Effects of acute and chronic ethanol treatment on pre- and postsynaptic responses to baclofen in rat hippocampus

Brain Research, 500 (1991) 84-91 © 1991 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/91/$03.50 ADONIS ()~)689939116975E

84

BRES 16975

Effects of acute and chronic ethanol treatment on pre- and postsynaptic responses to baclofen in rat hippocampus Gerald D. Frye, Lathrop Taylor, Jerome P. Trzeciakowski and William H. Griffith Department of Medical Pharmacology and Toxicology, Texas A&M University, College of Medicine, College Station, TX 77843-1114 (U.S.A.) (Accepted 30 April 1991) Key words: Ethanol; Tolerance; Dependence; Baclofen; 7-Aminobutyric acid B; Hippocampus; CA1 Pyramidal cell; Electrophysiological recording

Interactions between the GABA B receptor and acute or chronic ethanol treatment were studied using extracellular and intracellular electrophysiological recording techniques. Bath application of the GABA B receptor agonist, (-)-baclofen (0.1-100 #M) induced concentrationdependent inhibition of extracellularly recorded dendritic excitatory postsynaptic potentials (EPSPs) in the CA1 region of hippocampal slices. Responses to baclofen were unchanged relative to control either by simultaneous application of ethanol (10-60 mM) or by previous chronic ethanol exposure. The membrane potential of CA1 pyramidal neurons was reversibly hyperpolarized an average of 5 mV by pressure ejection of baelofen (1 mM). Bath application of ethanol (30 mM) alone occasionally caused a small depolarization of resting membrane potentials in CA1 neurons but failed to increase hyperpolarizing responses to pressure-ejected baclofen. However, in slices from chronic ethanoltreated animals hyperpolarizing responses to bath-applied baelofen (10/~M) were reduced by approximately 30% relative to controls. These results suggest that GABAB-mediated responses in CA1 hippocampat pyramidal neurons are relatively resistant to the acute effects of ethanol, but that continuous exposure to ethanol sufficient to induce physical dependence may evoke an adaptive reduction in some GABA B receptor mediated responses.

INTRODUCTION

cultured spinal cord neurons 2'3'47"49. Chronic ethanol

Simply put, the ' G A B A hypothesis of ethanol intoxication' proposes that ethanol increases central 7-aminobutyric acid ( G A B A ) neurotransmission to cause central nervous system impairment. This action, when sustained chronically, is p r o p o s e d to e v o k e a reversible, adaptive reduction in G A B A e r g i c inhibition leading to functional tolerance and physical d e p e n d e n c e on ethanol 2a2'26. Support for this overall hypothesis has come

treatment apparently leads to a loss of this facilitation of G A B A A receptor function by ethanol in brain membrane vesicles 2'3s and may also cause a reduction in G A B A A receptor responses to G A B A A agonists alone 38. Despite the relative consistency of this supporting biochemical evidence, findings from electrophysiological studies have been mixed b e t w e e n those which find that ethanol increases G A B A A - e v o k e d responses 1"7"42'5° and those that fail to find an effect 6'25'44'46. Thus the inter-

from behavioral studies in which G A B A agonists have been shown to increase ethanol intoxication 13'23 or to suppress certain signs of the ethanol withdrawal synd r o m e 8"17't8'2~. Presently the role G A B A receptors play in the actions of ethanol is receiving a great deal of attention 2, in part because ethanol seems unlikely to change presynaptic G A B A transmitter pools 14 and because ethanol can increase chloride u p t a k e during activation of G A B A A receptors in biochemical models. This latter effect may be due to an aUosteric interaction of ethanol with the G A B A A - g a t e d chloride channel in vesicles isolated from cerebral cortex and cerebellum or in

action between ethanol and G A B A A receptors may be more subtle and complex than first thought 44. A n o t h e r potential correlate of the G A B A hypothesis is the possibility that ethanol increases inhibitory G A B A e r g i c neurotransmission by increasing the activity of baclofen-activated, bicuculline-insensitive G A B A B receptors. Baclofen has been r e p o r t e d to increase ethanolinduced anesthesia in mice 33 or decrease ethanol preference in rats 9. The G A B A B receptor antagonist, phaclofen 29, has been shown to reduce several measures of ethanol-induced behavioral i m p a i r m e n t 4. The idea that ethanol withdrawal seizures might involve reduced

Correspondence: G.D. Frye, Department of Medical Pharmacology, Texas A&M University, College of Medicine, College Station, TX 77843-1114, U.S.A. Fax: (1) (409) 845-0699.

85 G A B A B receptor-mediated inhibition is also supported by the finding that microinjection of baclofen into the inferior collieulus of ethanol-dependent rats completely suppresses audiogenic seizures as does the G A B A A agonist, muscimo119. These indirect pharmacological studies are consistent with a potential role for G A B A B receptors in ethanol neuropharmacology. The present studies were undertaken to obtain a more direct assessment of the effects of ethanol treatment on the function of G A B A B receptors at the cellular level. Previously well characterized pre- and postsynaptic responses of hippocampal CA1 pyramidal cells to baclofen were used as models of G A B A B receptor function. The 'presynaptic' model utilized baclofen-induced inhibition of dendritic excitatory postsynaptic potentials (EPSPs)5" 28,30,43 while the 'postsynaptic' model was based on baclofen-induced hyperpolarization of pyramidal cells 27'28' 41 Changes in cellular responses to baclofen were examined to determine how acute and chronic ethanol treatment might change G A B A B receptor-mediated cellular functions. The results of these investigations have been presented in earlier preliminary reports 15'2°.

MATERIALS AND METHODS

Animals and chronic ethanol treatment Male Sprague-Dawley rats (125-150 g, Timco-Harlan Industries, Houston, TX, U.S.A.) were housed in pairs (22-25 °C; lights 7.0019.00 h). Some rats were used without prior treatment (untreated). Physical dependence on ethanol was induced in other animals by the previously well characterized liquid diet method of Frye et al. 16-19. These rats were housed singly. On the first day, in addition to rat chow and water, each animal was supplied with 35 ml of a nutritionally complete liquid diet prepared from purified materials which met or exceeded the nutrient requirements for rats of the National Research Council 4°. The composition of the diet was based on the 'low fat' diet described by Thompson and Reitz 48 with minor modification of the soluble mineral salts 45. On the second day, rat chow was removed but an additional 35 ml of liquid diet and water were again made available. Beginning the next day, 'ethanol-treated' animals were offered water, as well as a liquid diet in which ethanol replaced part of the dextrose, isocalorically (1 g of ethanol = 1.75 g of dextrose). This treatment was continued for 12 consecutive days. From the first 6 days to the last 6 days, ethanol concentrations were increased from 0.07 to 0.08 g/ml, respectively to compensate for the development of metabolic tolerance to ethanol. 'Ethanol-naive' controls were fed 35 ml of liquid diet without ethanol throughout the 14 day treatment period, an amount previously found to be calorically and nutritionally equivalent to the average daily consumption of liquid diet by ethanol-treated animals. Dextrose or ethanol liquid diets were provided fresh daily between 12.00 and 13.00 h in clean, graduated cylinders fitted with ballbearing drinking spouts (Wahmann Manufacturing Co., Timonium, MD). This regimen of ethanol diet feeding was found to induce daily ethanol consumption of 12-16 g/kg/24 h, maintain large amounts of ethanol in the blood stream (up to 2 mg/ml) and allow continued weight gain (1-2 g/day) without first having to fast the experimental animals 16. Following withdrawal of ethanol from treated animals, marked withdrawal signs including susceptibility to audiogenic seizures and forelimb tremor have been routinely observed within 3-5 h indicating the presence of physical dependence on ethanol ~6-19.

Drugs Absolute ethanol was purchased from the Warner-Graham Co. (CockeysviUe, MD) and included in the tissue bath superfusate. Racemic baclofen and (-)-baclofen (gifts from Ciba Pharmaceutical Co., Summit, NJ) were dissolved in physiological solution (see below) and applied in the tissue bath perfusate (0.1-100/~M) or by pressure ejection (1 mM).

Hippocampal slice preparation Following decapitation, the brain was rapidly removed and transverse hippocampal slices, 400-500/zm thick, were cut on a Mcllwain Tissue Chopper (Mickle Lab. Engineering Co. Ltd., Surrey, U.K.). Slices were held in an oxygenated (95% 0 2, 5% CO2) physiological solution (in mM: NaCI 120, KCI 3, CaC12 2.5, MgCI 1.2, NaH2PO 4 1.2, NaHCO 3 22.6, glucose 11.1) at approximately 30 °(2 for at least 1 h prior to use. Single slices used for electrical recordings were transferred to a plexiglass bath (2 ml) and completely submerged and perfused (4 ml/min, 34 °C) with the physiological solution. Although ethanol-treated animals were fully intoxicated at the time of sacrifice, no attempt was made to maintain the state by adding ethanol to the solutions bathing hippocampal slices. Slices essentially were free of ethanol for the 1-6 h during which extra- or intracellular recordings were completed.

Extracellular recordings Extracellularly recorded EPSPs were obtained as previously described 22 from the dendritic region of the CA1 pyramidal cell layer using glass recording pipettes (1.5 mm od; 1 M NaC1; 2-10 MQ). To evoke CA1 dendritic EPSPs the Schaffer collateraFcommissural pathway was stimulated with square wave pulses (1-70 V; 0.01-0.1 ms duration, 0.2 Hz). Three field EPSPs were averaged, then the slope of the rising phase was measured and plotted as an inputoutput (I-O) relationship against the stimulus intensity. The effects of ethanol and/or (-)-baclofen applied in the perfusion solution were determined by measuring changes in the rising slope of the EPSP.

Intracellular recordings Standard intracellular recording techniques were used to record from CA1 pyramidal cells. Electrodes filled with KAc (4 M) or KMeSO 4 (2 M) with tip resistances of 40-100 MQ were connected to an Axoclamp-2 amplifier for recording under bridge or switchcurrent clamp mode. Voltages and current were displayed on an ink chart recorder (Gould Electronics, Cleveland, OH) or digitized for later analysis. Acute effects of ethanol were studied in slices taken from untreated rats where ethanol (10-60 mM) was added directly to the perfusion solution. Postsynaptic effects of baclofen (1 mM) were tested by pressure ejection using a Picospritzer (General Valve Corp., Fairfield, N J). Glass micropipettes were broken back to a tip diameter of approximately 2-6/~M. Baclofen (1 mM) was released by applying pressure (13.8-34.5 kPa) for 0.1-1.0 s to micropipettes positioned just above the slice surface, over the cell being recorded. In other studies in which the effects of chronic ethanol treatment were examined, baclofen (10/~M) was bath-applied to tissue slices from ethanol-treated or ethanol-naive animals.

Statistical analysis The sample means of most data were compared by calculating an independent or a paired-t statistic, as appropriate, using 'Abstat release 6' (Anderson-Bell, Parker Co.). However, in 1 experiment the recordings of baclofen-induced hyperpolarizations at various membrane potentials were compared by linear regression analysis such that data from ethanol-treated and ethanol-naive cells were analyzed either alone or in combination. Calculation of an F statistic as described by Motulsky and Ransnas 39 was used to indicate whether the 2 data sets were best fit by the same or different linear relationships. Significance in all statistical tests was assumed when the two-tailed P for the 't' statistics or when the P for the F statistic was equal to, or less than 0.05.

86 }00

:@ .~

80

~

60

E

40

0.)

approximately 8 min caused a concentration-dependent inhibition of CA1 field EPSPs in slices from untreated rats. (-)-Baclofen ( 1 / t M ) reduced the EPSP by approximately 30%,

while

100 /~M essentially completely

blocked the response. Since the hypothesis being tested

g_ 2o

concerned a potential e n h a n c e m e n t of (-)-baclofen-induced inhibition by ethanol, this partial inhibitory effect

0

0,01

0. t

1

10

100

1000

(NM)

Boclofen

of (-)-baclofen (1 #M) was selected for use in subsequent tests designed to reveal any potential G A B A B enhancing effects of ethanol.

Fig. 1. Concentration dependence of inhibition of CA1 dendritic EPSPs by (-)-baclofen. (-)-Baclofen-induced inhibition was estimated by determining the magnitude of the reduction in the slope of the half maximum EPSP obtained under control conditions in slices from untreated rats. Points represent the mean _+ S.E.M. for n = 5-22 observations.

Effect of ethanol on baclofen inhibition of extracellularly recorded field EPSPs Ethanol in concentrations of 10-60 m M was applied

RESULTS

to hippocampal slices. These concentrations were used since similar blood ethanol concentrations were previ-

Effect of baclofen on extracellularly recorded field EPSPs

ously found to cause c o n c e n t r a t i o n - d e p e n d e n t increases in punished responding as well as impairment of motor coordination in intact Sprague-Dawley rats 16. Addition

Fig. 1 confirms previous reports (see Introduction) that bath application of (-)-baclofen (0.1-100/~M) for

(1) CONTROL

A

(2) BACLOFEN

(5) RINSE

(3) RINSE

(6) B A C L O F E N * E T O H

(4) ETOH

(7) RINSE

i 2mY 10 ms

B 25

T / O--

/ • ~ / A -~.u

0 CONTROL • 1 #M BACLOFEN ~ RINSE

~=/~ ~/o o ~o ~ ~o ~

E

2.5

"G'

2"0 t

E "~

1.5.

> vE

O--

0 CONTROL

• -O--D

•

• --

•

T

I U,M B A C L O F E N RINSE 1 ,~M BACLOFEN + 10 mM ETON

T

T [ o T i ~-¢ : j~'f!i 4.--'~'~/1

.

I

1,0.

S

e

05~ 0.0!

_/' -'~°

0

1/~

_i~t

i

I

S m

"

,

I

I

20

40

60

VOLTS STIMULUS

-• 80

0.5

0.0

0

20

40

60

80

VOLTS STIMULUS

Fig. 2. Effect of ethanol (10 mM) on inhibition of CA1 dendritic EPSPs by (-)-baclofen (1 #M). A: sequence of a complete experimental protocol recorded from a typical preparation are shown in the absence of drug (1), after 8 min superfusion with 1 #M (-)-baclofen alone (2), 10 mM ethanol alone (4), or the 2 drugs in combination (6). Rinse responses (3, 5 and 7) were recorded following a 20 min washout of the drugs. B: on the left is the I/O relationship recorded in slices from untreated rats before (control) during (-)-baclofen (1 gM) and after washing out the drug (rinse). On the fight, the I/O relationship recorded in a slice before drug (control), during (-)-baclofen (1 #M), or (-)-baciofen (1 #M) combined with ethanol (10 raM) and after washing out these drugs (rinse). Note the lack of change in the effect of (-)-baclofen when ethanol is present. Points represent the mean -+ S.E.M. for n = 9 observations, each from a different slice. Observations for ethanol alone and related rinses were not included for clarity.

87

•.t3 J:E-E(0

60. 50. 40. 30'

(J

20'

ta09

10

I

If/~j

0

rather than increase inhibition of the dendritic EPSP by (-)-baclofen (1 #M). However, there were no statistically significant differences at any concentration of ethanol.

] 1 /.LM Baclofen

liMiacl°fei

iETOH

Effect of baclofen on extracellulary recorded field EPSPs following chronic ethanol treatment

lOmM ETOH 30raMEl'OH 60ramETOH

Fig. 3. Effect of ethanol (10-60 mM) on inhibition of CA1 dendritic EPSPs by baclofen (1 /*M). The percent inhibition of the EPSP response by baclofen (1/,M) was determined by measuring the reduction of the slope of the half maximal EPSP untreated response. Inhibition was determined in the presence or absence of ethanol (10, 30 or 60 raM) applied simultaneously with baclofen in the same tissue slice. Only 1 concentration of ethanol was tested on each slice. Bars represent the mean -+ S.E.M. of 7, 9 or 10 observations for ethanol 10, 30 or 60 mM, respectively.

of these concentrations of ethanol alone for up to 8 min did not cause any apparent change in the I - O relationship of the EPSP and did not cause any apparent residual effects after ethanol was rinsed out of the preparation (data not shown). Fig. 2A shows the experimental protocol for evaluating ethanol effects on (-)-baclofeninduced inhibition. Drugs were bath-applied in the sequence shown in traces 1-7. Ethanol (10 mM) alone had no effect on the EPSPs. In the presence of ethanol (10 mM), superfusion of (-)-baclofen (1 #M) resulted in inhibition of the EPSP similar to that observed after (-)baclofen alone (Fig. 2A, B). Fig. 2B graphically displays the summarized results of several experiments similar to those described in Fig. 2A. Fig. 3 summarizes the results of a series of experiments that show that application of ethanol (10, 30 or 60 mM) actually appeared to decrease

g

lOO

"-~ 80 .-£ 60

The next set of experiments attempted to determine whether chronic ethanol intoxication would cause an adaptive reduction in GABAergic inhibition mediated by presynaptic GABA B receptors. To test this, the same protocol as described above for (-)-baclofen-induced inhibition of dendritic EPSPs was compared in hippocampal slices prepared from ethanoltreated or ethanol-naive animals. Fig. 4 shows that (-)baclofen (0.1-100 #M) caused a concentration-depen-

A

.5 sec Bac

.2 sec Bac

.1 sec Bac

CONTROL

3 0 m M ETOH

JSmV

RINSE

40s

B

CONTROL

ETOH

Z:x Untreated 0 Ethanol Treated

~Q~/'AIf

•

40

~, Q-

20

0

0.01

0.1

1

10

100

1000

Bactofen (#M)

Fig. 4. Effect of chronic ethanol treatment on the concentration dependence of inhibition of CA1 dendritic EPSPs by baclofen. Baclofen was applied by superfusion for 8 min before a series of 10-15 stimuli of increasing intensity were applied at 5 s intervals. Inhibition of the EPSP was caused by various concentrations of baclofen. Baclofen-induced inhibition was estimated by determining the magnitude of the reduction in the slope of the half maximum EPSP obtained under control conditions. Points represent the mean _+ S.E.M. for 3-10 or 4-8 observations recorded from slices obtained from ethanol-treated (open circles) or ethanol-naive (solid circles) animals, respectively. Open triangles are the same results as shown in Fig. 1.

Fig. 5. Effect of ethanol on baclofen-induced hyperpolarizing responses recorded in CA1 pyramidal neurons. A: pressure application of baclofen (1 mM) for 0.5, 0.2 or 0.1 s to a pyramidal cell resulted in hyperpolarizing responses that were relatively proportional to the duration of application (control). In the presence of superfused ethanol (30 mM ETOH) the same applications of baelofen produced similar hyperpolarizing reponses. Following washout of ethanol for 20 rain (rinse), reapplication of baclofen again produced similar hyperpolarizing responses. All responses were taken at --65 mV. B: bath application of ethanol (ETOH) had only small and inconsistent effects on membrane properties of the pyramidal neuron shown in A. Upper traces show membrane potential changes from -65 mV during positive current injection (lower trace).

88 A

B Ethanol Naive

Ethanol Naive

Bac 10

Bac 10 -60 III,

~

. . •10mV Ethanol Naive

2min

Ethanol Treated

Bac 30

Bac 10

-70

-60

C

- 1 ~0 ~ . ~

Membrone Potentiol (mV)

$90

-8o

-70

4

-60

2 0 --2

¢) 0

-6

I

E -10 ~ -12 ~Q ETHANOL NAIVE - -

~

ETHANOL TREATED - - - -

~

-14 ~ -16 ~ -18 < ~" -20

Fig. 6. Effect of chronic ethanol treatment on baclofen-induced hyperpolarization A: bath application of baclofen (10 or 30 aM) causes hyperpolarizing responses in a cell with the membrane potential held at -70 mV. Note that the size of the baclofen (30 aM) hyperpolarization is larger than that after 10 aM. B: from -60 mV, bath application of baclofen (10 aM) causes a larger hyperpolarizing response in the same ethanol-naive cell shown in A than in a cell from an ethanol-treated slice. C: plot of baclofen-induced hyperpolarization versus membrane potential for all cells. The mean -+ S.E.M. hyperpolarizing responses to baclofen (10 aM) at -60 mV (n = 15-16), -70 mV (n = 11-12) or -80 mV (n = 4-10) for cells from ethanol-naive (circles) or ethanol-treated (triangles) slices is shown along with the best fitting line for each set of data. *P < 0.01 when compared with control response at -60 mV.

dent inhibition of dendritic EPSPs in slices from ethanolnaive animals that was essentially identical to that observed earlier (Fig. 1) in slices from u n t r e a t e d animals. F u r t h e r m o r e , the inhibitory effects of (-)-baclofen in slices from e t h a n o l - t r e a t e d rats were essentially the same as o b s e r v e d for both control groups (Fig. 4). Surprisingly, there were no obvious differences in the general activity of slices from e t h a n o l - t r e a t e d animals relative to controls, despite the fact that these tissues were undergoing withdrawal from ethanol.

Effect of in vitro ethanol on intracellularly recorded baclofen-induced hyperpolarization In addition to inhibiting presynaptic transmitter release, baclofen also has been shown to act postsynaptically to induce m e m b r a n e hyperpolarization in CA1 py-

ramidal c e l l s 2 7 ' ~ ' 4 1 . Fig. 5 A shows that application of baclofen (1 m M ) by pressure ejection for 0.1-0.5 s caused a d u r a t i o n - d e p e n d e n t , g r a d e d hyperpolarization of the m e m b r a n e potential in cells from u n t r e a t e d rats. In order to examine the effects of acute ethanol exposure, the duration of baclofen applications was adjusted to give consistent hyperpolarizing responses of a p p r o x i m a t e l y 5 m V before beginning ethanol perfusion. In a few cells, following bath application of ethanol (30 m M ) there did a p p e a r to be a small e n h a n c e m e n t of the baclofen-induced hyperpolarization (Fig. 5A). H o w e v e r , when all cells were analyzed, ethanol (30 mM; 15-25 min) did not significantly enhance the response to baclofen (net baclofen-induced hyperpolarization: before ethanol = 5.2 --- 0.5 mV; during ethanol = 4.5 - 0.5 mV; n = 10 cells). F u r t h e r m o r e , bath application of ethanol (30 m M )

89 did not induce large consistent effects on the membrane properties in these pyramidal neurons (Fig. 5B). On occasion, a small 2-3 mV depolarization of the resting membrane potential and shortening of the afterhyperpolarization following a train of spikes was observed after ethanol application. However, input resistance of cells was not significantly altered in the presence of ethanol (input resistance: without ethanol, 43.7 - 5.6 Mf~; 30 mM ethanol, 41.6 --+ 4.6 MV~; n = 10).

Effect of chronic ethanol treatment on intracellularly recorded baclofen-induced hyperpolarization In the last experiment an attempt was made to determine whether chronic ethanol exposure might induce a change in postsynaptic G A B A B receptor-mediated inhibition. For these studies, baclofen (10 aM) was bath-applied rather than by pressure ejection in order to better quantitate the baclofen application across treatment groups. Baclofen (10/~M) was used in these studies since it induced a large consistent hyperpolarizing response that was somewhat less than the near maximum responses induced by higher concentrations of the drug (see Fig. 6A). When baclofen (10/,M) was bath-applied for 5-8 min to slices taken either from ethanol-treated or ethanol-naive rats, hyperpolarizing responses were qualitatively similar (Fig. 6B). Extrapolation from responses obtained when the membrane potential was held at approximately -60, -70 or -80 mV suggested that the hyperpolarization induced by baclofen had a reversal potential o f - 9 1 mV (Fig. 6C). This estimate is consistent with the widely held assumption that postsynaptic G A B A B receptors activate a K + conductance in CA1 pyramidal neurons 41. For results obtained from cells exposed to ethanol treatment, linear regression analysis indicated that the data were best fit by a different line from that fitting the ethanol-naive results (F = 9.33, df = 2, n = 67, P < 0.001; Fig. 6C). However, the estimated reversal potential o f - 9 0 . 5 mV was very similar. Point to point comparison showed that the apparent 30% reduction in the response to baclofen for ethanol-treated cells, current-clamped at -60 mV, was statistically significant. This suggested that a reduction in sensitivity of G A B A B receptors to baclofen might be present in cells from ethanol-treated animals. It seems unlikely that these differences were due to effects of ethanol treatment on the general membrane properties of the cells that were sampled since there were no significant differences in resting membrane potential (RMP (mV) ethanol-treated = 69.1 --- 1.9, n = 17; ethanol-naive = 67.4 --- 1.9, n = 21) or input resistance (RIR (Mf~) ethanol-treated = 47.0 ___ 8.4; ethanol-naive = 57.9 --- 4.8) between the groups. There were no obvious differences in the general activity of ceils from ethanol-treated animals rela-

tive to controls, despite the fact that these tissues were undergoing withdrawal from ethanol. DISCUSSION The present results suggest that pharmacologically relevant concentrations of ethanol do not acutely enhance either presynaptic or postsynaptic responses to baclofen recorded from pyramidal neurons in the CA1 region of the rat hippocampus. Despite the lack of an acute action of ethanol on these measures of G A B A B receptor function, the present results suggest that chronic in vivo ethanol treatment may evoke a reduction in the sensitivity of CA1 pyramidal cells to the hyperpolarizing effects of baclofen. Such an adaptive decrease in the activity of postsynaptic G A B A B receptors would be consistent with a role for reduced GABAergic inhibition in the development of ethanol tolerance and physical dependence. A similar adaptation in G A B A B receptor function following repeated ethanol treatment may be responsible for the reported reduction in baclofen-induced changes in [3H]GABA or [3H]norepinephrine release evoked by KC1 from mouse cerebral cortical slices 1°. However, no change in baclofen-induced inhibition of isoproterenol-stimulated cyclic AMP formation was observed in cortical slices from similarly treated mice 36. These results may provide partial support for the G A B A hypothesis (e.g. Introduction) with respect to down-regulation of some types of GABAB-mediated inhibition during chronic ethanol intoxication. At present there have been few studies of the interaction of ethanol with G A B A B receptor function. Failure to observe an increase in G A B A B receptor-mediated responses in the present study is consistent with earlier reports where acute ethanol treatment of hippocampal pyramidal cells did not consistently increase electrophysiologically recorded responses to G A B A 6'46. These studies did not attempt to differentiate between G A B A A and G A B A B receptor-mediated responses. However, it seems unlikely that activating G A B A A receptor responses would have masked an effect of ethanol to significantly increase GABAB-mediated hyperpolarization. In another model, we have also found that ethanol does not increase baclofen-induced inhibition of electricallyinduced contractions of the longitudinal muscle of the guinea pig ileum (unpublished observation). Despite these negative findings a recent report suggests that ethanol may increase the activity of G A B A g receptors by an interaction with G A B A B receptors in cerebellar microsacs 34, although this interaction was not confirmed in cultured spinal neurons 35. Because of the limited nature of the present results it would seem premature to rule out an acute interaction of ethanol with G A B A a recep-

90 tors in other brain areas at this time. Why chronic ethanol treatment should reduce the responsiveness of postsynaptic G A B A B receptors is not clear, especially since there is no evidence that ethanol acutely increases the activity of these receptors. It is possible that reduction of G A B A B receptors might be initiated in response to the p r o p o s e d involvement of these receptors in ethanol-induced increases in the activation of G A B A A receptors 34. H o w e v e r , this seems unlikely in light of the general failure to observe an action of ethanol to increase G A B A A r e c e p t o r function in the hippocampus (see Introduction). A n o t h e r possible mechanism that might drive an adaptive reduction of G A B A B receptors would be ethanol inhibition of N M D A recept o r - m e d i a t e d excitability in the hippocampus 31. A p p a r ently activation of postsynaptic G A B A B receptors mediates pair-pulse inhibition of N M D A - m e d i a t e d EPSPs in pyramidal neurons, while not altering presynaptic transmitter release 37. N M D A - m e d i a t e d EPSPs are also inhibited postsynaptically by acute t r e a t m e n t with ethanol although presynaptic release of transmitter does not a p p e a r to be altered 32. The a d d e d acute inhibitory effect of ethanol on N M D A - m e d i a t e d EPSPs in p y r a m i d a l neurons might lead to an offsetting down-regulation of postsynaptic G A B A B receptors during chronic ethanol treatment without evoking changes in presynaptic G A B A B receptors. The molecular basis for r e d u c e d G A B A B r e c e p t o r responses following chronic ethanol t r e a t m e n t is not yet clear. R e c e n t reports suggest that there m a y be at least 2 types of G A B A B receptors in the hippocampus. A p parently, G A B A B receptors that are anatomically pre-

REFERENCES 1 Aguayo, L.G., Ethanol potentiates the GABAA-activated CIcurrent in mouse hippocampal and cortical neurons, Eur. J. Pharmacol., 187 (1990) 127-130. 2 Allan, A.M. and Harris, R.A., Involvement of neuronal chloride channels in ethanol intoxication, tolerance and dependence. In M. Galanter (Ed.), Recent Developments in Alcoholism, Plenum, 1987, pp. 313-325. 3 Allan, A.M. and Harris, R.A., Acute and chronic ethanol treatments alter GABA receptor-operated chloride channels, Pharmacol. Biochem. Behav., 27 (1987) 665-670. 4 Allan, A.M. and Harris, R.A., A new alcohol antagonist: phaclofen, Life Sci., 45 (1989) 1771-1779. 5 Ault, B. and Nadler, J.V., Baclofen selectively inhibits transmission at synapses made by axons of CA3 pyramidal cells in the hippocampal slice, J. Pharm. Exp. Ther., 223 (1982) 291297. 6 Carlen, P.L., Gurevich, N. and Durand, D., Ethanol in low doses augments calcium-mediated mechanisms measured intracellularly in hippocampal neurons, Science, 215 (1982) 306-308. 7 Celentano, J.J., Gibbs, T.T. and Farb, D.H., Ethanol potentiates GABA- and glycine-induced chloride currents in chick spinal cord neurons, Brain Research, 455 (1988) 377-380. 8 Cooper, B.R., Viik, K., Ferris, R.M. and White, H.L., Antag-

synaptic relative to CA1 pyramidal cells differ pharmacologically from postsynaptic receptors on these neurons 1~'24. So-called postsynaptic G A B A B receptors which cause hyperpolarization a p p e a r to be blocked in the presence of the antagonist, phaclofen, and are inhibited by prior exposure of cells to pertussis toxin 11. These are the receptors which were altered by ethanol in the present study. O n the o t h e r hand, presynaptic G A B A B receptors a p p a r e n t l y regulate excitatory transmitter release onto pyramidal cells and are not inhibited by either phaclofen or pertussis toxin 11. W h e t h e r chronic ethanol t r e a t m e n t reduces the effects of baclofen postsynaptically by reducing the numbers of functional G A B A B receptors, or by reducing availability of associated pertussis toxin sensitive G T P - b i n d i n g proteins H and/or potassium channels 4~ that are thought to act as transducers for the postsynaptic G A B A B receptor, remains to be d e t e r m i n e d . In summary, acute ethanol application did not increase either pre- or postsynaptic G A B A B inhibitory actions in p y r a m i d a l cells of the CA1 region of the hippocampus. Treatment of rats with ethanol via a liquid diet in a m a n n e r known to induce severe physical dependence significantly decreased postsynaptic G A B A B inhibition without changing presynaptic responsiveness of p y r a m i d a l neurons.

Acknowledgements. This project was supported in part by Public Health Service grants AA06322 (G.D.E), RSDA AA00101 (G.D.E) and AG07805 (W.H.G.). The authors thank Ms. Annette Fincher for skillful technical assistance and Ms. Debbie Odstreil for expert word processing support.

onism of the enhanced susceptibility to audiogenic seizures during alcohol withdrawal in the rat by 7-aminobutyric acid (GABA) and 'GABA-mimetic' agents, J. Pharm. Exp. Ther., 209 (1979) 396-403. 9 Daoust, M., Saligaut, C., Lhuintre, J.P., Moore, N., Flipo, J.L. and Boismare, F., GABA transmission, but not benzodiazepine receptor stimulation, modulates ethanol intake by rats, Alcohol, 4 (1987) 469-472. 10 Daoust, M., Vadon, C., Henry, J.P. and Ollat, H., Ethanol antagonises the baclofen-induced modifications of [ 3]noradrenaline and [3H]GABA release in mice cortex slices, Alcohol Clin. Exp. Res., 14 (1990) 280 (abstract). 11 Dutar, P. and Nicoll, R.A., Pre- and postsynaptic GABA a receptors in the hippocampus have different pharmacological properties, Neuron, 1 (1988) 585-591. 12 Frye, G.D., GABA changes in alcohol withdrawal. In R.J. Porter, R.H. Mattson, J.A. Cramer and I. Diamond (Eds.), Alcohol and Seizures: Basic Mechanisms and Clinical Concepts, F.A. Davis Publishing Co., Philadelphia, PA, 1990, pp. 87-101. 13 Frye, G.D. and Breese, G.R., GABAergic modulation of ethanol-induced motor impairment, J. Pharm. Exp. Ther., 223 (1982) 750-756. 14 Frye, G.D. and Fincher, A.S., Effect of ethanol on gamma-vinyl GABA-induced GABA accumulation in the substantia nigra and on synaptosomal GABA content in six brain areas,

91 Brain Research, 449 (1988) 71-79. 15 Frye, G.D. and Griffith, W.H., Chronic ethanol treatment reduces postsynaptic baclofen responses in CA1 hippocampal neurons, Alcohol Clin. Exp. Res. 15 (1991)336. 16 Frye, G.D., Chapin, R.E., Vogel, R.A., Mailman, R.B., Kilts, C.D., Mueller, R.A. and Breese, G.R., Effects of acute and chronic 1,3-butanediol treatment on central nervous system function: a comparison with ethanol, J. Pharm. Exp. Ther., 216 (1981) 306-314. 17 Frye, G.D., McCown, T.J. and Breese, G.R., Characterization of susceptibility to audiogenic seizures in ethanol-dependent rats after microinjection of gamma-aminobutyric acid (GABA) agonists into the inferior colliculus, substantia nigra or medial septum, J. Pharm. Exp. Ther., 227 (1983) 663-670. 18 Frye, G.D., McCown, T.J. and Breese, G.R., Differential sensitivity of ethanol withdrawal signs in the rat to gamma-aminobutyric acid (GABA) mimetics: blockade of audiogenic seizures but not forelimb tremors, J. Pharm. Exp. Ther., 226 (1983) 720-725. 19 Frye, G.D., McCown, T.J., Breese, G.R. and Peterson, S.L., GABAergic modulation of inferior colliculus excitability: role in ethanol withdrawal audiogenic seizures, J. Pharmacol. Exp. Ther., 237 (1986) 478-485. 20 Frye, G.D., Taylor, L. and Griffith, W.H., Does ethanol enhance pre-postsynaptic effects of baclofen in vitro in rat hippocampal slices? Neurosci. Abstr., 15 (1989) 416. 21 Goldstein, D.B., Alcohol withdrawal reactions in mice: effects of drugs that modify neurotransmission, J. Pharm. Exp. Ther., 186 (1973) 1-9. 22 Griffith, W.H. and Taylor, L., Phenytoin reduces excitatory synaptic transmission and post-tetanic potentiation in the in vitro hippocampus, J. Pharm. Exp. Ther., 246 (1988) 851-858. 23 Hakkinen, H.-M. and Kulonen, E., Ethanol intoxication and gamma-aminobutyric acid, J. Neurochem., 27 (1976) 631-633. 24 Harrison, N.L., Lovinger, D.M., Lambert, N.A., Teyler, T.J., Prager, R., Ong, J. and Kerr, D.I.B., The actions of 2-hydroxysaclofen at presynaptic GABA a receptors in the rat hippocampus, Neurosci. Lett., 119 (1990) 272-276. 25 Huck, S., Gratzl, R. and Griesmayer, F., Ethanol has no effect on GABA-induced membrane current in cells cultured from dissociated embryonic rat hippocampus, Neurosci. Abstr., 13 (1987) 65. 26 Hunt, W.A., The effect of ethanol on GABAergic transmission, Neurosci. Biobehav. Rev., 7 (1983) 87-95. 27 Inoue, M., Matsuo, T. and Ogata, N., Baclofen activates voltage-dependent and 4-aminopyridine sensitive K + conductance in guinea-pig hippocampal pyramidal cells maintained in vitro, Br. J. Pharmacol., 84 (1985) 833-841. 28 Inoue, M., Matsuo, T. and Ogata, N., Characterization of preand postsynaptic actions of (-)-baclofen in the guinea-pig hippocampus in vitro, Br. J. Pharmacol., 84 (1985) 843-851. 29 Kerr, D.I.B., Ong, J., Prager, R.H., Gynther, B.D. and Curtis, D.R., Phaclofen: a peripheral and central baclofen antagonist, Brain Research, 405 (1987) 150-154. 30 Lanthorn, T.H. and Cotman, C.W., Baclofen selectively inhibits excitatory synaptic transmission in the hippocampus, Brain Research, 225 (1981) 171-178. 31 Lovinger, D.M., White, G. and Weight, F.E, Ethanol inhibits NMDA-activated ion current in hippocampal neurons, Science, 243 (1989) 1721-1724. 32 Lovinger, D.M., White, G. and Weight, EF., NMDA receptormediated synaptic excitation selectively inhibited by ethanol in hippocampal slice from adult rat, J. Neurosci., 10 (1990) 1372-

1379: 33 Martz, A., Deitrich, R.A. and Harris, R.A., Behavioral evidence for the involvement of gamma-aminobutyric acid in the actions of ethanol, Eur. J. Pharmacol., 89 (1983) 53-62. 34 McQuilkin, S.J. and Harris, R.A., Factors affecting actions of ethanol on GABA-activated chloride channels, Life Sci., 46 (1990) 527-541. 35 Mehta, A.K. and Ticku, M.K., Are GABA B receptors involved in the pharmacological effects of ethanol? Eur. J. Pharmacol., 182 (1990) 473-480. 36 Mizutani, H., Ucha, T. and Kuriyama, K., Functional alterations in cerebral GABA A and GABA a receptors during alcohol dependence, Alcohol Clin. Exp. Res., 14 (1990) 319. 37 Morrisett, R.A., Mott, D.D., Lewis, D.V., Swartzweider, H.S. and Wilson, W.A., GABAa-receptor-mediated inhibition of the N-methyl-D-aspartate component of synaptic transmission in the rat hippocampus, J. Neurosci., 11 (1991) 203-209. 38 Morrow, A.L., Suzdak, ED., Karanian, J.W. and Paul, S.M., Chronic ethanol administration alters ~,-aminobutyric acid, pentobarbital and ethanol-mediated 36C1- uptake in cerebral cortical synaptoneurosomes, J. Pharmacol. Exp. Ther., 246 (1988) 158. 39 Motulsky, H.J. and Ransnas, L.A., Fitting curves to data using nonlinear regression: a practical and nonmathematical review, FASEB J., 1 (1987) 365. 40 National Research Council, Nutrient requirements of the rat. In Nutrient Requirements of Laboratory Animals, Natl. Acad. Sci., Washington, D.C., 1972, pp. 56-117. 41 Newberry, N.R. and Nicoll, R.A., Direct hyperpolarizing action of baclofen on hippocampal pyramidal cells, Nature, 308 (1984) 450-452. 42 Nishio, M. and Narahashi, T., Ethanol enhances a component GABA-gated chloride channel currents in rat dorsal root ganglion neurons, Neurosci. Abstr., 14 (1988) 642. 43 Olpe, H.-R., Baudry, M., Fagni, L. and Lynch, G., The blocking action of baclofen on excitatory transmission in the rat hippocampal slice, J. Neurosci., 2 (1982) 698-703. 44 Osmanovic, S.S. and Shefner, S.A., Enhancement of current induced by superfusion of GABA in locus coeruleus neurons by pentobarbital, but not ethanol, Brain Research, 517 (1990) 324329. 45 Peng, T.C., Garner, S.C., Frye, G.D. and Crenshaw, M.A., Evidence of a toxic effect of ethanol on bone in rats, Alcohol Clin. Exp. Res., 6 (1982) 96-99. 46 Siggins, G.R., Pittman, Q.J. and French, E.D., Effects of ethanol on CA~ and CA 3 pyramidal cells in the hippocampal slice preparation: an intracellular study, Brain Research, 414 (1987) 22-34. 47 Suzdak, ED., Schwartz, R.D., Skolnick, P. and Paul, S.M., Ethanol stimulates gamma-aminobutyric acid receptor-mediated chloride transport in rat brain synaptoneurosomes, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 4071-4075. 48 Thompson, J.A. and Reitz, R.C., Effects of ethanol ingestion and dietary fat levels on mitochondrial lipids in male and female rats, Lipids, 13 (1978) 540-550. 49 Ticku, M.K., Lowrimore, E and Lehoullier, E, Ethanol enhances GABA-induced 36Cl-influx in primary spinal cord cultured neurons, Brain Res. Bull., 17 (1986) 123-126. 50 White, G., Lovinger, D.M. and Weight, F.E, Ethanol inhibits NMDA-activated current but does not alter GABA-activated current in an isolated adult mammalian neuron, Brain Research, 507 (1990) 332-336.