Electrophysiological studies on benzodiazepine antagonists

Brain Re.~earch, 295 (1984) 265 274 Elsevier

265

Electrophysiological Studies on Benzodiazepine Antagonists B. KRESPAN I. S. A. SPRINGFIELD I, H. HAAS-~and H. M. GELLER I.* ID~7~artrnent ~l'Phartnacolo*,,v, UMDNJ-Rutgers Medical School. Piscataway, NJ 08854 (U.S.A. ) and :Neuroph vsi~do*,,v Laboratory. Neurochirurgische Univer.~i#itsklinik, 8081 Ziirich (S~,it~erlaml)

(Accepted July 26th, 1983) Key words." benzodiazepines - - },-aminobutyric acid - - fl-carbolines - - benzodiazepine antagonists

The actions of the benzodiazepine (BDZ) antagonists 3-hydroxymethyl-fl-carboline (3-HMC), Ro 14-7437 and Ro 15-1788 wcrc tested on single cell activity of rat hypothalamic neurons in tissue cultures and on membrane properties of CA 1 hippocampal pyramidal neurons in transverse slices. In addition, we examined the interactions of some of these agents with inhibitions elicited by },-aminohutyric acid (GABA) as well as the ability of Ro 14-7437 to reverse the GABA-enhancing action of the BDZ agonist flurazepam BDZ antagonists did not alter patterns of spontaneous activity of hypothalamic neurons and did not affect resting membranc potential or membrane conductance in CAI pyramidal cells. Ro 14-7437 either partially or totally reversed the potentiation by flurazcpam of GABA-elicited depression of hypothalamic neuronal activity. Small and inconsistent actions on GABA-mediatcd inhibitions of hypothalamic neurons were noted. Electrically-elicited inhibitions of hypothalamic neurons were either not altered or slightly reduced. In the hippocampal slice, the frequency of spontaneous 1PSPs, the amplitude of stratum-radiatum evoked IPSPs and the conductance increase caused by stratum-radiatum stimulation were either not altered or slightly reduced. These findings demonstrate that non-convulsant BDZ antagonists block the action of BDZ agonists in facilitating GABA and further that the presence of a BDZ agonist is not required for these GABA-mediated events to occur. However, these experiments do not exclude a modulatory role for an endogenous BDZ agonist on GABA-mediated events. INTRODUCTION Recently, several c o m p o u n d s have b e e n characterized as putative ' b e n z o d i a z e p i n e a n t a g o n i s t s ' on the basis of results from behavioral and b i n d i n g studies. These c o m p o u n d s include 3-hydroxymethyl/3-carboline ( 3 - H M C ) a n d the i m i d a z o d i a z e p i n e s Ro 15-1788 ( e t h y l - 8 - f l u o r o - 5 , 6 - d i h y d r o - 5 - m e t h y l 6-oxo-4H-imidazo-[ 1,5][1,4]-benzodiazepine-3-carboxylate) and Ro 14-7437 (the desfluoro a n a l o g of Ro 15-1788). Both i m i d a z o d i a z e p i n e s competitively inhibit [3H]clonazepam a n d [3H]diazepam b i n d i n g in vitro with high affinity (IC50 = 2.3--4,5 × l(k '~ M)22,>,27 and 3 - H M C a n t a g o n i z e s [3HJdiazepam b i n d i n g in vitro ( K i = 1.5 × 10 -6 M) 6. Ro 15-1788 and 3 - H M C reverse the a n t i c o n v u l s a n t as well as other behavioral actions of d i a z e p a m 22,24,>,36. F u r t h e r more, Ro 15-1788 has b e e n d e m o n s t r a t e d to antagonize n e u r o c h e m i c a l effects of B D Z s such as the B D Z - i n d u c e d elevation of c G M P 26 and to block certain electrophysiological actions of d i a z e p a m includ* To whom requests for reprints should be addressed. 0006-8993/84/$03.110© 1984 Elsevier Science Publishers B.V.

ing those on p r e s y n a p t i c i n h i b i t i o n and polysynaptic reflexes in the spinal cord21,~1,34. At doses which antagonize the behavioral actions of B D Z agonists, these c o m p o u n d s do not display intrinsic activity, although at higher doses Ro 15-1788 has b e e n r e p o r t e d to have weak agonistic activity in behavioral tests ~,> M a n y actions of B D Z s are thought to result from a selective facilitation of }1-aminobutyric acid ( G A B A ] m e d i a t e d n e u r o t r a n s m i s s i o n which causes increased C1 influx~5. B D Z s e n h a n c e the inhibition of n e u r a l activity p r o d u c e d by e x o g e n o u s l y applied G A B A as well as G A B A - m e d i a t e d pre- and postsynaptic inhibition. A d d i t i o n a l l y , m a n y behavioral and anticonvulsant actions of B D Z s can be d i m i n i s h e d by reducing G A B A levels as well as by G A B A antagonists. These data suggest that 3 separate c o m p o n e n t s (the G A B A receptor, the b e n z o d i a z e p i n e receptor, and the CI i o n o p h o r e ) interact to p r o d u c e e n h a n c e m e n t of G A B A - m e d i a t e d events in the nervous system by B D Z s 15,-~2,33. As r e c e n t evidence has been p r e s e n t e d for the existence of e n d o g e n o u s ligands for the ben-

266 zodiazepine binding sitem, ~9, the question arises as to whether agonist activity at the BDZ binding site by an endogenous ligand may have a normal physiologic role in regulating G A B A neurotransmission. This question has been addressed indirectly by several recent electrophysiological studies utilizing a class of benzodiazepine ligands which, unlike the imidazodiazepines, are themselves proconvulsant and anxiogenic 2~,3°. Intravenous administration of ethyl fl-carboline-3-carboxylate (fl-CCE), potentiated the excitatory effects of fimbrial stimulation on CA1 pyramidal cells in the hippocampus and blocked the depressant actions of G A B A and BDZs on cells excited by fimbrial stimulation 35. The convulsant antagonist, methyl fl-carboline-3-carboxylate ~ - C C M ) , has also been reported to block the actions of G A B A in spinal cord cultures 8. In the present study, we have examined the actions of known concentrations of non-convulsant B D Z antagonists on intrinsic neuronal membrane properties as well as the interactions of these agents with GABA-mediated inhibition. In addition, we demonstrate at the cellular level the ability of the antagonist Ro 14-7437 to reverse agonist-induced increases in G A B A potency. These experiments utilized extracellular recording from hypothalamic cultures as well as intracellular recording from CA1 pyramidal cells in hippocampal slices, electrophysiological paradigms which we have used to document the GABA-enhancing actions of B D Z agonists I~->. Our present data support the conclusion that the presence of agonist BDZs is not necessary for G A B A transmission. METHODS Tissue cultures

Hypothalamic cultures were prepared from newborn Sprague-Dawley rats using techniques previously described in detail 14. Briefly, the tuberal region of the hypothalamus was removed by freehand dissection under a dissecting microscope. Pieces of the tuberal region were placed on collagen-coated coverslips, and thereafter maintained in a feeding medium containing Eagle's basal medium, horse serum, chick embryo extract and glucose, at 36-37 °C under an atmosphere of 95% air/5% CO 2. The cultures were rotated continually at 0.2 r.p.m, after the first day.

After 3-5 weeks in vitro, cultures displaying neuronal perikarya of 5-25 j4m in diameter with prominent nuclei and nucleoli were utilized for extracellular recording, as previously described L4. The cultures were placed in the recording chamber and perfused at 1-1.2 ml/min and 36--37 °C wilh Gev's balanced salt solution (GBSS). The composition of GBSS (m mM) is: NaC1, 119; KCI, 2.68: CaCI-,. 1.53: MgSOa, 0.28; MgCI 2, 1.03: NaeHPO~, tl.~4: Ktt:PO4, i).22"~ NaHCO 3, 27: and dextrose, 5.55, A single barrcl electrode filled with 3 M NaCI (5--15 M resistance) was moved through the culture under visual inspection until unit activity was recorded. Action potentials were separated from noise by a window discriminator (Model NSP-105, Labstar (7orp., Piscatawav. N J). The digitized output of the discriminator was used to generate and display peri-event histograms > on a Data General computer. Electrical stimulation was carried out by locating a bipolar electrical stimulating electrode within 100--300 ,um of the cell. Stimulating pulses (0.2 ms, 5-50 V) were delivered at 0.5 Hz. Multi-barreled micropipettes (3-6 Izm tip diameter) were used for drug application by ionophoresis. Most barrels were filled with various test drugs while one barrel contained 3 M NaCI which was used for c u r r e n t controls and t o balance drug ejection and holding currents. The drugs used were 1 M sodium glutamate, pH 8 (Sigma), 1 M G A B A , pH 4 (Sigma), IL5 M flurazepam HCI, pH 4 (Hoffman-La Roche). Drugs were applied as a series of pulses, 5-10 s in duration, repeated every 20--60 s. lonophoretic current was provided by a variable polarity constant current generator equipped with automatic current neutralization (Medical Systems, Inc., BH-2~. In some cases, G A B A (1-10 mM in GBSS) was applied by pressure ejection from a single barrel pipette as a 100 ms-1 s pulse every 5-10 s, Initially, a control histogram was constructed with drug or electrical stimulation repeated at fixed intervals at a fixed latency from the start of each sweep of the histogram. The average firing rate during a period of stimulus-bound activity (calculated by dividing the accumulated counts by the total time represented in that segment of the histogram) was compared to the firing rate during a period uninfluenced by the stimulus in the s a m e histogram and expressed as a normalized percent inhibition ~r as percent excita-

267 tion. To begin an experiment, the stimulus intensity was initially adjusted to produce an approximate

GBSS 8

50% change in stimulus-bound periods to ensure that shifts in the d o s e - r e s p o n s e curve would be reflected as changes in neuronal activity. GBSS containing known concentrations of B D Z antagonists was then superfused, and histograms constructed once again. The effect of ionophoretic drug application or electrical stimulation was again c o m p u t e d using the same two segments. The above calculation, by normalizing the effect of G A B A or electrical stimulation to baseline firing rate, permits this comparison despite possible changes in baseline firing rate. The culture was again perfused with GBSS alone, and return histograms were obtained in most cases.

lO-eM FLURAZEPAM 8

4

Hipp ocampal slices Eor intracellular experiments on h i p p o c a m p a l pyramidal cells, transverse slices of young adult (150-200 g) rat hippocampus were p r e p a r e d as previously described5 and placed in a chamber z~J which provided constant superfusion of the slices at 32 °C. A f t e r a preincubation period of 1-1.5 h, potassium chloride or potassium acetate-filled glass micropipettes (45-75 M ~ ) were utilized to impale single CA1 hippocampal pyramidal cells. A f t e r a stable impalement had been achieved (action potentials > 8 0 mV, m e m b r a n e resistance (R,,0 > 2 0 Mr2), the stratum radiatum, which contains afferents to CA1 pyramidal neurons, was stimulated by a bipolar electrode at a stimulus strength adequate to evoke an 1PSP. IPSPs were averaged on-line and displayed on a strip chart recorder. Continuous traces of m e m b r a n e potential and impedance (as produced by the voltage change to a constant-current hyperpolarizing pulse, 0.1-1.0 hA) were recorded on a penwriter. A f t e r stable records had been achieved, antagonists were added as a drop to the surface of the slice or as a continuous infusion into the superfusion fluid, and the measurements repeated. RESULTS

Interaction of benzodiazepine agonists and antagonists We first sought to demonstrate whether the B D Z antagonist, Ro 14-7437 could antagonize the G A B A enhancing actions of the B D Z agonist, flurazepam.

o

I,Z

GBSS 5-

O 0

o

I

Ill l Ill

IO'~M FLURAZEPAM / 10"5M RO14-7437 8-

o o J,

2.s SPIKE LATENCY

5 - SEC

Fig. 1. Interaction of flurazepam and Ro 14-7437 on GABAinduced depression of firing of a cultured hypothalamic neuron. GABA was applied as a 300 ms pulse by pressure ejection every 5 s producing a 68cf reduction in firing (top histogram). Superfusion of the culture with 10-a M flurazcpam enhanced the depression elicited by GABA to 95<~ and ~dso produced a reduction in spontaneous firing rate from 3.(~to 1.3 spikes/s (second histogram). Removal of flurazepam was followed bv a reduction in GABA potency to 75c÷ with no alteration in activity (1.1 spikes/s). The combination of flurazepam (lit" M) ~md Ro 14-7437 (ll~-~M) did not alter the potene} c~f GABA (,'~6c;-reduction) while the spontaneous firing rate ~ s increased to 2.2 spikes/s (bottom histogram ).

268 As we previously reported 13, the B D Z agonist flurazepam produced a pronounced enhancement of G A B A potency in depressing the activity of cultured hypothalamic neurons. In the present studies the potency of exogenously applied G A B A was enhanced in 11 of 12 cells by flurazepam (0.5-1.0 x 10-6 M). In the 7 cells tested, the GABA-enhancing action of flurazepam was partially (4 cells) or totally (2 cells) blocked or not affected (1 cell) by 10-5 M Ro 14-7437 (Fig. 1). In 3 other cells, flurazepam (10-6 M) slowed the firing rate and no recovery of activity was obtained in GBSS: Ro 14-7437 (10-~ M) alone was then superfused and the firing rate increased to previous control levels. The next series of experiments sought to determine whether the high concentration of Ro 14-7437 utilized to antagonize the actions of flurazepam could

by itself affect the potency of G A B A in depressing the activity of cultured hypothalamic neurons. Superfusion of 10-5 M Ro 14-7437 produced a mild enhancement of G A B A actions (5-20%) in 3 of 7 cells, and did not affect the actions of G A B A in the remaining 4 cells. The rate of spontaneous activity was not altered in 5 cells and increased in 2 cells during superfusion with Ro 14-7437. Fig. 2 illustrates an example of the maximal change in G A B A potency produced by Ro 14-7437. This enhancement of G A B A potency elicited by Ro :14-7437 was smaller than enhancements produced using the agonist flurazepam at 1/10 the concentration. The remainder of the experiments were performed using lower concentrations of BDZ antagonists more closely approximating the equilibrium constant in binding studies.

Effects on spontaneous activity

GBSS 10

Z m O:

The spontaneous activity of the vast majority of neurons was unaltered by the addition of either of the BDZ antagonists Ro 14-7437 (lir-s M) or 3-HMC (I(F 5 to 5 × 10-~ M) to the superfusion medium. Specifically, in the presence of 10-s M Ro 14-7437, the spontaneous firing rate of 34 neurons did not change, that of 2 neurons decreased, and I neuron increased. Likewise, during superfusion with 3-HMC, spontaneous activity was not altered in 15 neurons, decreased in 1 neuron, and increased in 3 neurons.

LU

n

Effects on G A BA-induced depression of activity

10-SM Ro 1 4 - 7 4 3 7

I--

Z

10

0 o

0

SPIKE LATENCY-SEC Fig. 2. Maximal enhancement of the GABA-elicited depression of firing of a hypothalamicneuron by 10-5M Ro 14-7437. GABA was applied by pressure ejection for 200 ms every 7 s at the time indicated by the arrow. Application of GABA completely suppressed firing for a period of 0.4 s (top histogram). During superfusion with 10-5 M Ro 14-7437, the period of complete suppression of firing was lengthened to 1 s (bottom histogram).

GABA, ionophoretically applied with currents of 0-30 nA, depressed the firing of all recorded neurons. The changes in G A B A potency elicited by superfusion of low doses of putative benzodiazepine antagonists were considered moderate as compared with the enhancement previously reported for benzodiazepine agonistsl3A6 or the antagonism observed using low concentrations of GABA-antagonists 17 During superfusion with Ro 14-7437 (10- ~ M) or 3HMC (10-5 M), no prevalent mode of interaction with GABA-induced depression was l;gted. Ro t47437 added to the superfusion medium had no effect on the depression induced by G A B A in 14 cells, enhanced (5-25%) the depression in 7 cells, and reduced (10-25%) it in 9 cells (Fig. 3). Similarly superfusion with 10--s M 3-HMC produced no consistent interaction with GABA-mediated depression: the depression of neuronal firing elicited by G A B A was un-

269

A

CONTROL 10.

GABA i

B

4

GABA 10

55-

-8

z

10

lOl

MRo14-7437

o

10.

Z

5

O

5

o

I

10

RECOVERY

o t

5¸

0 0

10

20

0

10

20

SPIKE L A T E N C Y (SEC) Fig. 3. Effect of 1(~s M Ro 14-7437 on GABA-elicited depression of firing of hypothalamic neurons illustrated by pcri-evcnt histograms. GABA was applied by diffusion out of an ionophoretic pipette by elimination of retaining current for a pcriod of 10 s every 20 s (indicated by the bar over each histogram). A: example of reduction of GABA-potency by IO s M Ro 14-7437. GABA applic~tion produced a 70(7. depression of firing rate (top histogram). During superfusion of 10-s M Ro 14-7437 the potency ol GABA was reduced to 2(Vi (middle histogram). After the removal of Ro 14-7437 the GABA potencv returned to 31(4 (bonom histogram). B: example of enhancement of GABA-elicited depression by 10-s M Ro 14-7437. Initially application of GABA induced a 60c; depression of firing (top histogram). This was enhanced to 83~74 (middle histogram) during superfusion with Rn 14-7437. The potency of depression by GABA returned to 5()G after removal of Ro 14-7437.

affected in 2 cells, a u g m e n t e d ( 5 - 3 0 % ) in 3 cells, and d e c r e a s e d ( 5 - 3 0 % ) in 4 cells• A t the highest concentration of 3 - H M C tested (5 × 105 M), a trend towards r e d u c t i o n in G A B A - i n d u c e d

d e p r e s s i o n was

o b s e r v e d . In the p r e s e n c e of 5 × 10-5 M 3 - H M C , the depression p r o d u c e d by G A B A

was u n a l t e r e d in a

m i n o r i t y of (4 of 9) cells and r e d u c e d in 5 of 9 cells

(the r e d u c t i o n in the p o t e n c y of G A B A cated several times in 3 of these 5 cells).

was repli-

Effects on electrically-evoked inhibition of activiO, E l e c t r i c a l l y - e v o k e d inhibition of n e u r o n a l firing in tissue cultures of M B H h a v e b e e n p r e v i o u s l y d e m o n strated to be blocked by picrotoxin r

thus suggesting

270 "FABLE I Effect of BDZ-antagonists on the efficacy of GA BA- and electrically-evoked inhibitions of central neurons in vitr~ Antagonist

Type of evoked inhibition

No change

Action of GABA * ....... Enhanced Decreased

10-5 M Ro 14-7437 10-8 M Ro 14-7437 l0 s M Ro 14-7437 10-6 M Ro 15-1788 10-s M 3-HMC 5 x 10-5 M 3-HMC

GABA GABA Electrical Electrical GABA GABA

3 13 4 3 2 4

4 u 1

U 7 7

(I

8

3 (I

-1 5

* Number of cells demonstrating no change, an enhancement, or a decrease in the potency of GABA-mediated inhibitions. they are mediated by e n d o g e n o u s G A B A . Superfu-

(10-6-10-8 M) did not produce consistent shifts in

sion with 10-8 M Ro 14-7437 tended to partially reduce (10-50%) such electrically-evoked depression.

resting m e m b r a n e potential, m e m b r a n e resistance or

The depressions were reduced in 7 cells (recovery was attained in 4/7 cells), e n h a n c e d in 1 cell, and unaffected in 4 cells. These effects of B D Z antagonists on G A B A - i n d u c e d and electrically-evoked inhibitions of hypothalamic n e u r o n s have been summa-

excitability, as measured by the m i n i m u m depolarizing current necessary to evoke an action potential. A typical record from such an experiment is illustrated in Fig. 4. Several parameters of G A B A - m e d i a t e d inhibition were examined before and during antagonist super-

rized in Table 1.

fusion on different cells. The amplitudes of stratum-

Effects on m e m b r a n e properties and G A B A - m e d i a t e d

radiatum evoked IPSPs were measured on 8 neurons, impaled with potassium acetate electrodes. Five of these demonstrated a moderate decrease (10-30%)

IPSPs

Recordings were made from 18 hippocampal pyramidal n e u r o n s during superfusion with Ro 15-1788 or Ro

A

14-7437.

Superfusion

of

these

antagonists

in IPSP amplitude during perfusion o f antagonists, while in 3 n e u r o n s IPSP amplitude remained constant during antagonist perfusion. The experiment illus-

IO-eM Ro15-1788 [11

I[

II

/

I

50

msec

B

50 msec Fig. 4. Contrast in effects of benzodiazepine antagonist Ro 15-1788 and benzodiazepine agonist midazolam on evoked IPSPs, evoked activity and membrane conductance in CA1 pyramidal neurons. In each record, upper trace is an average of 10 evoked IPSPs. Lower trace is an intracellular record demonstrating the IPSP evoked by stimulation of stratum radiatum (200 ,us, 50 ~A, 0.2 Hz) and membrane response to injected hyperpolarizing and depolarizing current pulses of 0.3 nA. A: initial record obtained during superfusion with Yamamoto's solution. Sequential records were taken after 5, 10 and 15 rain after superfusion of 10-6 M Ro 15-1788, showing slight reduction in the IPSP without any change in membrane resistance or excitability. B: the same experimental paradigm in another cell demonstrates an enhancement of the evoked IPSP by 10-6 midazolam, without any change in excitability or membrane resistance.

271 trated

in Fig. 4 demonstrates the

reduction of

4 and 5. As noted in these figures, these actions of

averaged evoked IPSPs by perfusion of Ro 15-1788

B D Z agonists on G A B A - m e d i a t e d inhibition were

at a time when m e m b r a n e resistance and excitability

observed at concentrations which did not alter membrane potential, conductance or excitability.

were unchanged. On 3 neurons, we were able to examine the effect of the benzodiazepine antagonists on the conductance decrease following stratum-radiatum stimulation by injection of a sine wave current through the membrane. This current was adjusted to produce an approximate 15 mV change in membrane potential, and usually elicited spike activity which was decreased following stimulation of stratum radiatum. One such experiment is illustrated in Fig. 5, where Ro 15-1788 slightly reduced both the

Six neurons were impaled with KCI electrodes in order to increase intracellular CI- concentration and examine the amplitude and frequency distribution of spontaneous IPSPs 2. On 4 of these the n u m b e r of IPSPs per unit time with amplitude greater than 2x baseline noise declined during antagonist perfusion. On 2 neurons no change in IPSP frequency was observed,

change in conductance and reduction of activity elic-

DISCUSSION

ited by focal (stratum radiatum) stimulation. Similar effects were noted in both of the other cells. In contrast to B D Z antagonists, the B D Z agonists, midazolain (I(F~, M) and diazepam (10-7 M) caused consis-

7437, Ro 15-1788 and 3-HMC, at concentrations which displace B D Z agonists from their binding sites,

tent increases in the amplitude of stratum-radiatum evoked IPSPs and also prolonged the reduction of activity following focal stimulation as illustrated in Figs.

A

Ir

Ii'!'

In the present study, the B D Z antagonists Ro 14-

did not consistently affect resting neuronal membrane potential or conductance or patterns of spontaneous neuronal activity. These same concentrations

IO-SM R o 1 5 - 1 7 8 8

r

TI

I,

E o

5 0 msec

B

10-7M DIAZEPAM

E

I

O

1 O0 msec

Fig. 5. Effect of benzodiazepine antagonist Ro 15-1788and diazepam on averaged IPSPs recorded in CAt pyramidal neurons by stimulation of stratum radiatum ( l(l(/!~s.401~A, 0.2 Hz). A continuous sine wave current was injected through the recording electrode with an amplitude adjusted to evoke action potentials on the depolarizing phase (2 nA). Stimulus presentation results in a conductance increase as indicated by the narrowing of the trace and an inhibition of spike generation as indicated by the pause in firing. The stimulus artifact is visible below each trace .just prior to the narrowing of the trace. Each record is the summation of 10 stimulus presentations. A: superfusion of l(t4~M Ro 15-1788does not produce an alteration in conductance change or the inhibition in spike generation. B: superfusion of 10 7 M diazepam on another cell results in an increase in conductance change and pause in spike generation.

272 of B D Z antagonists generally displayed moderate interactions with GABA-mediated events as compared with actions of benzodiazepine agonists. The present results should also be considered in the context of our previous work which demonstrated the consistent GABA-enhancing actions of agonist B D Z ligands in the same preparations t6,3'~. The results of the present experiments are summarized in Table I. Following addition of B D Z antagonists, no trend is observed when all cells are considered as a group. The potency of G A B A was unchanged in 40% of cells, enhanced in 20% and decreased in 40%. A trend is observed, however, when the experiments are analyzed by dose and type of antagonist. The interactions of 3-HMC or Ro 14-7437 with exogenously-applied G A B A did not show consistent changes. Moderate reduction or potentiation of G A B A was observed on about l/3 of the cells tested with these agents, while about 1/3 showed no change. The highest dose of 3-HMC did tend to reduce G A B A potency, while the highest dose of Ro 14-7437 predominantly enhanced G A B A actions. On the other hand, electrically-evoked depression of spontaneous activity in the hypothalamus and electrically-evoked recurrent inhibition of hippocampal pyramidal cells were generally reduced by the antagonists Ro 15-1788 and Ro 14-7437. A majority (65%) of the cells showed a reduction in inhibition, 30% showed no change, while on only 5% (1 cell) were inhibition enhanced. These results, demonstrating that G A B A actions are not strongly modulated by antagonist occupation of the B D Z receptor while BDZ-agonist enhancement of G A B A transmission is reduced, are in general agreement with both behavioral and biochemical studies. These studies showed that low doses of Ro 15-1788 can reverse the anticonvulsant actions of B D Z agonists but not G A B A agonists 24,29 and that Ro 15-1788 can reverse diazepam but not muscimol induced increases in cGMP levels in the cerebellum 26. These results are also in accord with evidence of GABA-benzodiazepine interactions from binding studies. The presence of G A B A does not affect the affinity of B D Z antagonists for B D Z receptors 4.25.27 whereas the affinity of B D Z agonists for B D Z binding sites is increased approximately two-fold by G A B A 4,23.37. Taken together, the results strongly support the conclusion that B D Z ligands

without efficacy cannot alter the configuration of the BDZ binding site in a way which permits an interaction with GABA. These results should also be considered in contrast to other electrophysiological investigations using B D Z antagonists. Polc 35 found that the B D Z antagonist ethyl fi-carboline-3-carboxylate (fl-CCE) blocked the suppressant actions of G A B A on population spikes in the CA1 region of the hippocampus. Another benzodiazepine ligand, /]-CCM, has been found to block the conductance increase induced by G A B A in cultured spinal neurons~. The difference between the present results and those with/3-CCE and fl-CCM parallel differences in behavioral and biochemical actions of these ligands. At moderate doses, Ro 15-1788 and 3-HMC are non-convulsant and non-axiogenic B D Z antagonists whereas/~-CCM is a convulsant 3.4 and/$-CCE has proconvulsant and anxiogenic actions 7,'~,29,3~,3s. Both the proconvulsant and the anxiogenic actions o f / # ( I C E are blocked by Ro 15-1788~2,2~,29 one of the antagonists employed in the present study. In addition, Braestrup et al? have recently shown that both ~-CCE and fi-CCM fall into a class of BDZ-ligands which demonstrate a negative shift in affinity for the BDZ receptor in the presence of G A B A whereas Ro 15-1788 does not. Thus the differences in results of electrophysiological, behavioral and biochemical experiments would indicate that the proconvulsant ligands arc not conventional antagonists as are the compounds used in this study. At high doses, Ro 15-1788 shows weak B D Z agonistic activity in some behavioral paradigms. The agonistic activity consists of weak antiaversive actions, potentiation of the anticonvulsant activity of the G A B A agonist progabide24, and anticonvulsant activity against pentylenetetrazol and bicuculline 29. A weak GABA-enhancing action might also be predicted from the fact that the binding of Ro 15-1788 to the BDZ receptor is weakly enhanced by G A B A ~-I:. However, mixed agonistic and antagonistic actions were noted in the effects of Ro 15-1788 on kindled amygdaloid seizures1, suggesting partial agonist-like activity. Whether Ro 14-7437 behaves similarly in these behavioral and chemical tests is not known. However, our data indicate that Ro 14-7437 might be expected to act similarly since, at lower concentrations, either enhancement or reduction of G A B A potency by Ro 14-7437 was observed, while at higher

273 concentrations,

a trend

towards

enhancement

of

G A B A by R o 14-7437 was o b s e r v e d . in h y p o t h a l a m u s

and

transmission. H o w e v e r , these

ligands are p r e s e n t , they play at most a small role in data also suggest that if e n d o g e n o u s B D Z - a g o n i s t

T h e t r e n d towards r e d u c t i o n of e l e c t r i c a l l y - e v o k e d inhibition

necessary for G A B A

hippocampus

by

Ro 14-7437 and R o 15-1788 may be a partial antagonism

m o d u l a t i n g the actions of G A B A

in the h y p o t h a l a -

mus or h i p p o c a m p u s following electrical stimulation.

of the actions of G A B A after release f r o m local circuit inhibitory neurons. A l t e r n a t i v e l y , it is possible that an e n d o g e n o u s

B D Z - l i k e substance

ACKNOWLEDGEMENTS

is co-re-

leased by electrical stimulation and its actions are

T h e e x p e r i m e n t s d e s c r i b e d in this publication w e r e

blocked selectively by B D Z antagonists. O u r data,

s u p p o r t e d in part by U.S. N I H G r a n t NS-14568 to

which indicate no t r e n d in the interactions of B D Z

H . M . G . , and travel award f r o m the U . S . - S w i s s Co-

antagonists with e x o g e n o u s l y a d m i n i s t e r e d G A B A ,

o p e r a t i v e Science P r o g r a m of the U.S. N a t i o n a l Sci-

but

ence F o u n d a t i o n . The a u t h o r s t h a n k Drs. F. Baldi-

a reduction

in the

potency

of e n d o g e n o u s

G A B A , could be e x p l a i n e d by such a m e c h a n i s m . A

no, Jr. and J. C h i s h o l m for helpful discussion and as-

similar hypothesis has b e e n i n v o k e d by File et al. 12 to

sistance and R o s e m a r i e P e s c h e k for her skilful typ-

explain the w e a k a n x i o g e n i c action of R o 15-1788.

ing. T h e authors also thank Dr. P. Skolnick for the

O u r results suggest that, in a m o d e l of the G A B A - r e -

gift of 3 - h y d r o x y m e t h y l - f l - c a r b o l i n e

c e p t o r B D Z r e c e p t o r C1 i o n o p h o r e c o m p l e x , occu-

H a e f e l y and H o f f m a n n - L a R o c h e for their gifts of

pation of a B D Z r e c e p t o r by a B D Z agonist is not

b e n z o d i a z e p i n e s and R o 15-1788 and Ro 14-7437.

REFERENCES 1 Albertson, T. E., Bowyer, .I.F. and Paule, M. G., Modification of the anticonvulsant efficacy of diazepam by Ro 151788 in the kindled amygdaloid seizure model, Lifo Sci., 31 (1982) 1597-1601. 2 Alger, B. E. and Nicoll, R. A., Spontaneous inhibitory post-synaptic potentials in hippocampus: mechanism for tonic inhibition, Brain Research, 200 (1980) 195-200. 3 Braestrup, C. and Nielsen, M., GABA reduces binding of ~H-methvl fl-carboline-3-carboxylate to brain benzodiazepine receptors. Nature (Lond.). 294 (1981) 472-474. 4 Braestrup, C., Sehmiechen, R., Neef, G., Nielsen, M. and Petersen, E. N., Interaction of convulsive ligands with benzodiazepi nc receptors, Science, 216 (1982) 1243-1251. 5 Buckle, P. ,1. and Haas, H. L., Enhancement of synaptic transmission by 4-aminopyridinc in hippocampal slices of the rat, J. Physiol. (Lond.). 326 (1982) 100-122. 6 Cain. M., Weber, R. W.. Guzman, F., Cook, J. M., Barker, S. A., Rice, K. (7., Crawley, J. N,, Paul, S. M. and Skolnick. P., /,kcarbolines: synthesis and neurochemical and pharmacological actions on brain benzodiazepine receptors, J. Med. Chem., 25 (1982) 1081-i091. 7 Cepeda, C., Tanaka, T., Besselievre, R., Potier, P., NaqueL R. and Rossier, J., Proconvulsant effects in baboons ol fi-carboline, a putative endogenous ligand for benzodiazepine receptors. Neurosci. l,ett.. 24 ( 1981 ) 53-57. 8 Chan. C. Y., Gibbs, T. T. and Farb, D. H., Action of betacarboline in flunitrazcpam-photolinked cultures, Soc. Neuro~ci. Ab.str.. 8 (1982) 572. 9 Cowen, P. ,1., Green. A. R., Nun, D. J. and Martin, I. L., Ethyl/~'-carboline carboxylate lowers seizure threshold and antagonizes flurazepan>induced sedation in rats, Nature (Lond.). 290 (1981) 54-55. 10 Davis, L. G., Is a peptide the natural ligand for the benzodiazcpine receptor? In E. Usdin, P. Skolnick, J. F. Tab

11

12

13

14

15

16

17

18 19

20

and

Dr.

W.

Imam Jr., D. Greenblatt and S. M. Paul (Eds.), Pharmacology of Benzodiazepines. Macmillan, London, 1982, pp. 537-547. Ehlert, F. J., Ragan, P., Chen, A., Roeske, W. R. and Yamamura, H. 1., Modulation of benzodiazepine receptor binding: insight into pharmacological efficacy, Europ. J, Pharmacol., 78 (1982) 249-253. File, S. E., Lister, R. G, and Nun, D. J., The anxiogenic action of benzodiazepine antagonists, Neuropharmacology. 2l (1982) 1033-1037. Geller, H. M., Water soluble benzodiazepines with agonistic and antagonistic actions on GABA-induccd inhibition in cultured hypothalamus, Neurosci. Lett., l 5 (1979) 313-318. Geller, H. M., Histamine actions on activity of cultured hypothalamic neurons: evidence for mediation by 1-1,-and H,histamine receptors, Develop. Brain Res.. I ( 1981 ) 89-101. Geller, H. M., Hoffer, B. J. and Taylor, D. A., Electrophysiological actions of bcnzodiazepines, t'~'d, l'roc., 39 (1980) 3016-3023. Geller, H. M., Taylor, D. A. and Hoffcr, B. J., Benzodiazepines and central inhibitory mechanisms, Naunvn-Schmiedeberg's Areh. Pharmacol., 304 ( 1978 ) 81-88. Geller, H. M. and Woodward, D. J., Synaptic organization of tuberal hypothalamus in tissue culture: effects of electrical stimulation and blockers of synaptic transmission, Exp. Neurol., 64 (1979) 535-552, Gottesfeld, Z., Kelly, J. S. and Renaud, L. P., Thc in vivo neuropharmacology of amino-oxyacctic acid in the cerebral cortex of the cat, Brain Research, 42 (1972) 311-318. Guidotti, A., Forchetti, (7. M., Ebstein, B. and Costa, E., Purification and characterization of an endogcnous pcpfidc putative effector for the benzodiazepine recognition site. In E. Usdin, P. Skolnick. J. F. Talhnan, .lr., D. Grecnblan and S. M. Paul (Eds.), Pharmacology of Benzodiaz~7~ine,~" Macmillan, London, 1982, pp. 529-535. Haas, H. l,., Schaerer, B. and Vosmanskv. M., A simple

274

21

22

23

24

25

26

27

28

29

perfusion chamber for the study of nervous tissue slices in vitro, J. Neurosci. Meth., 1 (1979)323-325. Haefely, W., Bonetti, E. P., Burkard, W. P., Cumin, R . Laurent, J.-P., Mohler, H., Pieri, L., Polc, P.. Richards. J. G., Schaffner, R. and Scherschlicht, R., Benzodiazepine antagonists, In E. Costa (Ed.), Benzodiuzepines - - f r o m Molecular Biology to Clinical Practice, Raven Press, New York, 1983. Hunkeler, W., M6hler, H., Pieri, L., Pole, P., Bonetti, E P., Cumin, R., Schaffner, R. and Haefely, W., Selective antagonists of benzodiazepines. Nature (Lond.), 290 (1981) 514-516. Karobath, M. and Sperk, G., Stimulation of benzodiazepine receptor binding by y-aminobutyric acid, Proc. nat, Acad. ScL U.S.A., 76 (1979) 1004-1007. Lloyd, K. G., Bovier, P., Broekkamp, C. L. and Worms. P., Reversal of the antiaversive and anticonvulsant actions of diazepam, but not of progabide, by a selective antagonist of benzodiazepine receptors, Europ. J. Pharmacol., 75 (1981) 77-78. Marangos, P. J. and Patel, J., Properties of 3H-fl-carboline3-carboxyl ethyl ester binding to the benzodiazepine receptor, Life Sci., 29 (1981) 1705-1714. M6hler, H,, Burkard, W. P., Keller, H. H., Richards, J. (5. and Haefely, W., Benzodiazepine antagonist Ro 15-1788: binding characteristics and interaction with drug-induced changes in dopamine turnover and cerebellar cGMP levels, J. Neurochem., 37 (1981) 714-722. M6hler, H. and Richards, J. G., Agonist and antagonist benzodiazepine receptor interaction in vitro, Nature (Lond.), 294 (1981) 763-765, Ninan, P. T., Insel, T. M., Cohen, R. M., Cook, J. M., Skolnick, P. and Paul, S. T., Benzodiazepine receptor-mediated experimental 'anxiety' in primates, Science. 218 (1982) 1132-1134. Nutt, D. J., Cowen, P. J. and Little. H. J., Unusual interac-

30

31

32 33

34

35

36

37

38

39

tions of benzodiazepine receptor antagonists. ~\altm ~ (Lond.), 2195 (1982) 436-438. Oakley, B. R. and Jones, B. J., The ptoconvulsant and diazepam-reversing effects of ethyl-[J-carboline-3-carboxylate, Europ. J, Pharmacol., 08 (198(t) 381-382. O'Brien, R. A., Schlosser, W., Spirt. N. M., Franco, S., Horst, W. D., Polc, P. and Bonetti, [:i. P,, Antagonism ~f benzodiazepine receptors by beta earbolines, LiCe Sci.. 2~ (1981) 75-82, Olsen, R. W., GABA-benzodiazepinc-barbiturate receptor interactions, J. Neurochem. 37 (1981) 1-13: Paul, S. M., Marangos, P. J. and Skolnick, P., I h e bcnzodiazepine-GABA-chloride ionophorc receptor complex: common site of minor tranquilizer actkm, Biol. Ps'ychiat,, 16 (1981) 213-229. Polc, P., Laurent, J.-P., Scherschlicht. R, and Haefclv, V~, Electrophysiological studies on the specific benzodiazepme antagonist Ro 15-1788, Naunyn-Schmiedeber~z's Arch, Pharmacol., 316 (1981) 317-325. Pole, P,, Ropert, N. and Wright, D, M., Ethyl-/#carboline3-earboxylate antagonizes the action of GABA and benzodiazepines in the hippocampus, Brain Research. 217 (1981) 216-220. Skolnick, P., Paul, S.. Crawley, J , Rice. K., Barker, S., Weber, R., Cain, M. and Cook, J.. 3-hydroxymethyl-fl-carboline antagonizes some pharmacological actions of diazcpam, Europ. J. Pharmacol.. 69 (1981) 525-527. Tallman. J. F., Thomas. J. W. and Gallaghcr, D. W., GABAergic modulation of benzodiazepinc binding site sensitivity, Nature (Lond.), 274 (1978) 383-385 Tenen, S. S and Hirsch, J. D., ,8-carboline-3-carboxylic acid ethyl ester antagonizes diazepam activity, Nature (Lond.). 288 (1980) 60%610, Wolf, P. and Haas, H., Effects oI diazepines and barbiturates on hippocampal recurrent inhibition, Naunyn-Schmiedeberg's Arch. Pharmacol., 299 (1977) 211-281.