Citation
The effects of chronic ethanol treatment on long term potentiation in the hippocampus

Material Information

Title:
The effects of chronic ethanol treatment on long term potentiation in the hippocampus
Creator:
Tremwel, Margaret Fairchild, 1954-
Publication Date:
Language:
English
Physical Description:
vii, 114 leaves : illustrations; 29 cm.

Subjects

Subjects / Keywords:
Research ( mesh )
Ethanol -- pharmacology ( mesh )
Ethanol -- toxicity ( mesh )
Long-Term Potentiation -- physiology ( mesh )
Long-Term Potentiation -- drug effects ( mesh )
Synapses -- drug effects ( mesh )
Pyramidal Cells -- drug effects ( mesh )
GABA Modulators ( mesh )
Receptors, N-Methyl-D-Aspartate ( mesh )
Hippocampus ( mesh )
Rats ( mesh )
Department of Neuroscience thesis Ph.D ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Neuroscience -- UF ( mesh )
Genre:
non-fiction ( marcgt )
Academic theses. ( lcgft )
Academic theses ( fast )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1993.
Bibliography:
Includes bibliographical references (leaves 103-113).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Margaret Fairchild Tremwel.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
028753236 ( ALEPH )
48574031 ( OCLC )

Downloads

This item has the following downloads:


Full Text
THE EFFECTS OF C¡TRONIC ETHANOL TREATMENT ON LONG TERM
POTENTIATION IN THE HIPPOCAMPUS
By
MARGARET FAIRCHILD TREMWEL
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1993


To Terry


ACKNOWLEDGEMENTS
I would like to thank my advisor, Bruce Hunter, for his contributions to
this thesis and sharing of his expertise during the course of this study. I also
thank the members of my committee, Kevin Anderson, Mike King, John
Middlebrooks, Phil Posner, Paul Reier, Floyd Thompson, and Don Walker, for
their helpful suggestions and critical evaluation of this manuscript. I would like
to especially thank Kevin Anderson and Joanna Peris for allowing me to work in
their labs.
Doug Dillard and Tim Cera provided invaluable programming assistance
during the physiology experiments. My research relied greatly on the technical
support of Pat Burnett. During the course of my stay at UF, I have been
supported financially by an NIH training grant, the American Paralysis Assoc.,
and an NIAAA predoctoral fellowship.
My mother, Dorothy Welch has provided much support and guidance.
Many fellow students have helped me to keep a healthy perspective on work.
Michele Davda and Laura Errante are especially talented in this respect.
Members of Westminster Presbyterian Church helped us to make Gainesville
home. Lastly, I thank my husband Terry whose loving support, patience and
sense of humor has sustained me.
Tis grace that brought me here this far and grace will lead me home.
m


TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS iii
ABSTRACT vii
CHAPTERS
1 INTRODUCTION 1
2 LITERATURE REVIEW 3
Behavioral Evidence of Memory Deficits in an Animal Model
of Long Term Alcohol Consumption 7
Anatomical Evidence of Hippocampal Pathology Following
Chronic Ethanol Treatment 8
Physiological Effects of Chronic Ethanol Toxicity in the
Hippocampus 9
The Effect of Duration of Exposure and Length of Abstinence
on the Substrates of Ethanol Toxicity 12
A Physiological Correlate of Memory Formation 12
Objectives of the Dissertation 17
3 GENERAL METHODS 19
4 THE EFFECTS OF CHRONIC ETHANOL TREATMENT ON
LONG TERM POTENTIATION IN THE HIPPOCAMPUS 21
Introduction 21
Methods 24
Treatment Methods 24
Electrophysiological Methods 25
Results 30
CET Actions on LTP Induction 30
LTP Induction in the Absence of GABAergicA
Synaptic Transmission 35
Discussion 36
IV


V
5 THE EFFECT OF A RECOVERY PERIOD FROM CHRONIC
ETHANOL TOXICITY ON LONG TERM POTENTIATION 43
Introduction 43
Methods 44
Treatment Methods 44
Electrophysiological Methods 45
Results 50
CET Effect on LTP Induction 50
LTP Induction in the Absence of GABAergicA
Synaptic Transmission 55
Discussion 58
6 GABAa RECEPTOR MEDIATED CHLORIDE UPTAKE IN
CORTEX AND HIPPOCAMPUS FOLLOWING CHRONIC
ETHANOL EXPOSURE 64
Introduction 64
Materials and Methods 66
Treatment Methods 66
Preparation of Microsacs 67
Measurement of 36C1 Uptake 67
Data Analysis 68
Results 68
Effect of Chronic Ethanol on GABA Stimulation of Cl
Uptake 68
Effect of Chronic Ethanol on Bicuculline Inhibition of
GABA Stimulated Cl Uptake 69
Discussion 74
7 QUANTITATIVE AUTORADIOGRAPHIC ANALYSIS OF
NMDA RECEPTOR BINDING WITH [3H]MK-801
FOLLOWING CHRONIC ETHANOL CONSUMPTION 78
Introduction 78
Methods 80
Treatment Methods 80
Autoradiographic Methods 81
Determination of NMDA Receptor Density 83
Results 83
Effects of CET on [3H]MK-801 Saturation Binding
Characteristics 83
Effects of CET on Glutamate Stimulation of [3H]MK-
801 Binding 89
Discussion 90


vi
8 SUMMARY AND DISCUSSION 96
Summary and Interpretation 96
Future Directions 99
REFERENCES 103
BIOGRAPHICAL SKETCH 114


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE EFFECTS OF CHRONIC ETHANOL TREATMENT ON LONG TERM
POTENTIATION IN THE HIPPOCAMPUS
By
Margaret Fairchild Tremwel
May, 1993
Chairman: Bruce E. Hunter
Major Department: Neuroscience
Previous work has demonstrated that chronic ethanol toxicity (CET)
produces severe and permanent alterations in hippocampal pyramidal cell
morphological, and functional characteristics. Additionally, memory formation,
a behavioral process thought to be at least in part attributed to hippocampal
functioning is severely attenuated following CET. The present study examined
the effect of CET on a well recognized physiological correlate of memory
formation, long term potentiation (LTP). Animals were fed a nutritionally
complete, ethanol containing diet for 28 weeks and then allowed a 48 hour or 5 -
7 month abstinence period. A control group was fed the same diet except
sucrose was isocalorically substituted for ethanol. Neurophysiological methods
measured the capacity of synaptic connections onto CA1 pyramidal cells to
support LTP in response to a variety of conditioning trains. At both periods
post-abstinence, the magnitude of LTP was reduced in CET animals as compared
with pair-fed controls. LTP induction is mediated by activation of the N-methyl-
d-aspartate (NMDA) receptor complex and is modulated by activation of y-
vii


viii
aminobutyric acid (GABA)ergic synaptic transmission. The effect of CET on LTP
magnitude might have been due to effects on the NMDA-induced induction itself
or on the GABAergic modulation of induction. To distinguish between these
possibilities, the induction of LTP was tested in the presence of bicuculline
methiodide (a specific antagonist to the GABAa receptor). Under these
conditions, there was no difference between CET and controls. This result
suggests that the reduction of LTP magnitude by CET observed in the absence of
bicuculline was due to a permanent alteration of GABAergic modulation of
induction.
GABAa function was next examined by measuring the efficacy of GABA
to stimulate and bicuculline to antagonize GABA-stimulated chloride uptake in
hippocampal and cortical membrane preparations. CET did not affect basal
uptake of chloride or the efficacy of either GABA or bicuculline at the GABAa
channel.
In addition to plasticity related events, NMDA receptor activation plays a
role in excitotoxicity and cell death. To investigate the hypothesis that NMDA
receptors may mediated cell loss following CET, the binding properties an
antagonist to the NMDA receptor channel, [3H](+)-5-methyl-10.11-dihydro-5H-
dibenzo[a,d]cyclohepten-5,10-imine ([3H]MK-801) were examined. Neither the
Kj nor the Bmax of [3H]MK-801 binding to the NMDA receptor or the ability of
glutamate to stimulate [3H]MK-801] binding was altered following CET.


CHAPTER 1
INTRODUCTION
Chronic abuse of alcohol results in deterioration of anatomical,
physiological and behavioral correlates of cognitive performance. The
neurotoxic damage extends into several brain structures including thalamus,
cerebellum, cerebral cortex, and hippocampus. Of particular consequence is the
pathology of anterograde amnesia, a profound and devastating symptom of
chronic abuse.
Interest in the types of anatomical and physiological deficits responsible
for the anterograde amnesia have directed the attention of researchers to the
hippocampal formation. This region is critical for acquisition of new memories
and is specifically and dramatically affected by all stages of early and chronic
ethanol abuse (Grant et al., 1990; Lovinger et al., 1990; Walker et al., 1981; Zola-
Morgan, et al., 1986; 1992). Two exciting areas of recent research have helped to
shed light on potential physiological substrates of ethanol-produced memory
deficits. The first area has focused on ethanol interactions with a possible
physiological correlate of memory formation, long term potentiation (Lovinger et
al., 1990; Blitzer et al., 1990; DeMontis, et al., 1991; Grant et al., 1990; Gulya et al.,
1991). Through this research, we have learned of a specific susceptibility of the
N-methyl-D-aspartate (NMDA) receptor complex to both acute and proldneed
exposure to low concentrations of ethanol. The effects of ethanol on NMDA
receptor function are most likely the cause of the observed deficits in LTP
induction following acute exposure. NMDA receptors also participate during
pathological states in the production of seizures and excitotoxic cell
1


2
death (Choi, 1988; Grant et al., 1990). Therefore, alterations in NMDA receptor
function may be involved with the cell loss which follows chronic ethanol abuse.
The second area of recent research has examined the interactions of
ethanol with specific subtypes of GABAa receptor function (Allan et al., 1991;
Luddens, et al., 1990; Wafford et al., 1991). For example, a specific subunit (y2l)
of the GABAa receptor is vulnerable to the effects of acute ethanol exposure
(Wafford et al., 1991). The regional specificity of ethanol actions on GABAergic
synaptic transmission may thereby reflect the regional distribution of GABAa
receptors which contain the subunit. GABAa receptors are well recognized as
potent modulators of neuronal excitability as well as plasticity-related functions
such as LTP induction.
The purpose of this dissertation work was twofold. First, I investigated
the possibility that chronic ethanol ingestion disrupts some component of LTP
induction. Secondly, I examined some potential mechanisms through which
chronic ethanol toxicity may mediate the disruption of LTP: 1) NMDA receptor
function and 2) GABAa receptor function.


CHAPTER 2
LITERATURE REVIEW
The first characterization of a neural disorder associated with chronic
alcohol abuse was described by Carl Wernicke in 1881. The major symptoms he
noted included global confusion, opthalmoplegia, nystagmus, ataxia, and
polyneuropathy of extremities (Butters and Cermak, 1980). Postmortem
examination of brain tissue revealed punctate hemorrhages located in the gray
matter around the third and fourth ventricles. It was later learned that treatment
with large doses of thiamine could reverse disease progression with eventual
dissipation of symptomatology. Several years after this initial description of
Wernicke's encephalopathy, S.S. Korsakoff described a state of amnesia that often
accompanied polyneuropathy disorders (Butters and Cermak, 1980). These
symptoms followed long term alcoholism, vomiting, and intestinal disorders. In
Korsakoff's syndrome, patients suffer from a profound anterograde amnesia of
both verbal and nonverbal information. No alterations in intellectual functioning
are associated with the memory loss. Additionally, Korsakoff's syndrome is
associated with a graded retrograde amnesia, that is, patients can recall events
from the distant past with greater ease than events from the recent past. When
asked questions that test recent memory function, Korsakoff's patients may tend
to confabulate or fill in the time period of which they have no memory with
events from the distant past. However, this is not a consistent feature of this
disease. Korsakoff patients tend to be apathetic, passive and in general
disinterested in alcohol, a marked distinction from their premorbid personality.
Eventually it was determined that Wernicke's encephalopathy and Korsakoff
3


4
syndrome occur as separate components of the chronic ethanol disease process.
Wernicke's encephalopathy is primarily caused by malnutrition associated with,
but not caused by chronic alcoholism. Following treatment with massive doses
of thiamine, the alcoholic patient may then enter into the chronic Korsakoff's
stage of this disease process. The neuropathology associated with each of these
disease processes is distinct. Wernicke's disease involves hemorrhagic injury to
midbrain structures and cerebellum whereas the pathology of Korsakoff's is
associated with lesions of thalamic nuclei and mammillary bodies.
In 1971, Ryback proposed the hypothesis that chronic alcohol abuse
produces a graded effect on memory impairment. That is, the alcoholic
Korsakoff patient, long term alcoholic and heavy social drinker all present with a
qualitatively equal memory disorder but the severity of the deficit progresses
with duration of drinking. Indeed, several studies have demonstrated that the
chronic alcoholic patient performs more poorly in short term memory tasks than
age-matched controls, but significantly better than age-matched alcoholic
Korsakoff's patients (Parker and Noble, 1977; Ryan et al., 1980). For example,
short term memory tests were given in which the patient was asked to associate
digit-symbol pairs so that upon presentation of a geometric symbol the patient
would name the number associated with it. Alcoholic Korsakoff's patients failed
to learn this task even over repeated trials. Chronic alcoholics performed better
than the Korsakoff's but never reached the level of performance of control
subjects. Similar results were obtained when patients were asked to remember
four unassociated words for durations of 15 to 30 seconds. In all cases, the deficit
of chronic alcoholics was most pronounced when the subject was asked to do
another cognitive task (e.g. counting backwards) during the time period between
presentation of the item to be remembered and recall (proactive interference).


5
There also are marked differences in the characteristics of amnesia
between the alcoholic Korsakoff's patient and chronic alcoholic. For example, the
memory disorder associated with chronic alcoholism is confined primarily to
short term memory with no retrograde amnesia. The alcoholic Korsakoff's
patient suffers from both anterograde and retrograde memory deficits.
Confabulation is a characteristic sometimes associated with the anterograde
amnesia of Korsakoff's but not seen in chronic alcoholism. The different
characteristics of memory deficits in Korsakoff's and chronic alcoholic patients
may reflect different or additional anatomical substrates responsible for the
amnesias.
The morphological substrates responsible for the alcoholic memory
disorder have been studied in both humans and animal models. In humans, cell
loss has been described in several brain structures including cerebellum,
thalamus, frontal cortex, basal forebrain, and hippocampus (Arendt et al., 1983;
Bengochea and Gonzalo, 1990). In order to determine the potential contributions
of these regions to the behavioral changes during alcoholism, researchers have
compared behavioral data from human clinical and animal temporal lobe and
medial thalamic lesion studies to that from chronic alcoholics. Severe
anterograde amnesia has been described in patients with a lesion somewhat
localized to the dorsal medial region of the thalamus or to area CA1 of the
hippocampus (Squire and Moore, 1979; Zola-Morgan et al., 1986).
Animal models with lesions restricted to either the dorsal thalamus or
hippocampus have helped to shed light on the specific contribution of these
regions to normal memory function. A commonly used task for testing short
term memory is the delayed non-matching to sample. In this test, the animal is
allowed to displace an object covering a food well with a raisin in it. Next, an
opaque wall is placed to block the monkey's view of the food well. After 8


6
seconds, the wall is removed and the monkey sees two objects, the previously
viewed object and a novel object. The monkey is trained to choose the novel
object to obtain the raisin reward. Once the monkey has learned this task, the
delay between initial presentation of the object and presentation of the two
objects is lengthened. Animals with lesions to either the dorsal medial region of
the thalamus or CA1 of the hippocampus have difficulty learning the task
initially and once criterion is reached, in performing the task correctly as delays
between presentations are increased to 60 sec or more (Aggleton and Mishkin,
1983; Zola-Morgan et al., 1992). These results demonstrate the deficits in new
memory acquisition as well as recall of recent information following lesions to
dorsal thalamus or hippocampus. In dorsal medial thalamus, lesions which
encompassed both anterior and posterior thalamus resulted in amnesia far
greater than following a lesion to either region alone. The notion of an increasing
memory deficit with increasing extent of the lesion holds true for the temporal
lobe as well. For example, lesions which include hippocampus,
parahippocampal gyrus and perirhinal cortex produce a more severe
anterograde amnesia than occurs following lesions restricted to the hippocampus
(Zola-Morgan et al., 1989). This suggests that the parahippocampal gyrus and
perirhinal cortex provide a contribution to memory in addition to the
contribution made by the hippocampus. These combined results indicate that a
number of separate brain regions are responsible for memory acquisition and
that a lesion to any one of these structures may result in short term memory loss.
Inferred from these studies is that the severity of amnesia may depend on the
extent of pathology to the memory circuit.
As stated earlier, the pathology most prominently associated with
alcoholic Korsakoff's disease is degeneration of the dorsomedial thalamus, which
is commonly thought responsible for the enduring memory loss associated with


7
this disease. There is only one study, though, that found a correlation between
damage to the dorsomedial thalamus and memory loss in alcoholic Korsakoff's
(Victor et al., 1971). Other studies have failed to consistently find such a
correlation (Butters and Cermak. 1980). The possibility remains then, that other
structures within the memory circuit may contribute or be solely responsible for
the memory loss following chronic alcoholism.
Because of the confounding variables of compromised nutritional status,
duration and quantity of drinking, as well as type of drinking (i.e., binge vs.
constant drinking) investigation of direct effects as well as the nature of the
effects of ethanol in humans is difficult. Several animal models of controlled
alcohol and nutritional intake have been developed to study the effects of chronic
alcohol alone.
Behavioral Evidence of Memory Deficits in an Animal Model of Long Term
Alcohol Consumption
In 1970, Freund described a rodent model of chronic ethanol consumption
in which animals received a nutritionally complete liquid diet with part of the
caloric intake in the form of ethanol. This diet served as the sole source of food
and liquid for the duration of the treatment period. A variety of behavioral tests
of memory acquisition demonstrated a permanent loss of memory acquisition
following a three to seven month exposure to ethanol (Freund and Walker, 1971;
Walker and Freund, 1973; Walker and Hunter, 1978; File and Mabbutt, 1990). In
one study, animals were trained to perform a temporal alternation task in which
bar presses were reinforced on alternate trials (Walker and Hunter, 1978).
Following 20 weeks of an ethanol containing diet and 2 months of abstinence,
animals were retrained on the same behavioral task. After the training period,
short term memory was tested by varying the duration between bar press trials.


8
Animals that received chronic ethanol treatment (CET) tested as well as pair-fed
controls when the intertrial interval was 20 seconds or less. The performance of
CET animals significantly declined as the intertrial interval increased to 60
seconds. The behavioral deficit demonstrated in this rodent model of CET
closely parallels the deficit in short term memory of chronic alcoholics and
alcoholic Korsakoff's patients, although a major difference was that the animal
model of memory loss occurred in the presence of a nutritionally complete diet.
These results demonstrate the nature of behavioral effect of ethanol alone.
Chronic ethanol consumption produced a toxic action to the neural substrate
underlying memory formation.
Anatomical Evidence of Hippocampal Pathology Following Chronic Ethanol
Treatment
Several studies have demonstrated loss of 10 to 40% hippocampal
pyramidal and dentate granule cells following a 20-week or longer exposure to a
chronic ethanol diet (Cadete-Leite, et al., 1988; Walker et al., 1980; Lescaudron
and Verna, 1985). Lescaudron and Verna (1985) went on to show that pyramidal
cell loss is greater in the ventral than the dorsal region of the hippocampus. In
addition, chronic ethanol abuse results in changes in spine density and decreases
in dendritic length (Riley and Walker, 1978; Me Mullen, et al., 1984; Goldstein et
al., 1983; Lescaudron et al., 1986; King et al., 1988; Cadete-Leite et al., 1989). In
the dentate gyrus, cell loss due to chronic ethanol toxicity is accompanied by an
increase in the dendritic extent of the surviving cells (Cadete-Leite, et al., 1988).
Following a 20-week recovery period, spine density returns to normal (King et
al., 1988).
Physiological studies have also been employed to examine dendritic
alterations following chronic ethanol consumption. Abraham et al. (1982)


9
described an alteration in the distribution of afferent synaptic connections in area
CA1. Specifically, current source density analysis of synaptic fields
demonstrated a 13.4% reduction in the Schaffer collateral and commissural
(Sch/Com) fiber synaptic field. The anatomical data taken together suggest that
concomitant processes of cell degeneration, adaptation, and a subsequent process
of recovery all contribute to the resulting behavioral and anatomical
manifestation of chronic ethanol toxicity.
In addition to the dramatic changes in principal cells of the hippocampus,
Lescaudron et al. (1986) described y-amino butyric acid (GABA)ergic cell loss in
the hippocampus. Using immunocytochemical analysis with an antibody to
GABA, they found decreased labeling intensity of immunopositive neurons and
fibers in both dorsal and ventral hippocampus, and decreased number of
immunopositive neurons in ventral hippocampus. These results may be
interpreted as a chronic ethanol-produced interneuronal cell loss in hippocampus
and/or a reduction in GABA content or antibody affinity in these neurons.
Mice specifically bred for sensitivity to the acute effects of ethanol were
used to examine the effects of a chronic ethanol diet on GABAergic cell number
in the dentate gyrus (Scheetz et al., 1987). Prior to ethanol exposure, there was no
difference in the number of dentate basket cells between ethanol sensitive and
resistant strains. Following a 3-month chronic ethanol diet, a significant
reduction in dentate granule layer basket cell number was observed only in
ethanol sensitive animals. These results again illustrate the specific neurotoxic
effects of ethanol on GABAergic interneurons.
Physiological Effects of Chronic Ethanol Toxicity in the Hippocampus
In the hippocampus, functional inhibition can be measured using a paired
pulse paradigm. When the first (conditioning) pulse is antidromic and the


10
second (test) pulse is orthodromic, the second response is reduced relative to the
first even with interpulse intervals of a few hundred milliseconds. The
conditioning pulse of this stimulus paradigm is thought to activate the recurrent
inhibitory pathway onto pyramidal cells. Comparison of the reduction of the test
pulse between diet treatment groups reflects the relative amount of functional
inhibition. Test pulse inhibition following chronic ethanol treatment is less
compared to pair-fed controls (Abraham et al., 1981; Durand and Carien, 1984a;
Rogers and Hunter, 1992). This suggests that recurrent inhibitory circuitry is
reduced after chronic ethanol treatment. Physiological alterations in the
hippocampus following chronic ethanol toxicity parallel the morphological
findings of decreased GABAergic interneurons.
Durand and Carien (1984a) demonstrated that following chronic ethanol
treatment the amplitude and duration of the K+ mediated afterhyperpolarization
and the orthodromically stimulated (GABA mediated) inhibitory postsynaptic
potential (IPSP) in CA1 pyramidal cells and dentate granule cells were reduced,
while other passive and active membrane characteristics of granule cells (input
resistance, resting membrane potential, excitatory postsynaptic potential (EPSP)
amplitude, etc.) remained unchanged. Rogers (1986) went on to test the efficacy
of GABA at the postsynaptic GABA receptor in response to iontophoresis of
GABA. Single test pulses delivered following iontophoresis of GABA were
reduced to a greater extent in chronic ethanol treated animals as compared to
pair-fed controls. He next tested single pulse responses in the presence of a
GABAa receptor antagonist, bicuculline methiodide. In this condition, responses
were larger in ethanol treated animals as compared to pair-fed controls. These
results suggested that chronic ethanol treatment resulted in an augmentation of
the GABAa receptor responses to its endogenous ligand, GABA. The efficacy of
bicuculline to reduce GABAergic inhibition was also enhanced. Taken together,


11
the above results suggested that CET produced a decrease in presynaptic release
along with either 1) an increase in postsynaptic GABAa receptor number along
with a decrease in presynaptic release or 2) an increase in efficacy of GABA at the
postsynaptic GABA receptor. The decrease in presynaptic release is most likely
due to a decrease in GABAergic innervation of principal cells.
In addition to the chronic ethanol associated changes in basal or single
pulse synaptic function, a few studies have examined ethanol actions on synaptic
function related to long term plasticity. In vitro measurements from area CA1
indicated that the number of hippocampal slices capable of expressing long term
potentiation (LTP) was decreased following chronic ethanol treatment but if LTP
was induced, the magnitude of potentiation was not different from controls
(Durand and Carien, 1984b). These experiments only measured the compound
action potential (PS) responses of pyramidal cells and were therefore more a
measure of LTP induced changes in cell excitability (i.e., threshold to activation
of an action potential and/or firing frequency) than LTP (which is synaptic in
origin). In vivo, there was no change in the early time course or magnitude of
LTP in area CA1. However, there was a trend toward a greater decay of the early
phase of LTP maintenance. In the dentate gyrus, LTP was unchanged (Abraham
et al., 1981; 1984). It should be noted that these studies recorded responses to
long duration conditioning trains which produced a maximal or asymptotic LTP
with a single train. Perhaps ethanol does not affect the maximal amount of
potentiation produced by a group of synapses, but rather the amount of
depolarization necessary to produce a given amount of potentiation. In other
words, chronic ethanol may affect the facility or ease of induction of LTP in
response to a given conditioning train. To test this, responses to short duration
conditioning trains which produce submaximal LTP must be measured.


12
The Effect of Duration of Exposure and Length of Abstinence on the Substrates of
Ethanol Toxicity
Throughout the alcohol literature it is widely accepted that differing
mechanisms are responsible for the pathology at each stage of neurotoxicity. The
particular stage of ethanol neurotoxicity reflects the duration of exposure or
withdrawal from ethanol. For example, acute exposure of hippocampal neurons
to low doses of ethanol results in a rapid block of the N-methyl-D-aspartate
(NMDA) receptor mediated EPSP and LTP induction (Lovinger et al., 1990;
Blitzer et al., 1990). Receptor binding studies revealed a coincident decrease in
the number of [3H]MK-801 (diclozipene, a specific antagonist to the NMDA
receptor channel) binding sites (DeMontis et al., 1991). Following a one or two
week exposure to a liquid diet containing ethanol, the number of [3H]MK-801
binding sites is upregulated as compared to controls (Gulya et al., 1991).
Behavioral evidence of heightened seizure susceptibility and reduced
responsiveness to the effects of acute ethanol (tolerance) are consistent with
increased NMDA receptor function (Grant et al., 1990). As the duration of
exposure lengthens to three to five months, hippocampal pyramidal cells and
interneurons undergo cell death (Walker et al., 1981; Lescaudron and Verna,
1985). The dendritic morphology of surviving cells is altered (King et al., 1988).
Following an abstinence period, dendritic morphology returns to normal. The
overall response of the animal or person to chronic ethanol, then, is likely to
reflect the entire ethanol history.
A Physiological Correlate of Memory Formation
In 1973, Bliss and Lomo demonstrated that following a high frequency
conditioning train to the perforant path fibers which synapse onto dentate
granule cell in the dentate gyrus, a long-lasting (up to 3 weeks) increase in


13
synaptic efficacy to subsequent single pulse stimuli occurred. The enhanced
synaptic efficacy was manifested as an increased amplitude and decreased
latency to onset of the PS of granule cells. The enhanced synaptic efficacy was
later termed long term potentiation and was described in a variety of central
nervous system (CNS) structures. Since its initial description, LTP has been
considered to be a physiological correlate of learning and memory because of its
enduring nature, the types of physiological stimuli which produce it (theta
pattern stimulation) and the anatomical substrates in which it is found (Teyler
and DiScenna, 1984). LTP is defined as an enduring or permanent enhancement
in the functional, biochemical and/or morphological elements of synaptic
transmission, which results in a long lasting enhancement of the efficacy of
synaptic transmission. The physiological consequence of LTP is an increase in
amplitude and slope of the extracellular recorded EPSP and a decrease in latency
to onset of the EPSP. The changes in EPSP reflect an increase in the amplitude
and/or decrease in threshold for activation of the intracellular recorded EPSP
(Schwartzkroin and Wester, 1975). LTP also results in an increase in the
amplitude of the extracellular PS, due to a decrease in threshold and increase in
firing frequency of pyramidal cells for a given stimulus (Chavez-Noriega et al.,
1990).
The mechanisms underlying LTP have been most thoroughly examined in
the hippocampus. Before discussing the results from these studies, it is
important to briefly describe the anatomical connections within this region (Fig.
2-1). The transverse plane of the hippocampus is generally arranged as a
trisynaptic circuit (Amaral and Witter, 1989). The input fibers from ipsilateral
and contralateral entorhinal cortex form the medial and lateral perforant path.
These fibers synapse onto the dendrites of the granule cells of the dentate gyrus.
The granule cells send projections (mossy fibers) to synapse on the CA3


14
pyramidal cells. CA3 pyramidal cells send projections ipsi- and contralaterally to
form the Sch/Com fibers in CA1. These fibers form synapses onto the CA1
pyramidal cells as well as inhibitory interneurons of CAI. CA1 pyramidal cells
also synapse on inhibitory interneurons (basket cells) which then synapse onto
CA1 pyramidal cells. The synaptic interconnections between CA1 pyramidal
cells and inhibitory interneurons mediate feedforward and feedback inhibition of
pyramidal cell function.
In response to single pulse stimulation of the Sch/Com fibers,
presynaptically released glutamate initially stimulates postsynaptic a-amino-3-
hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors. Subsequent flux
of Na+ and K+ through the AMPA channel results in a short onset and brief
duration postsynaptic depolarization (Hestrin et al., 1990). Postsynaptic NMDA
receptors are only minimally activated by glutamate under this condition due to
a voltage dependent channel block by Mg2+. In addition, single pulse
stimulation results in release of GABA from inhibitory interneurons, which
activates postsynaptic GABAa and GABAg receptors. Activation of GABAa
receptors results in an inward CP current with a short latency (3 ms or less) and
brief duration (225 ms) at 32 C (Davies, et al., 1990). Activation of the
postsynaptic GABAg receptor results in G-protein activation and subsequent
outward K+ conductance of longer latency (30 ms) and duration (700 800 ms) at
32 C.
Delivery of a high frequency train of pulses to the Sch/Com fibers results
in a greater postsynaptic depolarization due to repeated AMPA receptor
activation. The enhanced postsynaptic depolarization relieves the Mg2+ block of
the NMDA receptor, resulting in an additional inward Ca2+ and Na+
conductance through the glutamate activated NMDA receptor channel
(Dingledine, 1983; Harris et al., 1984; Mayer and Westbrook, 1984; Mayer et al.,


15
CAI
CA3
Figure 2-1. Schematic representation of the transverse plane of the hippocampus
to illustrate the connections of the trisynaptic circuit (Abbreviation are as follows:
CA1, corpus ammonis area 1 of hippocampus; EC, entorhinal cortex; DG, dentate
gyrus; MF, mossy fibers; PP, perforant path; SC, Schaffer collateral and
Commissural fibers).
1984; Nowak et al., 1984; Hestrin, et al., 1990; Regehr and Tank, 1990).
Postsynaptic GABAa and GABAb activation also occurs. Synaptically released
GABA can act back upon presynaptic GABAb receptors located on the terminals
of GABAergic interneurons, resulting in a G-protein mediated decrease in Ca2+
conductance and a subsequent decrease in GABA release (Davies et al., 1990;
1991). The GABAb mediated events occur with a latency of 200 400 ms. The
NMDA and GABAb mediated synaptic transmission play only a minimal role in
responses to single pulse stimulation. In the presence of plasticity inducing high
frequency conditioning trains, both the NMDA and GABAb receptor systems
produce a profound effect on the resulting pre- and postsynaptic response.
LTP can be conceived as a sequence of events which can be separated for
experimental purposes into 1) an induction process and 2) a process to maintain
synaptic enhancement. Induction of LTP requires cooperativity of pre- and


16
postsynaptic elements such that there is synchronous or near synchronous
presynaptic activation and postsynaptic depolarization (Gustafsson et al., 1987).
Current flows through the NMDA receptor channel only when synaptically
released glutamate binds to the postsynaptic receptor in combination with
postsynaptic depolarization. This allows relief of the voltage dependent Mg2+
block of the NMDA receptor and Ca2+ influx. Ca2+ entry can activate
postsynaptic protein kinases such as Ca2+/Calmodulin II, protein kinase C,
protein kinase A, and tyrosine kinases. These kinases are thought to provide the
substrates for maintenance of synaptic enhancement (Finn et al., 1980; Mody et
al., 1984; Hu et al., 1987; Lovinger and Routtenberg et al., 1988; Malinow et al.,
1988; Reyman et al., 1988; Malenka et al., 1989).
A consensus view of the nature of synaptic changes responsible for
maintenance of LTP has not yet been reached. Kauer et al. (1988) report
increased AMPA mediated postsynaptic current immediately following LTP
induction. This result was based on the evidence that following an LTP
conditioning train, the AMPA and not the NMDA response (in the presence of
antagonists to the AMPA channel and low Mg2+) to single pulse stimulation was
increased. Within 30 seconds, though, the NMDA response as well as the AMPA
response was increased. The increase in NMDA current returned to baseline
levels within four minutes after cessation of the LTP conditioning stimulus. On
the presynaptic side, several researchers have demonstrated an immediate and
persistent increase in glutamate release in response to LTP induction (Bekkers
and Stevens, 1990; Malinow and Tsein, 1990; Malinow, 1990). Using similar
techniques of quantal analysis, Kullmann and Nicoll (1992) and Manabe et al.
(1992) provided evidence supporting both pre- and postsynaptic changes. Using
in vivo LTP induction and in vitro (HPLC) measurement of neurotransmitter
release, Ghijsen et al. (1992) report that endogenous release of both glutamate


17
and GABA increase following LTP induction. Based on these combined results,
it is currently thought that both pre- and postsynaptic mechanisms contribute to
LTP maintenance.
The amount of potentiation of the intracellularly recorded EPSP is related
to the amount of postsynaptic depolarization during induction when the
postsynaptic depolarization is paired with presynaptic activation (Gustafsson et
al., 1987). With extracellular stimulation, the postsynaptic depolarization is
produced by a high frequency afferent stimulation. It appears that a minimum
number of afferent fibers or synaptic contacts must be activated to generate LTP.
Short duration conditioning trains will produce less postsynaptic depolarization
and therefore smaller amounts of LTP than will longer duration trains.
The expression of LTP is a result of combined potentiation of excitatory
and inhibitory synaptic transmission (Abraham et al., 1987; Morishita and Sastry,
1991). The relative contribution of excitatory synaptic transmission typically
predominates over the inhibitory contribution so that the resulting PSP reflects a
relative decrease in overall inhibitory influence.
Objectives of the Dissertation
Since LTP is a candidate substrate for memory acquisition, and the
hippocampus serves a vital role in memory formation, it is reasonable to
hypothesize that the hippocampal vulnerability to the effects of chronic ethanol
toxicity may contribute to the amnesic symptoms of alcoholism. A necessary
requisite for the behavioral disorder, though, is a parallel physiological disorder.
It is therefore critical to determine the plasticity related physiological effects of
chronic ethanol toxicity in the hippocampus.
The hypothesis tested by the experiments presented in this dissertation is
that chronic ethanol treatment is associated with transient and/or permanent


18
alterations in one or more components of the induction process of long term
potentiation in the hippocampus. To test this hypothesis, I investigated 1) the
effect of CET on induction of LTP in the presence and absence of GABAa
synaptic transmission; 2) the effect of recovery from CET on induction of low
levels and maximal magnitude LTP in the presence and absence of GABAa
mediated synaptic transmission; 3) changes in the binding characteristics of
[3H]MK-801 to the NMD A receptor channel following CET; 4) changes in the
efficacy of glutamate to enhance NMDA receptor channel function; 5) the effect
of CET on the efficacy of the endogenous ligand GABA to stimulate GABAa
receptor function and the efficacy of a GABAa receptor antagonist, bicuculline, to
inhibit GABA-stimulated GABAa receptor function.


CHAPTER 3
GENERAL METHODS
Male Long Evans hooded rats (200-250 g) were purchased from Charles
River. A two-week holding period was allowed following arrival to allow for
adjustment to the animal care facility. All animals were housed individually in
stainless steel cages in a colony room with an automatic 7 AM to 7 PM light cycle.
Animals within a particular shipment were paired by weight and assigned
to one of two liquid diet treatment groups: 1) An ethanol group (group E) in
which ethanol comprised 35-39% of the total caloric intake (8.1 9.4% v/v,
ethanol); and 2) A sucrose group (group S) which received an identical diet
except sucrose was isocalorically substituted for ethanol. The remaining calories
were supplied through Sustacal (Mead-Johnson Co.). Both diets were fortified
with Vitamin Diet Fortification Mixture, 0.3 g/100 ml and Salt Mixture XIV, 5.0
g/1 (ICN Nutritional Biochemicals, Cleveland). Both diets contained 1.3 kcal/ml
and provide several times the daily requirement of all essential vitamins and
nutrients (Walker and Freund, 1971; Walker, et al., 1980).
The amount of ethanol diet consumed per Group E animal each day was
measured and recorded. An equal volume of sucrose diet was given on the
following day to the pair-fed Group S animal. Pair-feeding group S and group E
animals ensured that each of a sucrose-ethanol pair received the same volume of
diet and thus the same caloric intake during the treatment period. The liquid
diets were prepared daily and animals received their daily allowance of diet
between 8 and 9 AM. The liquid diets were administered as the sole source of
food for a period of 28 weeks. The percentage of calories in the form of ethanol
19


20
or sucrose was increased by 1% every 4 weeks. A total of 34 sucrose-ethanol
pairs were used in this series of studies. The average daily intake of ethanol
using this paradigm was 12.30 0.17 g/kg/day. All rats were weighed weekly.
Rats maintained on these diets gain weight normally. At the end of the 28 week
period all animals received laboratory chow and water ad libitum for a period of
48 hours or 5-7 months, at which time the acute experiment was performed.


CHAPTER 4
THE EFFECTS OF CHRONIC ETHANOL TREATMENT ON LONG TERM
POTENTIATION IN THE HIPPOCAMPUS
Introduction
Chronic abuse of alcohol results in pathological deterioration of
anatomical, physiological and behavioral correlates of cognitive performance.
The neurotoxic damage extends into several brain structures including
thalamus, cerebellum, cerebral cortex and hippocampus. Of particular
consequence is the pathology of anterograde amnesia, a profound and
devastating symptom of chronic abuse (Freund, 1970; Freund and Walker,
1971; Walker and Freund, 1973; Walker and Hunter, 1978; Berocochea et al.,
1989). In order to better understand the underlying pathology responsible for
this anterograde amnesia, the permanent effects of chronic ethanol toxicity
(CET) on a physiological correlate of memory, long term potentiation (LTP)
was investigated. LTP is defined as an enduring or permanent enhancement
in the functional, biochemical and/or morphological elements of synaptic
transmission (Bliss and Lomo, 1973; Kennedy, 1989). Induction of LTP
requires influx of Ca2+ through the N-methyl-D-aspartate (NMDA) receptor
ionophore located in the postsynaptic membrane (Lynch et al., 1983; Mayer
and Westbrook, 1984; Mayer et al., 1984; MacDermott et al., 1986; Malinow and
Miller, 1986; Kauer et al., 1988; Malenka et al., 1988). During single pulse
excitatory synaptic transmission, activation of the hippocampal afferents fails
to open the NMDA receptor ionophore because of blockade by Mg2+ in a
voltage dependent manner. However, with high frequency stimulation of
21


22
the afferent fibers, the postsynaptic membrane is sufficiently depolarized to
remove the Mg2+ block. The subsequent influx of Ca2+ through the NMDA
receptor ionophore is thought to trigger a sequence of events leading to a
sustained enhancement of functional synaptic efficacy (Dingledine, 1983;
Harris et al., 1984; Kauer et al., 1988; Malenka et al., 1988; Mayer and
Westbrook, 1984; Mayer et al., 1984; Neuman et al., 1987; Nowak et al., 1984;
Regehr and Tank, 1990; Reyman et al., 1989; Thibault et al., 1989).
Recent research suggests that acute ethanol intoxication reduces
current flow through the NMDA channel (Lovinger et al., 1989; 1990), reduces
LTP induction (Sinclair and Lo, 1986; Blitzer et al., 1990) and reduces NMDA
receptor binding affinity (DeMontis et al., 1991). During ethanol withdrawal,
NMDA receptor binding increases (Grant et al., 1990). However, NMDA
receptor binding or function have not been examined in a period of
abstinence following chronic ethanol treatment and withdrawal
A few studies have examined the effects of chronic ethanol exposure
on LTP, as demonstrated by changes in neuronal excitability, but none have
studied synaptic plasticity. Using the hippocampal slice preparation, Durand
and Carien, (1984b) determined that in area CA1 the number of slices capable
of expressing LTP was decreased following chronic ethanol treatment but if
LTP was induced, the magnitude of potentiation was not different from
controls. These experiments only measured the compound action potential
(PS) responses of pyramidal cells and were therefore more a measure of LTP
induced changes in cell excitability (i.e., threshold to activation of an action
potential and/or firing frequency) than LTP (which is synaptic in origin). In
vivo, there was no change in the early time course or magnitude of LTP, but a
statistically nonsignificant trend toward early decay in area CA1. In the
dentate gyrus, LTP was unchanged (Abraham et al., 1981; 1984). Again, the


23
recordings in this study were taken from the cell body layer (PS) and therefore
not a direct examination of the synaptic component of LTP.
An alternate series of studies have suggested that chronic ethanol
ingestion may exert much of its effect by reducing the influence of synaptic
inhibition in the hippocampus. For example, 20 weeks of CET followed by 8
weeks of recovery results in decreased feedforward and recurrent inhibition
onto pyramidal cells of hippocampal area CA1 (Abraham et al., 1981; Rogers
and Hunter, 1992) Additionally, CET reduces the amplitude of the K+
mediated afterhyperpolarization and the GABA mediated inhibitory
postsynaptic potential (IPSP) in CA1 pyramidal cells (Durand and Carien,
1984a).
While NMDA receptor activation and subsequent Ca2+ entry are
necessary and sufficient for LTP induction to occur, GABAergic synaptic
transmission may modulate the magnitude of LTP (Gustafsson, et al., 1987;
Abraham et al., 1987; Morishita and Sastry, 1991). For example, in CA1
Abraham et al. (1987) described an enhancement of both excitatory and
inhibitory synaptic transmission following LTP induction, although
enhancement of excitatory synaptic transmission was greater than that of
inhibitory synaptic transmission. Morishita and Sastry (1991) recorded
intracellular events following LTP induction and confirmed these results.
Both the glutamate mediated EPSP and the GABAa mediated IPSP were
enhanced following LTP induction. Gustafsson and coworkers (1987) used
pairings of single pulse afferent stimulation with postsynaptic depolarization
to demonstrate that the amount of depolarization required to induce LTP was
reduced if GABAa receptors were blocked with the CL channel antagonist,
picrotoxin. Taken together, these results suggest that GABAergic synaptic
transmission can be involved in both the amount of glutamate stimulated


24
depolarization necessary to induce LTP and the relative contribution of
excitatory and inhibitory influences to the resultant synaptic potentiation.
Since NMDA receptor activation is required for LTP induction and
GABAergic synaptic transmission modulates both the magnitude of LTP
induction and the characteristics of the LTP produced, CET induced alteration
of either of these receptor systems could result in altered LTP.
The primary aim of this study was to examine the effect of CET on LTP
induction in area CA1 of the hippocampus and the potential modulatory
influence of GABAergic synaptic transmission on this process.
Methods
Treatment Methods
Male Long Evans hooded rats (200-250 g) were matched by weight and
age and assigned to one of two liquid diet treatment groups described in detail
in Chapter 3, General Methods. Briefly, the diet treatment consisted oftwo
groups: 1) An ethanol group (group E) in which ethanol comprised 35-39% of
the total caloric intake. The remaining calories were supplied by Sustacal
(Mead-Johnson Co.) 2) A sucrose group (group S) which received an identical
diet except sucrose was isocalorically substituted for ethanol. Group S
animals were individually pair-fed with group E animals such that each of a
sucrose-ethanol pair received the same volume of diet during the treatment
period. The liquid diets were administered as the sole source of food for a
period of 28 weeks. At the end of the 28 week period all animals received
laboratory chow and water for 48 hours. Food and water intake were
measured daily during the 48 hour period. A group of similar age animals
housed under similar conditions as group S and E animals but receiving
laboratory chow and water during the entire treatment period were included


25
in the study (group C). The preceding protocol of chronic ethanol treatment
(CET) has been previously used as a valid model for studying the behavioral,
morphological, and physiological consequences of chronic alcohol abuse
(Walker and Freund, 1971; 1973; Walker and Hunter, 1978; Walker et al., 1981;
1982).
Electrophysiological Methods
Slice preparation
Electrophysiological records were taken from multiple hippocampal
slices in each of 9 group S, 9 group E and 8 group C animals 50 60 hours
following cessation of the 28 week liquid diet treatment. All animals were
coded throughout data collection and analysis to avoid experimenter bias.
Transverse sections (400 pm) were cut through the ventral hippocampus
using a Mclllwain tissue chopper. Slices were incubated in a holding chamber
containing artificial cerebral-spinal fluid (aCSF) NaCl, 125 mM; KC1, 3.3 mM;
KH2PO4,1.25 mM; MgSCT*, 4.0 mM; CaCl2. 4.0 mM; NaHCC>3, 25 mM; glucose,
10 mM at room temperature. The pH was maintained at 7.4 with 95% O2/ 5%
CO2. Forty to sixty minutes prior to the onset of recording, a slice was
transferred to a Haas type interface recording chamber maintained at 32 C.
The slice was superfused at 1 ml/min with oxygenated medium of the same
composition as described for the holding chamber.
LTP induction protocol
At the beginning of each recording session, a stimulating electrode was
placed in stratum radiatum (SR) of CA1 midway between stratum pyramidal
and stratum lacunosum-moleculare. A glass recording micropipette filled
with 4 M NaCl was placed in SR of CA1, 800 1000 pm from the stimulating
electrode (Fig. 4-1A). Recordings of field EPSPs in SR were obtained in


26
B
Figure 4-1. Microelectrode placement and representative waveforms from
the hippocampal slice preparation.
(A) Schematic of a transverse hippocampal slice to illustrate electrode
placement. A bipolar stimulating electrode (S) was placed in stratum
radiatum (SR) of area CA1. An extracellular recording microelectrode (R) is
positioned in SR of area CA1 approximately 800 pm from the stimulating
electrode.
(B) Representative field EPSP waveforms recorded from CA1 SR in response
to 0.04 ms duration single pulse stimulation (a) before, and (b) 28 min
following a 100 Hz, 50 pulse conditioning train delivered to the Sch/Com
afferent fibers, (c) Field EPSP slope was measured between the time points at
10% and 90% of maximal EPSP amplitude. Calibration pulse is 2 ms duration,
2 mV amplitude.


27
response to 0.05 Hz test pulses of Schaffer Collateral and Commissural
(Sch/Com) fibers in SR (Fig. 4-1B). Although the conditioning trains which
induce LTP potentiate the field EPSP over a wide range of stimulus strengths,
we empirically determined that the maximal percent change in slope occurs
at stimulus strengths which produce a preconditioning field EPSP slope 75 to
100% of maximal. For this reason the single pulse stimulus strength was
adjusted to produce a field EPSP slope 75% of maximal.
Following an initial 12 min recording period, the stimulus strength
was adjusted to produce a maximal EPSP and a conditioning train of 50
pulses at 100 Hz was delivered. After this conditioning train, the stimulus
strength was returned to the previous level and recordings in response to
single pulse stimulation were made at 20 sec intervals for 28 min. A
subsequent conditioning train and test pulse recordings were taken repeating
this protocol except the duration of the train was lengthened to 100 pulses.
Just prior to each conditioning train and again at the end of the recording
session several test pulses covering a range of stimulus strengths were
delivered and the field EPSP slope recorded to demonstrate the magnitude of
potentiation over a range of stimulus strengths (input/output data)
LTP induction protocol in the absence of GABAergicA synaptic transmission
Experiments involving the removal of inhibitory synaptic
transmission were performed with 3.5 (iM bicuculline methiodide, a specific
GABAa receptor antagonist. This concentration of bicuculline was found to
produce maximal facilitation of LTP induction without producing bursting
and post-burst depression in response to a conditioning train. Bicuculline
was added to the superfusate 6 min after the onset of recording and
maintained until the end of the recording session. Under this condition,
EPSPs produced in response to single pulse stimulation reflect the activation


28
of primarily excitatory inputs. The treatment can result in longer burst
duration to single pulse stimulation. CA3 pyramidal cells normally fire in
bursts which compound the bursting activity in the postsynaptic CA1
pyramidal cell (Wong and Prince, 1979). To prevent uncontrollable
pyramidal cell bursting and resulting postburst depression, an acute
transection under microscopic control was made in each slice between area
CA1 and CA3. This disconnected the CA3 pyramidal cell bodies from the
Sch/Com fibers. In the CA1 region of the hippocampus, stimulus frequencies
of 0.1 Hz or less are normally ineffective in altering the amplitude, slope or
duration of the field EPSP (Alger and Teyler, 1976). In the presence of
bicuculline, single pulse stimulation at 10 sec intervals resulted in a small
increase in field EPSP slope and/or amplitude in rare instances. Increasing
the interpulse interval to 20 sec, when recording from slices treated with
bicuculline, prevented potentiation of the EPSP in response to single pulse
stimulation and resulted in a stable baseline recording. If EPSP slope did
change by more than 10%, the slice was not used in the LTP experiments.
The procedure for collecting single pulse data and induction of LTP was
the same as described for the previous LTP induction experiments except that
the conditioning train frequency was reduced to 50 Hz. In the presence of
bicuculline, a 100 Hz conditioning train can produce burst firing of CA1
pyramidal cells and post burst depression rather than LTP. By lowering the
conditioning train frequency to 50 Hz, LTP induction is reliably produced
with no evidence of prolonged postsynaptic cell firing or post burst
depression (Gustafsson and Wigstrom, 1987).
Blockade of LTP with D-amino-5-phosphonovalerate
A separate group of animals (S and E) were studied to determine
whether the LTP induction mechanism, either in the presence or absence of


29
bicuculline methiodide, is mediated via activation of the NMDA receptor
ionophore. To do this, an LTP inducing conditioning train was delivered in
the presence of 50 pM D-amino-5-phosphonovalerate (APV) to block the
NMDA receptor ionophore. Any potentiation requiring activation of the
NMDA receptor would be blocked by this treatment. A washout period of 40
to 50 min was allowed to remove the APV from the slice. Another
conditioning train was then delivered as described in the initial protocol.
Data analysis
Field EPSP data were passed through a Grass P-511 preamplifier,
digitized at 20 kHz and stored on an IBM AT computer for later analysis. Data
collection and analysis were performed using software developed in this
laboratory. Measurements were made of field EPSP slope (from 10% to 90% of
peak EPSP amplitude; Fig. 4-1B).
EPSP slope data were averaged over the last 10 min of the recording
period following a conditioning train. These data were grouped by duration
of pulse train and analyzed by treatment. Postconditioning train field EPSP
slope (Lt) was expressed as percent of preconditioning EPSP slope (Lo):
Lt
EPSP slope (% baseline) = lq x
and graphed as a function of time with the onset of the first conditioning
train being t=0.
Input/output data were expressed as EPSP slope (mV/ms).
Postconditioning train EPSP slope data were analyzed by a two-way analysis of
variance (ANOVA) with repeated measures with diet treatment as one factor
and duration of pulse train as the other factor.


Results
30
The purpose of these experiments was twofold. The first aim was to
determine the effects of CET on the magnitude of LTP produced in response
to conditioning trains of varying durations. Secondly, I examined the relative
effects of CET on NMDA versus GABAergic synaptic mechanisms.
CET Actions on LTP Induction
Recordings were taken from multiple slices in 9 group S, 9 group E and
8 group C animals 50 60 hours following the cessation of the liquid diet
treatment. Previous experiments in our laboratory have dealt with the effects
of CET following extended periods of recovery from the liquid diet treatment.
The experiments described in this chapter are among the first to test such
functional CET effects after such a short abstinence period. The experimental
paradigm used here allows one to determine substrates of chronic ethanol
toxicity without acute withdrawal effects and with minimal recovery. While
this provides valuable information about the substrates and timecourse of
ethanol effects, it was imperative to first ensure that despite a change in diet
from liquid to solids and water, animals would still consume food. Food and
water intake were measured every 24 hours for the two day abstinence period
and comparisons made between groups E and S (Table 4-1). There were no
differences in the quantity of intake between the two groups. A group of 8
chow-fed animals was included in these experiments to examine any
differences in responses which may be accounted for by the Sustacal diet
treatment alone.
Stimulating and recording electrode placement is illustrated in Figure
4-1A. Pyramidal cells from CA3 send projections along the transverse plane
of the hippocampus, into CA1 to form the Sch/Com fibers (Amaral and
Witter, 1989). These fibers form excitatory synapses onto the CA1 pyramidal


31
Table 4-1. Average daily chow and total water intake during the chronic diet
abstinence period.
Chow intake (g) + SEM
H20 intake (ml)/48 hr
SEM
Day 1
Day 2
Sucrose3
13.4 2.0
10.5 .75
66 12
Ethanol3
14.1 1.4
12.5 1.6
73 11
aANOVA with repeated measures of sucrose vs. ethanol was not significantly
different (P < 0.42).
cells as well as onto inhibitory interneurons of CA1. A bipolar stimulating
electrode was placed in SR of area CA1. Single test pulses of 0.04 ms duration
were delivered at 20 sec intervals to the Sch/Com fibers. The postsynaptic
dendritic response to single pulse stimulation of the Sch/Com fibers was
recorded in SR of CA1 (Fig. 4-1B). This extracellular EPSP represents the near
synchronous activation of multiple synapses. When measuring the
extracellular recorded EPSP, slope or amplitude measurements are most
commonly taken. If the stimulus pulse is suprathreshold for activation of
action potentials, the peak of the EPSP can be masked by the extracellular
recorded population spike (compound action potential from near
synchronous firing of multiple pyramidal cells). The slope of the EPSP,
though, is less vulnerable to interference from activation of a population
spike. For this reason measurements were taken of EPSP slope in the present
experiments. Average baseline (pretetanus) EPSP slope values + SEM were
0.66 + 0.06, 0.82 + 0.09 and 0.83 + 0.14 for groups C, S, and E respectively.
Following an LTP inducing conditioning train, the EPSP response to single
pulse stimulation increased in amplitude and slope (Fig 4-1B).


32
Figure 4-2. EPSP slope before and after two successive conditioning trains.
(A) Representative data expressed as percent of baseline. Baseline EPSP slope
recordings were obtained in response to 0.05 Hz single test pulses of Sch/Com
fibers in SR (t= -13 0 min). At t= 0, a 50 pulse, 100 Hz conditioning train was
delivered. Single test pulses were delivered and recordings taken at 0.05 Hz
for 28 min. A subsequent conditioning train of 100 pulses, 100 Hz was
delivered at t= 28 min. Single test pulse recordings were again taken for 28
min. Each conditioning train was followed by increased EPSP slope which
decayed to a stable potentiated level after several minutes (comparison of the
magnitude of potentiation within animals, ANOVA, P < 0.001).
(B) EPSP slope plotted as a function of stimulus strength. Baseline EPSP slope
measurements were taken prior to any conditioning train and 28 minutes
following the 50 and 100 pulse trains. EPSP slope increased as a function of
duration of the conditioning train over a range of stimulus strengths.


33
Initial extracellular field responses were recorded for 12 minutes.
Following this period of stable single pulse recordings, a conditioning train of
50 pulses at 100 Hz was delivered. In 9/9 group E, 9/9 group S and 7/8 group
C animals, the response consisted of an abrupt increase in EPSP slope
followed by a decay of several minutes to a stable potentiated level (Fig. 4-2A).
The subsequent 100 pulse, 100 Hz train further enhanced the potentiation of
EPSP slope in 7/9 group C, 7/9 group S and 6/8 group E animals. In instances
where the conditioning train did not produce LTP, post-train EPSP responses
consisted of an abrupt change in EPSP slope followed by a decay over several
minutes to near preconditioning levels. Responses to single pulses of
varying stimulus strengths taken 28 min following each conditioning train
demonstrate the EPSP potentiation over a range of stimulus strengths (Fig. 4-
2B). There were no apparent differences between groups C, S, and E with
respect to the early phase of decay of LTP or in the ability to maintain the
potentiation for the duration of the recording period.
All data were evaluated for inclusion in group statistical analysis based
on the following criteria: 1) Preconditioning single pulse responses must
vary with a SD< 10%. 2) Single pulse responses for the final 10 minutes of the
28 minute recording period following a conditioning train must vary with a
SD< 10%. For each animal and experimental condition, the first slice to meet
the criteria was included in the analysis.
Although the pattern of potentiation and the frequency with which
potentiation occurred among slices was similar in group S, E and C animals,
the magnitude of the potentiation in response to all conditioning trains was
reduced in group E as compared with group S (Fig. 4-3; P< 0.08). The average
magnitude of LTP for group C animals was similar to group S and larger than
group E (Table 4-2). The standard error for group C, though, was large enough


150-

O
Cf)
P+
£
w
100-
I
50
# PULSES
SUCROSE n= 9
ETHANOL n= 8
I
100
Figure 4-3. Mean EPSP slope expressed as percent of baseline following each of
the conditioning trains for group S and E animals. The magnitude of LTP of
group E animals was reduced following both the 50 and 100 pulse
conditioning trains. Data were analyzed by a two-way ANOVA with the
pulse trains as a within animal repeated measure. The treatment effect
approached statistical significance at P< .08.
Table 4-2. Mean EPSP slope SE expressed as percent of baseline following
each of the conditioning trains for groups C, S and E animals.
Number of pulses in
conditioning train
Chronic Treatment
Fifty
One Flundred
Laboratory chow & H2O
172 23
228 47
Sucrose
167 + 8
204 14
Ethanol
149 7
173 11


35
to prevent comparisons of it with group E from approaching any level of
statistical significance.
LTP Induction in the Absence of GABAergica Synaptic Transmission
Recordings were taken from multiple slices from the same group S, E
and C animals included in the first series of experiments. Stimulating and
recording electrode placement is again illustrated in Figure 4-1A.
Initial recordings were taken with aCSF as the superfusate. Following 6
minutes of stable single pulse recordings, 3.5 pM bicuculline methiodide was
added to the superfusate and single test pulse recordings again taken for 6-12
minutes. Blockade of GABAa receptors with bicuculline reduces the
functional inhibition of pyramidal cells. The resulting EPSP reflects primarily
the action of excitatory inputs. Bicuculline also facilitates the induction of
LTP. In the presence of bicuculline, a shorter duration or lower frequency
conditioning train can produce LTP.
LTP was recorded in each slice in response to two conditioning trains
which again differed only in duration. Following the initial 50 pulse, 50 Hz
conditioning train, EPSP slope and amplitude abruptly increased (Figs. 4-4 and
4-5). A decay in EPSP slope then occurred over the next several minutes and
stabilized at a new potentiated level in 8/8 group C, 8/9 group S and 9/9 group
E animals. A subsequent 100-pulse, 50 Hz conditioning train again produced
an immediate increase in EPSP slope which decayed to a level of higher
potentiation in 6/8 group C, 4/9 group S, and 6/9 group E animals. All
animals responded with LTP to at least one of the conditioning trains.
All data were evaluated for inclusion in group statistical analysis based
on the same criteria as for LTP induction in the presence of GABAergic
synaptic transmission. Neither the pattern nor the magnitude of potentiation


36
Figure 4-4. Representative field EPSP waveforms recorded in the presence of
3.5 pM bicuculline methiodide. Recordings were taken from SR of CA1 in
response to 0.04 ms duration single pulse stimulation (a) before, and (b) 28
min following a 50 Hz, 50 pulse conditioning train delivered to the Sch/Com
afferent fibers. Calibration pulse is 2 ms duration, 2 mV amplitude.
following each of the conditioning trains was altered in group E animals as
compared with groups S or C (Fig. 4-6; Table 4-3).
The dependence of LTP induction on activation of the NMDA receptor
ionophore is illustrated in Figure 4-7. In both group E and S animals, a 50
pulse, 50 Hz conditioning train failed to induce LTP when 50 pM APV, a
competitive antagonist to the NMDA receptor, was included in the
superfusate. Following a 25 minute washout period, a second train of 50
pulses, 50 Hz produced a sustained potentiation. APV was equally potent at
blocking LTP in the absence of bicuculline.
Discussion
Chronic ethanol exposure resulted in a reduction in the magnitude of
LTP under conditions in which excitatory and inhibitory synaptic
transmission contribute to LTP induction. This effect was seen following


37
Figure 4-5. EPSP slope before and after each of two successive conditioning
trains delivered in the presence of bicuculline methiodide.
(A) Representative data expressed as percent of baseline. Bicuculline (3.5 pM)
was added to the superfusate at t= -6 min and remained throughout the
experiment. Baseline EPSP slope recordings were obtained in response to 0.05
Hz single test pulses of Sch/Com fibers in SR (t= -13 0 min). At t= 0, a 50
pulse, 50 Hz conditioning train was delivered. Single test pulses were
delivered and recordings taken at 0.05 Hz for 28 min. A subsequent
conditioning train of 100 pulses, 50 Hz was delivered at t= 28 min. Single test
pulse recordings were again taken for 28 min. Each conditioning train was
followed by increased EPSP slope which decayed to a stable potentiated level
after several minutes.(comparison of the magnitude of potentiation within
animals, ANOVA, P < 0.001).
(B) EPSP slope plotted as a function of stimulus strength. EPSP slope
increased as a function of duration of the conditioning train over a range of
stimulus strengths.


38
250
OJ
5
cS
W
Ph
o
hJ
CD
Ph
cn
Ph
w
200-
150
100-

T"
50
# PULSES
SUCROSE n=9
ETHANOL n=8
1
100
Figure 4-6. Mean increase in EPSP slope following each of the conditioning
trains for group S and E animals. The magnitude of LTP in group S and E
animals was similar following both the 50 and 100 pulse conditioning trains.
Data were analyzed by a two way ANOVA with the pulse trains as a within
animal repeated measure (P< .80).
Table 4-3. Mean EPSP slope SE expressed as percent of baseline following
each of the conditioning trains for groups C, S and E animals.
Number of pulses in
conditioning train
Chronic Diet Treatment
Fifty a
One Hundred3
Laboratory chow & H20
167 11
191 10
Sucrose
169 17
184 20
Ethanol
168 20
201 24
a All values are in the presence of 3.5 pM bicuculline methiodide.


39
g 150-
(8
X)
PJ
Ph
O
hJ
1/5
C-c
£
w
100-
50-
TT
-10
50 |iM APV
r rrn i i 11 | i i 11 | n i i | i i 11 | i i i i | 11 i i | i i i 11 i i i i | 11 i i | i i i 11 i i i i | i
0 ^ 0 ^ 10 20 30 40 50
TIME (minutes)
washout
bicuculline 3.5 mM
remained in superfusate
Figure 4-7. Representative slice demonstrating the ability of APV to block
LTP induction in the presence of bicuculline methiodide. Bicuculline (3.5
pM) and 50 pM APV were added to the superfusate at t= 6 min. A 50 pulse,
50 Hz conditioning train at t= 0 min produced an abrupt increase in EPSP
slope which decayed to preconditioning levels within two minutes. A 25
minute washout period with bicuculline remaining in the superfusate was
begun at t= 3 min. Subsequently, a 50 pulse, 50 Hz conditioning train (t= 28
min) produced an abrupt increase in EPSP slope which decayed to a stable
potentiated level.
both the 50 and 100 pulse trains, that is, over a range of stimulus conditions.
Responses to conditioning trains appeared normal after removal of
GABAergicA influences, suggesting that CET had little or not effect on the
NMDA mediated component of LTP induction per se. Instead, it appears that
CET interferes with the mechanism by which GABAergic inhibition
modulates LTP induction. CET did not block the LTP induction process
entirely, because LTP induction occurred in response to at least one of the
conditioning trains in all animals. Rather, CET reduced the magnitude of
potentiation in response to each conditioning train. A tenable hypothesis is


40
that GABAergic modulation altered LTP induction by increasing the
threshold for LTP induction. The present study provided no evidence of a
CET effect on early stages of LTP maintenance. Therefore, it is most likely that
CET does not affect the pre- or postsynaptic substrates responsible for this
component of LTP.
The mechanism(s) through which inhibitory synaptic transmission
acts to modulate the induction of LTP are poorly understood. In area CA1,
single pulse activation of the Sch/Com fibers results in release of glutamate
from excitatory synaptic terminals. Some of these terminals synapse onto
inhibitory interneurons. The inhibitory interneurons in turn synapse onto
the CA1 pyramidal cells. Activation of inhibitory interneurons leads to
release of GABA. GABA acts upon postsynaptic GABAa and GABAb
receptors to produce a fast onset, short duration, CL mediated IPSP and a
slower onset, longer duration, K+ mediated IPSP (Alger and Nicoll, 1982).
Additionally, GABA can act upon presynaptic GABAb autoreceptors to reduce
subsequent release of GABA (Davies, et al., 1990; 1991). Since the response to
activation of the postsynaptic GABAb receptors and autoreceptors occurs with
a latency of 200 400 ms, these physiological events do not contribute to the
single pulse response. However, the duration of an LTP conditioning train
ranges anywhere from 200 ms to 1 second. Activation of the GABAb
autoreceptors reduces postsynaptic GABAa and GABAb currents through
decreases in GABA release with a latency of 200 ms or greater (Davies et al.,
1991; Pacelli et al., 1991). The GABAb produced reduction of GABA release is
critical for LTP induction to occur. Blockade of these autoreceptors with CGP
35348 during the conditioning train blocks LTP induction (Davies et al., 1991).
One of the most potent modulators of NMDA receptor activation and hence
LTP induction is depolarization of the postsynaptic membrane (Malinow and


41
Miller, 1986; Gustafsson et al., 1987). The most likely mechanism through
which GABAb receptor blockade prevents LTP induction is by sustained
GABA release and hence postsynaptic hyperpolarization. In the cerebellum,
postsynaptic GABAb receptor activation produces a reduction in GABAa
mediated Cb uptake (Hahner et al., 1991). A similar mechanism, if present in
hippocampus, would further enhance postsynaptic depolarization. Through
presynaptic and perhaps postsynaptic mechanisms, GABAb modulation of
GABAergic function has a powerful effect on LTP induction.
The data presented in this chapter can be explained in a number of
ways: 1) CET may produce an upregulation of postsynaptic GABAa receptor
number, 2) CET may produce an enhancement in the efficacy of GABA
and/or bicuculline at the GABAa receptor ; 3) CET may produce a decrease in
GABAb receptor function thus enhancing GABAa function, 4) CET may
produce an enhancement of presynaptic GABA release due to reduced
function of the presynaptic GABAb receptors, or hyperinnervation from
inhibitory interneurons. Abraham et al. (1981) and Rogers and Hunter (1992)
demonstrated that chronic ethanol exposure decreased functional recurrent
inhibition in CA1. Therefore, it seems unlikely the present results could be
due to increased GABA release, unless such increases occurred only during
high frequency conditioning trains. Alternatively, antagonist blockade of the
CET produced enhancement in GABAa receptor number or function would
effectively remove an abnormally strong and enduring hyperpolarizing block
of NMDA receptor activation. The hypothesis I have derived from this study,
is that CET increases either 1) the efficacy of GABA and/or bicuculline at the
GABAa channel or 2) the number of GABAa channels. Future work must
include measurements of current flow and chloride uptake through
postsynaptic GABAa channels, GABAa and B receptor binding studies, and


42
testing the ability of GABAg receptor agonist and antagonists to alter LTP
induction following CET.
Although NMDA receptor function does not seem to play a role in
producing the LTP deficit, it may be involved in CET produced excitotoxic cell
loss. Walker et al. (1981) showed pyramidal cell loss after prolonged exposure
to an ethanol diet. Short term exposure to ethanol (1-2 weeks) results in a
dramatic upregulation of NMDA receptor number during withdrawal which
returns to normal levels within 24 hours (Gulya et al., 1991). A further phase
of chronic ethanol toxicity may reveal enduring changes in receptor number
or function to coincide with the principal cell loss.
The large variability in results taken from chow-fed animals may have
been due to differences in age, shipment lot of animals, or housing
conditions. The chow-fed animals varied in age from 12 to 14 months old (2 -
4 months older than the liquid diet treated animals). Regardless of the cause,
it is clearly preferable to compare results from two groups which have
received the same treatment except for the variable (ethanol) one wishes to
test. It may be wise in the future, though, to include age-matched sucrose
treated animals in an initial test of protocol to adjust for differences in
responses.
The observed deficits in LTP induction provide evidence that an
alteration(s) in GABAergic function or receptor number is a potential
substrate for at least part of the pathology of memory loss associated with CET.
For an alteration of LTP to be involved in memory loss, it must be resistant to
any recovery induced by prolonged ethanol abstinence. In the next chapter, I
have tested this by examining the effects of a 5 7 month recovery period
from chronic ethanol toxicity on LTP induction and its modulation by
GABAergic synaptic transmission.


CHAPTER 5
THE EFFECT OF A RECOVERY PERIOD FROM CHRONIC ETHANOL
TOXICITY ON LONG TERM POTENTIATION
Introduction
Chronic ethanol toxicity is associated with a variety of central nervous
system (CNS) morphological, behavioral and functional deficits. During the
period of ethanol ingestion as well as with abstinence from a chronic ethanol
diet, adaptive and/or compensatory changes as well as recovery can occur.
For example, in the dentate gyrus, cell loss due to chronic ethanol treatment
(CET) is accompanied by increases in the dendritic extent of the surviving
granule cells (Durand et al., 1989), and increases in spine density which return
to normal following a 20 week recovery period from ethanol (King et al.,
1988). The concomitant morphological degeneration and recovery suggest
that hippocampal functional properties may also change as a consequence of
abstinence. It is therefore important to determine the permanence of the
functional losses accompanying CET.
The previous chapter described experiments demonstrating a reduction
of long term potentiation (LTP) in the hippocampus as a result of chronic
ethanol toxicity. The deficit was observed only when GABAergic modulation
of LTP induction was present. This suggested that CET produced an alteration
in GABAergic function which then acted to diminish the effectiveness of
conditioning trains to activate the NMDA receptor complex and thereby
induce LTP. It was hypothesized that this deficit may in part contribute to the
behavioral deficits in acquisition of new memories, which is a permanent,
43


44
nonrecoverable consequence of CET. In order for the deficit in LTP to be
involved with the deficits in memory acquisition, it too must be present
following extensive ethanol abstinence. In the following sets of experiments,
the permanent effects of CET on induction of LTP and the role of GABA
synaptic transmission in the manifestation of these deficits were examined.
To do this, animals previously treated for 28 weeks with an ethanol
containing diet were allowed a period of 5- 7 months abstinence. The capacity
of hippocampal synapses to support LTP both in the presence and absence of
GABAergic blockade was examined.
Methods
Treatment Methods
Male Long Evans hooded rats (200-250 g) were matched by weight and
age and assigned to one of two liquid diet treatment groups described in detail
in Chapter 3, General Methods. Briefly, the diet treatment groups consisted
of: 1) An ethanol group (group E) in which ethanol comprised 35-39% of the
total caloric intake. The remaining calories were supplied by Sustacal (Mead-
Johnson Co.); 2) A sucrose group (group S) which received an identical diet
except sucrose was isocalorically substituted for ethanol. Group S animals
were individually pair-fed with group E animals such that each of a sucrose-
ethanol pair received the same volume of diet during the treatment period.
The liquid diets were administered as the sole source of food for a period of 28
weeks (Walker and Freund, 1971; 1973; Walker and Hunter, 1978; Walker et
al., 1981; 1982). At the end of the 28 week period all animals received
laboratory chow and water ad libitum for a period of 5-7 months, at which
time the acute experiment was performed.


45
Electrophysiological Methods
Slice preparation
Electrophysiological data were collected from a total of 16 group S and
17 group E animals within a period 5-7 months following cessation of the 28
week liquid diet treatment. All animals were coded throughout data
collection and analysis to avoid experimenter bias. Transverse sections (400
pm) were cut through the ventral hippocampus using a Mclllwain tissue
chopper. Slices were incubated in a holding chamber containing NaCl, 125
mM; KC1, 3.3 mM; KH2P04/ 1.25 mM; MgS04, 2.0 mM; CaCl2, 1.9 mM;
NaHCCb, 25 mM; glucose, 10 mM at room temperature. The pH was
maintained at 7.4 with 95% 02/ 5%C02. Forty to sixty minutes prior to the
onset of recording, a slice was transferred to a Haas type interface recording
chamber maintained at 32 C. The slice was superfused at 1 ml/min with
oxygenated medium of the same composition as described for the holding
chamber.
LTP induction protocol
Electrophysiological records were taken from multiple slices in each of
9 group S and 8 group E animals. At the beginning of each recording session,
a stimulating electrode was placed in stratum radiatum (SR) of CA1 midway
between stratum pyramidal and stratum lacunosum-moleculare. A glass
recording micropipette filled with 4 M NaCl was placed in SR of area CA1, 800
- 1000 pm from the stimulating electrode (Fig. 5-1A). Recordings of field
EPSPs in SR were obtained in response to 0.1 Hz test pulses of Schaffer
Collateral and Commissural (Sch/Com) fibers in SR (Fig. 5-1B). Although the
conditioning trains which induce LTP potentiate the field EPSP over a wide
range of stimulus strengths, we empirically determined that the maximal


46
Figure 5-1. Microelectrode placement and representative waveforms from
the hippocampal slice preparation.
(A) Schematic of a transverse hippocampal slice to illustrate electrode
placement. A bipolar stimulating electrode (S) was placed in stratum
radiatum (SR) of area CA1. An extracellular recording microelectrode (R) was
positioned in SR of area CA1 approximately 800 pm from the stimulating
electrode.
(B) Representative field EPSP waveforms recorded from CA1 SR in response
to 0.04 ms duration single pulse stimulation (a) before, and (b) 15 min
following a 100 Hz, 30 pulse conditioning train delivered to the Schaffer
Collateral and Commissural (Sch/Com) afferent fibers, (c) Field EPSP slope
was measured between the time points at 10% and 90% of maximal EPSP
amplitude. Calibration pulse is 2 ms duration, 2 mV amplitude.


47
percent change in slope occurs at stimulus strengths which produced a
preconditioning field EPSP slope 75 to 100% of maximal. For this reason the
single pulse stimulus strength was adjusted to produce a field EPSP slope 75%
of maximal.
Following an initial 10 minute recording period, the stimulus strength
was adjusted to produce a maximal EPSP and a conditioning train of 30
pulses at 100 Hz was delivered. After this conditioning train, the stimulus
strength was returned to the previous level and recordings in response to
single pulse stimulation were taken at 0.1 Hz for 12 min. Subsequent
conditioning trains and test pulse recordings were taken repeating this
protocol except the duration of each train was lengthened in increments as
follows: t=10 min, 40 pulses; t=20 min, 50 pulses; t=30 min, 60 pulses. We
have previously determined that to obtain an incremental increase in the
magnitude of LTP the duration of the conditioning train must increase in an
incremental fashion. Repetition of the 30 pulse train will not result in an
incremental increase in potentiation. Just prior to each conditioning train
and again at the end of the recording session, several test pulses covering a
range of stimulus strengths were delivered and the field EPSP slope recorded
to demonstrate the magnitude of potentiation over a range of stimulus
strengths (input/output data).
In a subset of group S and E animals, test pulse recordings were taken at
0.1 Hz for 28 min following each of the conditioning trains to demonstrate
the long lasting nature of LTP.
LTP induction protocol in the absence of GABAergic synaptic transmission
Experiments involving the removal of inhibitory synaptic
transmission were performed with 5.0 pM bicuculline methiodide, a specific
GABAa receptor antagonist. This concentration of bicuculline was found to


48
produce maximal facilitation of LTP induction without producing bursting
and post-burst depression in response to a conditioning train. Bicuculline
was added to the superfusate 6 min after the onset of recording and remained
in the superfusate until the end of the recording session. Under this
condition, EPSPs produced in response to single pulse stimulation reflect the
activation of primarily excitatory inputs. The treatment can result in longer
burst duration to single pulse stimulation. CA3 pyramidal cells normally fire
in bursts which compound the bursting activity in the postsynaptic CA1
pyramidal cell. To prevent uncontrollable pyramidal cell bursting and
resulting postburst depression, a cut was made between area CA1 and CA3.
This disconnected the CA3 pyramidal cell bodies from the Sch/Com fibers.
Additionally, the concentration of Mg2+ in the superfusate was increased
from 2.0 to 4.0 mM to decrease activation of NMDA receptor channels in
response to single pulse stimulation.
In the CA1 region of the hippocampus, stimulus frequencies less than
0.1 Hz are normally ineffective in altering the amplitude, slope or duration of
the field EPSP (Alger and Teyler, 1976). In the presence of bicuculline, single
pulse stimulation at 0.1 Hz resulted in a small increase in field EPSP slope
and/or amplitude in rare instances. Reducing the stimulus frequency to 0.05
Hz when recording from slices treated with bicuculline prevented
potentiation of the EPSP in response to single pulse stimulation and resulted
in a stable baseline recording.
Electrophysiological records were taken from multiple slices in each of
9 group S and 8 group E animals. Electrode placement is as described in the
LTP induction protocol and Figure 1A. Test stimuli consisted of 0.04 ms
constant current (120 1000 pA) pulses. Field EPSPs were recorded in SR in
response to 0.05 Hz test pulses of Sch/Com fibers.


49
Single pulse stimulus strength was adjusted to produce a field EPSP
slope 75% of maximal. Following an initial 6 min baseline recording period
in the presence of bicuculline, the stimulus paradigm for LTP induction was
begun. The stimulus strength was adjusted to produce a maximal EPSP and a
conditioning train of 50 pulses at 50 Hz was delivered. After the conditioning
train, the stimulus strength was returned to the preconditioning level and
recordings in response to single pulse stimulation were taken at 0.05 Hz for 28
minutes. A subsequent conditioning train and test pulse recordings were
taken repeating this protocol except the duration of the train was lengthened
to 100 pulses. Just prior to each conditioning train and again at the end of the
recording session several test pulses covering a range of stimulus strengths
were delivered and the field EPSP slope recorded to demonstrate the
magnitude of potentiation over a range of stimulus strengths (input/output
data).
Blockade of LTP with D-amino-5-phosphonovalerate
A separate group of animals (S and E) were studied to determine
whether the LTP induction mechanism (under these stimulus conditions),
either in the presence or absence of bicuculline methiodide is mediated via
activation of the NMDA receptor ionophore. To do this we delivered an LTP
inducing conditioning train in the presence of 50 pM D-amino-5-
phosphonovalerate (APV) to block the NMDA receptor ionophore. Any
potentiation requiring activation of the NMDA receptor would be blocked by
this treatment. We then allowed a washout period of 40 to 50 min to remove
the APV from the slice. Another conditioning train was then delivered and
test pulse recordings taken.


50
Data analysis
Field EPSP data was passed through a Grass Model P511 preamplifier,
digitized at 20 kHz and stored on an IBM AT computer for later analysis. Data
collection and analysis was performed using software developed in this
laboratory. Measurements were made of field EPSP slope (from 10% to 90% of
peak EPSP amplitude; Fig IB).
EPSP slope data were averaged over the last 5 min of recording after a
conditioning train for experiments conducted in the absence of bicuculline
and the last 10 min in the presence of bicuculline. Data were grouped and
analyzed by treatment. Postconditioning train field EPSP slope (Lt) is
expressed as percent of preconditioning EPSP slope (Lo):
Lt
EPSP slope (% baseline) = j x 100
and graphed as a function of time with the onset of the first conditioning
train being t=0.
Input/output data were expressed as EPSP slope (mV/ms). Post
conditioning train EPSP slope data was analyzed by a two-way analysis of
variance (ANOVA) with repeated measures with diet treatment as one factor
and duration of conditioning train as the other factor.
Results
CET Effect on LTP Induction
The purpose of the present investigation was to determine the
permanent effects of CET on the capacity of hippocampal synapses to develop
LTP. Since many of the alterations associated with CET recover after a
sustained period of abstinence, we investigated the enduring nature of deficits
in LTP following a 5-7 month recovery period.


51
Stimulating and recording microelectrode placement within the CA1
subfield of the hippocampus is illustrated in Fig. 5-1A. The postsynaptic
dendritic response to single pulse stimulation of the Sch/Com fibers is
depicted in Fig. 5-1B. This extracellular EPSP represents the near synchronous
activation of multiple synapses. Following an LTP inducing conditioning
train, the EPSP response to single pulse stimulation increases in amplitude
and slope (Fig. 5-1B).
Field EPSP slope in response to 0.1 Hz single pulse stimulation was
recorded for 12 min prior to delivery of the first conditioning train. The
average baseline (pretetanus) EPSP slope values + SEM were 1.36 + 0.20 and
1.02+ 0.13 mV/ms for group E and S respectively. In 8/9 group S, and 7/8
group E animals, the response to a 30 pulse, 100 Hz conditioning train
consisted of an abrupt increase in EPSP slope followed by a decay of several
minutes to a stable potentiated level (Fig. 5-2A). A subsequent 40 pulse, 100
Hz conditioning train further enhanced the potentiation of EPSP slope in 7/9
group S, and 4/8 group E animals. By the 50 pulse, 100 Hz train, slices from
7/9 group S and 6/8 group E animals were maximally potentiated. EPSP
potentiation occurred in all group S and E animals in response to at least one
of the conditioning trains. Responses to single pulses of varying stimulus
strength taken 12 min following each conditioning train demonstrate the
EPSP potentiation over a wide range of stimulus strengths (Fig. 5-2B).
In a few animals from each group, postconditioning train recordings
were continued for an additional 15 minutes to examine the early stages of
LTP maintenance. Groups S and E were similar in their ability to maintain a
stable potentiated state for the duration of the 28 minute recording period
(Fig. 5-3). The pattern of potentiation following each of the conditioning


52
Figure 5-2. EPSP slope before and after each of four successive conditioning
trains.
(A) EPSP slope expressed as percent of baseline after successive conditioning
trains in a representative animal. Baseline field EPSP slope recordings were
obtained in response to 0.1 Hz single test pulses of Sch/Com fibers in SR (t =
-10 0 min). At t = 0, a 30 pulse 100 Flz conditioning train was delivered.
Single test pulses were delivered and recordings taken at 0.1 Hz for 12 min.
Subsequent conditioning trains were delivered as follows: t = 12 min, 40
pulses; t = 24 min, 50 pulses; t = 36 min, 60 pulses. Each of the conditioning
trains produced progressive, stable potentiation of the EPSP slope ( P< 0.001).
(B) EPSP slope plotted as a function of stimulus strength. The EPSP slope
increased as a function of duration of the conditioning train over a range of
stimulus strengths.


53
O!

5
X>
W
Ph
O
hJ
en
Ph
£
w
50-
^|ini|ini|ini|nii|nii|niipr-
-5 0 5 10 15 20 25
TIME (minutes)
Figure 5-3. EPSP slope expressed as a percent of baseline in a representative
group E animal. At t = 0 min a 50 pulse, 100 Hz conditioning train was
delivered to the Sch/Com afferent fibers. Single pulse recordings were
taken at 0.1 Hz for 28 min to demonstrate both the similar time course of
the early decay in EPSP slope and the long-lasting nature of LTP expression
in group E animals.
trains and the frequency with which potentiation occurred was similar in
groups S and E animals.
The response in slices lacking sustainable LTP consisted of an abrupt
increase in EPSP slope following the conditioning train which decreased to
the preconditioning baseline within several minutes.
All data were evaluated for inclusion in group statistical analysis based
on the following criteria: 1) Responses to preconditioning single pulses must
be sufficiently stable, as indicated by a standard deviation less than 15%. 2)
Responses to single pulses for the final 5 minutes of the 12 minute recording


54
Figure 5-4. Mean change in EPSP slope ( SEM) following each conditioning
train for group S and E animals. Both the minimal level for the induction of
LTP (30 pulses) and subsequent increments in the magnitude of LTP
induction including asymptotic levels were reduced in group E animals (P <
0.06).
period following a conditioning train must also vary with a standard
deviation less than 15%. For each animal and experimental condition, the
first slice to meet the criteria was included in the analysis. Figure 5-4
illustrates the magnitude of LTP in response to each conditioning train. All
pulse trains produced markedly less LTP in group E as compared with group S
( P < .06).
In most cases (7/9 group S and 6/8 group E), asymptotic potentiation
was reached by the 50 pulse train with no further potentiation upon delivery
of the 60 pulse train. To determine if the maximal capacity for potentiation is
reduced in group E animals, the percent potentiation for each animal at


55
asymptote was compared between groups without consideration of the
duration of the conditioning train needed to produce that level of
potentiation. Maximal potentiation was markedly reduced in group E
animals as compared with group S (mean + SEM, 203% + 17 group S and
172% + 10 group E; P< 0.08, ANOVA). These results indicate that not only
does CET permanently reduce long term synaptic efficacy in response to a
single conditioning train, but the maximal response that can be achieved is
reduced.
LTP Induction in the Absence of GABAergic^ Synaptic Transmission
The purpose of this set of experiments was to determine whether CET
exerts a direct influence on NMDA mediated induction of LTP or via an
alteration of GABAergic modulation of LTP induction.
Recordings were taken from multiple slices in 7 group S and 9 group E
animals. Stimulating and recording electrode placement is illustrated in Fig.
5-1A. Extracellular synaptic field responses were recorded in SR of CA1 in
response to the single test pulse stimulation. Initial recordings were taken
with aCSF as the superfusate. Following 6 min of stable recordings, 5.0 pM
bicuculline methiodide was added to the superfusate and single test pulse
recordings again taken for 6- 12 min (Fig. 5-5). Blockade of GABAa receptors
with bicuculline reduces the functional inhibition of pyramidal cells. The
resulting EPSP reflects primarily excitatory inputs. Bicuculline also facilitates
the induction of LTP such that a shorter duration or lower frequency
conditioning train is capable of inducing LTP. In initial pilot experiments, a
range of bicuculline concentrations and frequency of single test pulse
stimulation of the Sch/Com fibers were evaluated. A concentration of 5.0 pM
bicuculline and test pulse frequency of 0.05 Hz were determined empirically
to not significantly alter the slope or amplitude of the test pulse recordings in


56
Figure 5-5. Representative field EPSP waveforms recorded in the presence of
5.0 pM bicuculline methiodide. Recordings were taken from CA1 SR in
response to 0.04 ms duration single pulse stimulation (a) before, and (b) 28
min following a 50 Hz, 50 pulse conditioning train delivered to the Sch/Com
afferent fibers. Bicuculline methiodide (5.0 ¡iM) was added to the superfusate
6 min prior to the onset of recordings and remained throughout the
experiment. Calibration pulse is 2 ms duration, 2 mV amplitude.
most cases. If EPSP slope did change by more than 10%, the slice was not used
in the LTP experiments. The average baseline EPSP slope values + SEM were
0.53 + 0.08 mV/ms and 0.43 0.04 mV/ms for group S and E respectively.
LTP was recorded in each slice in response to two conditioning
stimulus trains which differed only in duration. The first, a 50 pulse, 50 Hz
train produced an increase in EPSP slope and amplitude. The timecourse of
the EPSP responses to single pulse stimulation following the initial train is
illustrated in Fig 5-6A. In 6/7 group S animals and 9/9 group E animals, the
increase in EPSP slope decayed over the course of several minutes to a stable
potentiated level. The initial change in EPSP slope returned to


57
TIME (minutes)
B
Figure. 5-6. EPSP slope before and after each of two successive conditioning
trains delivered in the presence of bicuculline methiodide.
(A) Representative data expressed as percent of baseline. Baseline EPSP slope
recordings were obtained in response to 0.05 Hz single test pulses of Sch/Com
fibers in SR (t= -6-0 min). At t= 0, a 50 pulse, 50 Hz conditioning train was
delivered. Single test pulses were delivered and recordings taken at 0.05 Hz
for 28 min. A subsequent conditioning train of 100 pulses, 50 Hz was
delivered at t= 28 min. Single pulse test recordings were again taken for 28
min. Each conditioning train was followed by increased EPSP slope which
decayed to a stable potentiated level after several minutes.
(B) EPSP slope plotted as a function of stimulus strength. The EPSP slope
increased as function of duration of the conditioning train over a range of
stimulus strengths.


58
preconditioning levels within several minutes in the remaining group S
animal. A subsequent 100 pulse, 50 Hz conditioning train further enhanced
EPSP slope potentiation in 5/7 group S and 6/9 group E animals. All animals
exhibited LTP in response to at least one conditioning train. EPSP responses
to single pulses of varying stimulus strength taken 28 min following each
conditioning train demonstrate the EPSP potentiation over a wide range of
stimulus strengths (Fig. 5-6B). Neither the pattern of potentiation following
each of the conditioning trains nor the magnitude of potentiation was altered
in group E animals as compared with group S (Fig. 5-7).
The dependence of LTP induction on activation of the NMDA receptor
ionophore is illustrated in Figure 5-8. Fifty micromolar APV added to the
superfusate 5 min prior to delivery of a conditioning train blocks the
induction of LTP in both group S and E animals. Following a 40 50 min
washout period, LTP induction occurs in response to the conditioning train.
Discussion
The main conclusions drawn from these experiments are: 1) the
magnitude of long term potentiation is reduced as a result of chronic ethanol
exposure; 2) the reduction of LTP is a permanent consequence of CET and not
dependent on sustained ethanol ingestion; 3) CET affects the reduction of
LTP by altering some aspect of GABAergic synaptic transmission on LTP
induction; 4) CET does not overtly affect early phases of LTP maintenance; 5)
CET does not alter the functional capacity of the NMDA receptor complex to
induce LTP.
In the previous chapter, evidence was provided that a deficit in LTP
induction appears as early as 48 hours following abstinence from the chronic
diet. This deficit was masked by blockade of GABAa receptors. Here,


59
240-1
QJ
% 200-
5 180-
O

SUCROSE n=7
ETHANOL n=9
100
50
100
# PULSES
Figure 5-7. Mean EPSP slope following each of the conditioning trains for
group S and E animals. The magnitude of LTP of group E animals was not
reduced following either the 50 or 100 pulse conditioning train. Data were
analyzed by a two-way ANOVA with the pulse trains as a within animal
repeated measure (P< 0.38).
evidence is provided that the reduction in LTP and the ability of GABAergic
blockade to mask the deficit remains after a prolonged recovery process.
Studies of GABAergic interneuronal cell number provide evidence
that chronic ethanol toxicity affects this population of neurons. Lescaudron,
et al. (1986) described a permanent reduction in GABA immunoreactive
interneurons in CA1 of ventral hippocampus. Physiological findings of CET
produced effects on GABAergic synaptic transmission are somewhat more
complex. Intracellular recordings have demonstrated a reduction of the
GABAa mediated IPSP amplitude and K+ mediated afterhyperpolarization
(Durand and Carien, 1984a) Other studies have demonstrated that CET


60
.w
p-q-yn //* i 1 > i > > i > i i 1 i
| 0 30 min 40 50 60 70 80 90
TIME (minutes)
Figure 5-8. Representative slice demonstrating the ability of APV to block
LTP induction. In the presence of 50 pM APV, a 50 pulse, 100 Hz conditioning
train produced an abrupt increase in EPSP slope which decayed to
preconditioning levels within several minutes. Following a 40 minute
washout period t= 48, a 50 pulse, 100 Hz conditioning train produced an
abrupt increase in EPSP slope which decayed to a stable potentiated level. A
second train t= 72 further increased the magnitude of LTP.
produced a reduction in recurrent and feedforward inhibition onto CA1
pyramidal cells (Abraham et al., 1981; Rogers and Hunter, 1992). While CET
appears to be result in a reduction of inhibitory input onto pyramidal cells,
other research has shown that the postsynaptic responsiveness to GABA is
enhanced following CET. Rogers (1986) used extracellular responses to single
pulse stimulation to test the effectiveness of iontophoretic application of
GABA. He found that exogenous GABA application resulted in a larger
reduction of population spike (PS) amplitude in CET animals than in pair-fed
controls. Following iontophoresis of the specific GABAa antagonist,
bicuculline methiodide, the PS response to subsequent single pulse
stimulation was greater than pair-fed controls. These combined results


61
suggest that the efficacy of GABA and bicuculline at the postsynaptic
membrane are enhanced following CET. Therefore, it appears that CET
results in a reduction of GABA release (most likely from interneuronal cell
loss) and enhanced GABA efficacy at the postsynaptic GABA receptor.
Recent studies have demonstrated a wide variety of GABAa receptor
subunit types, each with a specific regional distribution (Sato and Neale, 1989;
Verdoorn et al., 1990). This leaves open the possibility that CET may affect a
specific subtype of GABAa receptor. In fact, Wafford, et al. (1991) have
provided evidence that the Y2l subunit of the GABAa receptor is specifically
susceptible to the acute effects of ethanol. This particular subunit, though, is
not found in high concentration in the hippocampus. Further research may
provide additional clues to subunit function in the presence of protracted
ethanol toxicity. Another possibility is that basal functioning of GABAergic
synaptic transmission may be unchanged by CET while plasticity related
functioning, as tested in these experiments, may be altered. For example,
activation of the presynaptic GABAb autoreceptor by GABA and the
consequent reduction of GABA release only produces a postsynaptic effect
after 200 400 ms (Davies, et al., 1991). Postsynaptic conductances are reduced
with a similar time course by postsynaptic GABAb receptor activation
(Morrisett et al., 1991; Hahner et al., 1991). Single pulse stimulation,
therefore, would be unaffected by these autoreceptors (Pacelli, et al., 1991).
Under the plasticity related conditions described in these experiments,
conditioning trains would activate the autoreceptors which would in turn
enhance postsynaptic depolarization through reduction of GABA release. No
research to date has examined the effects of CET on plasticity-related function
of the GABAergic system in CA1.


62
Just as the actions of CET may be limited to plasticity-related
GABAergic function, NMDA receptor function may be altered under
conditions unrelated to LTP induction. For example, NMDA receptor
activation is thought to play a major role in excitotoxicity and cell death
(Choi, 1988; Abele et al., 1990). Ischemia induced cell loss or development of
seizures can be prevented by concomitant administration of the glycine
antagonist, kynurenic acid or competitive inhibitors of NMDA receptor
channels, APV or carboxypiperazinephosphonate (Uckele et al., 1989). There
is no evidence to date that demonstrates that an increase in NMDA receptors
is associated with increased LTP. The possibility remains that expression of
NMDA receptors may be increased by CET resulting in greater susceptibility of
neurons to excitotoxic cell death.
Recent studies have provided evidence that localized lesions
involving only the hippocampus or more specifically, area CA1 result in
permanent inability to acquire memories while leaving other aspects of
cognitive performance unaltered (Zola-Morgan et al, 1986; 1992; Squire and
Zola-Morgan, 1991). Several unimodal and polymodal sensory areas send
reciprocal projections to the perirhinal, parahippocampal, and entorhinal
cortex (Insausti et al., 1987). Perirhinal and parahippocampal cortex also send
reciprocal connections to entorhinal cortex which in turn sends a major
connection to the dentate gyrus of the hippocampus. Removal of either
hippocampus or surrounding cortex results in short term memory loss (Zola-
Morgan et al., 1989). Combined lesions of hippocampus, perirhinal, and
parahippocampal cortex results in amnesia of greater severity than removal
of hippocampus or adjacent cortex alone. This suggests that the perirhinal,
parahippocampal cortex and hippocampus each contribute to memory
formation. Loss of any or all of these structures, though, does not impair


63
recall of past events or retrograde memory. When new memories are
formed, recall of these memories is dependent on temporal lobe functioning
for only a limited period of time. Thereafter, the memory is apparently stored
in the neocortex and retrieval of the memory is independent of temporal lobe
function. Certain characteristics of LTP induction in the hippocampus
including induction by physiological stimuli and relatively long duration but
eventual decay both correlate well with the hypothesis that LTP is a
physiological correlate of memory formation in the hippocampus (Squire and
Zola-Morgan, 1991).
The enduring nature of the LTP deficit makes it a candidate substrate
for at least part of the pathology responsible for the memory impairments
associated with CET. It is possible, then, that functional disturbances
produced by CET in the hippocampus may be sufficient to produce
anterograde amnesia.


CHAPTER 6
GABAa RECEPTOR MEDIATED CHLORIDE UPTAKE IN CORTEX AND
HIPPOCAMPUS FOLLOWING CHRONIC ETHANOL EXPOSURE
Introduction
Previous chapters have described a phenomenon in hippocampus in
which chronic ethanol treatment acts to diminish the magnitude of long term
potentiation (LTP). When the hippocampal tissue is exposed to a selective
GABAa antagonist, bicuculline methiodide, the CET induced reduction of LTP is
masked. That is, hippocampal slices from chronic ethanol treated animals
respond as pair-fed controls to a given conditioning train only when that train is
delivered in the presence of bicuculline. A hypothesis to explain these results is
that chronic ethanol treatment (CET) produces an enduring increase in
GABAergic synaptic transmission. The postsynaptic action of enhanced
GABAergic responses counteracts the depolarizing effects of the LTP inducing,
high frequency conditioning train and subsequent NMDA receptor activation.
Experimental blockade of GABAergic synaptic transmission by bicuculline
would allow sufficient postsynaptic depolarization to produce LTP of equal
magnitude to that of controls.
Previous studies have demonstrated that CET reduces recurrent inhibition
onto CA1 pyramidal cells as well as the amplitude and duration of the GABA
mediated inhibitory postsynaptic potential (IPSP) and CA1 GABAergic
interneuron cell number (Abraham et al., 1981; Durand and Carien 1984a;
Lescaudron et al., 1986; Rogers and Hunter, 1992). These data suggest that
GABAergic inputs are actually reduced following CET. Therefore an
64


65
enhancement in inhibitory synaptic transmission would most likely be from a
postsynaptic source.
Rogers (1986) demonstrated that the postsynaptic response to single pulse
afferent stimulation following iontophoretic application of either GABA or
bicuculline was exaggerated following CET. This suggests an enhancement in
the efficacy of GABA and bicuculline at the GABAa receptor. Therefore, the
simplest explanation for this chronic ethanol action is that CET enhances the
efficacy of GABA stimulation of a GABAa mediated chloride current and/or the
efficacy of bicuculline antagonism of GABA stimulated GABAa receptor
activation. Several factors govern the GABAergic contribution to the
postsynaptic membrane potential. The GABAa receptor mediates a short
latency, brief duration hyperpolarization in response to single pulse stimulation
(Curtis et al., 1970; Newberry and Nicoll, 1985; Thalmann, 1988). The GABAa
receptor is linked to a Ch ion channel in the same protein complex. GABAb
receptors mediate a slower latency and longer duration hyperpolarization in
response to single pulse stimulation. GABAb activity is G-protein mediated and
linked to a K+ conductance (Andrade et al., 1986; Dutar and Nicoll, 1988). In
addition to the postsynaptic effects, synaptically released GABA activates
presynaptic GABAb autoreceptors to produce decreased release of GABA in a
time dependent manner (200 400 ms) following the single pulse (Davies et al.,
1990; 1991). This effect is also G-protein mediated and thought to result from a
decrease in presynaptic Ca2+ conductance. In response to high frequency
conditioning trains (as in LTP induction), GABA exerts the above mentioned pre-
and postsynaptic effects during delivery of the train. A diminution in GABA
release in response to high frequency trains has been hypothesized to facilitate
postsynaptic depolarization, thereby enhancing relief of the voltage dependent
Mg2+ block of the NMDA receptor channel and subsequent NMDA receptor


66
activation (Davies et al., 1990; 1991). Subsequent to delivery of a conditioning
train, the AMPA mediated EPSP and GABAa and GABAg mediated IPSPs are
enhanced for prolonged periods (Morishita and Sastry, 1991). The magnitude of
the resulting LTP reflects summation of the potentiation of both excitatory and
inhibitory postsynaptic potentials (Abraham et al., 1987).
As a first step toward understanding the mechanism by which chronic
ethanol exposure disrupts LTP, the efficacy of GABA and bicuculline at the
GABAa receptor was examined by measuring uptake of 36C1 in cortical and
hippocampal membrane preparations.
Materials and Methods
Treatment Methods
Male Long Evans hooded rats (200-250 g) were matched by weight and
age and assigned to one of two liquid diet treatment groups described in detail in
Chapter 3, General Methods. Briefly, the diet treatment consisted of: 1) An
ethanol group (group E) in which ethanol comprised 35-39% of the total caloric
intake. The remaining calories were supplied by Sustacal (Mead-Johnson Co.); 2)
A sucrose group (group S) which received an identical diet except sucrose was
isocalorically substituted for ethanol. Group S animals were individually pair-
fed with group E animals such that each of a sucrose-ethanol pair received the
same volume of diet during the treatment period. The liquid diets were
administered as the sole source of food for a period of 28 weeks. At the end of
the 28 week period all animals received laboratory chow and water ad lib. for
either 2 or 6 months. The preceding protocol of chronic ethanol treatment (CET)
has been previously used as a valid model for studying the behavioral,
morphological, and physiological consequences of chronic alcohol abuse (Walker
and Freund, 1971; 1973; Walker and Hunter, 1978; Walker et al., 1981; 1982).


67
Preparation of Microsacs
Each experiment included one each of an ethanol/sucrose pair. Brains
were removed from decapitated animals and placed on ice. Whole hippocampus
and one-half of the cortex (randomly chosen) were dissected from brains and
placed in 5 ml of ice-cold buffer (145 mM NaCl, 5.0 mM KC1,1.0 mM Mg CI2,1.0
mM CaCl2, 10 mM glucose, 10 mM N-2-hydroxyethylpiperazine-n2-
ethanesulfonic acid, pH 7.5 with Tris). Tissue was homogenized by hand (10
strokes) with a glass-Teflon homogenizer. The homogenate was centrifuged at
900 g for 15 min at 0 C. The supernatant was decanted and the pellet
resuspended in 5 ml of fresh buffer, homogenized (10 strokes) and centrifuged at
900 g for an additional 15 min at 0 C. The supernatant was decanted and the
pellet resuspended in fresh buffer (5 ml for cortex and 4 ml for hippocampus),
homogenized (10 strokes) and placed on ice. Protein content was determined by
the method of Lowry (1951).
Measurement of 36C1 Uptake
The procedure for measurement of 36C1 uptake in microsacs is taken from
Allan and Harris (1986). Microsacs (200 pi) were pre-incubated in a water bath at
34 C for 5 minutes in the presence or absence of bicuculline (see below). Next
200 pi of a solution containing 36C1 (New England Nuclear, Boston, MA, U.S.A.)
and varying concentrations of GABA (0 300 pM) or a constant concentration of
GABA (100 or 300 pM) and varying concentrations of bicuculline methiodide (0 -
100 pM) were added during constant vortexing. In experiments in which the
efficacy of bicuculline to inhibit GABA-stimulated Cl uptake was tested,
bicuculline was added during the preincubation and incubation. After 3 sec, ^Cl
uptake was terminated by the addition of 4 ml ice-cold buffer containing 100 pM
picrotoxin and rapid filtration onto a premoistened (GF/C) Whatman filter,
using a Hoefer manifold (Hoefer Scientific, San Francisco, CA). Filters were


68
washed with an additional 8 ml of ice-cold buffer containing 100 (iM picrotoxin.
The amount of radioactivity on the filters was measured by liquid scintillation
spectrometry. The amount of 36C1 bound to the filters in the absence of
membranes was subtracted from all values. Total Cl uptake was calculated using
36C1 as a tracer. GABA-dependent Cl uptake was calculated by subtracting the Cl
uptake in the absence of agonist (GABA-independent Cl uptake) from the Cl
uptake in the presence of agonist (total Cl uptake).
Data Analysis
Statistical analysis was performed by a two-way analysis of variance
(ANOVA) with repeated measures (treatment vs. agonist or antagonist
concentration). ANOVA was used to compare the difference between two
means.
Results
Effect of Chronic Ethanol on GABA Stimulation of Cl Uptake
All comparisons were first made between the two (n =3 ethanol and 3
sucrose) and four (n = 5 ethanol and 5 sucrose) month abstinence groups. Since
the duration of abstinence did not significantly affect basal or stimulated Cl
uptake within each diet treatment (data not shown), animals from the two
abstinence periods were grouped together.
The objective of this set of experiments was to determine the efficacy of
GABA at the GABAa receptor following chronic ethanol exposure. Basal uptake
was first measured to determine differences in GABA-independent Cl uptake.
Mean values of group S and E animals are listed in Table 6-1. Comparison
between groups failed to demonstrate any difference in basal uptake.


69
Table 6-1. Basal Cl uptake in hippocampus and cortex of chronic ethanol and
sucrose treated animals.
Basal Cl Uptake (nmol/mg protein/3 s) ;SEM
Diet Treatment
Hippocampus
Cortex
Sucrose3
27.2 5.7
24.7 2.3
Ethanol3
21.5 2.2
24.5 3.1
3 Data were analysed by an ANOVA (P < 0.37 for hippocampus and 0.95 for
cortex).
The effects of chronic ethanol on GABA-dependent uptake of Cl was next
examined. Figure 6-1 shows the concentration-response curve for GABA (0 300
(iM) in microsacs prepared from hippocampus and cortex of ethanol and pair-fed
sucrose treated animals. Chronic treatment with ethanol had no effect on the
concentration-response curve for GABA stimulated uptake of Cl. EC50 values for
group E and S are listed in Table 6-2. Comparison of EC50 using an ANOVA
revealed no difference between diet treatment groups (P < 0.37 for hippocampus
and 0.87 for cortex).
Effect of Chronic Ethanol on Bicuculline Inhibition of GABA Stimulated Cl
Uptake
The objective of the next set of experiments was to examine the efficacy of
the GABAa antagonist, bicuculline methiodide, to inhibit maximal GABA
stimulation of Cl uptake in hippocampus and cortex. It has been previously
shown that bicuculline blocks the chronic ethanol induced reduction of LTP.
This capacity of bicuculline to prevent the disruption of LTP induction following
chronic ethanol exposure, may be mediated by an enhanced ability to block
GABA stimulated Cl uptake.
Chloride uptake was first examined in chow-fed control animals to
determine a concentration of GABA which produced maximal stimulation and


70
in hippocampal (A), (B) and cortical (C), (D) microsac preparations. (A) and (C)
represent Cl uptake as a percent of maximal (100 |iM GABA) Cl uptake. (B) and (D)
represent actual values of Cl uptake expressed as nmol/mg protein/3 s of
exposure.


71
Table 6-2. Effect of chronic ethanol treatment on GABA stimulated and
bicuculline inhibition of GABA stimulated Cl uptake.
Hippocampus
Cortex
Sucrose
Ethanol
Sucrose
Ethanol
EC50 GABAa
19.8 4.1
31.7 10.3
29.3 5.0
30.8 7.2
IC50 Bicucullineb
10.7 3.6
16.9 7.5
16.6 6.5
12.7 3.3
a P< 0.37 for hippocampus and 0.87 for cortex
b P< 0.44 for hippocampus and 0.61 for cortex; IC50 was determined under
conditions of maximal GABA stimulation (100 pM).
also exhibited a range of responses to a range of bicuculline concentrations.
Figure 6-2 shows the effects of bicuculline (0 -100 pM) in the presence of either
100 or 300 pM GABA, concentrations which produce maximal stimulation of Cl
uptake. Since the bicuculline antagonism of Cl uptake was complete in 100 pM
GABA but not in 300 pM GABA, all subsequent experiments were performed
with 100 pM GABA.
The efficacy of bicuculline to block GABA-stimulated Cl uptake in
hippocampal and cortical microsacs from animals in groups S and E was then
examined. Comparison of group S and E revealed no difference in the
concentration-response curve for bicuculline (Fig. 6-3; P < 0.48). Table 6-2 lists
the IC50 for bicuculline in hippocampus and cortex from animals in groups S and
E. Again, between-group comparisons revealed no significant differences in
either measure (P < 0.44 for hippocampus and 0.61 for cortex).
Lastly, the effect of bicuculline on basal Cl uptake was determined to test
if bicuculline alone produced an effect on Cl uptake. To do this, uptake in the
presence of 100 pM bicuculline and no GABA in the incubation solution was
measured. This value measures the amount of basal Cl uptake resistant to


72
Figure 6-2. Bicuculline inhibition of GABA stimulated Cl uptake in cortical and
hippocampal microsacs from chow-fed control animals.
Microsacs from chow-fed control animals were tested for the effectiveness of
bicuculline methiodide to block maximal GABA (100 or 300 pM) stimulated Cl
uptake in the hippocampus. Cl uptake was measured at bicuculline
concentrations of 0 -100 pM in (A) hippocampal and (B) cortical microsacs. In the
presence of 100 pM GABA, Cl uptake was blocked by bicuculline concentrations
of 30 pM or higher. When microsacs were incubated with 300 pM GABA,
blockade of Cl uptake was incomplete for bicuculline doses up to 100 pM.
bicuculline blockade. Bicuculline blocked virtually all basal Cl uptake in both
groups S and E. A two-way ANOVA with diet treatment as one factor and total
and bicuculline sensitive basal Cl uptake as the second factor showed no
significant differences (P < 0.89 for cortex and for hippocampus). Measurements
of bicuculline sensitive basal Cl uptake are listed in Table 6-3.


73
Figure 6-3. Chronic ethanol treatment failed to affect bicuculline inhibition of
GABA stimulated Cl uptake in hippocampal (A), (B) and cortical (C), (D)
microsacs. The concentration of GABA was kept constant at 100 pM and
bicuculline methiodide concentration ranged from 0 100 pM. (A) and (C) Data
are expressed as percent of maximal GABA stimulation. (B) and (D) represent
actual values of Cl uptake expressed as nmol/mg protein/3 sec exposure time.


74
Table 6-3. Bicuculline-sensitive basal uptake of Cl in hippocampal and cortical
microsacs.
Cl Uptake (nmol/mg protein/3 s) ;SEM
Diet Treatment
Hippocampus
Cortex
Sucrose3
25.5 6.6
28.3 3.2
Ethanol3
20.2 7.8
29.8 4.8
a Data were analyzed by a two-way ANOVA (P < 0.65 for hippocampus and 0.80
for cortex).
Discussion
The principal aim of these sets of experiments was to measure GABAa
receptor function following chronic ethanol treatment. GABAa receptor function
was assessed by quantifying GABA-dependent uptake of Cl in membrane
preparations. Increased Cl uptake reflects enhanced GABA receptor function to
open the Cl ionophore. No chronic ethanol effects were found in either GABA-
stimulated uptake or bicuculline-inhibition of GABA-stimulated Cl uptake.
Additionally, chronic ethanol exposure did not affect basal (GABA-independent)
Cl uptake. Taken together, these results indicate that chronic ethanol exposure
does not directly affect GABAa receptor channel activation by GABA or
inhibition of GABA activation by bicuculline.
Several previous studies using electrophysiological methods have
demonstrated that chronic ethanol decreases functional inhibition and increases
postsynaptic responsiveness to GABAergic inhibition. For example, Abraham et
al. (1981) and Rogers and Hunter (1992) found that animals chronically treated
with ethanol demonstrated an enduring reduction in functional recurrent
inhibition onto CA1 pyramidal cells. It was hypothesized that the reduction in
functional inhibition resulted from chronic ethanol produced interneuronal cell
loss in CA1. Rogers (1986) went on to test responsiveness of CA1 pyramidal cells
to iontophoretic application of GABA. Excitatory single pulse responses


75
produced in the presence of exogenous GABA were reduced by a greater amount
in chronic ethanol treated animals as compared to pair-fed controls. This
indicated that chronic ethanol exposure produced a heightened sensitivity of
CA1 pyramidal cells to exogenous application of GABA. Excitatory single pulse
responses in the presence of bicuculline were next examined. In this case,
responses were larger in ethanol treated animals as compared with pair-fed
controls, suggesting that chronic ethanol produced a heightened sensitivity of the
postsynaptic membrane to bicuculline. Therefore, chronic ethanol treatment
appears to enhance either postsynaptic GABAa receptor number in CA1 or the
efficacy of both GABA and bicuculline.
In the present experiments, though, no apparent enhancement in GABA
stimulation or bicuculline inhibition of Cl uptake was detected. An important
difference between these two experimental paradigms was that the physiological
studies were localized to the CA1 region of the hippocampus whereas the Cl
uptake experiments described here measured whole hippocampal function.
Perhaps GABAergic susceptibility to the effects of chronic ethanol exposure is
limited to distinct hippocampal subfields such as CA1 which go undetected
because of a dilution from GABAa receptors of dentate gyrus and CA3 regions.
The number of [3H]bicuculline binding sites is greater in the dentate gyrus than
CA1 (Olsen, et al., 1990) as is the magnitude of functional inhibition and its
ability to suppress LTP induction (Steward et al., 1990). In order for chronic
ethanol to have an effect limited to CA1, there must be a distinct element in the
affected region susceptible to chronic ethanol. Such selectivity in ethanol
susceptibility may occur via regional differences in GABAa subunit composition.
There are five different GABA receptor subunit types (a, p, y, 8, and p) and
several variants of each type expressed in the CNS (Tobin et al., 1991). The
various combinations of these subunit variants confer distinct functional


76
properties on the resulting GABA receptor (Verdoorn et al., 1990). The regional
distribution of GABAa receptor subunit variants throughout the CNS may-
produce a diversity of GABA responses, such that each functional type is specific
for the unique function that brain region subserves. Just as GABA receptor
subunit variants may determine specificity of receptor function, responsiveness
of subunit variants to various environmental toxins may be diverse. For
example, the 72-subunit which is thought to affect the size of the open channel
configuration and confers enhanced efficacy of GABA by benzodiazepines
(Pritchett et al., 1989; Sigel et al., 1990), is also sensitive to the acute action of
ethanol (Wafford et al., 1991). Perhaps the basic principle that a neurotoxic effect
can be directed toward a specific receptor subunit type may apply to chronic
ethanol toxicity.
The physiological effect of CET to reduce LTP induction was necessarily
recorded following a high frequency conditioning train (Chapters 4 and 5). The
present results of Cl uptake were measured from animals that had not received
prior LTP induction. It is possible the effect of CET is restricted to plasticity-
related events and therefore not a direct effect on the single pulse functioning of
the GABAa receptor channel. For example, LTP induction results in potentiation
of both the glutamate mediated EPSP and the GABAa mediated IPSP (Abraham
et al., 1987; Morishita and Sastry, 1991). CET could act to enhance the relative
contribution of GABAergicA potentiation to LTP. Such an enhancement of
GABAa LTP would effectively mask LTP of excitatory synaptic transmission.
Only when the GABAa mediated component was blocked by bicuculline, would
LTP of excitatory responses be seen. The CET produced enhancement of
GABAergic responses reported by Rogers (1986) were recorded following a 5
second iontophoretic application of GABA. Here too, it is possible that the


77
prolonged exposure of the postsynaptic membrane to neurotransmitter produced
an effect that a shorter duration of exposure would not.
Future experiments should be aimed at investigating the single cell
responsiveness of GABA and receptor agonists and antagonists both in response
to single pulse stimulation and during LTP conditioning trains. This will provide
detailed information about postsynaptic GABAergic function both during single
pulse and plasticity-related synaptic transmission. The characteristics of the
GABA mediated IPSPs following an LTP conditioning train should be examined
to determine if LTP results in a specific enhancement of IPSP relative to EPSP
following CET. In addition, the numbers and affinities of GABAa and GABAp
receptors measured under equilibrium conditions will provide added
information about GABA receptor responses to CET.


CHAPTER 7
QUANTITATIVE AUTORADIOGRAPHIC ANALYSIS OF NMDA
RECEPTOR BINDING WITH [3H]MK-801 FOLLOWING CHRONIC
ETHANOL CONSUMPTION
Introduction
Recent research suggests that acute exposure to low doses of ethanol
functionally inhibits the N-methyl-D-aspartate (NMDA) receptor channel
while the a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)
receptor remains little affected (Lovinger et al., 1990). Additionally, long term
potentiation (LTP), which is produced by NMDA receptor activation, is
severely reduced by low doses of ethanol (Sinclair and Lo, 1986; Blitzer et al.,
1990). Few studies have examined the enduring or long term effects of
ethanol on NMDA receptor function. In humans, the number of 3-(()-2-
carboxypiperazin-4-yl)-[l,23H]propyl-l-phosphonic acid ([3H]CPP, a
competitive NMDA receptor antagonist) binding sites in hippocampus is
decreased with chronic alcoholism while receptor affinity is increased
(Michaelis et al., 1990). Conversely, the number of [3H]glutamate binding sites
is increased with chronic alcoholism relative to age-matched control patients.
Experiments described in chapters 4 and 5 of this dissertation demonstrate an
enduring reduction in the magnitude of LTP following a 28 week exposure to
ethanol in rats. This disruption in LTP, though, was masked when GABAa
receptors were blocked with the specific antagonist, bicuculline methiodide.
These results suggest that chronic ethanol abuse impairs the mechanism by
which GABAergic inhibition modulates LTP induction rather than through a
direct action on NMDA receptor function.
78


79
A wealth of research has demonstrated that the NMDA receptor not
only functions under normal physiological states as a mediator of long term
synaptic plasticity, but also plays a critical role in certain pathological
conditions such as excitotoxic mediated cell death. For example, upon
withdrawal from a 1 to 2 week period of exposure to low doses of ethanol,
mice and rats demonstrate a heightened susceptibility to seizures as well as
enhanced tolerance to ataxia induced by ethanol (Khanna, et al., 1991). Both
behavioral tolerance and susceptibility to seizures can be blocked by prior
administration of blockers of the NMDA receptor complex. In mice, a 7 day
exposure to low doses of ethanol results in upregulation of [3H]diclozipene
([3H]MK-801) binding sites within the hippocampus (Gulya et al., 1991 and
Grant et al., 1990). The upregulation of the NMDA receptor complex is
thought to be at least partially responsible for the heightened susceptibility to
seizures during ethanol withdrawal as well as behavioral tolerance to
repeated exposure to ethanol (Khanna et al., 1991). In this model, it is not
known what alteration of the NMDA receptor complex occurs following long
term (more than 2 weeks) exposure to ethanol. Neurotoxic effects of ethanol
following a 7 or 14 day exposure are typically transient in nature (Gulya et al.,
1991), while 20 to 28 weeks of exposure results in permanent effects such as
cell loss and alteration of dendritic density and morphology as well as
adaptive responses such as increases in the dendritic extent of surviving cells
(Lescaudron and Verna, 1985; Cadete-Leite et al., 1988; Durand et al., 1989;
Walker et al., 1981).
The aim of the present study was twofold. First, the effect of long term
(5 month) ethanol exposure on NMDA receptor binding was measured to
quantify changes in receptor number and binding affinity. Second, the ability


80
of L-glutamate to enhance [3H]MK-801 binding was investigated as a means of
evaluating NMDA receptor function.
MK-801 binds the NMDA receptor only under conditions in which the
channel is open (Huettner and Bean, 1988; MacDonald and Nowak, 1990).
MK-801 binding to the NMDA receptor channel blocks current flow through
the channel and it is through this mechanism that MK-801 acts as a specific
noncompetitive NMDA antagonist. L-Glutamate activation of the NMDA
receptor produces channel opening, and therefore, MK-801 binding. In
experimental conditions, as the concentration of L-glutamate increases, so
does binding of [3H]-MK-801. The efficacy with which L-glutamate produces
receptor activation and channel opening can be measured by determining
pH]MK-801 binding in the presence of a range of glutamate concentrations.
Since excessive activation of the NMDA receptor complex may be excitotoxic
to neurons, an alteration of NMDA receptor number could be responsible for
morphological alterations such as neurotoxic cell loss following chronic
ethanol treatment. In addition, changes in glutamate stimulation of [3H]-MK-
801 binding may reflect an alteration of the efficacy of glutamate to open the
NMDA receptor channel.
Methods
Treatment Methods
Male Long Evans hooded rats (200-250 g) were matched by weight and
age and assigned to one of two liquid diet treatment groups described in detail
in Chapter 3, General Methods. Briefly, the diet treatment consisted of: 1) An
ethanol group (group E) in which ethanol comprised 35-39% of the total
caloric intake. The remaining calories were supplied by Sustacal (Mead-
Johnson Co.); 2) A sucrose group (group S) which received an identical diet
except sucrose was isocalorically substituted for ethanol. Group S animals


81
were individually pair-fed with group E animals such that each of a sucrose-
ethanol pair received the same volume of diet during the treatment period.
The liquid diets were administered as the sole source of food for a period of 28
weeks. At the end of the 28 week period all animals received laboratory chow
and water ad lib. for 48 hours. Two similar age animals housed under
similar conditions as group S and E animals but receiving laboratory chow
and water during the entire treatment period were included in the study. The
preceding protocol of chronic ethanol treatment (CET) has been previously
used as a valid model for studying the behavioral, morphological, and
physiological consequences of chronic alcohol abuse (Walker and Freund,
1971; 1973; Walker and Hunter, 1978; Walker et al., 1981; 1982).
Autoradiographic Methods
Data were collected from 8 group S, 8 group E and 2 chow-fed control
animals 48 hours following cessation of the 28 week liquid diet treatment.
All animals were coded throughout data collection and analysis to avoid
experimenter bias. Rats were killed by decapitation, the brains removed and
frozen on powdered dry ice and maintained at -70 C. A midsagittal cut was
made through each brain. Each brain half was randomly chosen for
sectioning on the horizontal or coronal plane. Six micron sections were cut
on a cryostat and thaw-mounted on acid washed and gelatin-subbed slides.
Brains were sectioned such that each slide contained one coronal and one
horizontal section from each of an ethanol, sucrose pair. Sections from each
of the two chow treated animals were mounted with one of the sucrose-
ethanol pairs. Sections were refrozen and stored at -20 C for no more than 24
hr prior to use.


82
PH1MK-801 saturation binding assay
The procedure for [3H]MK-801 binding represented a modified protocol
that has been previously described (Monaghan, 1991). Slides were warmed to
room temperature and dried with an air stream prior to preincubation with
buffer. The sections were then preincubated in 50 mM Tris acetate with 0.1%
saponin and 1.0 mM EDTA, (pH 7.7) at room temperature for 10 minutes.
Sections were then rinsed in 50 mM Tris-acetate buffer for 60 min at 30C.
This treatment removes endogenous glutamate, glycine and various ions.
Sections were next incubated with varying concentrations of [3H]MK-801
(0.0005 1 pM; (30 pCi/mM) New England Nuclear, Boston, MA, U.S.A.) in 50
mM Tris buffer containing 25 pM L-glutamate, 20 pM D-(-)-2-amino-7-
phosphonoheptanoic acid (D-AP7), 20 pM glycine, and 250 pM spermine for
60 minutes at room temperature. Non-specific binding was determined in
the presence of 50 pM MK-801.
Following incubation, slides were rinsed in ice-cold buffer containing
20 pM D-AP7 for 60 min. Sections were dried under an air stream and placed
into x-ray cassettes with tritium-sensitive film (Hyperfilm, Amersham,
Arlington Heights, IL). Standards were included along with radioactive tissue
during exposure (microscales, Amersham). Film exposure was 4 weeks at
4C, followed by standard film development in Kodak D-19 at 20C.
Glutamate stimulation of [3H1MK-801 binding
Another group of sections were pre-incubated in 50 mM Tris acetate
with 0.1% saponin and 1.0 mM EDTA, pH 7.7 at room temperature for 10
minutes followed by a rinse in 50 mM Tris-acetate buffer for 60 min at 30 C.
Sections were next incubated in varying concentrations of glutamate (0 25
pM) and 10 nM [3H]MK-801 in 50 mM Tris buffer containing 20 pM D-AP7, 20
pM glycine, and 250 pM spermine for 60 minutes at 0 4 C. Non-specific


83
binding was determined in the presence of 50 (iM MK-801. Sections were
rinsed, dried and processed as described above.
Determination of NMDA Receptor Density
Autoradiograms were analyzed by computer assisted densitometry
with a MCID (Microcomputer Imaging Device, Imaging Research, Inc.) image
analysis system. Densitometric measurements were converted during
analysis to pmol/mg protein binding. Statistical analysis was performed by a
two-way analysis of variance (ANOVA) with repeated measures (treatment
vs. brain region).
Results
The goal of this study was to determine the effects of protracted
exposure to ethanol on NMDA receptor complex number and function in the
central nervous system (CNS).
Effects of CET on (3H1MK-801 Saturation Binding Characteristics
In the first series of experiments, the number of receptor binding sites
and the affinity of [3H]MK-801 for the NMDA receptor channel was
determined in a variety of brain regions. MK-801 binds only to the open state
of the NMDA receptor channel (Huettner and Bean, 1988; MacDonald and
Nowak, 1990). Glutamate is thought to enhance MK-801 binding by
increasing the channel open time. Glutamate is also the endogenous
transmitter for the NMDA receptor channel. Glycine and spermine each bind
to specific and distinct sites on the NMDA receptor (Ransom and Stec, 1988).
Glycine and spermine, in isolation, do not cause NMDA channel opening.
Instead, these ligands act to allosterically enhance the ability of glutamate to
produce channel opening. Prior work has demonstrated that maximal MK-
801 binding occurs when assay conditions include glutamate, glycine and


84
spermine in the concentrations used in these experiments (Ransom and Stec,
1988; Monaghan, 1991).
Autoradiograms reveal a high concentration of [3H]MK-801 binding
sites in hippocampal dendritic fields, dentate gyrus, layer I-III of cortex,
septum and several thalamic structures (Fig. 7-1). Within the hippocampus
and dentate, [3H]MK-801 binding predominates in stratum radiatum and
oriens, and the dentate molecular layer.
Specific [3H]MK-801 binding measurements from each of several brain
regions were taken and data from each brain region of each animal were
plotted versus the concentration of [3H]MK-801 in the incubation solution
(Fig. 7-2). The Kd and Bmax were calculated from the best fit of the binding
isotherm. Table 7-1 lists the values of Kd and Bmax of [3H]MK-801 binding
for several brain regions. The Kd values varied little among brain regions.
The highest affinity binding was found in hippocampus stratum radiatum
and dentate molecular layer. Bmax values were similar to those reported in
the literature (Grant et al., 1990). Among the sucrose and chow treated
animals, regional variations in Bmax were found between dorsal and ventral
stratum radiatum of CA1 (1.961 + 0.05 ventral; 2.331 + 0.07 pmol/mg protein
dorsal; P < 0.0001) and dentate molecular layer (1.756 0.09 ventral; 2.085
0.08 pmol/mg protein, dorsal; P < 0.0007). and Bmax values for sucrose and
ethanol treated animals are listed in Table 7-1. Statistical analysis across brain
regions failed to demonstrate an effect of chronic ethanol exposure on either
Kd or Bmax (ANOVA, P< 0.74, Kd and 0.1, Bmax ).
Previous work has demonstrated regional selectivity of the neurotoxic
effects of chronic ethanol (Walker et al., 1981; Lescaudron and Verna, 1985;


Figure 7-1. NMDA receptor distribution as determined by [3H]MK-801
binding. (A) Non-specific binding was measured with 0.1 pM [3H]MK-801 and
50 pM MK-801. (B) Total binding in the presence of 0.1 pM [3H]MK-801.
Binding density increases from black to white.




87
Figure 7-2. Representative average specific [3H]MK-801 saturation binding.
Binding isotherms depicted were from (A) ventral stratum radiatum of CA1
and (B) parietal cortex layers I III in chronic ethanol treated and sucrose
control animals. Each data point represents the mean + SEM of eight ethanol
or sucrose treated animals. Curves represent the optimal non-linear fitting to
the rectangular hyperbola equation. Binding constants are listed in Table 7-1.
(Inset), Scatchard plot of [3H]MK-801 binding. Line was fit by linear
regression.


88
Table 7-1. Effect of chronic ethanol on [3H]MK-801 binding in various brain
structures.
Treatment
Sucrose
Ethanol
Sucrose
Ethanol
Brain region
Kd (nM) + SEM
Bmax SEM
(pmol/mg protein)
Dentate Gyms
d. Molecular Layer
5.93 0.72
5.77 0.59
2.063 0.09
1.930 0.05
v. Molecular Layer
6.07 1.19
5.61 0.61
1.837 0.08
1.902 0.07
Hippocampus
d. Stratum Radiatum
4.26 0.54
4.61 0.41
2.318 0.09
2.494 0.08
v. Stratum Radiatum
5.48 0.89
5.38 0.58
1.983 0.06
2.059 0.06
v. Subiculum
7.81 2.2
8.67 1.1
0.726 0.07
0.728 0.04
Cortex
d. Cortex (layers I-III)
7.62 1.34
8.82 1.51
1.323 0.05
1.471 0.11
d. Cortex (layers IV- VI)
7.25 1.02
6.59 1.28
0.895 0.04
1.031 0.09
v. Parietal Cortex
7.30 0.64
7.30 0.65
1.455 0.07
1.477 0.05
(layers I-III)
v. Parietal Cortex
5.94 0.62
6.55 1.05
0.856 0.07
0.935 0.07
(layers IV-VI)
Entorhinal Cortex
5.29 0.95
6.64 1.40
0.993 0.08
0.982 0.07
Septum
v. lateral Septum
9.09 1.21
11.6 1.66
1.004 0.05
1.182 0.11
Thalamus
v. Medial Thalamus
9.67 2.2
13.0 3.1
0.777 0.04
0.925 0.06
v. Lateral Thalamus
9.59 1.12
11.9 1.56
0.915 0.06
1.031 0.03
d. Thalamus
14.1 2.08
12.2 2.29
0.863 0.04
0.835 0.02
Data was analyzed by a two-way ANOVA (P < 0.74
for Kd, and 0.1 for Bmax).


89
King et al., 1988; Durand et al., 1989). Hippocampus and dentate gyrus are
among the most affected. When the Bmax from these regions only were
grouped as a repeated measure in an ANOVA, an interaction effect of liquid
diet treatment with brain region was found (P < 0.02). No effect was found
when comparing the Kj among these brain regions.
Effects of CET on Glutamate Stimulation of [3H]MK-801 Binding
As stated earlier, [3H]MK-801 binding is maximally enhanced when
glutamate, glycine and spermine are included in the incubation solution.
Glutamate's enhancement of MK-801 binding occurs in a dose dependent
fashion. Since glutamate is also the endogenous transmitter for the NMDA
receptor channel, one method of evaluating the functional capacity of the
NMDA channel is to examine the ability of glutamate to enhance [3H]MK-801
binding. In the next set of experiments, [3H]MK-801 binding was measured
with various concentrations of glutamate included in the incubation
solution. Quantifying [3H]MK-801 binding as a function of glutamate
concentration measures the ability of glutamate to stimulate channel
opening.
Representative autoradiograms of [3H]MK-801 binding under a range of
L-glutamate concentrations are shown in Figure 7-3. When no glutamate was
included in the incubation solution, only background levels of binding were
detected. Low but detectable levels of binding were found in most brain
regions studied when a low concentration of glutamate (0.05 pM) was used.
pH]MK-801 binding increased in a dose dependent manner with maximal
binding most typically at 10 pM glutamate. Measurements of [3H]MK-801
binding in different concentrations of glutamate from ventral stratum
radiatum and parietal cortex are represented graphically in Figure 7-4. EC50


90
values calculated from curve fitting of these data demonstrate regional
differences among sucrose controls (Table 7-2). These results are consistent
with those previously reported in the literature (Monaghan, 1991). The
regional differences in EC50 are thought to reflect pharmacological diversity of
NMDA receptors in the CNS. A comparison of ECso's between diet treatment
groups across brain regions failed to demonstrate an effect of chronic ethanol
exposure (P < 0.86).
Discussion
The aim of these experiments was to determine the effect of prolonged
ethanol exposure on NMDA receptor number and function in the CNS.
Chronic ethanol treatment did not produce a global change in either the
affinity of [3H]MK-801 for the NMDA receptor channel, the number of
binding sites or the ability of glutamate to activate the receptor and thus cause
channel opening. These results contrast with human studies in which the
number of [3H]CPP binding sites decreased and receptor affinity increased with
chronic alcoholism (Michaelis et al., 1990). Several factors could explain this
discrepancy. First, the nutritionally supplemented diet used in the present
study controlled for any potential effects from alcoholic malnutrition. Such
controls are impossible in human studies. Also, it is difficult to eliminate
influencing factors such as prior drug exposure or disease states in human
studies. By the very nature of the disease of alcoholism, accurate and
complete histories are difficult to obtain.
In addition to the physiologic role of the NMDA receptor as mediator
of long term plasticity, several studies have implicated the NMDA receptor as
partially responsible for seizure generation associated with ethanol
withdrawal and excitotoxic cell death. Several studies have implicated


Figure 7-3. Total [3H]MK-801 binding in a range of concentrations of L-
glutamate and a constant concentration of [3H]MK-801 (10 nM).
Concentrations of L-glutamate in each of the depicted examples were as
follows: (A), 0.05 pM; (B), 0.1 pM; (C), 5.0 pM; (D), 25.0 pM. (A) Minimal
levels of binding were measured from sections incubated in the presence of a
low concentration of L-glutamate. (B D) As the concentration of L-
glutamate increased, [3H]MK-801 binding increased in a dose dependent
manner. Binding density increases from black to white.


92
C
D


Full Text
THE EFFECTS OF C¡TRONIC ETHANOL TREATMENT ON LONG TERM
POTENTIATION IN THE HIPPOCAMPUS
By
MARGARET FAIRCHILD TREMWEL
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1993

To Terry

ACKNOWLEDGEMENTS
I would like to thank my advisor, Bruce Hunter, for his contributions to
this thesis and sharing of his expertise during the course of this study. I also
thank the members of my committee, Kevin Anderson, Mike King, John
Middlebrooks, Phil Posner, Paul Reier, Floyd Thompson, and Don Walker, for
their helpful suggestions and critical evaluation of this manuscript. I would like
to especially thank Kevin Anderson and Joanna Peris for allowing me to work in
their labs.
Doug Dillard and Tim Cera provided invaluable programming assistance
during the physiology experiments. My research relied greatly on the technical
support of Pat Burnett. During the course of my stay at UF, I have been
supported financially by an NIH training grant, the American Paralysis Assoc.,
and an NIAAA predoctoral fellowship.
My mother, Dorothy Welch has provided much support and guidance.
Many fellow students have helped me to keep a healthy perspective on work.
Michele Davda and Laura Errante are especially talented in this respect.
Members of Westminster Presbyterian Church helped us to make Gainesville
home. Lastly, I thank my husband Terry whose loving support, patience and
sense of humor has sustained me.
Tis grace that brought me here this far and grace will lead me home.
m

TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS iii
ABSTRACT vii
CHAPTERS
1 INTRODUCTION 1
2 LITERATURE REVIEW 3
Behavioral Evidence of Memory Deficits in an Animal Model
of Long Term Alcohol Consumption 7
Anatomical Evidence of Hippocampal Pathology Following
Chronic Ethanol Treatment 8
Physiological Effects of Chronic Ethanol Toxicity in the
Hippocampus 9
The Effect of Duration of Exposure and Length of Abstinence
on the Substrates of Ethanol Toxicity 12
A Physiological Correlate of Memory Formation 12
Objectives of the Dissertation 17
3 GENERAL METHODS 19
4 THE EFFECTS OF CHRONIC ETHANOL TREATMENT ON
LONG TERM POTENTIATION IN THE HIPPOCAMPUS 21
Introduction 21
Methods 24
Treatment Methods 24
Electrophysiological Methods 25
Results 30
CET Actions on LTP Induction 30
LTP Induction in the Absence of GABAergicA
Synaptic Transmission 35
Discussion 36
IV

V
5 THE EFFECT OF A RECOVERY PERIOD FROM CHRONIC
ETHANOL TOXICITY ON LONG TERM POTENTIATION 43
Introduction 43
Methods 44
Treatment Methods 44
Electrophysiological Methods 45
Results 50
CET Effect on LTP Induction 50
LTP Induction in the Absence of GABAergicA
Synaptic Transmission 55
Discussion 58
6 GABAa RECEPTOR MEDIATED CHLORIDE UPTAKE IN
CORTEX AND HIPPOCAMPUS FOLLOWING CHRONIC
ETHANOL EXPOSURE 64
Introduction 64
Materials and Methods 66
Treatment Methods 66
Preparation of Microsacs 67
Measurement of 36C1 Uptake 67
Data Analysis 68
Results 68
Effect of Chronic Ethanol on GABA Stimulation of Cl
Uptake 68
Effect of Chronic Ethanol on Bicuculline Inhibition of
GABA Stimulated Cl Uptake 69
Discussion 74
7 QUANTITATIVE AUTORADIOGRAPHIC ANALYSIS OF
NMDA RECEPTOR BINDING WITH [3H]MK-801
FOLLOWING CHRONIC ETHANOL CONSUMPTION 78
Introduction 78
Methods 80
Treatment Methods 80
Autoradiographic Methods 81
Determination of NMDA Receptor Density 83
Results 83
Effects of CET on [3H]MK-801 Saturation Binding
Characteristics 83
Effects of CET on Glutamate Stimulation of [3H]MK-
801 Binding 89
Discussion 90

vi
8 SUMMARY AND DISCUSSION 96
Summary and Interpretation 96
Future Directions 99
REFERENCES 103
BIOGRAPHICAL SKETCH 114

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE EFFECTS OF CHRONIC ETHANOL TREATMENT ON LONG TERM
POTENTIATION IN THE HIPPOCAMPUS
By
Margaret Fairchild Tremwel
May, 1993
Chairman: Bruce E. Hunter
Major Department: Neuroscience
Previous work has demonstrated that chronic ethanol toxicity (CET)
produces severe and permanent alterations in hippocampal pyramidal cell
morphological, and functional characteristics. Additionally, memory formation,
a behavioral process thought to be at least in part attributed to hippocampal
functioning is severely attenuated following CET. The present study examined
the effect of CET on a well recognized physiological correlate of memory
formation, long term potentiation (LTP). Animals were fed a nutritionally
complete, ethanol containing diet for 28 weeks and then allowed a 48 hour or 5 -
7 month abstinence period. A control group was fed the same diet except
sucrose was isocalorically substituted for ethanol. Neurophysiological methods
measured the capacity of synaptic connections onto CA1 pyramidal cells to
support LTP in response to a variety of conditioning trains. At both periods
post-abstinence, the magnitude of LTP was reduced in CET animals as compared
with pair-fed controls. LTP induction is mediated by activation of the N-methyl-
d-aspartate (NMDA) receptor complex and is modulated by activation of y-
vii

viii
aminobutyric acid (GABA)ergic synaptic transmission. The effect of CET on LTP
magnitude might have been due to effects on the NMDA-induced induction itself
or on the GABAergic modulation of induction. To distinguish between these
possibilities, the induction of LTP was tested in the presence of bicuculline
methiodide (a specific antagonist to the GABAa receptor). Under these
conditions, there was no difference between CET and controls. This result
suggests that the reduction of LTP magnitude by CET observed in the absence of
bicuculline was due to a permanent alteration of GABAergic modulation of
induction.
GABAa function was next examined by measuring the efficacy of GABA
to stimulate and bicuculline to antagonize GABA-stimulated chloride uptake in
hippocampal and cortical membrane preparations. CET did not affect basal
uptake of chloride or the efficacy of either GABA or bicuculline at the GABAa
channel.
In addition to plasticity related events, NMDA receptor activation plays a
role in excitotoxicity and cell death. To investigate the hypothesis that NMDA
receptors may mediated cell loss following CET, the binding properties an
antagonist to the NMDA receptor channel, [3H](+)-5-methyl-10.11-dihydro-5H-
dibenzo[a,d]cyclohepten-5,10-imine ([3H]MK-801) were examined. Neither the
Kj nor the Bmax of [3H]MK-801 binding to the NMDA receptor or the ability of
glutamate to stimulate [3H]MK-801] binding was altered following CET.

CHAPTER 1
INTRODUCTION
Chronic abuse of alcohol results in deterioration of anatomical,
physiological and behavioral correlates of cognitive performance. The
neurotoxic damage extends into several brain structures including thalamus,
cerebellum, cerebral cortex, and hippocampus. Of particular consequence is the
pathology of anterograde amnesia, a profound and devastating symptom of
chronic abuse.
Interest in the types of anatomical and physiological deficits responsible
for the anterograde amnesia have directed the attention of researchers to the
hippocampal formation. This region is critical for acquisition of new memories
and is specifically and dramatically affected by all stages of early and chronic
ethanol abuse (Grant et al., 1990; Lovinger et al., 1990; Walker et al., 1981; Zola-
Morgan, et al., 1986; 1992). Two exciting areas of recent research have helped to
shed light on potential physiological substrates of ethanol-produced memory
deficits. The first area has focused on ethanol interactions with a possible
physiological correlate of memory formation, long term potentiation (Lovinger et
al., 1990; Blitzer et al., 1990; DeMontis, et al., 1991; Grant et al., 1990; Gulya et al.,
1991). Through this research, we have learned of a specific susceptibility of the
N-methyl-D-aspartate (NMDA) receptor complex to both acute and proldneed
exposure to low concentrations of ethanol. The effects of ethanol on NMDA
receptor function are most likely the cause of the observed deficits in LTP
induction following acute exposure. NMDA receptors also participate during
pathological states in the production of seizures and excitotoxic cell
1

2
death (Choi, 1988; Grant et al., 1990). Therefore, alterations in NMDA receptor
function may be involved with the cell loss which follows chronic ethanol abuse.
The second area of recent research has examined the interactions of
ethanol with specific subtypes of GABAa receptor function (Allan et al., 1991;
Luddens, et al., 1990; Wafford et al., 1991). For example, a specific subunit (y2l)
of the GABAa receptor is vulnerable to the effects of acute ethanol exposure
(Wafford et al., 1991). The regional specificity of ethanol actions on GABAergic
synaptic transmission may thereby reflect the regional distribution of GABAa
receptors which contain the subunit. GABAa receptors are well recognized as
potent modulators of neuronal excitability as well as plasticity-related functions
such as LTP induction.
The purpose of this dissertation work was twofold. First, I investigated
the possibility that chronic ethanol ingestion disrupts some component of LTP
induction. Secondly, I examined some potential mechanisms through which
chronic ethanol toxicity may mediate the disruption of LTP: 1) NMDA receptor
function and 2) GABAa receptor function.

CHAPTER 2
LITERATURE REVIEW
The first characterization of a neural disorder associated with chronic
alcohol abuse was described by Carl Wernicke in 1881. The major symptoms he
noted included global confusion, opthalmoplegia, nystagmus, ataxia, and
polyneuropathy of extremities (Butters and Cermak, 1980). Postmortem
examination of brain tissue revealed punctate hemorrhages located in the gray
matter around the third and fourth ventricles. It was later learned that treatment
with large doses of thiamine could reverse disease progression with eventual
dissipation of symptomatology. Several years after this initial description of
Wernicke's encephalopathy, S.S. Korsakoff described a state of amnesia that often
accompanied polyneuropathy disorders (Butters and Cermak, 1980). These
symptoms followed long term alcoholism, vomiting, and intestinal disorders. In
Korsakoff's syndrome, patients suffer from a profound anterograde amnesia of
both verbal and nonverbal information. No alterations in intellectual functioning
are associated with the memory loss. Additionally, Korsakoff's syndrome is
associated with a graded retrograde amnesia, that is, patients can recall events
from the distant past with greater ease than events from the recent past. When
asked questions that test recent memory function, Korsakoff's patients may tend
to confabulate or fill in the time period of which they have no memory with
events from the distant past. However, this is not a consistent feature of this
disease. Korsakoff patients tend to be apathetic, passive and in general
disinterested in alcohol, a marked distinction from their premorbid personality.
Eventually it was determined that Wernicke's encephalopathy and Korsakoff
3

4
syndrome occur as separate components of the chronic ethanol disease process.
Wernicke's encephalopathy is primarily caused by malnutrition associated with,
but not caused by chronic alcoholism. Following treatment with massive doses
of thiamine, the alcoholic patient may then enter into the chronic Korsakoff's
stage of this disease process. The neuropathology associated with each of these
disease processes is distinct. Wernicke's disease involves hemorrhagic injury to
midbrain structures and cerebellum whereas the pathology of Korsakoff's is
associated with lesions of thalamic nuclei and mammillary bodies.
In 1971, Ryback proposed the hypothesis that chronic alcohol abuse
produces a graded effect on memory impairment. That is, the alcoholic
Korsakoff patient, long term alcoholic and heavy social drinker all present with a
qualitatively equal memory disorder but the severity of the deficit progresses
with duration of drinking. Indeed, several studies have demonstrated that the
chronic alcoholic patient performs more poorly in short term memory tasks than
age-matched controls, but significantly better than age-matched alcoholic
Korsakoff's patients (Parker and Noble, 1977; Ryan et al., 1980). For example,
short term memory tests were given in which the patient was asked to associate
digit-symbol pairs so that upon presentation of a geometric symbol the patient
would name the number associated with it. Alcoholic Korsakoff's patients failed
to learn this task even over repeated trials. Chronic alcoholics performed better
than the Korsakoff's but never reached the level of performance of control
subjects. Similar results were obtained when patients were asked to remember
four unassociated words for durations of 15 to 30 seconds. In all cases, the deficit
of chronic alcoholics was most pronounced when the subject was asked to do
another cognitive task (e.g. counting backwards) during the time period between
presentation of the item to be remembered and recall (proactive interference).

5
There also are marked differences in the characteristics of amnesia
between the alcoholic Korsakoff's patient and chronic alcoholic. For example, the
memory disorder associated with chronic alcoholism is confined primarily to
short term memory with no retrograde amnesia. The alcoholic Korsakoff's
patient suffers from both anterograde and retrograde memory deficits.
Confabulation is a characteristic sometimes associated with the anterograde
amnesia of Korsakoff's but not seen in chronic alcoholism. The different
characteristics of memory deficits in Korsakoff's and chronic alcoholic patients
may reflect different or additional anatomical substrates responsible for the
amnesias.
The morphological substrates responsible for the alcoholic memory
disorder have been studied in both humans and animal models. In humans, cell
loss has been described in several brain structures including cerebellum,
thalamus, frontal cortex, basal forebrain, and hippocampus (Arendt et al., 1983;
Bengochea and Gonzalo, 1990). In order to determine the potential contributions
of these regions to the behavioral changes during alcoholism, researchers have
compared behavioral data from human clinical and animal temporal lobe and
medial thalamic lesion studies to that from chronic alcoholics. Severe
anterograde amnesia has been described in patients with a lesion somewhat
localized to the dorsal medial region of the thalamus or to area CA1 of the
hippocampus (Squire and Moore, 1979; Zola-Morgan et al., 1986).
Animal models with lesions restricted to either the dorsal thalamus or
hippocampus have helped to shed light on the specific contribution of these
regions to normal memory function. A commonly used task for testing short
term memory is the delayed non-matching to sample. In this test, the animal is
allowed to displace an object covering a food well with a raisin in it. Next, an
opaque wall is placed to block the monkey's view of the food well. After 8

6
seconds, the wall is removed and the monkey sees two objects, the previously
viewed object and a novel object. The monkey is trained to choose the novel
object to obtain the raisin reward. Once the monkey has learned this task, the
delay between initial presentation of the object and presentation of the two
objects is lengthened. Animals with lesions to either the dorsal medial region of
the thalamus or CA1 of the hippocampus have difficulty learning the task
initially and once criterion is reached, in performing the task correctly as delays
between presentations are increased to 60 sec or more (Aggleton and Mishkin,
1983; Zola-Morgan et al., 1992). These results demonstrate the deficits in new
memory acquisition as well as recall of recent information following lesions to
dorsal thalamus or hippocampus. In dorsal medial thalamus, lesions which
encompassed both anterior and posterior thalamus resulted in amnesia far
greater than following a lesion to either region alone. The notion of an increasing
memory deficit with increasing extent of the lesion holds true for the temporal
lobe as well. For example, lesions which include hippocampus,
parahippocampal gyrus and perirhinal cortex produce a more severe
anterograde amnesia than occurs following lesions restricted to the hippocampus
(Zola-Morgan et al., 1989). This suggests that the parahippocampal gyrus and
perirhinal cortex provide a contribution to memory in addition to the
contribution made by the hippocampus. These combined results indicate that a
number of separate brain regions are responsible for memory acquisition and
that a lesion to any one of these structures may result in short term memory loss.
Inferred from these studies is that the severity of amnesia may depend on the
extent of pathology to the memory circuit.
As stated earlier, the pathology most prominently associated with
alcoholic Korsakoff's disease is degeneration of the dorsomedial thalamus, which
is commonly thought responsible for the enduring memory loss associated with

7
this disease. There is only one study, though, that found a correlation between
damage to the dorsomedial thalamus and memory loss in alcoholic Korsakoff's
(Victor et al., 1971). Other studies have failed to consistently find such a
correlation (Butters and Cermak. 1980). The possibility remains then, that other
structures within the memory circuit may contribute or be solely responsible for
the memory loss following chronic alcoholism.
Because of the confounding variables of compromised nutritional status,
duration and quantity of drinking, as well as type of drinking (i.e., binge vs.
constant drinking) investigation of direct effects as well as the nature of the
effects of ethanol in humans is difficult. Several animal models of controlled
alcohol and nutritional intake have been developed to study the effects of chronic
alcohol alone.
Behavioral Evidence of Memory Deficits in an Animal Model of Long Term
Alcohol Consumption
In 1970, Freund described a rodent model of chronic ethanol consumption
in which animals received a nutritionally complete liquid diet with part of the
caloric intake in the form of ethanol. This diet served as the sole source of food
and liquid for the duration of the treatment period. A variety of behavioral tests
of memory acquisition demonstrated a permanent loss of memory acquisition
following a three to seven month exposure to ethanol (Freund and Walker, 1971;
Walker and Freund, 1973; Walker and Hunter, 1978; File and Mabbutt, 1990). In
one study, animals were trained to perform a temporal alternation task in which
bar presses were reinforced on alternate trials (Walker and Hunter, 1978).
Following 20 weeks of an ethanol containing diet and 2 months of abstinence,
animals were retrained on the same behavioral task. After the training period,
short term memory was tested by varying the duration between bar press trials.

8
Animals that received chronic ethanol treatment (CET) tested as well as pair-fed
controls when the intertrial interval was 20 seconds or less. The performance of
CET animals significantly declined as the intertrial interval increased to 60
seconds. The behavioral deficit demonstrated in this rodent model of CET
closely parallels the deficit in short term memory of chronic alcoholics and
alcoholic Korsakoff's patients, although a major difference was that the animal
model of memory loss occurred in the presence of a nutritionally complete diet.
These results demonstrate the nature of behavioral effect of ethanol alone.
Chronic ethanol consumption produced a toxic action to the neural substrate
underlying memory formation.
Anatomical Evidence of Hippocampal Pathology Following Chronic Ethanol
Treatment
Several studies have demonstrated loss of 10 to 40% hippocampal
pyramidal and dentate granule cells following a 20-week or longer exposure to a
chronic ethanol diet (Cadete-Leite, et al., 1988; Walker et al., 1980; Lescaudron
and Verna, 1985). Lescaudron and Verna (1985) went on to show that pyramidal
cell loss is greater in the ventral than the dorsal region of the hippocampus. In
addition, chronic ethanol abuse results in changes in spine density and decreases
in dendritic length (Riley and Walker, 1978; Me Mullen, et al., 1984; Goldstein et
al., 1983; Lescaudron et al., 1986; King et al., 1988; Cadete-Leite et al., 1989). In
the dentate gyrus, cell loss due to chronic ethanol toxicity is accompanied by an
increase in the dendritic extent of the surviving cells (Cadete-Leite, et al., 1988).
Following a 20-week recovery period, spine density returns to normal (King et
al., 1988).
Physiological studies have also been employed to examine dendritic
alterations following chronic ethanol consumption. Abraham et al. (1982)

9
described an alteration in the distribution of afferent synaptic connections in area
CA1. Specifically, current source density analysis of synaptic fields
demonstrated a 13.4% reduction in the Schaffer collateral and commissural
(Sch/Com) fiber synaptic field. The anatomical data taken together suggest that
concomitant processes of cell degeneration, adaptation, and a subsequent process
of recovery all contribute to the resulting behavioral and anatomical
manifestation of chronic ethanol toxicity.
In addition to the dramatic changes in principal cells of the hippocampus,
Lescaudron et al. (1986) described y-amino butyric acid (GABA)ergic cell loss in
the hippocampus. Using immunocytochemical analysis with an antibody to
GABA, they found decreased labeling intensity of immunopositive neurons and
fibers in both dorsal and ventral hippocampus, and decreased number of
immunopositive neurons in ventral hippocampus. These results may be
interpreted as a chronic ethanol-produced interneuronal cell loss in hippocampus
and/or a reduction in GABA content or antibody affinity in these neurons.
Mice specifically bred for sensitivity to the acute effects of ethanol were
used to examine the effects of a chronic ethanol diet on GABAergic cell number
in the dentate gyrus (Scheetz et al., 1987). Prior to ethanol exposure, there was no
difference in the number of dentate basket cells between ethanol sensitive and
resistant strains. Following a 3-month chronic ethanol diet, a significant
reduction in dentate granule layer basket cell number was observed only in
ethanol sensitive animals. These results again illustrate the specific neurotoxic
effects of ethanol on GABAergic interneurons.
Physiological Effects of Chronic Ethanol Toxicity in the Hippocampus
In the hippocampus, functional inhibition can be measured using a paired
pulse paradigm. When the first (conditioning) pulse is antidromic and the

10
second (test) pulse is orthodromic, the second response is reduced relative to the
first even with interpulse intervals of a few hundred milliseconds. The
conditioning pulse of this stimulus paradigm is thought to activate the recurrent
inhibitory pathway onto pyramidal cells. Comparison of the reduction of the test
pulse between diet treatment groups reflects the relative amount of functional
inhibition. Test pulse inhibition following chronic ethanol treatment is less
compared to pair-fed controls (Abraham et al., 1981; Durand and Carien, 1984a;
Rogers and Hunter, 1992). This suggests that recurrent inhibitory circuitry is
reduced after chronic ethanol treatment. Physiological alterations in the
hippocampus following chronic ethanol toxicity parallel the morphological
findings of decreased GABAergic interneurons.
Durand and Carien (1984a) demonstrated that following chronic ethanol
treatment the amplitude and duration of the K+ mediated afterhyperpolarization
and the orthodromically stimulated (GABA mediated) inhibitory postsynaptic
potential (IPSP) in CA1 pyramidal cells and dentate granule cells were reduced,
while other passive and active membrane characteristics of granule cells (input
resistance, resting membrane potential, excitatory postsynaptic potential (EPSP)
amplitude, etc.) remained unchanged. Rogers (1986) went on to test the efficacy
of GABA at the postsynaptic GABA receptor in response to iontophoresis of
GABA. Single test pulses delivered following iontophoresis of GABA were
reduced to a greater extent in chronic ethanol treated animals as compared to
pair-fed controls. He next tested single pulse responses in the presence of a
GABAa receptor antagonist, bicuculline methiodide. In this condition, responses
were larger in ethanol treated animals as compared to pair-fed controls. These
results suggested that chronic ethanol treatment resulted in an augmentation of
the GABAa receptor responses to its endogenous ligand, GABA. The efficacy of
bicuculline to reduce GABAergic inhibition was also enhanced. Taken together,

11
the above results suggested that CET produced a decrease in presynaptic release
along with either 1) an increase in postsynaptic GABAa receptor number along
with a decrease in presynaptic release or 2) an increase in efficacy of GABA at the
postsynaptic GABA receptor. The decrease in presynaptic release is most likely
due to a decrease in GABAergic innervation of principal cells.
In addition to the chronic ethanol associated changes in basal or single
pulse synaptic function, a few studies have examined ethanol actions on synaptic
function related to long term plasticity. In vitro measurements from area CA1
indicated that the number of hippocampal slices capable of expressing long term
potentiation (LTP) was decreased following chronic ethanol treatment but if LTP
was induced, the magnitude of potentiation was not different from controls
(Durand and Carien, 1984b). These experiments only measured the compound
action potential (PS) responses of pyramidal cells and were therefore more a
measure of LTP induced changes in cell excitability (i.e., threshold to activation
of an action potential and/or firing frequency) than LTP (which is synaptic in
origin). In vivo, there was no change in the early time course or magnitude of
LTP in area CA1. However, there was a trend toward a greater decay of the early
phase of LTP maintenance. In the dentate gyrus, LTP was unchanged (Abraham
et al., 1981; 1984). It should be noted that these studies recorded responses to
long duration conditioning trains which produced a maximal or asymptotic LTP
with a single train. Perhaps ethanol does not affect the maximal amount of
potentiation produced by a group of synapses, but rather the amount of
depolarization necessary to produce a given amount of potentiation. In other
words, chronic ethanol may affect the facility or ease of induction of LTP in
response to a given conditioning train. To test this, responses to short duration
conditioning trains which produce submaximal LTP must be measured.

12
The Effect of Duration of Exposure and Length of Abstinence on the Substrates of
Ethanol Toxicity
Throughout the alcohol literature it is widely accepted that differing
mechanisms are responsible for the pathology at each stage of neurotoxicity. The
particular stage of ethanol neurotoxicity reflects the duration of exposure or
withdrawal from ethanol. For example, acute exposure of hippocampal neurons
to low doses of ethanol results in a rapid block of the N-methyl-D-aspartate
(NMDA) receptor mediated EPSP and LTP induction (Lovinger et al., 1990;
Blitzer et al., 1990). Receptor binding studies revealed a coincident decrease in
the number of [3H]MK-801 (diclozipene, a specific antagonist to the NMDA
receptor channel) binding sites (DeMontis et al., 1991). Following a one or two
week exposure to a liquid diet containing ethanol, the number of [3H]MK-801
binding sites is upregulated as compared to controls (Gulya et al., 1991).
Behavioral evidence of heightened seizure susceptibility and reduced
responsiveness to the effects of acute ethanol (tolerance) are consistent with
increased NMDA receptor function (Grant et al., 1990). As the duration of
exposure lengthens to three to five months, hippocampal pyramidal cells and
interneurons undergo cell death (Walker et al., 1981; Lescaudron and Verna,
1985). The dendritic morphology of surviving cells is altered (King et al., 1988).
Following an abstinence period, dendritic morphology returns to normal. The
overall response of the animal or person to chronic ethanol, then, is likely to
reflect the entire ethanol history.
A Physiological Correlate of Memory Formation
In 1973, Bliss and Lomo demonstrated that following a high frequency
conditioning train to the perforant path fibers which synapse onto dentate
granule cell in the dentate gyrus, a long-lasting (up to 3 weeks) increase in

13
synaptic efficacy to subsequent single pulse stimuli occurred. The enhanced
synaptic efficacy was manifested as an increased amplitude and decreased
latency to onset of the PS of granule cells. The enhanced synaptic efficacy was
later termed long term potentiation and was described in a variety of central
nervous system (CNS) structures. Since its initial description, LTP has been
considered to be a physiological correlate of learning and memory because of its
enduring nature, the types of physiological stimuli which produce it (theta
pattern stimulation) and the anatomical substrates in which it is found (Teyler
and DiScenna, 1984). LTP is defined as an enduring or permanent enhancement
in the functional, biochemical and/or morphological elements of synaptic
transmission, which results in a long lasting enhancement of the efficacy of
synaptic transmission. The physiological consequence of LTP is an increase in
amplitude and slope of the extracellular recorded EPSP and a decrease in latency
to onset of the EPSP. The changes in EPSP reflect an increase in the amplitude
and/or decrease in threshold for activation of the intracellular recorded EPSP
(Schwartzkroin and Wester, 1975). LTP also results in an increase in the
amplitude of the extracellular PS, due to a decrease in threshold and increase in
firing frequency of pyramidal cells for a given stimulus (Chavez-Noriega et al.,
1990).
The mechanisms underlying LTP have been most thoroughly examined in
the hippocampus. Before discussing the results from these studies, it is
important to briefly describe the anatomical connections within this region (Fig.
2-1). The transverse plane of the hippocampus is generally arranged as a
trisynaptic circuit (Amaral and Witter, 1989). The input fibers from ipsilateral
and contralateral entorhinal cortex form the medial and lateral perforant path.
These fibers synapse onto the dendrites of the granule cells of the dentate gyrus.
The granule cells send projections (mossy fibers) to synapse on the CA3

14
pyramidal cells. CA3 pyramidal cells send projections ipsi- and contralaterally to
form the Sch/Com fibers in CA1. These fibers form synapses onto the CA1
pyramidal cells as well as inhibitory interneurons of CAI. CA1 pyramidal cells
also synapse on inhibitory interneurons (basket cells) which then synapse onto
CA1 pyramidal cells. The synaptic interconnections between CA1 pyramidal
cells and inhibitory interneurons mediate feedforward and feedback inhibition of
pyramidal cell function.
In response to single pulse stimulation of the Sch/Com fibers,
presynaptically released glutamate initially stimulates postsynaptic a-amino-3-
hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors. Subsequent flux
of Na+ and K+ through the AMPA channel results in a short onset and brief
duration postsynaptic depolarization (Hestrin et al., 1990). Postsynaptic NMDA
receptors are only minimally activated by glutamate under this condition due to
a voltage dependent channel block by Mg2+. In addition, single pulse
stimulation results in release of GABA from inhibitory interneurons, which
activates postsynaptic GABAa and GABAg receptors. Activation of GABAa
receptors results in an inward CP current with a short latency (3 ms or less) and
brief duration (225 ms) at 32 C (Davies, et al., 1990). Activation of the
postsynaptic GABAg receptor results in G-protein activation and subsequent
outward K+ conductance of longer latency (30 ms) and duration (700 800 ms) at
32 C.
Delivery of a high frequency train of pulses to the Sch/Com fibers results
in a greater postsynaptic depolarization due to repeated AMPA receptor
activation. The enhanced postsynaptic depolarization relieves the Mg2+ block of
the NMDA receptor, resulting in an additional inward Ca2+ and Na+
conductance through the glutamate activated NMDA receptor channel
(Dingledine, 1983; Harris et al., 1984; Mayer and Westbrook, 1984; Mayer et al.,

15
CAI
CA3
Figure 2-1. Schematic representation of the transverse plane of the hippocampus
to illustrate the connections of the trisynaptic circuit (Abbreviation are as follows:
CA1, corpus ammonis area 1 of hippocampus; EC, entorhinal cortex; DG, dentate
gyrus; MF, mossy fibers; PP, perforant path; SC, Schaffer collateral and
Commissural fibers).
1984; Nowak et al., 1984; Hestrin, et al., 1990; Regehr and Tank, 1990).
Postsynaptic GABAa and GABAb activation also occurs. Synaptically released
GABA can act back upon presynaptic GABAb receptors located on the terminals
of GABAergic interneurons, resulting in a G-protein mediated decrease in Ca2+
conductance and a subsequent decrease in GABA release (Davies et al., 1990;
1991). The GABAb mediated events occur with a latency of 200 400 ms. The
NMDA and GABAb mediated synaptic transmission play only a minimal role in
responses to single pulse stimulation. In the presence of plasticity inducing high
frequency conditioning trains, both the NMDA and GABAb receptor systems
produce a profound effect on the resulting pre- and postsynaptic response.
LTP can be conceived as a sequence of events which can be separated for
experimental purposes into 1) an induction process and 2) a process to maintain
synaptic enhancement. Induction of LTP requires cooperativity of pre- and

16
postsynaptic elements such that there is synchronous or near synchronous
presynaptic activation and postsynaptic depolarization (Gustafsson et al., 1987).
Current flows through the NMDA receptor channel only when synaptically
released glutamate binds to the postsynaptic receptor in combination with
postsynaptic depolarization. This allows relief of the voltage dependent Mg2+
block of the NMDA receptor and Ca2+ influx. Ca2+ entry can activate
postsynaptic protein kinases such as Ca2+/Calmodulin II, protein kinase C,
protein kinase A, and tyrosine kinases. These kinases are thought to provide the
substrates for maintenance of synaptic enhancement (Finn et al., 1980; Mody et
al., 1984; Hu et al., 1987; Lovinger and Routtenberg et al., 1988; Malinow et al.,
1988; Reyman et al., 1988; Malenka et al., 1989).
A consensus view of the nature of synaptic changes responsible for
maintenance of LTP has not yet been reached. Kauer et al. (1988) report
increased AMPA mediated postsynaptic current immediately following LTP
induction. This result was based on the evidence that following an LTP
conditioning train, the AMPA and not the NMDA response (in the presence of
antagonists to the AMPA channel and low Mg2+) to single pulse stimulation was
increased. Within 30 seconds, though, the NMDA response as well as the AMPA
response was increased. The increase in NMDA current returned to baseline
levels within four minutes after cessation of the LTP conditioning stimulus. On
the presynaptic side, several researchers have demonstrated an immediate and
persistent increase in glutamate release in response to LTP induction (Bekkers
and Stevens, 1990; Malinow and Tsein, 1990; Malinow, 1990). Using similar
techniques of quantal analysis, Kullmann and Nicoll (1992) and Manabe et al.
(1992) provided evidence supporting both pre- and postsynaptic changes. Using
in vivo LTP induction and in vitro (HPLC) measurement of neurotransmitter
release, Ghijsen et al. (1992) report that endogenous release of both glutamate

17
and GABA increase following LTP induction. Based on these combined results,
it is currently thought that both pre- and postsynaptic mechanisms contribute to
LTP maintenance.
The amount of potentiation of the intracellularly recorded EPSP is related
to the amount of postsynaptic depolarization during induction when the
postsynaptic depolarization is paired with presynaptic activation (Gustafsson et
al., 1987). With extracellular stimulation, the postsynaptic depolarization is
produced by a high frequency afferent stimulation. It appears that a minimum
number of afferent fibers or synaptic contacts must be activated to generate LTP.
Short duration conditioning trains will produce less postsynaptic depolarization
and therefore smaller amounts of LTP than will longer duration trains.
The expression of LTP is a result of combined potentiation of excitatory
and inhibitory synaptic transmission (Abraham et al., 1987; Morishita and Sastry,
1991). The relative contribution of excitatory synaptic transmission typically
predominates over the inhibitory contribution so that the resulting PSP reflects a
relative decrease in overall inhibitory influence.
Objectives of the Dissertation
Since LTP is a candidate substrate for memory acquisition, and the
hippocampus serves a vital role in memory formation, it is reasonable to
hypothesize that the hippocampal vulnerability to the effects of chronic ethanol
toxicity may contribute to the amnesic symptoms of alcoholism. A necessary
requisite for the behavioral disorder, though, is a parallel physiological disorder.
It is therefore critical to determine the plasticity related physiological effects of
chronic ethanol toxicity in the hippocampus.
The hypothesis tested by the experiments presented in this dissertation is
that chronic ethanol treatment is associated with transient and/or permanent

18
alterations in one or more components of the induction process of long term
potentiation in the hippocampus. To test this hypothesis, I investigated 1) the
effect of CET on induction of LTP in the presence and absence of GABAa
synaptic transmission; 2) the effect of recovery from CET on induction of low
levels and maximal magnitude LTP in the presence and absence of GABAa
mediated synaptic transmission; 3) changes in the binding characteristics of
[3H]MK-801 to the NMD A receptor channel following CET; 4) changes in the
efficacy of glutamate to enhance NMDA receptor channel function; 5) the effect
of CET on the efficacy of the endogenous ligand GABA to stimulate GABAa
receptor function and the efficacy of a GABAa receptor antagonist, bicuculline, to
inhibit GABA-stimulated GABAa receptor function.

CHAPTER 3
GENERAL METHODS
Male Long Evans hooded rats (200-250 g) were purchased from Charles
River. A two-week holding period was allowed following arrival to allow for
adjustment to the animal care facility. All animals were housed individually in
stainless steel cages in a colony room with an automatic 7 AM to 7 PM light cycle.
Animals within a particular shipment were paired by weight and assigned
to one of two liquid diet treatment groups: 1) An ethanol group (group E) in
which ethanol comprised 35-39% of the total caloric intake (8.1 9.4% v/v,
ethanol); and 2) A sucrose group (group S) which received an identical diet
except sucrose was isocalorically substituted for ethanol. The remaining calories
were supplied through Sustacal (Mead-Johnson Co.). Both diets were fortified
with Vitamin Diet Fortification Mixture, 0.3 g/100 ml and Salt Mixture XIV, 5.0
g/1 (ICN Nutritional Biochemicals, Cleveland). Both diets contained 1.3 kcal/ml
and provide several times the daily requirement of all essential vitamins and
nutrients (Walker and Freund, 1971; Walker, et al., 1980).
The amount of ethanol diet consumed per Group E animal each day was
measured and recorded. An equal volume of sucrose diet was given on the
following day to the pair-fed Group S animal. Pair-feeding group S and group E
animals ensured that each of a sucrose-ethanol pair received the same volume of
diet and thus the same caloric intake during the treatment period. The liquid
diets were prepared daily and animals received their daily allowance of diet
between 8 and 9 AM. The liquid diets were administered as the sole source of
food for a period of 28 weeks. The percentage of calories in the form of ethanol
19

20
or sucrose was increased by 1% every 4 weeks. A total of 34 sucrose-ethanol
pairs were used in this series of studies. The average daily intake of ethanol
using this paradigm was 12.30 0.17 g/kg/day. All rats were weighed weekly.
Rats maintained on these diets gain weight normally. At the end of the 28 week
period all animals received laboratory chow and water ad libitum for a period of
48 hours or 5-7 months, at which time the acute experiment was performed.

CHAPTER 4
THE EFFECTS OF CHRONIC ETHANOL TREATMENT ON LONG TERM
POTENTIATION IN THE HIPPOCAMPUS
Introduction
Chronic abuse of alcohol results in pathological deterioration of
anatomical, physiological and behavioral correlates of cognitive performance.
The neurotoxic damage extends into several brain structures including
thalamus, cerebellum, cerebral cortex and hippocampus. Of particular
consequence is the pathology of anterograde amnesia, a profound and
devastating symptom of chronic abuse (Freund, 1970; Freund and Walker,
1971; Walker and Freund, 1973; Walker and Hunter, 1978; Berocochea et al.,
1989). In order to better understand the underlying pathology responsible for
this anterograde amnesia, the permanent effects of chronic ethanol toxicity
(CET) on a physiological correlate of memory, long term potentiation (LTP)
was investigated. LTP is defined as an enduring or permanent enhancement
in the functional, biochemical and/or morphological elements of synaptic
transmission (Bliss and Lomo, 1973; Kennedy, 1989). Induction of LTP
requires influx of Ca2+ through the N-methyl-D-aspartate (NMDA) receptor
ionophore located in the postsynaptic membrane (Lynch et al., 1983; Mayer
and Westbrook, 1984; Mayer et al., 1984; MacDermott et al., 1986; Malinow and
Miller, 1986; Kauer et al., 1988; Malenka et al., 1988). During single pulse
excitatory synaptic transmission, activation of the hippocampal afferents fails
to open the NMDA receptor ionophore because of blockade by Mg2+ in a
voltage dependent manner. However, with high frequency stimulation of
21

22
the afferent fibers, the postsynaptic membrane is sufficiently depolarized to
remove the Mg2+ block. The subsequent influx of Ca2+ through the NMDA
receptor ionophore is thought to trigger a sequence of events leading to a
sustained enhancement of functional synaptic efficacy (Dingledine, 1983;
Harris et al., 1984; Kauer et al., 1988; Malenka et al., 1988; Mayer and
Westbrook, 1984; Mayer et al., 1984; Neuman et al., 1987; Nowak et al., 1984;
Regehr and Tank, 1990; Reyman et al., 1989; Thibault et al., 1989).
Recent research suggests that acute ethanol intoxication reduces
current flow through the NMDA channel (Lovinger et al., 1989; 1990), reduces
LTP induction (Sinclair and Lo, 1986; Blitzer et al., 1990) and reduces NMDA
receptor binding affinity (DeMontis et al., 1991). During ethanol withdrawal,
NMDA receptor binding increases (Grant et al., 1990). However, NMDA
receptor binding or function have not been examined in a period of
abstinence following chronic ethanol treatment and withdrawal
A few studies have examined the effects of chronic ethanol exposure
on LTP, as demonstrated by changes in neuronal excitability, but none have
studied synaptic plasticity. Using the hippocampal slice preparation, Durand
and Carien, (1984b) determined that in area CA1 the number of slices capable
of expressing LTP was decreased following chronic ethanol treatment but if
LTP was induced, the magnitude of potentiation was not different from
controls. These experiments only measured the compound action potential
(PS) responses of pyramidal cells and were therefore more a measure of LTP
induced changes in cell excitability (i.e., threshold to activation of an action
potential and/or firing frequency) than LTP (which is synaptic in origin). In
vivo, there was no change in the early time course or magnitude of LTP, but a
statistically nonsignificant trend toward early decay in area CA1. In the
dentate gyrus, LTP was unchanged (Abraham et al., 1981; 1984). Again, the

23
recordings in this study were taken from the cell body layer (PS) and therefore
not a direct examination of the synaptic component of LTP.
An alternate series of studies have suggested that chronic ethanol
ingestion may exert much of its effect by reducing the influence of synaptic
inhibition in the hippocampus. For example, 20 weeks of CET followed by 8
weeks of recovery results in decreased feedforward and recurrent inhibition
onto pyramidal cells of hippocampal area CA1 (Abraham et al., 1981; Rogers
and Hunter, 1992) Additionally, CET reduces the amplitude of the K+
mediated afterhyperpolarization and the GABA mediated inhibitory
postsynaptic potential (IPSP) in CA1 pyramidal cells (Durand and Carien,
1984a).
While NMDA receptor activation and subsequent Ca2+ entry are
necessary and sufficient for LTP induction to occur, GABAergic synaptic
transmission may modulate the magnitude of LTP (Gustafsson, et al., 1987;
Abraham et al., 1987; Morishita and Sastry, 1991). For example, in CA1
Abraham et al. (1987) described an enhancement of both excitatory and
inhibitory synaptic transmission following LTP induction, although
enhancement of excitatory synaptic transmission was greater than that of
inhibitory synaptic transmission. Morishita and Sastry (1991) recorded
intracellular events following LTP induction and confirmed these results.
Both the glutamate mediated EPSP and the GABAa mediated IPSP were
enhanced following LTP induction. Gustafsson and coworkers (1987) used
pairings of single pulse afferent stimulation with postsynaptic depolarization
to demonstrate that the amount of depolarization required to induce LTP was
reduced if GABAa receptors were blocked with the CL channel antagonist,
picrotoxin. Taken together, these results suggest that GABAergic synaptic
transmission can be involved in both the amount of glutamate stimulated

24
depolarization necessary to induce LTP and the relative contribution of
excitatory and inhibitory influences to the resultant synaptic potentiation.
Since NMDA receptor activation is required for LTP induction and
GABAergic synaptic transmission modulates both the magnitude of LTP
induction and the characteristics of the LTP produced, CET induced alteration
of either of these receptor systems could result in altered LTP.
The primary aim of this study was to examine the effect of CET on LTP
induction in area CA1 of the hippocampus and the potential modulatory
influence of GABAergic synaptic transmission on this process.
Methods
Treatment Methods
Male Long Evans hooded rats (200-250 g) were matched by weight and
age and assigned to one of two liquid diet treatment groups described in detail
in Chapter 3, General Methods. Briefly, the diet treatment consisted oftwo
groups: 1) An ethanol group (group E) in which ethanol comprised 35-39% of
the total caloric intake. The remaining calories were supplied by Sustacal
(Mead-Johnson Co.) 2) A sucrose group (group S) which received an identical
diet except sucrose was isocalorically substituted for ethanol. Group S
animals were individually pair-fed with group E animals such that each of a
sucrose-ethanol pair received the same volume of diet during the treatment
period. The liquid diets were administered as the sole source of food for a
period of 28 weeks. At the end of the 28 week period all animals received
laboratory chow and water for 48 hours. Food and water intake were
measured daily during the 48 hour period. A group of similar age animals
housed under similar conditions as group S and E animals but receiving
laboratory chow and water during the entire treatment period were included

25
in the study (group C). The preceding protocol of chronic ethanol treatment
(CET) has been previously used as a valid model for studying the behavioral,
morphological, and physiological consequences of chronic alcohol abuse
(Walker and Freund, 1971; 1973; Walker and Hunter, 1978; Walker et al., 1981;
1982).
Electrophysiological Methods
Slice preparation
Electrophysiological records were taken from multiple hippocampal
slices in each of 9 group S, 9 group E and 8 group C animals 50 60 hours
following cessation of the 28 week liquid diet treatment. All animals were
coded throughout data collection and analysis to avoid experimenter bias.
Transverse sections (400 pm) were cut through the ventral hippocampus
using a Mclllwain tissue chopper. Slices were incubated in a holding chamber
containing artificial cerebral-spinal fluid (aCSF) NaCl, 125 mM; KC1, 3.3 mM;
KH2PO4,1.25 mM; MgSCT*, 4.0 mM; CaCl2. 4.0 mM; NaHCC>3, 25 mM; glucose,
10 mM at room temperature. The pH was maintained at 7.4 with 95% O2/ 5%
CO2. Forty to sixty minutes prior to the onset of recording, a slice was
transferred to a Haas type interface recording chamber maintained at 32 C.
The slice was superfused at 1 ml/min with oxygenated medium of the same
composition as described for the holding chamber.
LTP induction protocol
At the beginning of each recording session, a stimulating electrode was
placed in stratum radiatum (SR) of CA1 midway between stratum pyramidal
and stratum lacunosum-moleculare. A glass recording micropipette filled
with 4 M NaCl was placed in SR of CA1, 800 1000 pm from the stimulating
electrode (Fig. 4-1A). Recordings of field EPSPs in SR were obtained in

26
B
Figure 4-1. Microelectrode placement and representative waveforms from
the hippocampal slice preparation.
(A) Schematic of a transverse hippocampal slice to illustrate electrode
placement. A bipolar stimulating electrode (S) was placed in stratum
radiatum (SR) of area CA1. An extracellular recording microelectrode (R) is
positioned in SR of area CA1 approximately 800 pm from the stimulating
electrode.
(B) Representative field EPSP waveforms recorded from CA1 SR in response
to 0.04 ms duration single pulse stimulation (a) before, and (b) 28 min
following a 100 Hz, 50 pulse conditioning train delivered to the Sch/Com
afferent fibers, (c) Field EPSP slope was measured between the time points at
10% and 90% of maximal EPSP amplitude. Calibration pulse is 2 ms duration,
2 mV amplitude.

27
response to 0.05 Hz test pulses of Schaffer Collateral and Commissural
(Sch/Com) fibers in SR (Fig. 4-1B). Although the conditioning trains which
induce LTP potentiate the field EPSP over a wide range of stimulus strengths,
we empirically determined that the maximal percent change in slope occurs
at stimulus strengths which produce a preconditioning field EPSP slope 75 to
100% of maximal. For this reason the single pulse stimulus strength was
adjusted to produce a field EPSP slope 75% of maximal.
Following an initial 12 min recording period, the stimulus strength
was adjusted to produce a maximal EPSP and a conditioning train of 50
pulses at 100 Hz was delivered. After this conditioning train, the stimulus
strength was returned to the previous level and recordings in response to
single pulse stimulation were made at 20 sec intervals for 28 min. A
subsequent conditioning train and test pulse recordings were taken repeating
this protocol except the duration of the train was lengthened to 100 pulses.
Just prior to each conditioning train and again at the end of the recording
session several test pulses covering a range of stimulus strengths were
delivered and the field EPSP slope recorded to demonstrate the magnitude of
potentiation over a range of stimulus strengths (input/output data)
LTP induction protocol in the absence of GABAergicA synaptic transmission
Experiments involving the removal of inhibitory synaptic
transmission were performed with 3.5 (iM bicuculline methiodide, a specific
GABAa receptor antagonist. This concentration of bicuculline was found to
produce maximal facilitation of LTP induction without producing bursting
and post-burst depression in response to a conditioning train. Bicuculline
was added to the superfusate 6 min after the onset of recording and
maintained until the end of the recording session. Under this condition,
EPSPs produced in response to single pulse stimulation reflect the activation

28
of primarily excitatory inputs. The treatment can result in longer burst
duration to single pulse stimulation. CA3 pyramidal cells normally fire in
bursts which compound the bursting activity in the postsynaptic CA1
pyramidal cell (Wong and Prince, 1979). To prevent uncontrollable
pyramidal cell bursting and resulting postburst depression, an acute
transection under microscopic control was made in each slice between area
CA1 and CA3. This disconnected the CA3 pyramidal cell bodies from the
Sch/Com fibers. In the CA1 region of the hippocampus, stimulus frequencies
of 0.1 Hz or less are normally ineffective in altering the amplitude, slope or
duration of the field EPSP (Alger and Teyler, 1976). In the presence of
bicuculline, single pulse stimulation at 10 sec intervals resulted in a small
increase in field EPSP slope and/or amplitude in rare instances. Increasing
the interpulse interval to 20 sec, when recording from slices treated with
bicuculline, prevented potentiation of the EPSP in response to single pulse
stimulation and resulted in a stable baseline recording. If EPSP slope did
change by more than 10%, the slice was not used in the LTP experiments.
The procedure for collecting single pulse data and induction of LTP was
the same as described for the previous LTP induction experiments except that
the conditioning train frequency was reduced to 50 Hz. In the presence of
bicuculline, a 100 Hz conditioning train can produce burst firing of CA1
pyramidal cells and post burst depression rather than LTP. By lowering the
conditioning train frequency to 50 Hz, LTP induction is reliably produced
with no evidence of prolonged postsynaptic cell firing or post burst
depression (Gustafsson and Wigstrom, 1987).
Blockade of LTP with D-amino-5-phosphonovalerate
A separate group of animals (S and E) were studied to determine
whether the LTP induction mechanism, either in the presence or absence of

29
bicuculline methiodide, is mediated via activation of the NMDA receptor
ionophore. To do this, an LTP inducing conditioning train was delivered in
the presence of 50 pM D-amino-5-phosphonovalerate (APV) to block the
NMDA receptor ionophore. Any potentiation requiring activation of the
NMDA receptor would be blocked by this treatment. A washout period of 40
to 50 min was allowed to remove the APV from the slice. Another
conditioning train was then delivered as described in the initial protocol.
Data analysis
Field EPSP data were passed through a Grass P-511 preamplifier,
digitized at 20 kHz and stored on an IBM AT computer for later analysis. Data
collection and analysis were performed using software developed in this
laboratory. Measurements were made of field EPSP slope (from 10% to 90% of
peak EPSP amplitude; Fig. 4-1B).
EPSP slope data were averaged over the last 10 min of the recording
period following a conditioning train. These data were grouped by duration
of pulse train and analyzed by treatment. Postconditioning train field EPSP
slope (Lt) was expressed as percent of preconditioning EPSP slope (Lo):
Lt
EPSP slope (% baseline) = lq x
and graphed as a function of time with the onset of the first conditioning
train being t=0.
Input/output data were expressed as EPSP slope (mV/ms).
Postconditioning train EPSP slope data were analyzed by a two-way analysis of
variance (ANOVA) with repeated measures with diet treatment as one factor
and duration of pulse train as the other factor.

Results
30
The purpose of these experiments was twofold. The first aim was to
determine the effects of CET on the magnitude of LTP produced in response
to conditioning trains of varying durations. Secondly, I examined the relative
effects of CET on NMDA versus GABAergic synaptic mechanisms.
CET Actions on LTP Induction
Recordings were taken from multiple slices in 9 group S, 9 group E and
8 group C animals 50 60 hours following the cessation of the liquid diet
treatment. Previous experiments in our laboratory have dealt with the effects
of CET following extended periods of recovery from the liquid diet treatment.
The experiments described in this chapter are among the first to test such
functional CET effects after such a short abstinence period. The experimental
paradigm used here allows one to determine substrates of chronic ethanol
toxicity without acute withdrawal effects and with minimal recovery. While
this provides valuable information about the substrates and timecourse of
ethanol effects, it was imperative to first ensure that despite a change in diet
from liquid to solids and water, animals would still consume food. Food and
water intake were measured every 24 hours for the two day abstinence period
and comparisons made between groups E and S (Table 4-1). There were no
differences in the quantity of intake between the two groups. A group of 8
chow-fed animals was included in these experiments to examine any
differences in responses which may be accounted for by the Sustacal diet
treatment alone.
Stimulating and recording electrode placement is illustrated in Figure
4-1A. Pyramidal cells from CA3 send projections along the transverse plane
of the hippocampus, into CA1 to form the Sch/Com fibers (Amaral and
Witter, 1989). These fibers form excitatory synapses onto the CA1 pyramidal

31
Table 4-1. Average daily chow and total water intake during the chronic diet
abstinence period.
Chow intake (g) + SEM
H20 intake (ml)/48 hr
SEM
Day 1
Day 2
Sucrose3
13.4 2.0
10.5 .75
66 12
Ethanol3
14.1 1.4
12.5 1.6
73 11
aANOVA with repeated measures of sucrose vs. ethanol was not significantly
different (P < 0.42).
cells as well as onto inhibitory interneurons of CA1. A bipolar stimulating
electrode was placed in SR of area CA1. Single test pulses of 0.04 ms duration
were delivered at 20 sec intervals to the Sch/Com fibers. The postsynaptic
dendritic response to single pulse stimulation of the Sch/Com fibers was
recorded in SR of CA1 (Fig. 4-1B). This extracellular EPSP represents the near
synchronous activation of multiple synapses. When measuring the
extracellular recorded EPSP, slope or amplitude measurements are most
commonly taken. If the stimulus pulse is suprathreshold for activation of
action potentials, the peak of the EPSP can be masked by the extracellular
recorded population spike (compound action potential from near
synchronous firing of multiple pyramidal cells). The slope of the EPSP,
though, is less vulnerable to interference from activation of a population
spike. For this reason measurements were taken of EPSP slope in the present
experiments. Average baseline (pretetanus) EPSP slope values + SEM were
0.66 + 0.06, 0.82 + 0.09 and 0.83 + 0.14 for groups C, S, and E respectively.
Following an LTP inducing conditioning train, the EPSP response to single
pulse stimulation increased in amplitude and slope (Fig 4-1B).

32
Figure 4-2. EPSP slope before and after two successive conditioning trains.
(A) Representative data expressed as percent of baseline. Baseline EPSP slope
recordings were obtained in response to 0.05 Hz single test pulses of Sch/Com
fibers in SR (t= -13 0 min). At t= 0, a 50 pulse, 100 Hz conditioning train was
delivered. Single test pulses were delivered and recordings taken at 0.05 Hz
for 28 min. A subsequent conditioning train of 100 pulses, 100 Hz was
delivered at t= 28 min. Single test pulse recordings were again taken for 28
min. Each conditioning train was followed by increased EPSP slope which
decayed to a stable potentiated level after several minutes (comparison of the
magnitude of potentiation within animals, ANOVA, P < 0.001).
(B) EPSP slope plotted as a function of stimulus strength. Baseline EPSP slope
measurements were taken prior to any conditioning train and 28 minutes
following the 50 and 100 pulse trains. EPSP slope increased as a function of
duration of the conditioning train over a range of stimulus strengths.

33
Initial extracellular field responses were recorded for 12 minutes.
Following this period of stable single pulse recordings, a conditioning train of
50 pulses at 100 Hz was delivered. In 9/9 group E, 9/9 group S and 7/8 group
C animals, the response consisted of an abrupt increase in EPSP slope
followed by a decay of several minutes to a stable potentiated level (Fig. 4-2A).
The subsequent 100 pulse, 100 Hz train further enhanced the potentiation of
EPSP slope in 7/9 group C, 7/9 group S and 6/8 group E animals. In instances
where the conditioning train did not produce LTP, post-train EPSP responses
consisted of an abrupt change in EPSP slope followed by a decay over several
minutes to near preconditioning levels. Responses to single pulses of
varying stimulus strengths taken 28 min following each conditioning train
demonstrate the EPSP potentiation over a range of stimulus strengths (Fig. 4-
2B). There were no apparent differences between groups C, S, and E with
respect to the early phase of decay of LTP or in the ability to maintain the
potentiation for the duration of the recording period.
All data were evaluated for inclusion in group statistical analysis based
on the following criteria: 1) Preconditioning single pulse responses must
vary with a SD< 10%. 2) Single pulse responses for the final 10 minutes of the
28 minute recording period following a conditioning train must vary with a
SD< 10%. For each animal and experimental condition, the first slice to meet
the criteria was included in the analysis.
Although the pattern of potentiation and the frequency with which
potentiation occurred among slices was similar in group S, E and C animals,
the magnitude of the potentiation in response to all conditioning trains was
reduced in group E as compared with group S (Fig. 4-3; P< 0.08). The average
magnitude of LTP for group C animals was similar to group S and larger than
group E (Table 4-2). The standard error for group C, though, was large enough

150-

O
Cf)
P+
£
w
100-
I
50
# PULSES
SUCROSE n= 9
ETHANOL n= 8
I
100
Figure 4-3. Mean EPSP slope expressed as percent of baseline following each of
the conditioning trains for group S and E animals. The magnitude of LTP of
group E animals was reduced following both the 50 and 100 pulse
conditioning trains. Data were analyzed by a two-way ANOVA with the
pulse trains as a within animal repeated measure. The treatment effect
approached statistical significance at P< .08.
Table 4-2. Mean EPSP slope SE expressed as percent of baseline following
each of the conditioning trains for groups C, S and E animals.
Number of pulses in
conditioning train
Chronic Treatment
Fifty
One Flundred
Laboratory chow & H2O
172 23
228 47
Sucrose
167 + 8
204 14
Ethanol
149 7
173 11

35
to prevent comparisons of it with group E from approaching any level of
statistical significance.
LTP Induction in the Absence of GABAergica Synaptic Transmission
Recordings were taken from multiple slices from the same group S, E
and C animals included in the first series of experiments. Stimulating and
recording electrode placement is again illustrated in Figure 4-1A.
Initial recordings were taken with aCSF as the superfusate. Following 6
minutes of stable single pulse recordings, 3.5 pM bicuculline methiodide was
added to the superfusate and single test pulse recordings again taken for 6-12
minutes. Blockade of GABAa receptors with bicuculline reduces the
functional inhibition of pyramidal cells. The resulting EPSP reflects primarily
the action of excitatory inputs. Bicuculline also facilitates the induction of
LTP. In the presence of bicuculline, a shorter duration or lower frequency
conditioning train can produce LTP.
LTP was recorded in each slice in response to two conditioning trains
which again differed only in duration. Following the initial 50 pulse, 50 Hz
conditioning train, EPSP slope and amplitude abruptly increased (Figs. 4-4 and
4-5). A decay in EPSP slope then occurred over the next several minutes and
stabilized at a new potentiated level in 8/8 group C, 8/9 group S and 9/9 group
E animals. A subsequent 100-pulse, 50 Hz conditioning train again produced
an immediate increase in EPSP slope which decayed to a level of higher
potentiation in 6/8 group C, 4/9 group S, and 6/9 group E animals. All
animals responded with LTP to at least one of the conditioning trains.
All data were evaluated for inclusion in group statistical analysis based
on the same criteria as for LTP induction in the presence of GABAergic
synaptic transmission. Neither the pattern nor the magnitude of potentiation

36
Figure 4-4. Representative field EPSP waveforms recorded in the presence of
3.5 pM bicuculline methiodide. Recordings were taken from SR of CA1 in
response to 0.04 ms duration single pulse stimulation (a) before, and (b) 28
min following a 50 Hz, 50 pulse conditioning train delivered to the Sch/Com
afferent fibers. Calibration pulse is 2 ms duration, 2 mV amplitude.
following each of the conditioning trains was altered in group E animals as
compared with groups S or C (Fig. 4-6; Table 4-3).
The dependence of LTP induction on activation of the NMDA receptor
ionophore is illustrated in Figure 4-7. In both group E and S animals, a 50
pulse, 50 Hz conditioning train failed to induce LTP when 50 pM APV, a
competitive antagonist to the NMDA receptor, was included in the
superfusate. Following a 25 minute washout period, a second train of 50
pulses, 50 Hz produced a sustained potentiation. APV was equally potent at
blocking LTP in the absence of bicuculline.
Discussion
Chronic ethanol exposure resulted in a reduction in the magnitude of
LTP under conditions in which excitatory and inhibitory synaptic
transmission contribute to LTP induction. This effect was seen following

37
Figure 4-5. EPSP slope before and after each of two successive conditioning
trains delivered in the presence of bicuculline methiodide.
(A) Representative data expressed as percent of baseline. Bicuculline (3.5 pM)
was added to the superfusate at t= -6 min and remained throughout the
experiment. Baseline EPSP slope recordings were obtained in response to 0.05
Hz single test pulses of Sch/Com fibers in SR (t= -13 0 min). At t= 0, a 50
pulse, 50 Hz conditioning train was delivered. Single test pulses were
delivered and recordings taken at 0.05 Hz for 28 min. A subsequent
conditioning train of 100 pulses, 50 Hz was delivered at t= 28 min. Single test
pulse recordings were again taken for 28 min. Each conditioning train was
followed by increased EPSP slope which decayed to a stable potentiated level
after several minutes.(comparison of the magnitude of potentiation within
animals, ANOVA, P < 0.001).
(B) EPSP slope plotted as a function of stimulus strength. EPSP slope
increased as a function of duration of the conditioning train over a range of
stimulus strengths.

38
250
OJ
5
cS
W
Ph
o
hJ
CD
Ph
cn
Ph
w
200-
150
100-

T"
50
# PULSES
SUCROSE n=9
ETHANOL n=8
1
100
Figure 4-6. Mean increase in EPSP slope following each of the conditioning
trains for group S and E animals. The magnitude of LTP in group S and E
animals was similar following both the 50 and 100 pulse conditioning trains.
Data were analyzed by a two way ANOVA with the pulse trains as a within
animal repeated measure (P< .80).
Table 4-3. Mean EPSP slope SE expressed as percent of baseline following
each of the conditioning trains for groups C, S and E animals.
Number of pulses in
conditioning train
Chronic Diet Treatment
Fifty a
One Hundred3
Laboratory chow & H20
167 11
191 10
Sucrose
169 17
184 20
Ethanol
168 20
201 24
a All values are in the presence of 3.5 pM bicuculline methiodide.

39
g 150-
(8
X)
PJ
Ph
O
hJ
1/5
C-c
£
w
100-
50-
TT
-10
50 |iM APV
r rrn i i 11 | i i 11 | n i i | i i 11 | i i i i | 11 i i | i i i 11 i i i i | 11 i i | i i i 11 i i i i | i
0 ^ 0 ^ 10 20 30 40 50
TIME (minutes)
washout
bicuculline 3.5 mM
remained in superfusate
Figure 4-7. Representative slice demonstrating the ability of APV to block
LTP induction in the presence of bicuculline methiodide. Bicuculline (3.5
pM) and 50 pM APV were added to the superfusate at t= 6 min. A 50 pulse,
50 Hz conditioning train at t= 0 min produced an abrupt increase in EPSP
slope which decayed to preconditioning levels within two minutes. A 25
minute washout period with bicuculline remaining in the superfusate was
begun at t= 3 min. Subsequently, a 50 pulse, 50 Hz conditioning train (t= 28
min) produced an abrupt increase in EPSP slope which decayed to a stable
potentiated level.
both the 50 and 100 pulse trains, that is, over a range of stimulus conditions.
Responses to conditioning trains appeared normal after removal of
GABAergicA influences, suggesting that CET had little or not effect on the
NMDA mediated component of LTP induction per se. Instead, it appears that
CET interferes with the mechanism by which GABAergic inhibition
modulates LTP induction. CET did not block the LTP induction process
entirely, because LTP induction occurred in response to at least one of the
conditioning trains in all animals. Rather, CET reduced the magnitude of
potentiation in response to each conditioning train. A tenable hypothesis is

40
that GABAergic modulation altered LTP induction by increasing the
threshold for LTP induction. The present study provided no evidence of a
CET effect on early stages of LTP maintenance. Therefore, it is most likely that
CET does not affect the pre- or postsynaptic substrates responsible for this
component of LTP.
The mechanism(s) through which inhibitory synaptic transmission
acts to modulate the induction of LTP are poorly understood. In area CA1,
single pulse activation of the Sch/Com fibers results in release of glutamate
from excitatory synaptic terminals. Some of these terminals synapse onto
inhibitory interneurons. The inhibitory interneurons in turn synapse onto
the CA1 pyramidal cells. Activation of inhibitory interneurons leads to
release of GABA. GABA acts upon postsynaptic GABAa and GABAb
receptors to produce a fast onset, short duration, CL mediated IPSP and a
slower onset, longer duration, K+ mediated IPSP (Alger and Nicoll, 1982).
Additionally, GABA can act upon presynaptic GABAb autoreceptors to reduce
subsequent release of GABA (Davies, et al., 1990; 1991). Since the response to
activation of the postsynaptic GABAb receptors and autoreceptors occurs with
a latency of 200 400 ms, these physiological events do not contribute to the
single pulse response. However, the duration of an LTP conditioning train
ranges anywhere from 200 ms to 1 second. Activation of the GABAb
autoreceptors reduces postsynaptic GABAa and GABAb currents through
decreases in GABA release with a latency of 200 ms or greater (Davies et al.,
1991; Pacelli et al., 1991). The GABAb produced reduction of GABA release is
critical for LTP induction to occur. Blockade of these autoreceptors with CGP
35348 during the conditioning train blocks LTP induction (Davies et al., 1991).
One of the most potent modulators of NMDA receptor activation and hence
LTP induction is depolarization of the postsynaptic membrane (Malinow and

41
Miller, 1986; Gustafsson et al., 1987). The most likely mechanism through
which GABAb receptor blockade prevents LTP induction is by sustained
GABA release and hence postsynaptic hyperpolarization. In the cerebellum,
postsynaptic GABAb receptor activation produces a reduction in GABAa
mediated Cb uptake (Hahner et al., 1991). A similar mechanism, if present in
hippocampus, would further enhance postsynaptic depolarization. Through
presynaptic and perhaps postsynaptic mechanisms, GABAb modulation of
GABAergic function has a powerful effect on LTP induction.
The data presented in this chapter can be explained in a number of
ways: 1) CET may produce an upregulation of postsynaptic GABAa receptor
number, 2) CET may produce an enhancement in the efficacy of GABA
and/or bicuculline at the GABAa receptor ; 3) CET may produce a decrease in
GABAb receptor function thus enhancing GABAa function, 4) CET may
produce an enhancement of presynaptic GABA release due to reduced
function of the presynaptic GABAb receptors, or hyperinnervation from
inhibitory interneurons. Abraham et al. (1981) and Rogers and Hunter (1992)
demonstrated that chronic ethanol exposure decreased functional recurrent
inhibition in CA1. Therefore, it seems unlikely the present results could be
due to increased GABA release, unless such increases occurred only during
high frequency conditioning trains. Alternatively, antagonist blockade of the
CET produced enhancement in GABAa receptor number or function would
effectively remove an abnormally strong and enduring hyperpolarizing block
of NMDA receptor activation. The hypothesis I have derived from this study,
is that CET increases either 1) the efficacy of GABA and/or bicuculline at the
GABAa channel or 2) the number of GABAa channels. Future work must
include measurements of current flow and chloride uptake through
postsynaptic GABAa channels, GABAa and B receptor binding studies, and

42
testing the ability of GABAg receptor agonist and antagonists to alter LTP
induction following CET.
Although NMDA receptor function does not seem to play a role in
producing the LTP deficit, it may be involved in CET produced excitotoxic cell
loss. Walker et al. (1981) showed pyramidal cell loss after prolonged exposure
to an ethanol diet. Short term exposure to ethanol (1-2 weeks) results in a
dramatic upregulation of NMDA receptor number during withdrawal which
returns to normal levels within 24 hours (Gulya et al., 1991). A further phase
of chronic ethanol toxicity may reveal enduring changes in receptor number
or function to coincide with the principal cell loss.
The large variability in results taken from chow-fed animals may have
been due to differences in age, shipment lot of animals, or housing
conditions. The chow-fed animals varied in age from 12 to 14 months old (2 -
4 months older than the liquid diet treated animals). Regardless of the cause,
it is clearly preferable to compare results from two groups which have
received the same treatment except for the variable (ethanol) one wishes to
test. It may be wise in the future, though, to include age-matched sucrose
treated animals in an initial test of protocol to adjust for differences in
responses.
The observed deficits in LTP induction provide evidence that an
alteration(s) in GABAergic function or receptor number is a potential
substrate for at least part of the pathology of memory loss associated with CET.
For an alteration of LTP to be involved in memory loss, it must be resistant to
any recovery induced by prolonged ethanol abstinence. In the next chapter, I
have tested this by examining the effects of a 5 7 month recovery period
from chronic ethanol toxicity on LTP induction and its modulation by
GABAergic synaptic transmission.

CHAPTER 5
THE EFFECT OF A RECOVERY PERIOD FROM CHRONIC ETHANOL
TOXICITY ON LONG TERM POTENTIATION
Introduction
Chronic ethanol toxicity is associated with a variety of central nervous
system (CNS) morphological, behavioral and functional deficits. During the
period of ethanol ingestion as well as with abstinence from a chronic ethanol
diet, adaptive and/or compensatory changes as well as recovery can occur.
For example, in the dentate gyrus, cell loss due to chronic ethanol treatment
(CET) is accompanied by increases in the dendritic extent of the surviving
granule cells (Durand et al., 1989), and increases in spine density which return
to normal following a 20 week recovery period from ethanol (King et al.,
1988). The concomitant morphological degeneration and recovery suggest
that hippocampal functional properties may also change as a consequence of
abstinence. It is therefore important to determine the permanence of the
functional losses accompanying CET.
The previous chapter described experiments demonstrating a reduction
of long term potentiation (LTP) in the hippocampus as a result of chronic
ethanol toxicity. The deficit was observed only when GABAergic modulation
of LTP induction was present. This suggested that CET produced an alteration
in GABAergic function which then acted to diminish the effectiveness of
conditioning trains to activate the NMDA receptor complex and thereby
induce LTP. It was hypothesized that this deficit may in part contribute to the
behavioral deficits in acquisition of new memories, which is a permanent,
43

44
nonrecoverable consequence of CET. In order for the deficit in LTP to be
involved with the deficits in memory acquisition, it too must be present
following extensive ethanol abstinence. In the following sets of experiments,
the permanent effects of CET on induction of LTP and the role of GABA
synaptic transmission in the manifestation of these deficits were examined.
To do this, animals previously treated for 28 weeks with an ethanol
containing diet were allowed a period of 5- 7 months abstinence. The capacity
of hippocampal synapses to support LTP both in the presence and absence of
GABAergic blockade was examined.
Methods
Treatment Methods
Male Long Evans hooded rats (200-250 g) were matched by weight and
age and assigned to one of two liquid diet treatment groups described in detail
in Chapter 3, General Methods. Briefly, the diet treatment groups consisted
of: 1) An ethanol group (group E) in which ethanol comprised 35-39% of the
total caloric intake. The remaining calories were supplied by Sustacal (Mead-
Johnson Co.); 2) A sucrose group (group S) which received an identical diet
except sucrose was isocalorically substituted for ethanol. Group S animals
were individually pair-fed with group E animals such that each of a sucrose-
ethanol pair received the same volume of diet during the treatment period.
The liquid diets were administered as the sole source of food for a period of 28
weeks (Walker and Freund, 1971; 1973; Walker and Hunter, 1978; Walker et
al., 1981; 1982). At the end of the 28 week period all animals received
laboratory chow and water ad libitum for a period of 5-7 months, at which
time the acute experiment was performed.

45
Electrophysiological Methods
Slice preparation
Electrophysiological data were collected from a total of 16 group S and
17 group E animals within a period 5-7 months following cessation of the 28
week liquid diet treatment. All animals were coded throughout data
collection and analysis to avoid experimenter bias. Transverse sections (400
pm) were cut through the ventral hippocampus using a Mclllwain tissue
chopper. Slices were incubated in a holding chamber containing NaCl, 125
mM; KC1, 3.3 mM; KH2P04/ 1.25 mM; MgS04, 2.0 mM; CaCl2, 1.9 mM;
NaHCCb, 25 mM; glucose, 10 mM at room temperature. The pH was
maintained at 7.4 with 95% 02/ 5%C02. Forty to sixty minutes prior to the
onset of recording, a slice was transferred to a Haas type interface recording
chamber maintained at 32 C. The slice was superfused at 1 ml/min with
oxygenated medium of the same composition as described for the holding
chamber.
LTP induction protocol
Electrophysiological records were taken from multiple slices in each of
9 group S and 8 group E animals. At the beginning of each recording session,
a stimulating electrode was placed in stratum radiatum (SR) of CA1 midway
between stratum pyramidal and stratum lacunosum-moleculare. A glass
recording micropipette filled with 4 M NaCl was placed in SR of area CA1, 800
- 1000 pm from the stimulating electrode (Fig. 5-1A). Recordings of field
EPSPs in SR were obtained in response to 0.1 Hz test pulses of Schaffer
Collateral and Commissural (Sch/Com) fibers in SR (Fig. 5-1B). Although the
conditioning trains which induce LTP potentiate the field EPSP over a wide
range of stimulus strengths, we empirically determined that the maximal

46
Figure 5-1. Microelectrode placement and representative waveforms from
the hippocampal slice preparation.
(A) Schematic of a transverse hippocampal slice to illustrate electrode
placement. A bipolar stimulating electrode (S) was placed in stratum
radiatum (SR) of area CA1. An extracellular recording microelectrode (R) was
positioned in SR of area CA1 approximately 800 pm from the stimulating
electrode.
(B) Representative field EPSP waveforms recorded from CA1 SR in response
to 0.04 ms duration single pulse stimulation (a) before, and (b) 15 min
following a 100 Hz, 30 pulse conditioning train delivered to the Schaffer
Collateral and Commissural (Sch/Com) afferent fibers, (c) Field EPSP slope
was measured between the time points at 10% and 90% of maximal EPSP
amplitude. Calibration pulse is 2 ms duration, 2 mV amplitude.

47
percent change in slope occurs at stimulus strengths which produced a
preconditioning field EPSP slope 75 to 100% of maximal. For this reason the
single pulse stimulus strength was adjusted to produce a field EPSP slope 75%
of maximal.
Following an initial 10 minute recording period, the stimulus strength
was adjusted to produce a maximal EPSP and a conditioning train of 30
pulses at 100 Hz was delivered. After this conditioning train, the stimulus
strength was returned to the previous level and recordings in response to
single pulse stimulation were taken at 0.1 Hz for 12 min. Subsequent
conditioning trains and test pulse recordings were taken repeating this
protocol except the duration of each train was lengthened in increments as
follows: t=10 min, 40 pulses; t=20 min, 50 pulses; t=30 min, 60 pulses. We
have previously determined that to obtain an incremental increase in the
magnitude of LTP the duration of the conditioning train must increase in an
incremental fashion. Repetition of the 30 pulse train will not result in an
incremental increase in potentiation. Just prior to each conditioning train
and again at the end of the recording session, several test pulses covering a
range of stimulus strengths were delivered and the field EPSP slope recorded
to demonstrate the magnitude of potentiation over a range of stimulus
strengths (input/output data).
In a subset of group S and E animals, test pulse recordings were taken at
0.1 Hz for 28 min following each of the conditioning trains to demonstrate
the long lasting nature of LTP.
LTP induction protocol in the absence of GABAergic synaptic transmission
Experiments involving the removal of inhibitory synaptic
transmission were performed with 5.0 pM bicuculline methiodide, a specific
GABAa receptor antagonist. This concentration of bicuculline was found to

48
produce maximal facilitation of LTP induction without producing bursting
and post-burst depression in response to a conditioning train. Bicuculline
was added to the superfusate 6 min after the onset of recording and remained
in the superfusate until the end of the recording session. Under this
condition, EPSPs produced in response to single pulse stimulation reflect the
activation of primarily excitatory inputs. The treatment can result in longer
burst duration to single pulse stimulation. CA3 pyramidal cells normally fire
in bursts which compound the bursting activity in the postsynaptic CA1
pyramidal cell. To prevent uncontrollable pyramidal cell bursting and
resulting postburst depression, a cut was made between area CA1 and CA3.
This disconnected the CA3 pyramidal cell bodies from the Sch/Com fibers.
Additionally, the concentration of Mg2+ in the superfusate was increased
from 2.0 to 4.0 mM to decrease activation of NMDA receptor channels in
response to single pulse stimulation.
In the CA1 region of the hippocampus, stimulus frequencies less than
0.1 Hz are normally ineffective in altering the amplitude, slope or duration of
the field EPSP (Alger and Teyler, 1976). In the presence of bicuculline, single
pulse stimulation at 0.1 Hz resulted in a small increase in field EPSP slope
and/or amplitude in rare instances. Reducing the stimulus frequency to 0.05
Hz when recording from slices treated with bicuculline prevented
potentiation of the EPSP in response to single pulse stimulation and resulted
in a stable baseline recording.
Electrophysiological records were taken from multiple slices in each of
9 group S and 8 group E animals. Electrode placement is as described in the
LTP induction protocol and Figure 1A. Test stimuli consisted of 0.04 ms
constant current (120 1000 pA) pulses. Field EPSPs were recorded in SR in
response to 0.05 Hz test pulses of Sch/Com fibers.

49
Single pulse stimulus strength was adjusted to produce a field EPSP
slope 75% of maximal. Following an initial 6 min baseline recording period
in the presence of bicuculline, the stimulus paradigm for LTP induction was
begun. The stimulus strength was adjusted to produce a maximal EPSP and a
conditioning train of 50 pulses at 50 Hz was delivered. After the conditioning
train, the stimulus strength was returned to the preconditioning level and
recordings in response to single pulse stimulation were taken at 0.05 Hz for 28
minutes. A subsequent conditioning train and test pulse recordings were
taken repeating this protocol except the duration of the train was lengthened
to 100 pulses. Just prior to each conditioning train and again at the end of the
recording session several test pulses covering a range of stimulus strengths
were delivered and the field EPSP slope recorded to demonstrate the
magnitude of potentiation over a range of stimulus strengths (input/output
data).
Blockade of LTP with D-amino-5-phosphonovalerate
A separate group of animals (S and E) were studied to determine
whether the LTP induction mechanism (under these stimulus conditions),
either in the presence or absence of bicuculline methiodide is mediated via
activation of the NMDA receptor ionophore. To do this we delivered an LTP
inducing conditioning train in the presence of 50 pM D-amino-5-
phosphonovalerate (APV) to block the NMDA receptor ionophore. Any
potentiation requiring activation of the NMDA receptor would be blocked by
this treatment. We then allowed a washout period of 40 to 50 min to remove
the APV from the slice. Another conditioning train was then delivered and
test pulse recordings taken.

50
Data analysis
Field EPSP data was passed through a Grass Model P511 preamplifier,
digitized at 20 kHz and stored on an IBM AT computer for later analysis. Data
collection and analysis was performed using software developed in this
laboratory. Measurements were made of field EPSP slope (from 10% to 90% of
peak EPSP amplitude; Fig IB).
EPSP slope data were averaged over the last 5 min of recording after a
conditioning train for experiments conducted in the absence of bicuculline
and the last 10 min in the presence of bicuculline. Data were grouped and
analyzed by treatment. Postconditioning train field EPSP slope (Lt) is
expressed as percent of preconditioning EPSP slope (Lo):
Lt
EPSP slope (% baseline) = j x 100
and graphed as a function of time with the onset of the first conditioning
train being t=0.
Input/output data were expressed as EPSP slope (mV/ms). Post
conditioning train EPSP slope data was analyzed by a two-way analysis of
variance (ANOVA) with repeated measures with diet treatment as one factor
and duration of conditioning train as the other factor.
Results
CET Effect on LTP Induction
The purpose of the present investigation was to determine the
permanent effects of CET on the capacity of hippocampal synapses to develop
LTP. Since many of the alterations associated with CET recover after a
sustained period of abstinence, we investigated the enduring nature of deficits
in LTP following a 5-7 month recovery period.

51
Stimulating and recording microelectrode placement within the CA1
subfield of the hippocampus is illustrated in Fig. 5-1A. The postsynaptic
dendritic response to single pulse stimulation of the Sch/Com fibers is
depicted in Fig. 5-1B. This extracellular EPSP represents the near synchronous
activation of multiple synapses. Following an LTP inducing conditioning
train, the EPSP response to single pulse stimulation increases in amplitude
and slope (Fig. 5-1B).
Field EPSP slope in response to 0.1 Hz single pulse stimulation was
recorded for 12 min prior to delivery of the first conditioning train. The
average baseline (pretetanus) EPSP slope values + SEM were 1.36 + 0.20 and
1.02+ 0.13 mV/ms for group E and S respectively. In 8/9 group S, and 7/8
group E animals, the response to a 30 pulse, 100 Hz conditioning train
consisted of an abrupt increase in EPSP slope followed by a decay of several
minutes to a stable potentiated level (Fig. 5-2A). A subsequent 40 pulse, 100
Hz conditioning train further enhanced the potentiation of EPSP slope in 7/9
group S, and 4/8 group E animals. By the 50 pulse, 100 Hz train, slices from
7/9 group S and 6/8 group E animals were maximally potentiated. EPSP
potentiation occurred in all group S and E animals in response to at least one
of the conditioning trains. Responses to single pulses of varying stimulus
strength taken 12 min following each conditioning train demonstrate the
EPSP potentiation over a wide range of stimulus strengths (Fig. 5-2B).
In a few animals from each group, postconditioning train recordings
were continued for an additional 15 minutes to examine the early stages of
LTP maintenance. Groups S and E were similar in their ability to maintain a
stable potentiated state for the duration of the 28 minute recording period
(Fig. 5-3). The pattern of potentiation following each of the conditioning

52
Figure 5-2. EPSP slope before and after each of four successive conditioning
trains.
(A) EPSP slope expressed as percent of baseline after successive conditioning
trains in a representative animal. Baseline field EPSP slope recordings were
obtained in response to 0.1 Hz single test pulses of Sch/Com fibers in SR (t =
-10 0 min). At t = 0, a 30 pulse 100 Flz conditioning train was delivered.
Single test pulses were delivered and recordings taken at 0.1 Hz for 12 min.
Subsequent conditioning trains were delivered as follows: t = 12 min, 40
pulses; t = 24 min, 50 pulses; t = 36 min, 60 pulses. Each of the conditioning
trains produced progressive, stable potentiation of the EPSP slope ( P< 0.001).
(B) EPSP slope plotted as a function of stimulus strength. The EPSP slope
increased as a function of duration of the conditioning train over a range of
stimulus strengths.

53
O!

5
X>
W
Ph
O
hJ
en
Ph
£
w
50-
^|ini|ini|ini|nii|nii|niipr-
-5 0 5 10 15 20 25
TIME (minutes)
Figure 5-3. EPSP slope expressed as a percent of baseline in a representative
group E animal. At t = 0 min a 50 pulse, 100 Hz conditioning train was
delivered to the Sch/Com afferent fibers. Single pulse recordings were
taken at 0.1 Hz for 28 min to demonstrate both the similar time course of
the early decay in EPSP slope and the long-lasting nature of LTP expression
in group E animals.
trains and the frequency with which potentiation occurred was similar in
groups S and E animals.
The response in slices lacking sustainable LTP consisted of an abrupt
increase in EPSP slope following the conditioning train which decreased to
the preconditioning baseline within several minutes.
All data were evaluated for inclusion in group statistical analysis based
on the following criteria: 1) Responses to preconditioning single pulses must
be sufficiently stable, as indicated by a standard deviation less than 15%. 2)
Responses to single pulses for the final 5 minutes of the 12 minute recording

54
Figure 5-4. Mean change in EPSP slope ( SEM) following each conditioning
train for group S and E animals. Both the minimal level for the induction of
LTP (30 pulses) and subsequent increments in the magnitude of LTP
induction including asymptotic levels were reduced in group E animals (P <
0.06).
period following a conditioning train must also vary with a standard
deviation less than 15%. For each animal and experimental condition, the
first slice to meet the criteria was included in the analysis. Figure 5-4
illustrates the magnitude of LTP in response to each conditioning train. All
pulse trains produced markedly less LTP in group E as compared with group S
( P < .06).
In most cases (7/9 group S and 6/8 group E), asymptotic potentiation
was reached by the 50 pulse train with no further potentiation upon delivery
of the 60 pulse train. To determine if the maximal capacity for potentiation is
reduced in group E animals, the percent potentiation for each animal at

55
asymptote was compared between groups without consideration of the
duration of the conditioning train needed to produce that level of
potentiation. Maximal potentiation was markedly reduced in group E
animals as compared with group S (mean + SEM, 203% + 17 group S and
172% + 10 group E; P< 0.08, ANOVA). These results indicate that not only
does CET permanently reduce long term synaptic efficacy in response to a
single conditioning train, but the maximal response that can be achieved is
reduced.
LTP Induction in the Absence of GABAergic^ Synaptic Transmission
The purpose of this set of experiments was to determine whether CET
exerts a direct influence on NMDA mediated induction of LTP or via an
alteration of GABAergic modulation of LTP induction.
Recordings were taken from multiple slices in 7 group S and 9 group E
animals. Stimulating and recording electrode placement is illustrated in Fig.
5-1A. Extracellular synaptic field responses were recorded in SR of CA1 in
response to the single test pulse stimulation. Initial recordings were taken
with aCSF as the superfusate. Following 6 min of stable recordings, 5.0 pM
bicuculline methiodide was added to the superfusate and single test pulse
recordings again taken for 6- 12 min (Fig. 5-5). Blockade of GABAa receptors
with bicuculline reduces the functional inhibition of pyramidal cells. The
resulting EPSP reflects primarily excitatory inputs. Bicuculline also facilitates
the induction of LTP such that a shorter duration or lower frequency
conditioning train is capable of inducing LTP. In initial pilot experiments, a
range of bicuculline concentrations and frequency of single test pulse
stimulation of the Sch/Com fibers were evaluated. A concentration of 5.0 pM
bicuculline and test pulse frequency of 0.05 Hz were determined empirically
to not significantly alter the slope or amplitude of the test pulse recordings in

56
Figure 5-5. Representative field EPSP waveforms recorded in the presence of
5.0 pM bicuculline methiodide. Recordings were taken from CA1 SR in
response to 0.04 ms duration single pulse stimulation (a) before, and (b) 28
min following a 50 Hz, 50 pulse conditioning train delivered to the Sch/Com
afferent fibers. Bicuculline methiodide (5.0 ¡iM) was added to the superfusate
6 min prior to the onset of recordings and remained throughout the
experiment. Calibration pulse is 2 ms duration, 2 mV amplitude.
most cases. If EPSP slope did change by more than 10%, the slice was not used
in the LTP experiments. The average baseline EPSP slope values + SEM were
0.53 + 0.08 mV/ms and 0.43 0.04 mV/ms for group S and E respectively.
LTP was recorded in each slice in response to two conditioning
stimulus trains which differed only in duration. The first, a 50 pulse, 50 Hz
train produced an increase in EPSP slope and amplitude. The timecourse of
the EPSP responses to single pulse stimulation following the initial train is
illustrated in Fig 5-6A. In 6/7 group S animals and 9/9 group E animals, the
increase in EPSP slope decayed over the course of several minutes to a stable
potentiated level. The initial change in EPSP slope returned to

57
TIME (minutes)
B
Figure. 5-6. EPSP slope before and after each of two successive conditioning
trains delivered in the presence of bicuculline methiodide.
(A) Representative data expressed as percent of baseline. Baseline EPSP slope
recordings were obtained in response to 0.05 Hz single test pulses of Sch/Com
fibers in SR (t= -6-0 min). At t= 0, a 50 pulse, 50 Hz conditioning train was
delivered. Single test pulses were delivered and recordings taken at 0.05 Hz
for 28 min. A subsequent conditioning train of 100 pulses, 50 Hz was
delivered at t= 28 min. Single pulse test recordings were again taken for 28
min. Each conditioning train was followed by increased EPSP slope which
decayed to a stable potentiated level after several minutes.
(B) EPSP slope plotted as a function of stimulus strength. The EPSP slope
increased as function of duration of the conditioning train over a range of
stimulus strengths.

58
preconditioning levels within several minutes in the remaining group S
animal. A subsequent 100 pulse, 50 Hz conditioning train further enhanced
EPSP slope potentiation in 5/7 group S and 6/9 group E animals. All animals
exhibited LTP in response to at least one conditioning train. EPSP responses
to single pulses of varying stimulus strength taken 28 min following each
conditioning train demonstrate the EPSP potentiation over a wide range of
stimulus strengths (Fig. 5-6B). Neither the pattern of potentiation following
each of the conditioning trains nor the magnitude of potentiation was altered
in group E animals as compared with group S (Fig. 5-7).
The dependence of LTP induction on activation of the NMDA receptor
ionophore is illustrated in Figure 5-8. Fifty micromolar APV added to the
superfusate 5 min prior to delivery of a conditioning train blocks the
induction of LTP in both group S and E animals. Following a 40 50 min
washout period, LTP induction occurs in response to the conditioning train.
Discussion
The main conclusions drawn from these experiments are: 1) the
magnitude of long term potentiation is reduced as a result of chronic ethanol
exposure; 2) the reduction of LTP is a permanent consequence of CET and not
dependent on sustained ethanol ingestion; 3) CET affects the reduction of
LTP by altering some aspect of GABAergic synaptic transmission on LTP
induction; 4) CET does not overtly affect early phases of LTP maintenance; 5)
CET does not alter the functional capacity of the NMDA receptor complex to
induce LTP.
In the previous chapter, evidence was provided that a deficit in LTP
induction appears as early as 48 hours following abstinence from the chronic
diet. This deficit was masked by blockade of GABAa receptors. Here,

59
240-1
QJ
% 200-
5 180-
O

SUCROSE n=7
ETHANOL n=9
100
50
100
# PULSES
Figure 5-7. Mean EPSP slope following each of the conditioning trains for
group S and E animals. The magnitude of LTP of group E animals was not
reduced following either the 50 or 100 pulse conditioning train. Data were
analyzed by a two-way ANOVA with the pulse trains as a within animal
repeated measure (P< 0.38).
evidence is provided that the reduction in LTP and the ability of GABAergic
blockade to mask the deficit remains after a prolonged recovery process.
Studies of GABAergic interneuronal cell number provide evidence
that chronic ethanol toxicity affects this population of neurons. Lescaudron,
et al. (1986) described a permanent reduction in GABA immunoreactive
interneurons in CA1 of ventral hippocampus. Physiological findings of CET
produced effects on GABAergic synaptic transmission are somewhat more
complex. Intracellular recordings have demonstrated a reduction of the
GABAa mediated IPSP amplitude and K+ mediated afterhyperpolarization
(Durand and Carien, 1984a) Other studies have demonstrated that CET

60
.w
p-q-yn //* i 1 > i > > i > i i 1 i
| 0 30 min 40 50 60 70 80 90
TIME (minutes)
Figure 5-8. Representative slice demonstrating the ability of APV to block
LTP induction. In the presence of 50 pM APV, a 50 pulse, 100 Hz conditioning
train produced an abrupt increase in EPSP slope which decayed to
preconditioning levels within several minutes. Following a 40 minute
washout period t= 48, a 50 pulse, 100 Hz conditioning train produced an
abrupt increase in EPSP slope which decayed to a stable potentiated level. A
second train t= 72 further increased the magnitude of LTP.
produced a reduction in recurrent and feedforward inhibition onto CA1
pyramidal cells (Abraham et al., 1981; Rogers and Hunter, 1992). While CET
appears to be result in a reduction of inhibitory input onto pyramidal cells,
other research has shown that the postsynaptic responsiveness to GABA is
enhanced following CET. Rogers (1986) used extracellular responses to single
pulse stimulation to test the effectiveness of iontophoretic application of
GABA. He found that exogenous GABA application resulted in a larger
reduction of population spike (PS) amplitude in CET animals than in pair-fed
controls. Following iontophoresis of the specific GABAa antagonist,
bicuculline methiodide, the PS response to subsequent single pulse
stimulation was greater than pair-fed controls. These combined results

61
suggest that the efficacy of GABA and bicuculline at the postsynaptic
membrane are enhanced following CET. Therefore, it appears that CET
results in a reduction of GABA release (most likely from interneuronal cell
loss) and enhanced GABA efficacy at the postsynaptic GABA receptor.
Recent studies have demonstrated a wide variety of GABAa receptor
subunit types, each with a specific regional distribution (Sato and Neale, 1989;
Verdoorn et al., 1990). This leaves open the possibility that CET may affect a
specific subtype of GABAa receptor. In fact, Wafford, et al. (1991) have
provided evidence that the Y2l subunit of the GABAa receptor is specifically
susceptible to the acute effects of ethanol. This particular subunit, though, is
not found in high concentration in the hippocampus. Further research may
provide additional clues to subunit function in the presence of protracted
ethanol toxicity. Another possibility is that basal functioning of GABAergic
synaptic transmission may be unchanged by CET while plasticity related
functioning, as tested in these experiments, may be altered. For example,
activation of the presynaptic GABAb autoreceptor by GABA and the
consequent reduction of GABA release only produces a postsynaptic effect
after 200 400 ms (Davies, et al., 1991). Postsynaptic conductances are reduced
with a similar time course by postsynaptic GABAb receptor activation
(Morrisett et al., 1991; Hahner et al., 1991). Single pulse stimulation,
therefore, would be unaffected by these autoreceptors (Pacelli, et al., 1991).
Under the plasticity related conditions described in these experiments,
conditioning trains would activate the autoreceptors which would in turn
enhance postsynaptic depolarization through reduction of GABA release. No
research to date has examined the effects of CET on plasticity-related function
of the GABAergic system in CA1.

62
Just as the actions of CET may be limited to plasticity-related
GABAergic function, NMDA receptor function may be altered under
conditions unrelated to LTP induction. For example, NMDA receptor
activation is thought to play a major role in excitotoxicity and cell death
(Choi, 1988; Abele et al., 1990). Ischemia induced cell loss or development of
seizures can be prevented by concomitant administration of the glycine
antagonist, kynurenic acid or competitive inhibitors of NMDA receptor
channels, APV or carboxypiperazinephosphonate (Uckele et al., 1989). There
is no evidence to date that demonstrates that an increase in NMDA receptors
is associated with increased LTP. The possibility remains that expression of
NMDA receptors may be increased by CET resulting in greater susceptibility of
neurons to excitotoxic cell death.
Recent studies have provided evidence that localized lesions
involving only the hippocampus or more specifically, area CA1 result in
permanent inability to acquire memories while leaving other aspects of
cognitive performance unaltered (Zola-Morgan et al, 1986; 1992; Squire and
Zola-Morgan, 1991). Several unimodal and polymodal sensory areas send
reciprocal projections to the perirhinal, parahippocampal, and entorhinal
cortex (Insausti et al., 1987). Perirhinal and parahippocampal cortex also send
reciprocal connections to entorhinal cortex which in turn sends a major
connection to the dentate gyrus of the hippocampus. Removal of either
hippocampus or surrounding cortex results in short term memory loss (Zola-
Morgan et al., 1989). Combined lesions of hippocampus, perirhinal, and
parahippocampal cortex results in amnesia of greater severity than removal
of hippocampus or adjacent cortex alone. This suggests that the perirhinal,
parahippocampal cortex and hippocampus each contribute to memory
formation. Loss of any or all of these structures, though, does not impair

63
recall of past events or retrograde memory. When new memories are
formed, recall of these memories is dependent on temporal lobe functioning
for only a limited period of time. Thereafter, the memory is apparently stored
in the neocortex and retrieval of the memory is independent of temporal lobe
function. Certain characteristics of LTP induction in the hippocampus
including induction by physiological stimuli and relatively long duration but
eventual decay both correlate well with the hypothesis that LTP is a
physiological correlate of memory formation in the hippocampus (Squire and
Zola-Morgan, 1991).
The enduring nature of the LTP deficit makes it a candidate substrate
for at least part of the pathology responsible for the memory impairments
associated with CET. It is possible, then, that functional disturbances
produced by CET in the hippocampus may be sufficient to produce
anterograde amnesia.

CHAPTER 6
GABAa RECEPTOR MEDIATED CHLORIDE UPTAKE IN CORTEX AND
HIPPOCAMPUS FOLLOWING CHRONIC ETHANOL EXPOSURE
Introduction
Previous chapters have described a phenomenon in hippocampus in
which chronic ethanol treatment acts to diminish the magnitude of long term
potentiation (LTP). When the hippocampal tissue is exposed to a selective
GABAa antagonist, bicuculline methiodide, the CET induced reduction of LTP is
masked. That is, hippocampal slices from chronic ethanol treated animals
respond as pair-fed controls to a given conditioning train only when that train is
delivered in the presence of bicuculline. A hypothesis to explain these results is
that chronic ethanol treatment (CET) produces an enduring increase in
GABAergic synaptic transmission. The postsynaptic action of enhanced
GABAergic responses counteracts the depolarizing effects of the LTP inducing,
high frequency conditioning train and subsequent NMDA receptor activation.
Experimental blockade of GABAergic synaptic transmission by bicuculline
would allow sufficient postsynaptic depolarization to produce LTP of equal
magnitude to that of controls.
Previous studies have demonstrated that CET reduces recurrent inhibition
onto CA1 pyramidal cells as well as the amplitude and duration of the GABA
mediated inhibitory postsynaptic potential (IPSP) and CA1 GABAergic
interneuron cell number (Abraham et al., 1981; Durand and Carien 1984a;
Lescaudron et al., 1986; Rogers and Hunter, 1992). These data suggest that
GABAergic inputs are actually reduced following CET. Therefore an
64

65
enhancement in inhibitory synaptic transmission would most likely be from a
postsynaptic source.
Rogers (1986) demonstrated that the postsynaptic response to single pulse
afferent stimulation following iontophoretic application of either GABA or
bicuculline was exaggerated following CET. This suggests an enhancement in
the efficacy of GABA and bicuculline at the GABAa receptor. Therefore, the
simplest explanation for this chronic ethanol action is that CET enhances the
efficacy of GABA stimulation of a GABAa mediated chloride current and/or the
efficacy of bicuculline antagonism of GABA stimulated GABAa receptor
activation. Several factors govern the GABAergic contribution to the
postsynaptic membrane potential. The GABAa receptor mediates a short
latency, brief duration hyperpolarization in response to single pulse stimulation
(Curtis et al., 1970; Newberry and Nicoll, 1985; Thalmann, 1988). The GABAa
receptor is linked to a Ch ion channel in the same protein complex. GABAb
receptors mediate a slower latency and longer duration hyperpolarization in
response to single pulse stimulation. GABAb activity is G-protein mediated and
linked to a K+ conductance (Andrade et al., 1986; Dutar and Nicoll, 1988). In
addition to the postsynaptic effects, synaptically released GABA activates
presynaptic GABAb autoreceptors to produce decreased release of GABA in a
time dependent manner (200 400 ms) following the single pulse (Davies et al.,
1990; 1991). This effect is also G-protein mediated and thought to result from a
decrease in presynaptic Ca2+ conductance. In response to high frequency
conditioning trains (as in LTP induction), GABA exerts the above mentioned pre-
and postsynaptic effects during delivery of the train. A diminution in GABA
release in response to high frequency trains has been hypothesized to facilitate
postsynaptic depolarization, thereby enhancing relief of the voltage dependent
Mg2+ block of the NMDA receptor channel and subsequent NMDA receptor

66
activation (Davies et al., 1990; 1991). Subsequent to delivery of a conditioning
train, the AMPA mediated EPSP and GABAa and GABAg mediated IPSPs are
enhanced for prolonged periods (Morishita and Sastry, 1991). The magnitude of
the resulting LTP reflects summation of the potentiation of both excitatory and
inhibitory postsynaptic potentials (Abraham et al., 1987).
As a first step toward understanding the mechanism by which chronic
ethanol exposure disrupts LTP, the efficacy of GABA and bicuculline at the
GABAa receptor was examined by measuring uptake of 36C1 in cortical and
hippocampal membrane preparations.
Materials and Methods
Treatment Methods
Male Long Evans hooded rats (200-250 g) were matched by weight and
age and assigned to one of two liquid diet treatment groups described in detail in
Chapter 3, General Methods. Briefly, the diet treatment consisted of: 1) An
ethanol group (group E) in which ethanol comprised 35-39% of the total caloric
intake. The remaining calories were supplied by Sustacal (Mead-Johnson Co.); 2)
A sucrose group (group S) which received an identical diet except sucrose was
isocalorically substituted for ethanol. Group S animals were individually pair-
fed with group E animals such that each of a sucrose-ethanol pair received the
same volume of diet during the treatment period. The liquid diets were
administered as the sole source of food for a period of 28 weeks. At the end of
the 28 week period all animals received laboratory chow and water ad lib. for
either 2 or 6 months. The preceding protocol of chronic ethanol treatment (CET)
has been previously used as a valid model for studying the behavioral,
morphological, and physiological consequences of chronic alcohol abuse (Walker
and Freund, 1971; 1973; Walker and Hunter, 1978; Walker et al., 1981; 1982).

67
Preparation of Microsacs
Each experiment included one each of an ethanol/sucrose pair. Brains
were removed from decapitated animals and placed on ice. Whole hippocampus
and one-half of the cortex (randomly chosen) were dissected from brains and
placed in 5 ml of ice-cold buffer (145 mM NaCl, 5.0 mM KC1,1.0 mM Mg CI2,1.0
mM CaCl2, 10 mM glucose, 10 mM N-2-hydroxyethylpiperazine-n2-
ethanesulfonic acid, pH 7.5 with Tris). Tissue was homogenized by hand (10
strokes) with a glass-Teflon homogenizer. The homogenate was centrifuged at
900 g for 15 min at 0 C. The supernatant was decanted and the pellet
resuspended in 5 ml of fresh buffer, homogenized (10 strokes) and centrifuged at
900 g for an additional 15 min at 0 C. The supernatant was decanted and the
pellet resuspended in fresh buffer (5 ml for cortex and 4 ml for hippocampus),
homogenized (10 strokes) and placed on ice. Protein content was determined by
the method of Lowry (1951).
Measurement of 36C1 Uptake
The procedure for measurement of 36C1 uptake in microsacs is taken from
Allan and Harris (1986). Microsacs (200 pi) were pre-incubated in a water bath at
34 C for 5 minutes in the presence or absence of bicuculline (see below). Next
200 pi of a solution containing 36C1 (New England Nuclear, Boston, MA, U.S.A.)
and varying concentrations of GABA (0 300 pM) or a constant concentration of
GABA (100 or 300 pM) and varying concentrations of bicuculline methiodide (0 -
100 pM) were added during constant vortexing. In experiments in which the
efficacy of bicuculline to inhibit GABA-stimulated Cl uptake was tested,
bicuculline was added during the preincubation and incubation. After 3 sec, ^Cl
uptake was terminated by the addition of 4 ml ice-cold buffer containing 100 pM
picrotoxin and rapid filtration onto a premoistened (GF/C) Whatman filter,
using a Hoefer manifold (Hoefer Scientific, San Francisco, CA). Filters were

68
washed with an additional 8 ml of ice-cold buffer containing 100 (iM picrotoxin.
The amount of radioactivity on the filters was measured by liquid scintillation
spectrometry. The amount of 36C1 bound to the filters in the absence of
membranes was subtracted from all values. Total Cl uptake was calculated using
36C1 as a tracer. GABA-dependent Cl uptake was calculated by subtracting the Cl
uptake in the absence of agonist (GABA-independent Cl uptake) from the Cl
uptake in the presence of agonist (total Cl uptake).
Data Analysis
Statistical analysis was performed by a two-way analysis of variance
(ANOVA) with repeated measures (treatment vs. agonist or antagonist
concentration). ANOVA was used to compare the difference between two
means.
Results
Effect of Chronic Ethanol on GABA Stimulation of Cl Uptake
All comparisons were first made between the two (n =3 ethanol and 3
sucrose) and four (n = 5 ethanol and 5 sucrose) month abstinence groups. Since
the duration of abstinence did not significantly affect basal or stimulated Cl
uptake within each diet treatment (data not shown), animals from the two
abstinence periods were grouped together.
The objective of this set of experiments was to determine the efficacy of
GABA at the GABAa receptor following chronic ethanol exposure. Basal uptake
was first measured to determine differences in GABA-independent Cl uptake.
Mean values of group S and E animals are listed in Table 6-1. Comparison
between groups failed to demonstrate any difference in basal uptake.

69
Table 6-1. Basal Cl uptake in hippocampus and cortex of chronic ethanol and
sucrose treated animals.
Basal Cl Uptake (nmol/mg protein/3 s) ;SEM
Diet Treatment
Hippocampus
Cortex
Sucrose3
27.2 5.7
24.7 2.3
Ethanol3
21.5 2.2
24.5 3.1
3 Data were analysed by an ANOVA (P < 0.37 for hippocampus and 0.95 for
cortex).
The effects of chronic ethanol on GABA-dependent uptake of Cl was next
examined. Figure 6-1 shows the concentration-response curve for GABA (0 300
(iM) in microsacs prepared from hippocampus and cortex of ethanol and pair-fed
sucrose treated animals. Chronic treatment with ethanol had no effect on the
concentration-response curve for GABA stimulated uptake of Cl. EC50 values for
group E and S are listed in Table 6-2. Comparison of EC50 using an ANOVA
revealed no difference between diet treatment groups (P < 0.37 for hippocampus
and 0.87 for cortex).
Effect of Chronic Ethanol on Bicuculline Inhibition of GABA Stimulated Cl
Uptake
The objective of the next set of experiments was to examine the efficacy of
the GABAa antagonist, bicuculline methiodide, to inhibit maximal GABA
stimulation of Cl uptake in hippocampus and cortex. It has been previously
shown that bicuculline blocks the chronic ethanol induced reduction of LTP.
This capacity of bicuculline to prevent the disruption of LTP induction following
chronic ethanol exposure, may be mediated by an enhanced ability to block
GABA stimulated Cl uptake.
Chloride uptake was first examined in chow-fed control animals to
determine a concentration of GABA which produced maximal stimulation and

70
in hippocampal (A), (B) and cortical (C), (D) microsac preparations. (A) and (C)
represent Cl uptake as a percent of maximal (100 |iM GABA) Cl uptake. (B) and (D)
represent actual values of Cl uptake expressed as nmol/mg protein/3 s of
exposure.

71
Table 6-2. Effect of chronic ethanol treatment on GABA stimulated and
bicuculline inhibition of GABA stimulated Cl uptake.
Hippocampus
Cortex
Sucrose
Ethanol
Sucrose
Ethanol
EC50 GABAa
19.8 4.1
31.7 10.3
29.3 5.0
30.8 7.2
IC50 Bicucullineb
10.7 3.6
16.9 7.5
16.6 6.5
12.7 3.3
a P< 0.37 for hippocampus and 0.87 for cortex
b P< 0.44 for hippocampus and 0.61 for cortex; IC50 was determined under
conditions of maximal GABA stimulation (100 pM).
also exhibited a range of responses to a range of bicuculline concentrations.
Figure 6-2 shows the effects of bicuculline (0 -100 pM) in the presence of either
100 or 300 pM GABA, concentrations which produce maximal stimulation of Cl
uptake. Since the bicuculline antagonism of Cl uptake was complete in 100 pM
GABA but not in 300 pM GABA, all subsequent experiments were performed
with 100 pM GABA.
The efficacy of bicuculline to block GABA-stimulated Cl uptake in
hippocampal and cortical microsacs from animals in groups S and E was then
examined. Comparison of group S and E revealed no difference in the
concentration-response curve for bicuculline (Fig. 6-3; P < 0.48). Table 6-2 lists
the IC50 for bicuculline in hippocampus and cortex from animals in groups S and
E. Again, between-group comparisons revealed no significant differences in
either measure (P < 0.44 for hippocampus and 0.61 for cortex).
Lastly, the effect of bicuculline on basal Cl uptake was determined to test
if bicuculline alone produced an effect on Cl uptake. To do this, uptake in the
presence of 100 pM bicuculline and no GABA in the incubation solution was
measured. This value measures the amount of basal Cl uptake resistant to

72
Figure 6-2. Bicuculline inhibition of GABA stimulated Cl uptake in cortical and
hippocampal microsacs from chow-fed control animals.
Microsacs from chow-fed control animals were tested for the effectiveness of
bicuculline methiodide to block maximal GABA (100 or 300 pM) stimulated Cl
uptake in the hippocampus. Cl uptake was measured at bicuculline
concentrations of 0 -100 pM in (A) hippocampal and (B) cortical microsacs. In the
presence of 100 pM GABA, Cl uptake was blocked by bicuculline concentrations
of 30 pM or higher. When microsacs were incubated with 300 pM GABA,
blockade of Cl uptake was incomplete for bicuculline doses up to 100 pM.
bicuculline blockade. Bicuculline blocked virtually all basal Cl uptake in both
groups S and E. A two-way ANOVA with diet treatment as one factor and total
and bicuculline sensitive basal Cl uptake as the second factor showed no
significant differences (P < 0.89 for cortex and for hippocampus). Measurements
of bicuculline sensitive basal Cl uptake are listed in Table 6-3.

73
Figure 6-3. Chronic ethanol treatment failed to affect bicuculline inhibition of
GABA stimulated Cl uptake in hippocampal (A), (B) and cortical (C), (D)
microsacs. The concentration of GABA was kept constant at 100 pM and
bicuculline methiodide concentration ranged from 0 100 pM. (A) and (C) Data
are expressed as percent of maximal GABA stimulation. (B) and (D) represent
actual values of Cl uptake expressed as nmol/mg protein/3 sec exposure time.

74
Table 6-3. Bicuculline-sensitive basal uptake of Cl in hippocampal and cortical
microsacs.
Cl Uptake (nmol/mg protein/3 s) ;SEM
Diet Treatment
Hippocampus
Cortex
Sucrose3
25.5 6.6
28.3 3.2
Ethanol3
20.2 7.8
29.8 4.8
a Data were analyzed by a two-way ANOVA (P < 0.65 for hippocampus and 0.80
for cortex).
Discussion
The principal aim of these sets of experiments was to measure GABAa
receptor function following chronic ethanol treatment. GABAa receptor function
was assessed by quantifying GABA-dependent uptake of Cl in membrane
preparations. Increased Cl uptake reflects enhanced GABA receptor function to
open the Cl ionophore. No chronic ethanol effects were found in either GABA-
stimulated uptake or bicuculline-inhibition of GABA-stimulated Cl uptake.
Additionally, chronic ethanol exposure did not affect basal (GABA-independent)
Cl uptake. Taken together, these results indicate that chronic ethanol exposure
does not directly affect GABAa receptor channel activation by GABA or
inhibition of GABA activation by bicuculline.
Several previous studies using electrophysiological methods have
demonstrated that chronic ethanol decreases functional inhibition and increases
postsynaptic responsiveness to GABAergic inhibition. For example, Abraham et
al. (1981) and Rogers and Hunter (1992) found that animals chronically treated
with ethanol demonstrated an enduring reduction in functional recurrent
inhibition onto CA1 pyramidal cells. It was hypothesized that the reduction in
functional inhibition resulted from chronic ethanol produced interneuronal cell
loss in CA1. Rogers (1986) went on to test responsiveness of CA1 pyramidal cells
to iontophoretic application of GABA. Excitatory single pulse responses

75
produced in the presence of exogenous GABA were reduced by a greater amount
in chronic ethanol treated animals as compared to pair-fed controls. This
indicated that chronic ethanol exposure produced a heightened sensitivity of
CA1 pyramidal cells to exogenous application of GABA. Excitatory single pulse
responses in the presence of bicuculline were next examined. In this case,
responses were larger in ethanol treated animals as compared with pair-fed
controls, suggesting that chronic ethanol produced a heightened sensitivity of the
postsynaptic membrane to bicuculline. Therefore, chronic ethanol treatment
appears to enhance either postsynaptic GABAa receptor number in CA1 or the
efficacy of both GABA and bicuculline.
In the present experiments, though, no apparent enhancement in GABA
stimulation or bicuculline inhibition of Cl uptake was detected. An important
difference between these two experimental paradigms was that the physiological
studies were localized to the CA1 region of the hippocampus whereas the Cl
uptake experiments described here measured whole hippocampal function.
Perhaps GABAergic susceptibility to the effects of chronic ethanol exposure is
limited to distinct hippocampal subfields such as CA1 which go undetected
because of a dilution from GABAa receptors of dentate gyrus and CA3 regions.
The number of [3H]bicuculline binding sites is greater in the dentate gyrus than
CA1 (Olsen, et al., 1990) as is the magnitude of functional inhibition and its
ability to suppress LTP induction (Steward et al., 1990). In order for chronic
ethanol to have an effect limited to CA1, there must be a distinct element in the
affected region susceptible to chronic ethanol. Such selectivity in ethanol
susceptibility may occur via regional differences in GABAa subunit composition.
There are five different GABA receptor subunit types (a, p, y, 8, and p) and
several variants of each type expressed in the CNS (Tobin et al., 1991). The
various combinations of these subunit variants confer distinct functional

76
properties on the resulting GABA receptor (Verdoorn et al., 1990). The regional
distribution of GABAa receptor subunit variants throughout the CNS may-
produce a diversity of GABA responses, such that each functional type is specific
for the unique function that brain region subserves. Just as GABA receptor
subunit variants may determine specificity of receptor function, responsiveness
of subunit variants to various environmental toxins may be diverse. For
example, the 72-subunit which is thought to affect the size of the open channel
configuration and confers enhanced efficacy of GABA by benzodiazepines
(Pritchett et al., 1989; Sigel et al., 1990), is also sensitive to the acute action of
ethanol (Wafford et al., 1991). Perhaps the basic principle that a neurotoxic effect
can be directed toward a specific receptor subunit type may apply to chronic
ethanol toxicity.
The physiological effect of CET to reduce LTP induction was necessarily
recorded following a high frequency conditioning train (Chapters 4 and 5). The
present results of Cl uptake were measured from animals that had not received
prior LTP induction. It is possible the effect of CET is restricted to plasticity-
related events and therefore not a direct effect on the single pulse functioning of
the GABAa receptor channel. For example, LTP induction results in potentiation
of both the glutamate mediated EPSP and the GABAa mediated IPSP (Abraham
et al., 1987; Morishita and Sastry, 1991). CET could act to enhance the relative
contribution of GABAergicA potentiation to LTP. Such an enhancement of
GABAa LTP would effectively mask LTP of excitatory synaptic transmission.
Only when the GABAa mediated component was blocked by bicuculline, would
LTP of excitatory responses be seen. The CET produced enhancement of
GABAergic responses reported by Rogers (1986) were recorded following a 5
second iontophoretic application of GABA. Here too, it is possible that the

77
prolonged exposure of the postsynaptic membrane to neurotransmitter produced
an effect that a shorter duration of exposure would not.
Future experiments should be aimed at investigating the single cell
responsiveness of GABA and receptor agonists and antagonists both in response
to single pulse stimulation and during LTP conditioning trains. This will provide
detailed information about postsynaptic GABAergic function both during single
pulse and plasticity-related synaptic transmission. The characteristics of the
GABA mediated IPSPs following an LTP conditioning train should be examined
to determine if LTP results in a specific enhancement of IPSP relative to EPSP
following CET. In addition, the numbers and affinities of GABAa and GABAp
receptors measured under equilibrium conditions will provide added
information about GABA receptor responses to CET.

CHAPTER 7
QUANTITATIVE AUTORADIOGRAPHIC ANALYSIS OF NMDA
RECEPTOR BINDING WITH [3H]MK-801 FOLLOWING CHRONIC
ETHANOL CONSUMPTION
Introduction
Recent research suggests that acute exposure to low doses of ethanol
functionally inhibits the N-methyl-D-aspartate (NMDA) receptor channel
while the a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)
receptor remains little affected (Lovinger et al., 1990). Additionally, long term
potentiation (LTP), which is produced by NMDA receptor activation, is
severely reduced by low doses of ethanol (Sinclair and Lo, 1986; Blitzer et al.,
1990). Few studies have examined the enduring or long term effects of
ethanol on NMDA receptor function. In humans, the number of 3-(()-2-
carboxypiperazin-4-yl)-[l,23H]propyl-l-phosphonic acid ([3H]CPP, a
competitive NMDA receptor antagonist) binding sites in hippocampus is
decreased with chronic alcoholism while receptor affinity is increased
(Michaelis et al., 1990). Conversely, the number of [3H]glutamate binding sites
is increased with chronic alcoholism relative to age-matched control patients.
Experiments described in chapters 4 and 5 of this dissertation demonstrate an
enduring reduction in the magnitude of LTP following a 28 week exposure to
ethanol in rats. This disruption in LTP, though, was masked when GABAa
receptors were blocked with the specific antagonist, bicuculline methiodide.
These results suggest that chronic ethanol abuse impairs the mechanism by
which GABAergic inhibition modulates LTP induction rather than through a
direct action on NMDA receptor function.
78

79
A wealth of research has demonstrated that the NMDA receptor not
only functions under normal physiological states as a mediator of long term
synaptic plasticity, but also plays a critical role in certain pathological
conditions such as excitotoxic mediated cell death. For example, upon
withdrawal from a 1 to 2 week period of exposure to low doses of ethanol,
mice and rats demonstrate a heightened susceptibility to seizures as well as
enhanced tolerance to ataxia induced by ethanol (Khanna, et al., 1991). Both
behavioral tolerance and susceptibility to seizures can be blocked by prior
administration of blockers of the NMDA receptor complex. In mice, a 7 day
exposure to low doses of ethanol results in upregulation of [3H]diclozipene
([3H]MK-801) binding sites within the hippocampus (Gulya et al., 1991 and
Grant et al., 1990). The upregulation of the NMDA receptor complex is
thought to be at least partially responsible for the heightened susceptibility to
seizures during ethanol withdrawal as well as behavioral tolerance to
repeated exposure to ethanol (Khanna et al., 1991). In this model, it is not
known what alteration of the NMDA receptor complex occurs following long
term (more than 2 weeks) exposure to ethanol. Neurotoxic effects of ethanol
following a 7 or 14 day exposure are typically transient in nature (Gulya et al.,
1991), while 20 to 28 weeks of exposure results in permanent effects such as
cell loss and alteration of dendritic density and morphology as well as
adaptive responses such as increases in the dendritic extent of surviving cells
(Lescaudron and Verna, 1985; Cadete-Leite et al., 1988; Durand et al., 1989;
Walker et al., 1981).
The aim of the present study was twofold. First, the effect of long term
(5 month) ethanol exposure on NMDA receptor binding was measured to
quantify changes in receptor number and binding affinity. Second, the ability

80
of L-glutamate to enhance [3H]MK-801 binding was investigated as a means of
evaluating NMDA receptor function.
MK-801 binds the NMDA receptor only under conditions in which the
channel is open (Huettner and Bean, 1988; MacDonald and Nowak, 1990).
MK-801 binding to the NMDA receptor channel blocks current flow through
the channel and it is through this mechanism that MK-801 acts as a specific
noncompetitive NMDA antagonist. L-Glutamate activation of the NMDA
receptor produces channel opening, and therefore, MK-801 binding. In
experimental conditions, as the concentration of L-glutamate increases, so
does binding of [3H]-MK-801. The efficacy with which L-glutamate produces
receptor activation and channel opening can be measured by determining
pH]MK-801 binding in the presence of a range of glutamate concentrations.
Since excessive activation of the NMDA receptor complex may be excitotoxic
to neurons, an alteration of NMDA receptor number could be responsible for
morphological alterations such as neurotoxic cell loss following chronic
ethanol treatment. In addition, changes in glutamate stimulation of [3H]-MK-
801 binding may reflect an alteration of the efficacy of glutamate to open the
NMDA receptor channel.
Methods
Treatment Methods
Male Long Evans hooded rats (200-250 g) were matched by weight and
age and assigned to one of two liquid diet treatment groups described in detail
in Chapter 3, General Methods. Briefly, the diet treatment consisted of: 1) An
ethanol group (group E) in which ethanol comprised 35-39% of the total
caloric intake. The remaining calories were supplied by Sustacal (Mead-
Johnson Co.); 2) A sucrose group (group S) which received an identical diet
except sucrose was isocalorically substituted for ethanol. Group S animals

81
were individually pair-fed with group E animals such that each of a sucrose-
ethanol pair received the same volume of diet during the treatment period.
The liquid diets were administered as the sole source of food for a period of 28
weeks. At the end of the 28 week period all animals received laboratory chow
and water ad lib. for 48 hours. Two similar age animals housed under
similar conditions as group S and E animals but receiving laboratory chow
and water during the entire treatment period were included in the study. The
preceding protocol of chronic ethanol treatment (CET) has been previously
used as a valid model for studying the behavioral, morphological, and
physiological consequences of chronic alcohol abuse (Walker and Freund,
1971; 1973; Walker and Hunter, 1978; Walker et al., 1981; 1982).
Autoradiographic Methods
Data were collected from 8 group S, 8 group E and 2 chow-fed control
animals 48 hours following cessation of the 28 week liquid diet treatment.
All animals were coded throughout data collection and analysis to avoid
experimenter bias. Rats were killed by decapitation, the brains removed and
frozen on powdered dry ice and maintained at -70 C. A midsagittal cut was
made through each brain. Each brain half was randomly chosen for
sectioning on the horizontal or coronal plane. Six micron sections were cut
on a cryostat and thaw-mounted on acid washed and gelatin-subbed slides.
Brains were sectioned such that each slide contained one coronal and one
horizontal section from each of an ethanol, sucrose pair. Sections from each
of the two chow treated animals were mounted with one of the sucrose-
ethanol pairs. Sections were refrozen and stored at -20 C for no more than 24
hr prior to use.

82
PH1MK-801 saturation binding assay
The procedure for [3H]MK-801 binding represented a modified protocol
that has been previously described (Monaghan, 1991). Slides were warmed to
room temperature and dried with an air stream prior to preincubation with
buffer. The sections were then preincubated in 50 mM Tris acetate with 0.1%
saponin and 1.0 mM EDTA, (pH 7.7) at room temperature for 10 minutes.
Sections were then rinsed in 50 mM Tris-acetate buffer for 60 min at 30C.
This treatment removes endogenous glutamate, glycine and various ions.
Sections were next incubated with varying concentrations of [3H]MK-801
(0.0005 1 pM; (30 pCi/mM) New England Nuclear, Boston, MA, U.S.A.) in 50
mM Tris buffer containing 25 pM L-glutamate, 20 pM D-(-)-2-amino-7-
phosphonoheptanoic acid (D-AP7), 20 pM glycine, and 250 pM spermine for
60 minutes at room temperature. Non-specific binding was determined in
the presence of 50 pM MK-801.
Following incubation, slides were rinsed in ice-cold buffer containing
20 pM D-AP7 for 60 min. Sections were dried under an air stream and placed
into x-ray cassettes with tritium-sensitive film (Hyperfilm, Amersham,
Arlington Heights, IL). Standards were included along with radioactive tissue
during exposure (microscales, Amersham). Film exposure was 4 weeks at
4C, followed by standard film development in Kodak D-19 at 20C.
Glutamate stimulation of [3H1MK-801 binding
Another group of sections were pre-incubated in 50 mM Tris acetate
with 0.1% saponin and 1.0 mM EDTA, pH 7.7 at room temperature for 10
minutes followed by a rinse in 50 mM Tris-acetate buffer for 60 min at 30 C.
Sections were next incubated in varying concentrations of glutamate (0 25
pM) and 10 nM [3H]MK-801 in 50 mM Tris buffer containing 20 pM D-AP7, 20
pM glycine, and 250 pM spermine for 60 minutes at 0 4 C. Non-specific

83
binding was determined in the presence of 50 (iM MK-801. Sections were
rinsed, dried and processed as described above.
Determination of NMDA Receptor Density
Autoradiograms were analyzed by computer assisted densitometry
with a MCID (Microcomputer Imaging Device, Imaging Research, Inc.) image
analysis system. Densitometric measurements were converted during
analysis to pmol/mg protein binding. Statistical analysis was performed by a
two-way analysis of variance (ANOVA) with repeated measures (treatment
vs. brain region).
Results
The goal of this study was to determine the effects of protracted
exposure to ethanol on NMDA receptor complex number and function in the
central nervous system (CNS).
Effects of CET on (3H1MK-801 Saturation Binding Characteristics
In the first series of experiments, the number of receptor binding sites
and the affinity of [3H]MK-801 for the NMDA receptor channel was
determined in a variety of brain regions. MK-801 binds only to the open state
of the NMDA receptor channel (Huettner and Bean, 1988; MacDonald and
Nowak, 1990). Glutamate is thought to enhance MK-801 binding by
increasing the channel open time. Glutamate is also the endogenous
transmitter for the NMDA receptor channel. Glycine and spermine each bind
to specific and distinct sites on the NMDA receptor (Ransom and Stec, 1988).
Glycine and spermine, in isolation, do not cause NMDA channel opening.
Instead, these ligands act to allosterically enhance the ability of glutamate to
produce channel opening. Prior work has demonstrated that maximal MK-
801 binding occurs when assay conditions include glutamate, glycine and

84
spermine in the concentrations used in these experiments (Ransom and Stec,
1988; Monaghan, 1991).
Autoradiograms reveal a high concentration of [3H]MK-801 binding
sites in hippocampal dendritic fields, dentate gyrus, layer I-III of cortex,
septum and several thalamic structures (Fig. 7-1). Within the hippocampus
and dentate, [3H]MK-801 binding predominates in stratum radiatum and
oriens, and the dentate molecular layer.
Specific [3H]MK-801 binding measurements from each of several brain
regions were taken and data from each brain region of each animal were
plotted versus the concentration of [3H]MK-801 in the incubation solution
(Fig. 7-2). The Kd and Bmax were calculated from the best fit of the binding
isotherm. Table 7-1 lists the values of Kd and Bmax of [3H]MK-801 binding
for several brain regions. The Kd values varied little among brain regions.
The highest affinity binding was found in hippocampus stratum radiatum
and dentate molecular layer. Bmax values were similar to those reported in
the literature (Grant et al., 1990). Among the sucrose and chow treated
animals, regional variations in Bmax were found between dorsal and ventral
stratum radiatum of CA1 (1.961 + 0.05 ventral; 2.331 + 0.07 pmol/mg protein
dorsal; P < 0.0001) and dentate molecular layer (1.756 0.09 ventral; 2.085
0.08 pmol/mg protein, dorsal; P < 0.0007). and Bmax values for sucrose and
ethanol treated animals are listed in Table 7-1. Statistical analysis across brain
regions failed to demonstrate an effect of chronic ethanol exposure on either
Kd or Bmax (ANOVA, P< 0.74, Kd and 0.1, Bmax ).
Previous work has demonstrated regional selectivity of the neurotoxic
effects of chronic ethanol (Walker et al., 1981; Lescaudron and Verna, 1985;

Figure 7-1. NMDA receptor distribution as determined by [3H]MK-801
binding. (A) Non-specific binding was measured with 0.1 pM [3H]MK-801 and
50 pM MK-801. (B) Total binding in the presence of 0.1 pM [3H]MK-801.
Binding density increases from black to white.


87
Figure 7-2. Representative average specific [3H]MK-801 saturation binding.
Binding isotherms depicted were from (A) ventral stratum radiatum of CA1
and (B) parietal cortex layers I III in chronic ethanol treated and sucrose
control animals. Each data point represents the mean + SEM of eight ethanol
or sucrose treated animals. Curves represent the optimal non-linear fitting to
the rectangular hyperbola equation. Binding constants are listed in Table 7-1.
(Inset), Scatchard plot of [3H]MK-801 binding. Line was fit by linear
regression.

88
Table 7-1. Effect of chronic ethanol on [3H]MK-801 binding in various brain
structures.
Treatment
Sucrose
Ethanol
Sucrose
Ethanol
Brain region
Kd (nM) + SEM
Bmax SEM
(pmol/mg protein)
Dentate Gyms
d. Molecular Layer
5.93 0.72
5.77 0.59
2.063 0.09
1.930 0.05
v. Molecular Layer
6.07 1.19
5.61 0.61
1.837 0.08
1.902 0.07
Hippocampus
d. Stratum Radiatum
4.26 0.54
4.61 0.41
2.318 0.09
2.494 0.08
v. Stratum Radiatum
5.48 0.89
5.38 0.58
1.983 0.06
2.059 0.06
v. Subiculum
7.81 2.2
8.67 1.1
0.726 0.07
0.728 0.04
Cortex
d. Cortex (layers I-III)
7.62 1.34
8.82 1.51
1.323 0.05
1.471 0.11
d. Cortex (layers IV- VI)
7.25 1.02
6.59 1.28
0.895 0.04
1.031 0.09
v. Parietal Cortex
7.30 0.64
7.30 0.65
1.455 0.07
1.477 0.05
(layers I-III)
v. Parietal Cortex
5.94 0.62
6.55 1.05
0.856 0.07
0.935 0.07
(layers IV-VI)
Entorhinal Cortex
5.29 0.95
6.64 1.40
0.993 0.08
0.982 0.07
Septum
v. lateral Septum
9.09 1.21
11.6 1.66
1.004 0.05
1.182 0.11
Thalamus
v. Medial Thalamus
9.67 2.2
13.0 3.1
0.777 0.04
0.925 0.06
v. Lateral Thalamus
9.59 1.12
11.9 1.56
0.915 0.06
1.031 0.03
d. Thalamus
14.1 2.08
12.2 2.29
0.863 0.04
0.835 0.02
Data was analyzed by a two-way ANOVA (P < 0.74
for Kd, and 0.1 for Bmax).

89
King et al., 1988; Durand et al., 1989). Hippocampus and dentate gyrus are
among the most affected. When the Bmax from these regions only were
grouped as a repeated measure in an ANOVA, an interaction effect of liquid
diet treatment with brain region was found (P < 0.02). No effect was found
when comparing the Kj among these brain regions.
Effects of CET on Glutamate Stimulation of [3H]MK-801 Binding
As stated earlier, [3H]MK-801 binding is maximally enhanced when
glutamate, glycine and spermine are included in the incubation solution.
Glutamate's enhancement of MK-801 binding occurs in a dose dependent
fashion. Since glutamate is also the endogenous transmitter for the NMDA
receptor channel, one method of evaluating the functional capacity of the
NMDA channel is to examine the ability of glutamate to enhance [3H]MK-801
binding. In the next set of experiments, [3H]MK-801 binding was measured
with various concentrations of glutamate included in the incubation
solution. Quantifying [3H]MK-801 binding as a function of glutamate
concentration measures the ability of glutamate to stimulate channel
opening.
Representative autoradiograms of [3H]MK-801 binding under a range of
L-glutamate concentrations are shown in Figure 7-3. When no glutamate was
included in the incubation solution, only background levels of binding were
detected. Low but detectable levels of binding were found in most brain
regions studied when a low concentration of glutamate (0.05 pM) was used.
pH]MK-801 binding increased in a dose dependent manner with maximal
binding most typically at 10 pM glutamate. Measurements of [3H]MK-801
binding in different concentrations of glutamate from ventral stratum
radiatum and parietal cortex are represented graphically in Figure 7-4. EC50

90
values calculated from curve fitting of these data demonstrate regional
differences among sucrose controls (Table 7-2). These results are consistent
with those previously reported in the literature (Monaghan, 1991). The
regional differences in EC50 are thought to reflect pharmacological diversity of
NMDA receptors in the CNS. A comparison of ECso's between diet treatment
groups across brain regions failed to demonstrate an effect of chronic ethanol
exposure (P < 0.86).
Discussion
The aim of these experiments was to determine the effect of prolonged
ethanol exposure on NMDA receptor number and function in the CNS.
Chronic ethanol treatment did not produce a global change in either the
affinity of [3H]MK-801 for the NMDA receptor channel, the number of
binding sites or the ability of glutamate to activate the receptor and thus cause
channel opening. These results contrast with human studies in which the
number of [3H]CPP binding sites decreased and receptor affinity increased with
chronic alcoholism (Michaelis et al., 1990). Several factors could explain this
discrepancy. First, the nutritionally supplemented diet used in the present
study controlled for any potential effects from alcoholic malnutrition. Such
controls are impossible in human studies. Also, it is difficult to eliminate
influencing factors such as prior drug exposure or disease states in human
studies. By the very nature of the disease of alcoholism, accurate and
complete histories are difficult to obtain.
In addition to the physiologic role of the NMDA receptor as mediator
of long term plasticity, several studies have implicated the NMDA receptor as
partially responsible for seizure generation associated with ethanol
withdrawal and excitotoxic cell death. Several studies have implicated

Figure 7-3. Total [3H]MK-801 binding in a range of concentrations of L-
glutamate and a constant concentration of [3H]MK-801 (10 nM).
Concentrations of L-glutamate in each of the depicted examples were as
follows: (A), 0.05 pM; (B), 0.1 pM; (C), 5.0 pM; (D), 25.0 pM. (A) Minimal
levels of binding were measured from sections incubated in the presence of a
low concentration of L-glutamate. (B D) As the concentration of L-
glutamate increased, [3H]MK-801 binding increased in a dose dependent
manner. Binding density increases from black to white.

92
C
D

93
LOG [GLUTAMATE] (uM)
Figure 7-4. Representative average specific [3H]MK-801 binding in the
presence of varying concentrations of L-glutamate. [3H]MK-801 binding
increased with increased concentration of L-glutamate in a dose dependent
manner. Semi-log plots depicted were from (A) ventral stratum radiatum of
CA1 and (B) parietal cortex layers I HI in chronic ethanol treated and sucrose
control animals. Each data point represents the mean + SEM of eight ethanol
or sucrose treated animals. Curves represent the optimal non-linear fit to a
sigmoid equation. ECso's are listed in Table 7-2.

94
Table 7-2. The effect of chronic ethanol on EC50 of glutamate at the NMDA
receptor complex in various brain regions.
REGION
EC50 (pM) SEM
Sucrose
Ethanol
Dentate Gyrus
d. Molecular Layer
1.28 0.40
1.37 0.48
v. Molecular Layer
0.98 0.02
1.20 0.28
Hippocampus
d. Stratum Radiatum
1.21 0.26
1.00 0.24
v. Stratum Radiatum
0.89 0.28
0.88 0.17
v. Subiculum
1.59 0.30
1.24 0.20
Cortex
d. Cortex (layers I-III)
1.05 0.19
1.09 0.25
d Cortex (layers IV- VI)
1.73 0.31
1.60 0.38
v. Parietal Cortex
1.25 0.29
1.05 0.16
(layers I-III)
v. Parietal Cortex
2.42 0.76
2.24 0.84
(layers IV-VI)
v. Entorhinal Cortex
1.85 0.56
1.42 0.42
Septum
v. lateral Septum
0.51 0.13
0.60 0.20
Thalamus
v. Medial Thalamus
0.98 0.25
1.15 0.12
v. Lateral Thalamus
0.90 0.20
1.04 0.25
d. Thalamus
1.05 0.22
0.96 0.21
v. Striatum
0.64 0.14
0.66 0.11
Data were analyzed with two-way ANOVA (P< 0.86).

95
increased calcium influx through voltage-gated or NMDA receptor channels
to be at least partly responsible for ethanol withdrawal seizure activity (Grant
et ah, 1990; Khanna et al., 1991). During the withdrawal period, NMDA
receptor binding increases with no change in receptor affinity. It is possible
that the enhanced calcium flux also contributes to excitotoxic death of
pyramidal cells. Binding properties of [3H]MK-801 were examined in brain
regions particularly susceptible to the excitotoxic effects of ethanol
withdrawal. Comparison of Bmax values across dorsal and ventral stratum
radiatum of CA1 and molecular layer of dentate revealed a significant
interaction effect of treatment with brain region (P < 0.02). Bmax values in
dorsal and ventral stratum radiatum of CA1 and ventral dentate molecular
layer were increased in chronic ethanol treated animals. Bmax decreased in
dorsal dentate molecular layer. Several studies have described a 15% or
greater principal cell loss from hippocampus and dentate gyrus following
chronic ethanol toxicity (Walker et al., 1981; Lescaudron and Verna, 1985).
The observed trend toward increased number of binding sites found in the
present study may suggest an increase in the number of NMDA receptors per
pyramidal cell. Correlation though, must be made between chronic ethanol
induced cell loss and NMDA receptor density within the same animal.
In summary, the dramatic increases in [3H]MK-801 binding observed
coincident with ethanol withdrawal were not immediately apparent as a
chronic feature of ethanol abuse although a trend toward an increased
number of binding sites in hippocampus and perhaps dentate gyrus may be
present. Analysis of receptor binding as a function of cell number could
determine whether there is an upregulation in NMDA receptor number per
neuron.

CHAPTER 8
SUMMARY AND DISCUSSION
Summary and Interpretation
The hypothesis tested by the experiments presented in this dissertation
was that chronic ethanol treatment is associated with transient and/or
permanent alterations in one or more components of the induction process of
long term potentiation (LTP) in the hippocampus. Several observations can be
drawn from the results of experiments performed to test this hypothesis.
1. The magnitude of LTP is reduced as a result of chronic ethanol
exposure (P < 0.08). After 48 hr of abstinence from a 28 week exposure to a diet
in which ethanol comprised 35 39% of the total caloric intake, the magnitude of
LTP was reduced by 18% and 31% following a 50 or 100 pulse conditioning train
respectively.
2. The reduction of LTP was a permanent consequence of CET and not
dependent on sustained ethanol ingestion (P < 0.06). Following a 5 to 7 month
recovery period from the chronic ethanol diet, the magnitude of LTP remained
markedly reduced. The CET produced reduction in the magnitude of LTP was
present over a range of stimulus conditions, and therefore not limited to a
specific set of stimulus parameters.
3. Experimental blockade of GABAa receptors with the specific
antagonist, bicuculline methiodide (3.5 5.0 pM) eliminated the reduction in LTP
produced by CET. This effect was observed after both a 48 hr and 5-7 month
abstinence period.
4. CET did not overtly affect the early phases of LTP maintenance.
96

97
5. Quantitative autoradiography showed that CET did not alter the Bmax
or IQ of [3H]MK-801 binding to the NMDA receptor channel. Additionally, CET
did not affect the ability of L-glutamate to stimulate [3H]MK-801 binding.
6. Neither GABA-stimulated nor bicuculline inhibition of GABA-
stimulated Cl uptake in microsac preparations from hippocampus or cortex was
altered by CET.
Taken together, these results indicate that CET produced a lasting
reduction of LTP induction in the hippocampus. The reduction was not
produced by a direct alteration of NMDA receptor number or the efficacy of
glutamate to produce NMDA receptor channel opening. Rather, CET altered
some aspect of GABAergic synaptic transmission and its impact on NMDA
receptor function so as to reduce the ease with which LTP could be induced. The
effect of GABAergic synaptic transmission was not due to a direct alteration in
the efficacy of GABA at the GABAa channel or to the efficacy of the
experimentally used antagonist, bicuculline to block GABA-stimulated receptor
activation.
There are several potential mechanisms through which CET may impart
its effect on LTP induction.
1. Basal functioning of GABAergic synaptic transmission may be
unchanged by CET (as evidenced by the lack of CET effect on GABA-stimulated
Cl uptake in hippocampal and cortical microsac preparations) while plasticity
related functioning, as tested in the physiology experiments, may be altered. For
example, activation of the presynaptic GABAg autoreceptor by GABA and the
consequent reduction of GABA release only produces a postsynaptic effect after
200 400 ms (Davies, et al., 1990; 1991). Postsynaptic conductances are reduced
with a similar time course by GABAg receptor activation (Morrisett et al., 1991;
Hahner et al., 1991). Single pulse stimulation, therefore, would be unaffected by

98
activation of autoreceptors (Pacelli, et al., 1991). Under the plasticity related
conditions described in these experiments, conditioning trains would activate the
autoreceptors as well as produce the postsynaptic depolarizing effect of the
reduction of GABA release. No research to date has examined the effects of CET
on the activation of GABAergic plasticity during high frequency stimulation.
One possibility consistent with what is currently understood of CET effects on
GABAergic function is that CET may decrease presynaptic GABAb receptor
number or function. During an LTP inducing conditioning train, the reduced
GABAb influence would fail to mediated the normally occurring decrease in
GABA release. This in turn would result in greater postsynaptic
hyperpolarization, making relief of the Mg2+ block on the NMDA receptor and
LTP induction more difficult.
2) CET may act to enhance LTP of GABAergic inputs relative to excitatory
inputs onto pyramidal cells. The mechanisms through which LTP of GABAergic
synaptic transmission occurs are poorly understood. It is clear that in control
animals, a lasting increase in GABA release occurs following an LTP
conditioning train and the amplitude of the postsynaptic IPSP increases (Ghijsen
et al., 1992; Morishita and Sastry, 1991). Whether there are postsynaptic
mechanisms responsible for enhanced GABAergic synaptic transmission is as yet
unknown. No study to date has examined any aspect of GABAergic synaptic
transmission following LTP in CET animals.
3) CET may exert its effect on a specific subtype of the GABAa receptor.
Alternatively, CET may restrict its effect on the GABAa receptor to the CA1
region of the hippocampus. Recent studies have demonstrated five GABAa
receptor subunit types, each with a specific regional distribution (Sato and Neale,
1989 and Verdoorn et al., 1990). Wafford, et al. (1991) have provided evidence
that the 2yL subunit of the GABAa receptor is vulnerable to the acute effects of

99
ethanol. This particular subunit, though, is not found in high concentration in
the hippocampus. Further research may provide additional clues to specific
GABAa subunit function in the presence of protracted ethanol toxicity.
Future Directions
Based on the findings of this study, future research should be aimed at
understanding how GABAa and GABAb receptors function following CET.
Since both receptors types play an important role during single pulse and
plasticity related synaptic transmission, it will be important to carefully
investigate GABA receptor function associated with both types of stimulus
conditions.
Basal function of the GABAa receptor in cortex and hippocampus was
found in the present study to be unaffected by CET. The next step then would be
to examine GABAa function under conditions that induce plasticity. This could
be examined in two ways. First, the mechanisms activated during a conditioning
train could be studied using a hippocampal microsac preparation. The aim
would be to measure the effect of GABAb receptor activation on the efficacy of
GABA to stimulate Cl uptake in hippocampus. In cerebellum and cortex, GABAb
receptor activation reduces Cl uptake through GABAa receptor channels.
Specifically, one would measure the IC50 of baclofen to inhibit GABA stimulated
Cl uptake and the ability of a GABAb antagonist (e.g. phaclofen) to block the
baclofen effect. This experiment would provide information about potential
postsynaptic modulatory effects of GABAb receptor activation following CET.
Second, the postsynaptic response during an LTP conditioning train could be
examined with intracellular recordings from pyramidal cells. In this study one
would examine the plasticity of the GABAa and GABAb mediated IPSP in

100
response to either conditioning trains of 200 400 ms duration (typical for
activation of GABAb receptor systems) or in response to paired pulse stimuli
with interpulse intervals of 200 400 ms. These data would provide information
about the effects of CET on physiological responses to endogenous release of
GABA under conditions that would typically be involved in plasticity. One
could next test the efficacy of antagonists and agonists to the GABAa and
GABAp receptor to measure the function of specific receptor subtypes and
binding sites. Because pre- and postsynaptic GABAb receptors are
pharmacologically different, their respective contribution to LTP induction could
be measured. For example, phaclofen is a more potent antagonist at the
postsynaptic receptor and 2-hydroxy-saclofen is more potent at the presynaptic
receptor in the hippocampus (Dutar and Nicoll, 1988). If CET alters the
presynaptic GABAb receptor, then the alteration in release would be manifested
in both the postsynaptic GABAa and GABAb responses. If CET produced its
effect via alteration of postsynaptic GABAa or GABAb function, then the
abnormal response would be limited to the effected receptor type.
In addition to investigating the function of GABA receptors during an LTP
inducing conditioning train, it is important to investigate the long lasting effects
of plasticity on GABAergic synaptic transmission. To do this the characteristics
of the intracellular GABAergic IPSP would be measured both before and after
LTP induction. If CET reduces LTP through an enhancement of GABAergic LTP
relative to glutaminergic LTP, then one would measure a greater increase in LTP
of the IPSP than the EPSP. This study may also be performed using extracellular
recordings. The magnitude of LTP following a conditioning train would first be
measured. Once a stable potentiated EPSP was measured, bicuculline
methiodide would be added and single pulse responses again measured. Both
measurements would be recorded within the same slice and in response to the

101
same conditioning train. Again, if CET produces its effect on LTP induction by
enhancing LTP of the IPSP relative to the EPSP, then comparisons between
sucrose and ethanol treated animals should reveal reduced LTP in ethanol
treated animals in the absence of bicuculline. Ethanol and sucrose treated
animals would produce similar magnitude LTP in the presence of bicuculline.
Recent studies have revealed interactions of ethanol with specific GABAa
receptor subunits (Wafford et al., 1991). To investigate this as a potential
mechanism for the effect of CET in the hippocampus, the binding properties of
GABAa and GABAb receptors could be examined. To do this, quantitative
receptor autoradiography using a series of radioactive agonists and antagonists
to the binding sites of the GABA receptors would be measured. In particular, it
would be interesting to examine the binding properties of GABAb agonists as
this has not even been examined in control animals. The effect of CET on
receptor number or affinity would shed light on the mechanism by which CET
alters GABAergic synaptic transmission.
Several anatomical and behavioral as well as human clinical studies have
demonstrated the critical role of the hippocampus in memory formation (Squire
and Zola-Morgan, 1991). LTP is at present a candidate physiological substrate
for memory acquisition for several reasons including: 1) it can be induced in
several anatomical substrates of the memory system, 2) LTP is induced by
physiological stimuli, 3) LTP has a long duration of expression, and 4) LTP
eventually decays. The hippocampus is selectively damaged both
morphologically and physiologically as a result of chronic ethanol exposure. In
particular, the enduring nature of the LTP deficit makes it a candidate substrate
for at least part of the pathology responsible for the memory impairments
associated with CET.

102
In summary, CET reduces the magnitude of long term potentiation. The
reduction of LTP is a permanent consequence of CET and not dependent on
sustained ethanol ingestion. CET affects the reduction of LTP by altering some
component of GABAergic synaptic transmission and not via alteration of NMDA
receptor number or function. The effect of CET on GABAergic synaptic
transmission is not via a direct alteration of the efficacy of GABA at the GABAa
receptor. The CET effect may be restricted to plasticity related events, or be via
an indirect action of GABAb receptor activation and subsequent changes in
GABA release or GABAa function.

REFERENCES
Abele, A.E., Scholz, K.P., Scholz, W.K., and Miller, R.J. (1990). Excitotoxicity
induced by enhanced excitatory neurotransmission in cultured hippocampal
pyramidal neurons. Neuron 2, 413 419.
Abraham, W.C., Gustafsson, B., and Wigstrom, H. (1987). Long-term
potentiation involves enhanced synaptic excitation relative to synaptic
inhibition in guinea-pig hippocampus. J. Physiol. 394, 367 380.
Abraham, W.C., Hunter, B.E., Zornetzer, S.F., and Walker, D.W. (1981).
Augmentation of short-term plasticity in CA1 of rat hippocampus after
chronic ethanol treatment. Br. Res. 222, 271 287.
Abraham, W.C., Manis, P.B., Hunter, B.E., Zornester, S.F., and Walker, D.W.
(1982). Electrophysiological analysis of synaptic distribution in CA1 of rat
hippocampus after chronic ethanol exposure. Br. Res. 237, 91 105.
Abraham, W.C., Rogers, C.J., and Hunter, B.E. (1984). Chronic ethanol-
induced decreases in the response of dentate granule cells to perforant path
input in the rat. Exp. Br. Res. 54, 406 414.
Aggleton, J.P. and Mishkin, M. (1983). Memory impairments following
restricted medial thalamic lesions in monkeys. Exp. Br. Res. 52, 199 209.
Alger, B.E. and Nicoll, R.A. (1982). Pharmacological evidence for two kinds
of GABA receptor on rat hippocampal pyramidal cells studied in vitro. J.
Physiol. 328,125- 141.
Alger, B.E. and Teyler, T.J. (1976). Long-term and short-term plasticity in the
CAI, CA3, and dentate regions of the rat hippocampal slice. Br. Res. 110, 463 -
480.
Allan, A.M., Burnett, D., and Harris, R.A. (1991). Ethanol-induced changes in
chloride flux are mediated by both GABAa and GABAb receptors. Ale. Clin.
Exp. Res. 15, 233 237.
Allan, A.M. and Harris, R.A. (1986). y-Aminobutyric acid agonists and
antagonists alter chloride flux across brain membranes. Mol. Pharmacol. 29,
497 505.
103

104
Amaral, D.G. and Witter, M.P. (1989). The three-dimemsional organization
of the hippocampal formation: a review of anatomical data. Neurosci. 31, 571
-591.
Andrade, R., Malenka, R.C., and Nicoll, R.A. (1986). A G-protein couples
serotonin and GABAb to the same channels in hippocampus. Science 234,
1261 1265.
Arendt, T., Bigl, V., Arendt, A., and Tennstedt, A. (1983). Loss of neurons in
the nucleus basalis of Meynert in Alzheimer's disease, paralysis agitans and
Korsakoff's disease. Acta Neuropathol. 61,101 -108.
Bekkers, J.M. and Stevens, C.F. (1990). Presynaptic mechanism for long-term
potentiation in the hippocampus. Nature 346, 724 728.
Bengochea, O. and Gonzalo, L.M. (1990). Effect of chronic alcoholism on the
human hippocampus. Histol. Histopath. 5, 349 357.
Beracochea, D.J., Tako, A.N., and Jaffard, R. (1989). Accelerated rates of
forgetting of spatial information during aging and long-term ethanol
consumption in mice: Evidence for two distinct forms of amnesia.
Psychobiology 17, 358 362.
Bliss, T.V.P. and Lomo, T. (1973). Long-lasting potentiation of synaptic
transmission in the dentate area of the anaesthetized rabbit following
stimulation of the perforant path. J. Physiol. 232, 331 356.
Blitzer, R.D., Gil, O., and Landau, E.M. (1990). Long-term potentiation in rat
hippocampus is inhibited by low concentrations of ethanol. Br. Res. 536, 203 -
208.
Butters, N. and Cermak, L. (1980). Alcoholic Korsakoff's syndrome: an
information-procession approach to amnesia. Academic Press, NY, NY pp.
188.
Cadete-Leite, A., Tavares, M.A., Uylings, H.B.M., and Paula-Barbosa, M (1988).
Granule cell loss and dendritic regrowth in the hippocampal dentate gyrus of
the rat after chronic alcohol consumption. Br. Res. 473, 1 14.
Cadete-Leite, A., Tavares, M.A., Pacheco, M.M., Volk, B., and Paula-Barbosa,
M.M. (1989). Hippocampal mossy fiber-CA3 synapses after chronic alcohol
consumption and withdrawal. Alcohol 6, 303 310.
Chavez-Noriega, L.E., Halliwell, V., and Bliss, T.V.P. (1990). A decrease in
firing threshold observed after inductio of the EPSP-spike (E-S) component of
long-term potentiation in rat hippocampal slices. Exp. Br. Res. 79, 633 641.

105
Choi, D.W. (1988). Glutamate neurotoxicity and diseases of the nervous
system. Neuron 1, 623 634.
Curtis, D.R., Duggan, A.W., Felix, D., and Johnson, G.A.R. (1970). GABA,
bicuculline and central inhibition. Nature 266, 1222 1224.
Davies, C.H., Davies, S.N., and Collingridge, G.L. (1990). Paired-pulse
depression of monosynaptic GABA-mediated inhibitory postsynaptic
responses in rat hippocampus. J. Physiol. 424, 513 531.
Davies, C.H., Starkey, S.J., Pozza, M.F., and Collingridge, G.L. (1991). GABAb
autoreceptors regulate the induction of LTP. Nature 349, 609 611.
Deisz, R.A. and Prince, D.A. (1989). Frequency-dependent depression os
inhibition in guinea-pig neocortex in vitro by GABAb receptor feed-back on
GABA release. J. Physiol. 412, 513 541.
DeMontis, G., Devoto, P., Giorgi, G., Tagliamonte, A., and Gessa, G.L. (1991).
Ethanol, at micromolar concentration, increases the affinity of [3H]MK-801
binding in rat brain. Eur. J. Pharmacol. 199, 139 140.
Dingledine, R (1983). N-methyl aspartate activates voltage-dependent
calcium conductance in rat hippocampus: localization and frequency
dependency. J. Physiol. 343,385 405.
Durand, D. and Carien, P.L. (1984a). Decreased neuronal inhibition in vitro
after long-term administration of ethanol. Science 224, 1359 1361.
Durand, D., and Carien, P.L. (1984b). Impairment of long-term potentiation
in rat hippocampus following chronic ethanol treatment. Br. Res. 308, 325 -
332.
Durand, D., Saint-Cyr, J.A., Gurevich, N., and Carien, P.L. (1989). Ethanol-
induced dendritic alterations in hippocampal granule cells. Br. Res. 477, 373 -
377.
Dutar, P. and Nicoll, R.A. (1988). Pre- and postsynaptic GABAb receptors in
the hippocampus have different pharmacological profiles. Neuron 1, 585 -
591.
File, S.E. and Mabbutt, P.S. (1990). Long-lasting effects on habituation and
passive avoidance performance of a period of chronic ethanol administration
in the rat. Behav. Br. Res. 36, 171 178.

106
Finn, R.C., Browning, M., and Lynch, G. (1980). Trifluoperazine inhibits
hippocampal long-term potentiation and the phosphorylation of a 40,000
dalton protein. Neurosci. Lett. 19,103 108.
Freund, G. (1970). Impairment of shock avoidance learning after long-term
alcohol ingestion in mice. Science 160, 1599 1601.
Freund, G. and Walker, D.W. (1971). Impairment of avoidance learning by
prolonged ethanol consumption in mice. J. Pharmacol. Exp. Ther. 179, 284 -
292.
Ghijsen, W.E.J.M., Besselsen, E., Geukers, V., Kamphuis, W., and Lopes da
Silva, F.H. (1992). Enhancement of endogenous release of glutamate and y-
aminobutyric acid from hippocampus CA1 slices after in vivo long-term
potentiation. J. of Neurochem. 59, 482 486.
Goldstein, B., Maxwell, D.S., Ellison, G., Flammer, R.P. (1983). Dendritic
vacuolization in the central nervous system of rats after long-term voluntary
consumption of ethanol. J. Neuropath. Exp. Neurol. 42, 579 589.
Grant, K.A., Valverius, P., Hudspith, M., and Tabakoff, B. (1990). Ethanol
withdrawal seizures and the NMDA receptor complex. Eur. J. Pharmacol. 176,
289 296.
Gulya, K., Grant, K.A., Valverius, P., Hoffman, P.L., and Tabakoff, B. (1991).
Brain regional specificity and time-course of changes in the NMDA receptor-
ionophore complex during ethanol withdrawal. Brain Res. 547, 129 134.
Gustafsson, B., Wigstrom, J., Abraham, W.C., and Huang, Y.-Y. (1987). Long
term potentiation in the hippocampus using depolarizing current pulses as
the conditioning stimulus to single volley synaptic potentials. J. Neurosci. 7,
774 780.
Hahner, L., McQuilkin, S., and Harris, R.A. (1991). Cerebellar GABAb
receptors modulate function of GABAa receptors. FASEB J. 5, 2466 2472.
Harris, E.V., Ganong, A.H., and Cotman, C.W. (1984). Long-term
potentiation in the hippocampus involves activation of NMDA receptors.
Br. Res. 323,132 137.
Harrison, N.L. (1990). On the presynaptic action of baclofen at inhibitory
synapses between cultured rat hippocampal neurones. J. Physiol. 422, 433 -
446.

107
Hess, G., Kuhnt, U., and Voronin, L.L. (1987). Quantal analysis of paired-
pulse facilitation in guinea pig hippocampal slices. Neurosci. Lett. 77, 187 -
192.
Hestrin, S., Nicoll, R.A., Perkel, D.J., and Sah, P. (1990). Analysis of excitatory
synaptic action in the rat hippocampus using whole cell recording from thin
slices. J. Physiol. 422, 203 225.
Hu, G.-Y., Hvalby, O., Walaas, S.I., Albert, K.A., Skjeflo, P., Andersen, P., and
Greengard, P. (1987). Protein kinase C injection into hippocampal pyramidal
cells elicits features of long term potentiation. Nature 328, 426 429.
Huettner, J.E. and Bean, B.P. (1988). Block of N-methyl-D-aspartate-activated
current by the anticonvulsant MK-801: Selective binding to open channels.
Proc. Natl. Acad. Sci. 85,1307 1311.
Insausti, R., Amaral, D.G., and Cowan, W.M. (1987). The entorhinal cortex of
the monkey: II. Cortical afferents. J. Comp. Neurol. 264, 356 395.
Kauer, J.A. Malenka, R.C., and Nicoll, R.A. (1988). NMDA application
potentiates synaptic transmission in the hippocampus. Nature 334, 250 252.
Kennedy, M.B. (1989). Regulation of synaptic transmission in the central
nervous system: long-term potentiation. Cell 59, 777 787.
Khanna, J.M., Wu, P.H., Weiner, J., and Kalant, H. (1991). NMDA antagonist
inhibits rapid tolerance to ethanol. Br. Res. Bull. 26, 643 645.
King, M.A., Hunter, B.E., and Walker, D.W. (1988). Alterations and recovery
of dendritic spine density in rat hippocampus following long-term ethanol
ingestion. Br. Res. 459, 381 385.
Kullmann, D.M. and Nicoll, R.A. (1992). Long-term potentiation is
associated with increases in quantal content and quantal amplitude. Nature
357, 240.
Lescaudron, L., Seguela, P., Geffard, M., and Verna, A. (1986). Effects of long
term ethanol consumption on GABAergic neurons in the mouse
hippocampus: A quantitative immunochemical study. Drug and Ale. Dep. 18,
377 384.
Lescaudron, L. and Verna, A. (1985). Effects of chronic ethanol consumption
on pyramidal neurons of the mouse dorsal and ventral hippocampus: A
quantitative histological analysis. Exp. Br. Res. 58, 362 367.

108
Lovinger, D.M., and Routtenberg, A. (1988). Synapse-specific protein kinase C
activation inhances maintenance of long-term potentiation in rat
hippocampus. J. Physiol. 400, 321 333.
Lovinger, D.M., White, G., and Weight, F.F. (1989). Ethanol inhibits NMDA-
activated ion current in hippocampal neurons. Science 243, 1721 1724.
Lovinger, D.M., White, G., and Weight, F.F. (1990). NMDA receptor-
mediated synaptic excitation selectively inhibited by ethanol in hippocampal
slice from adult rat. J. Neurosci. 10,1372 1379.
Lowry, O.H., Rosebrough, A.L., Farr, A.L., and Randall, R.J. (1951). Protein
measurement with the folin phenol reagent. J. Biol. Chem. 193, 265 275.
Luddens, H., Pritchett, D.B., Kohler, M., Killisch, L, Keinanen, K., Monyer, FI.,
Sprengel, R., and Seeburg, P.PL (1990). Cerebellar GABAa receptor selective
for a behavioral alcohol antagonist. Nature 346, 648 651.
Lynch, G., Larson, J., Kelso, S., Barrionuevo, G., and Schottler, F. (1983).
Intracellular injections of EGTA block induction of hippocampal long-term
potentiation. Nature 305, 719 721.
MacDermott, A.B., Mayer, M.L., Westbrook, G.L., Smith, S.J., and Barker, J.L.
(1986). NMDA-receptor activation increases cytoplasmic calcium
concentration in cultured spinal cord neurones. Nature 321, 519 522.
MacDonald, J.F., and Nowak, L.M. (1990). Mechanisms of blockade of
excitatory amino acid receptor channels. Trends Pharmacol. Sci. 11, 167 172.
Malenka, R.C., Kauer, J.A., Zucker, R.S., and Nicoll, R.A. (1988). Postsynaptic
calcium is sufficient for potentiation of hippocampal synaptic transmission.
Science 242, 81 84.
Malenka, R.C., Kauer, J.A., Perkel, D.J., Mauk, M.D., Kelly, P.T., Nicoll, R.A.,
and Waxham, M.N. (1989). An essential role for postsynaptic calmodulin
and protein kinase activity in long-term potentiation. Nature 340, 554 557.
Malinow, R. (1990). Transmission between pairs of hippocampal slice
neurons: Quantal levels, oscillations, and LTP. Science 252, 722 724.
Malinow, R., Madison, D.V., and Tsein, R.W. (1988). Persistent protein
kinase activity underlying long-term potentiation. Nature 335, 820 824.
Malinow, R. and Miller, J.P. (1986). Postsynaptic hyperpolarization during
conditioning reversibly blocks induction of long-term potentiation. Nature
320, 529 530.

109
Malinow, R., and Tsein, R.W. (1990). Presynaptic enhancement shown by
whole-cell recordings of long-term potentiation in hippocampal slices.
Nature 346, 177 180.
Manabe, T., Renner, P., and Nicoll, R.A. (1992). Postsynaptic contribution to
long-term potentiation by the analysis of miniature synaptic currents. Nature
355, 50.
Mayer, M.L. and Westbrook, G.L. (1984). Permeation and block of M-methyl-
D-aspartatic acid receptor channels by divalent cations in mouse central
neurones. J. Physiol. 394, 501 528.
Mayer, M.L., Westbrook, G.L., and Guthrie, P.B. (1984). Voltage-dependent
block by Mg2+ of NMDA responses in spinal cord neurones. Nature 309, 261 -
263.
McMullen, P.A., Saint-Cyr, J.A., and Carien, P.L. (1984). Morphological
alterations in the rat CA1 hippocampal pyramidal cell dendrites resulting
from chronic ethanol consumption and withdrawal. J. Comp. Neurol. 225,
111-118.
Michaelis, E.K., Freed, W.J., Galton, N., Foye, J., Michaelis, M.L., Phillips, I.,
and Kleinman, J.E. (1990). Glutamate receptor changes in brain synaptic
membranes from human alcoholics. Neurochem. Res. 11, 1055 1063.
Mody, I., Baimbridge, K.G., and Miller, J.J. (1984). Blockade of tetanic and
calcium-induced long-term potentiation in the hippocampal slice preparation
by neuroleptics. Neuropharmacology 23, 625 631.
Monaghan, D.T. (1991). Differential stimulation of [3H]MK-801 binding to
subpopulations of NMDA receptors. Neurosci. Lett. 122, 21 24.
Morishita, W. and Sastry, B.R. (1991). Chelation of postsynaptic Ca2 +
facilitates long-term potentiation of hippocampal IPSPs. NeuroReport 2, 533 -
536.
Morrisett, R.A., Mott, D.D., Lewis, D.V., Swartzwelder, H.S., and Wilson,
W.A. (1991). GABAb-receptor-mediated inhibition of the N-methyl-D-
aspartate component of synaptic transmission in the rat hippocampus. J.
Neurosci. 11, 203 209.
Neuman, R., Cherubini, E., and Ben-Ari, Y. (1987). Is activation of N-methyl-
D-aspartate receptor gated channels sufficient to induce long term
potentiation? Neurosci. Lett. 80, 283 288.

110
Newberry, N.R. and Nicoll, R.A. (1985). Comparison of the action of baclofen
with y-amino butyric acid on rat hippocampal pyramidal cells in vitro. J.
Physiol. 360,161 185.
Nowak, L., Bregestovski, P., Ascher, P., Herbet, A., and Prochiantz, A. (1984).
Magnesium gates glutamate-activated channels in mouse central neurones.
Nature 307, 462 465.
Olsen, R.W., McCabe, R.T., and Wamsley, J.K. (1990). GABAa receptor
subtypes: autoradiographic comparison of GABA, benzodiazepine, and
convulsant binding sites in the rat central nervous system. J. Chem.
Neuroanat. 3, 59 76.
Pacelli, G.J., Su, W., and Kelso, S.R. (1991). Activity-induced decrease in early
and late inhibitory synaptic conductances in hippocampus. Synapse 7, 1 -13.
Parker, E.S. and Noble, E. (1977). Alcohol consumption and cognitive
functioning in social drinkers. J. of Studies of Alcohol 38,1224 1232.
Pritchett, D.B., Sontheimer, G., Shivers, B.D., Ymer, S., Kettenmann, EL,
Schofield, P.R., and Seeburg, P.H. (1989). Importance of a novel GABAa
receptor subunit for benzodiazepine pharmacology. Nature 338, 582 585.
Ransom, R.W. and Stec, N.L. (1988). Cooperative modulation of [3H]MK-801
binding to the N-methly-D-aspartate receptor-ion channel complex by L-
glutamate, glycine, and polyamines. J. Neurochem. 51, 830 836.
Regehr, W.G. and Tank, D.W. (1990). Postsynaptic NMDA receptor-mediated
calcium accumulation in hippocampal CA1 pyramidal cell dendrites. Nature
345, 807 810.
Reyman, K.G., Brodemann, R., Kase, El., and Matthies, El. (1988). Inhibitors
of calmodulin and protein kinase C block different phases of hippocampal
long-term potentiation. Brain Res. 461, 388 392.
Reyman, K.G., Matthies, H.K., Schulzeck, K., and Matthies, H. (1989). N-
methyl-D-aspartate receptor activation is required for the induction of both
early and late phases of long-term potentiation in rat hippocampal slices.
Neurosci. Lett. 96, 96 101.
Riley, J.N. and Walker, D.W. (1978). Morphological alteration in
hippocampus after long-term alcohol consumption in mice. Science 201, 646 -
648.
Rogers, C.J. (1986). The effect of chronic ethanol injestion on synaptic
inhibition in CA1 of the rat. Dissertation, University of Florida.

Ill
Rogers, C.J. and Hunter, B.E. (1992). Chronic ethanol treatment reduces
inhibition in CA1 of the rat hippocampus. Br. Res. Bull. 28, 587 592.
Ryan, C, Butters, N., Montgomery, K., Adinolfi, A., and Didario, B. (1980).
Memory deficits in chronic alcoholics: Continuities between the "intact"
alcoholic and the alcoholic Korsakoff's patient. In: Biological effects of
alcohol, (H. Begleiter, Ed.), pp. 701 718.
Ryback, R. (1971). The continuim and specificity of the effects of alcohol on
memory. Quar. J. Stud. Ale. 32, 995 1016.
Sato, T.N. and Neale, J.H. (1989). Immunological identification of multiple
a-like subunits of the y-aminobutyric acidA receptor complex purified from
neonatal rat cortex. J. Neurochem. 53, 1089 1095.
Scheetz A.J., Markham, J.A., and Fifkova, E. (1987). Changes in the frequency
of basket cells in the dentate fascia following chronic ethanol administration
in mice. Br. Res. 403, 151 154.
Schwartzkroin, R.A. and Wester, K. (1975). Long-lasting facilitation of
synaptic potential following tetanization in the in vitro hippocampal slice.
Br. Res. 89, 107 119.
Sigel, E., R. Baur, Trube, G., Mohler, H., and Malherbe, P. (1990). The effect of
subunit composition of rat brain GABAa receptors on channel function.
Neuron 5, 703 711.
Sinclair, J.G. and G.F. Lo (1986). Ethanol blocks tetanic and calcium-induced
long-term potentiation in the hippocampal slice. Gen. Pharmac. 17, 321 233.
Squire, L.R. and Moore, R.Y. (1979). Dorsal thalamic lesion in a noted case of
human memory dysfunction. Ann. Neurol. 6, 503 506.
Squire, L.R. and Zola-Morgan, S. (1991). The medial temporal lobe memory
system. Science 253,1380 1386.
Steward, S, Tomasulo, R., and Levy, W.B. (1990). Blockade of inhibition in a
pathway with dual excitatory and inhibitory action unmasks a capability for
LTP that is otherwise not expressed. Brain Res. 516, 292 300.
Teyler, T.J. and DiScenna, P. (1984). Long-term potentiation as a candidate
mnemonic device. Br. Res. Rev. 7, 15 28.

112
Thalmann, R.H. (1988). Blockade of a late inhibitory postsynaptic potential in
hippocampal CA3 neurons in vitro reveals a late depolarizing potential that
is augmented by pentobarbital. Neurosci. Lett. 95, 155 160.
Thibault, O., Joly, M., Muller, D., Schottler, F., Dudek, S., and Lynch, G. (1989).
Long-lasting physiological effects of bath applied N-methyl-D-aspartate. Br.
Res. 476,170 -173.
Tobin, A.J., Khrestchatisky, M., MacLennan, A.J., Chiang, M.-Y., Tillakaratne,
N.J.K., Xu, W., Jackson, M.B., Brecha, N., Sternine, C., and Olsen, R.W. (1991).
Structural, developmental and functional heterogeneity of rat GABAa
receptors. In: Neuroreceptor mechanisms in brain, (S. Kito, Ed.) pp. 365 374.
Uckele, J.E., McDonald, J.W., Johnston, M.V., and Silverstein, F.S., (1989).
Effect of glycine and glycine receptor antagonists on NMDA-induced brain
injury. Neurosci. Lett. 107, 279 283.
Verdoorn, T.A., Draguhn, A., Ymer, S., Seeburg, P., and Sakmann, B. (1990).
Functional properties of recombinant rat GABAa receptors depend upon
subunit composition. Neuron 4, 919 928.
Victor, M., Adams, R.D., and Collins, G.H. (1971). The Wernicke-Korsakoff
syndrome. (F.A. Davis, Ed.) Phil. PA.
Wafford, K.A., Burnett, D.M., Leidenheimer, N.J., Burt, D.R., Wang, J.B.,
Kofuji, P., Dunwiddie, T.V., Harris, R.A., and Sikela, J.M. (1991). Ethanol
sensitivity of the GABAa receptor expressed in Xenopus oocytes requires 8
amino acids contained in the y2L subunit. Neuron 7, 27 33.
Walker, D.W., Barnes, D.E., Zornester, S.F., Hunter, B.E., and Kubanis, P.
(1980). Neuronal loss in hippocampus induced by prolonged ethanol
consumption in rats. Science 209, 711 713.
Walker, D.W. and Freund, G. (1971). Impairment of shuttle box avoidance
learning following prolonged alcohol consumption in rats. Physiol. Behav. 7,
773 778.
Walker, D.W. and Freund, G. (1973). Impairment of timing behavior after
prolonged alcohol consumption in rats. Science 182, 597 598.
Walker, D.W. and Hunter, B.E. (1978). Short-term memory impairment
following chronic alcohol consumption in rats. Neuropsych. 16, 545 553.
Walker, D.W., Hunter, B.E., and Abraham, W.C. (1981). Neuroanatomical
and functional deficits subsequent to chronic ethanol administration in
animals. Alcoholism: Clin. Exp. Res. 5, 267 282.

113
Walker, D.W., Hunter, B.E., Barnes, D.E., and Riley, J.N. (1982). An animal
model of alcohol-induced brain damage: a behavioral and anatomical
analysis. In D.A. Wilkinson (Ed.), Cerebral Deficits in Alcoholism. Addiction
Research Foundation, Toronto, pp. 123 147.
Wong, R.K.S. and Prince, D.A. (1979). Dendritic mechanisms underlying
pennicillin-induced epileptiform activity. Science 204, 1228 1231.
Zola-Morgan, S., Squire, L.R., and Amaral, D.G. (1986). Human amnesia and
the medial temporal region: enduring memory impairment following a
bilateral lesion limited to field CA1 of the hippocampus. J. Neurosci. 6, 2950 -
2967.
Zola-Morgan, S., Squire, L.R., Amaral, D.G., and Suzuki, W.A. (1989).
Lesions of perirhinal and parahippocampal cortex that spare the amygdala
and hippocampal formation produce severe memory impairment. J.
Neurosci. 9, 4355 4370.
Zola-Morgan, S., Squire, L.R., Rempel, N.L., Clower, R.P., and Amaral, D.G.
(1992). Enduring memory impairment in monkeys after ischemic damage to
the hippocampus. J. Neurosci. 12, 2582 2596.

BIOGRAPHICAL SKETCH
Margaret Fairchild Welch was born of Dorothy Fairchild and Edward
Michel Welch on November 2, 1954 in Utica, NY, "Gateway to the
Adirondack^." Her early childhood years were spent investigating nature and
surviving Catholic schools. She began study of nursing at Niagara University in
1972 and graduated in 1975 with visions of life in the "big city." She moved to
Boston, MA and worked as a staff nurse at Massachusettes General Hospital for
the next six years. When she had had quite enough of the "big city," Margaret
moved to Ames, IA to study zoology at Iowa State University. There she
received her first exposure to the field of neurobiology and developed a long-
lasting interest in neurophysiological research. During this same time, Margaret
met Terry K. Tremper, fell in love and married. They moved to Esteli,
Nicaragua, where she was an instructor of physiology and pathology at La
Escuela de Agricultura y Ganaderia de Esteli. She also worked on health related
community development projects in Esteli. On October 31, 1986, Terry and
Margaret became parents as she gave birth to Martin Mohandas Tremwel. They
returned to the United States in 1987 when she began graduate work at the
University of Florida.
114

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
£
Bruce E. Hunter, Chair
Associate Professor of Neuroscience
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
Kevin J. Anderson
Assistant Professor of Neuroscience
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
Michael A. King )
Assistant Research Scientist of Neuroscience
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
//John Middleb^ooks
// Associate Professor of Neuroscience

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
o C
(J
Philip Posnetf
Professor of Physiology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
Paul Reier
Mark F. Overstreet Professor of Neurological
Surgery and Neuroscience
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
Associate Professor of Neuroscience
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
2^-r? u/-
Don W. Walker
Professor of Neuroscience

This dissertation was submitted to the Graduate Faculty of the College of
Medicine and to the Graduate School and was accepted as partial fulfillment of
the requirements for the degree of Doctor of Philosophy.
May, 1993
cx/>
Dean, College of Medicine
u
Dean, Graduate School