increase in NR2B
subunit message to 147 ± 16% on average (n = 15 control, n = 16 ethanol slices, both experiments
pooled). Statistically significant differences were obtained for both
CE groups in one of the two independent experiments (35 mM, P < .05; 75 mM, P < .01) whereas pooling of data from both
experiments resulted in no overall significant difference. Thus, there
was a trend which suggests that the level of NR2B message was higher in
ethanol-exposed explants than in controls, consistent with previous
reports of changes in NR2B subunit mRNA after chronic ethanol exposure
(Hu et al., 1996).
| |
Discussion |
NMDAR-mediated synaptic responses and the acute effects of
ethanol.
NMDAR-mediated EPSCs recorded from CA1 pyramidal cells in
the explants showed characteristics similar to NMDAR-mediated currents isolated in acute slice preparations (Hestrin et al., 1990;
Randall et al., 1990). The mean time course of the EPSCs was
similar to that observed in recordings from acute slices from mature
rats (Hestrin et al., 1990; Randall et al.,
1990). The similarity between the NMDAR-mediated EPSCs recorded in the
explants and those recorded in acute slices of hippocampus from mature
brain provides additional evidence to suggest that NMDAR function
develops normally in the explants (Bahr, 1995).
Prior to examining the effects of chronic ethanol exposure on NMDAR
function in the explants, it was important to establish whether this
receptor subtype was sensitive to acutely applied ethanol. Ethanol has
been shown to inhibit NMDAR-mediated synaptic responses recorded in
acute hippocampal slices (Lovinger et al., 1990; Morrisett
and Swartzwelder, 1993) and NMDA-induced currents in dissociated
neurons (Lovinger et al., 1989). However, recent evidence
suggests that the sensitivity of NMDARs to ethanol may decrease with
developmental age (Swartzwelder et al., 1995; Lovinger, 1995). With this possible confound in mind, we used a concentration of
ethanol (75 mM) demonstrated to maximally inhibit NMDAR-mediated responses in mature hippocampal tissue (Lovinger et al.,
1990; Morrisett et al., 1991). The level of inhibition of
NMDA EPSCs by 75 mM ethanol acutely applied to the explants was close
to that seen for CA1 population synaptic potentials recorded in acute slices from adult rats (Lovinger et al., 1990; Morrisett
et al., 1991). Additionally, these results provide the first
demonstration of an acute inhibitory effect of ethanol on
NMDAR-dependent synaptic currents in hippocampal explants.
Linkage of NMDAR synaptic responses and ictal events.
Field
recordings demonstrated the selective enhancement of a slow,
NMDAR-dependent component of the evoked PSP in explant slices after
in vitro withdrawal from chronic ethanol exposure. This
conclusion regarding the specificity of the enhancement to the NMDA
vs. the non-NMDA component of transmission is supported by
the following evidence. The threshold stimulus amplitude (reflecting presynaptic fiber excitability) was unaltered after chronic ethanol exposure and withdrawal. The average peak amplitude of the field PSP,
reflecting the non-NMDAR mediated component of excitatory transmission,
was not significantly different in ethanol-exposed explants and was
unaffected by NMDAR antagonists. Conversely, Molleman and Little
(1995), using acute slices prepared from animals chronically exposed to
ethanol, have observed increases in isolated fast excitatory synaptic
potentials. However, the increases were not observed until several
hours after withdrawal, which was taken as the time of slice
preparation. There are technical and theoretical reasons which might
account for this difference; however, we would suggest that a delayed
upregulation of the non-NMDA component may occur as a consequence of an
earlier enhanced NMDAR activity during withdrawal.
Our finding of an increase in an NMDAR-mediated synaptic component in
the hippocampal explants was statistically significant only after
withdrawal (however there was a trend for enhancement during chronic 35 mM exposure, possibly reflecting less acute inhibition by ethanol).
Indeed, this points up the major thrust of this study: to understand
the role of NMDAR function in withdrawal seizure expression. Previous
investigators have reported alterations in NMDAR channel antagonist
binding or NMDAR-dependent calcium flux in native tissue after chronic
exposure but before withdrawal (Grant et al., 1990; Iorio
et al., 1992) and in electrophysiological assessment of
native tissue prepared from animals after withdrawal (Ripley and
Little, 1995; Whittington et al., 1995). The present work
extends these findings but is distinct in one important respect: we
have electrophysiologically monitored the alterations in NMDAR function
during the withdrawal period and correlated the observed changes with
electrographic seizure event occurrence. This is a particular technical
advantage afforded by the explant model system that enables the linkage
of synaptic mechanisms to the pathological phenomenon of interest, with
real time assays for both measures.
The most significant finding of this study is the direct correlation
between the enhanced NMDAR activity observed after the washout of
ethanol and the subsequent expression of electrographic seizures.
Several lines of evidence indicate that the expression of the ictal
events was dependent on activation of NMDARs. First, the induction of
EGS events was temporally correlated with the expression of the
NMDAR-dependent component of the field PSP. Thus, EGSs were only
initiated after evoked or spontaneous PSPs that exhibited a large slow
component. Second, the EGSs were completely abolished in the presence
of the NMDAR antagonist, D-APV. Further, the expression of seizure
activity coincided with the reappearance of a large slow PSP component
during the washout of the receptor antagonist. Third, ictal events were
not observed before the withdrawal of ethanol from the recording media,
a period where the NMDAR-dependent PSP component was substantially
reduced. Therefore, we conclude that NMDAR activation is required for
the expression of postwithdrawal ictal activity in hippocampus.
The ability of an in vitro model system to exhibit EGS
events is of obvious advantage for the study of cellular mechanisms of
withdrawal hyperexcitability. However, we recognize that it is
important to distinguish between hyperexcitability that may develop due
to the effects of chronic ethanol exposure and that described
previously in long-term explant preparations. Explants cultured using
the roller-tube method (Gähwiler, 1981) have been reported to
display an increased level of excitability with increasing time in
culture (McBain et al., 1989). These explants display a
progressive onset of spontaneous epileptiform bursts and can exhibit
long-lasting ictal events, particularly after 30 days in culture
(McBain et al., 1989). In the present study, explants were
cultured using the interface method (Stoppini et al., 1991) which preserves more three-dimensional structure of the hippocampal network. The slices were maintained in vitro for 1 week
before exposure to ethanol, a period of time that allowed for recovery of synaptic responses (Muller et al., 1993). Recordings were
thus performed on tissue that had been in culture for 2 to 3 weeks, and
we observed ictal activity in only 1 of 14 control slices. We conclude
that slices cultured using the interface technique display more
acceptable characteristics for a study such as this than slices
cultured using the roller-tube technique.
One potential confound is related to the possibility that neurotoxicity
may develop due to our exposure and/or withdrawal paradigm, and it is
this neurotoxicity that may be responsible for the hyperexcitability
observed. We have assessed this possibility in another extensive study
through the use of propidium iodide fluorescence techniques (Thomas and
Morrisett, 1997; Thomas and Morrisett, submitted). In that study, we
found no evidence for a role of neurotoxicity mechanisms in the
induction or expression of the withdrawal hyperexcitability.
Furthermore, those studies substantiated the present results in which
increased NMDA receptor function was observed after withdrawal from
chronic ethanol exposure.
NMDAR subunit composition and the effects of chronic ethanol
exposure.
NMDA receptor ionophores are thought to exist as
heteromeric complexes consisting of subunits derived from two related
gene families, designated NMDAR1 (or NR1) and NMDAR2 (NR2) for rat brain (Moriyoshi et al., 1991; Monyer et al.,
1992). Developmental changes in subunit expression have been described
for rodent brain (Watanabe et al., 1992; Monyer et
al., 1994; Sheng et al., 1994; Wenzel et
al., 1997). In particular, NR2A subunit expression in hippocampus
is low at birth and peaks during the first 3 postnatal weeks (Wenzel
et al., 1997), whereas NR2B expression peaks perinatally before declining to adult levels (Watanabe et al., 1992).
The presence of both NR2A and NR2B subunit mRNA in explants cultured for several weeks is thus consistent with the maintenance of a relatively mature NMDAR subunit composition in organotypic hippocampal cultures.
Changes in NMDAR subunit expression have been reported after chronic
ethanol exposure (Trevisan et al., 1994; Hu et
al., 1996). Hu and coworkers (1996), using an RNase protection
assay, reported that in cultured dissociated cortical neurons after
chronic ethanol exposure the expression of mRNA encoding the NR2B
subunit was selectively increased. We observed an increase in NR2B
subunit mRNA levels in the explants after chronic ethanol exposure
using in situ hybridization techniques, consistent with the
results of Hu et al. (1996). These data thus support the
hypothesis that an increase in NR2B mRNA expression occurs after
exposure of central neurons to chronic ethanol and may represent one
mechanism contributing to the observed enhancement of NMDAR function.
Summary.
We have demonstrated a marked enhancement of NMDAR
function in hippocampal explants which is apparent after withdrawal
from chronic (but not acute) ethanol exposure, which is also strongly correlated with the expression of NMDAR-dependent epileptiform events.
These findings appear to satisfy criteria that support a causative
relationship between enhancement of NMDAR function and the induction
and subsequent expression of ethanol withdrawal hyperexcitability. The
primary caveat to this argument is that it assumes that the expression
of withdrawal hyperexcitability exhibited by the explanted hippocampus
represents a model system that is relevant to behavioral seizure
expression during ethanol withdrawal. Obviously, other brain regions,
most notably the inferior colliculus (McCown et al., 1995),
also represent important sites for withdrawal hyperexcitability.
The role that this alteration in NMDAR function plays in the generation
of the withdrawal events in the whole animal in terms of the generation
of the paroxysmal depolarizing shift remains to be explored,
particularly with respect to other cellular alterations contributing to
withdrawal hyperexcitability. Taken together, these different threads
should help begin to tie together a more cogent and thorough
understanding of the alterations in NMDAR function which may underlie
ethanol withdrawal hyperexcitability and seizures.
The authors would like to thank Dr. Ben Bahr for technical
advice on culturing hippocampal explants and Dr. Pat Randall for assistance with the statistical analyses.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
