Brain Research, 93 (1975) 77-90

77

© Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands

INHIBITORY AND EXCITATORY EFFECTS OF CNS DEPRESSANTS ON INVERTEBRATE SYNAPSES

JEFFERY L. BARKER

Behavioral Biology Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Md. 20014 (U.S.A.) (Accepted February 17th, 1975)

SUMMARY

(1) The effects of pentobarbital were studied on the membrane properties and synaptic activity of crustacean neuromuscular junction preparations and molluscan neurons. (2) Pentobarbital selectively depressed in a dose-dependent, reversible manner the excitatory postsynaptic potentials (EPSPs) recorded at crustacean neuromuscular junctions without altering either inhibitory postsynaptic potentials (IPSPs) or postsynaptic membrane properties. (3) Pentobarbital depressed cholinergic EPSPs recorded in an identified molluscan neuron and depressed the depolarizing phase of a biphasic PSP without affecting the hyperpolarizing phase of the BPSP on the same cell. Facilitation of the EPSP was not affected. (4) Pentobarbital did not appreciably alter the reversal potentials of the EPSP and IPSP. (5) Low concentrations of pentobarbital did not alter the appearance of spontaneously occurring IPSPs, while high concentrations changed the pattern of regular IPSP input to an irregular, burst-like pattern. (6) Pentobarbital and 5 other CNS depressants (chloralose, chloroform, ethanol, and urethane) increased the excitability and altered the current-voltage relations of a cell whose membrane properties have been proposed as a model of presynaptic terminal membranes. The effects were dependent on the species of external divalent cation present. (7) The results in these invertebrate systems may provide insight into the cellular basis of the depressant and excitatory effects of these agents.

78 INTRODUCTION

The mechanisms underlying general anesthesia and the associated depression of neuronal excitability of the central nervous system (CNS) have received considerable attention over the past century (for recent review, see Seemana2). Depression of synaptic events under conditions of general anesthesia is common to much of the previous research and has prompted a number of theories relating depression of CNS functioning to depression of axonal and synaptic physiology. The depression of neuronal functioning has been related to the effects of general anesthetics on nerve cell metabolism43, but the concentrations required to produce biochemical changes are greater than those which affect axonal conduction and synaptic transmission18,3~,a3 and which are present in blood and brainl3, 5t during general anesthesia. In fact, general anesthetics protect against anoxia68. More recent theories have attempted to account for the phenomenon primarily in terms of alterations in membrane excitability, either in terms of a blockade of axonal conduction9,e7,52,62,64 or in terms of a disturbance of synaptic physiology. Anesthetics are also known to transiently produce varying degrees of CNS excitation. | have used several different invertebrate preparations with easily observable membrane properties and synaptic physiology to study the effects of pentobarbital and other CNS depressants on these parameters. In this paper evidence is presented to indicate that (1) pentobarbital selectively depresses excitatory transmission without reducing inhibitory transmission, and that (2) pentobarbital and a number of other CNS depressants tested can also directly increase neuronal membrane excitability. Coupled with previously published work showing that pentobarbital selectively depresses postsynaptic excitatory responses (coupled primarily to Na ~) of the same preparationsSa,sb, 7, the results suggest that the selective depression of synaptic excitation by pentobarbital is mediated through a postsynaptic mechanism. METHODS

Preparations Lobsters (Hornarus americanus) were used immediately after purchase. Crayfish (Orconectes virilis) were kept in a plastic container filled with water to a depth of 1 in. Aplysia californica were obtained from Pacific-Biomarine Co. (Venice, Calif.) and kept in a well-aerated seawater aquarium. Snails (Otata lactea) were obtained from Scozzaro, Inc. (Brooklyn, N.Y.) and were kept either in an active state on wet cellulose absorbant paper in a plastic container or in an inactive (dormant) state in a dry plastic container at room temperature. The walking leg neuromuscular junction preparations (either extensor or flexor muscles in the arthropodite) of both the crayfish and lobster were prepared and placed in the appropriate saline-filled chambers. Ganglia were removed from either Aplysia or snails, the connective tissue sheath enveloping the cells carefully incised and the preparation placed in the appropriate saline-filled chamber.

79 Solutions The salines contained the following salts (in mM): lobster - - 450 NaCI, 20 CaC12, 10 KC1, and 10 Tris-C1 (pH 7.8); crayfish - - 160 NaC1, 15 CaC12, 10 MgC12, 5 KC1, 5 Tris-C1 (pH 7.8); Aplysia - - 5130NaC1, 10 KC1, 50 MgC12, 10 CaC12 and 10 Tris-C1 (pH 7.8); snail - - 1130NaC1, 10 CaC12, 10 Tris-C1 (pH 7.8), 4 KCI, 5 MgClz. In many experiments K + was removed from the saline to enhance the amplitude of both the inhibitory and excitatory responses and to make current-voltage relations linear over a wider range of membrane potentials in molluscan neurons. Drugs The following drugs were obtained from the following laboratories: acetylcholine (Sigma, St. Louis, Mo.), sodium pentobarbital (Abbott Laboratories, North Chicago, Ill.), 5,5 diphenylhydantoin (K and K Laboratories, Plainview, N.Y.), ethanol, chloroform, a-chloralose and urethane (Fisher Scientific, Fairlawn, N.J.). The agents to be tested were dissolved in the saline just prior to perfusion. Twenty chamber volumes containing the desired drug concentration were used to wash the preparation and change the concentration of the drug in the bath. Some uncertainty concerning final concentrations of chloroform and diphenylhydantoin exists since the former agent is volatile while the latter is somewhat insoluble at the pH of the experimental salines. In several experiments acetylcholine (ACh) was bath applied to cell 11 in the snail. Recording One or two glass micropipettes having resistances of 10-20 M ~ were placed in the crustacean muscle fiber or an identified molluscan neuron. Cells were identified according to the system of Frazier et al. 19 in Aplysia and of Gainer 2° in Otala. One electrode was used to record the membrane potential while the other allowed passage of transmembrane current. The membrane potential and voltage response to applied current were recorded using conventional methods, displayed on an oscillescope and recorded on a rectilinear pen recorder. Stimulation of the appropriate neural input to the various preparations was accomplished using bipolar electrodes coupled to a stimulus isolation unit. Stimulation of the right connective and genito-pericardial nerves was used to elicit synaptic potentials in R15 of Aplysia, while stimulation of the left parietal nerve provoked synaptic potentials in cell 11 of Otala. RESULTS

Depressant effects on synaptic potentials Pentobarbital depressed the EPSPs recorded in lobster muscle fibers in a dosedependent, reversible manner but did not measurably alter the IPSPs recorded from the same fibers (Fig. 1). Little effect of pentobarbital on the resting membrane properties of the muscle fibers in the membrane potential range over which the synaptic potentials were recorded was observed. The concentration of pentobarbital required to depress EPSP amplitude by 50 % in 8 muscle fibers was 0.1-0.2 mM. Usually 20--40

80 COrTeX

WASH

8"4 ~L A

0-'~1

I

I

I

0

I

I I I II

I

I

I

I

Q1 m , rroe~um'At (,,~)

Fig. 1. Pentobarbital selectively antagonizes EPSPs at the lobster neuromuscular junction. EPSP amplitude graphed as a function of pentobarbital concentration after 30 min washing in drug-free solution. Specimen membrane potential traces of EPSPs and IPSPs above data points. Little effect on IPSPs observed at 0.5 m M drug. Resting membrane potential: - 8 6 mV. Calibration: 10 mV, 5 sec.

min of continuous washing with drug-free saline was required to re-establish the control EPSP amplitude following this depression. Pentobarbital reversibly depressed the depolarizing phase of a biphasic postsynaptic potential (BPSP) recorded in R15 in Aplysia 7 and cell 11 in Otala without affecting the hyperpolarizing phase (Fig. 2). At the concentrations tested (up to 0.2 mM) and under the conditions employed pentobarbital did not affect the membrane properties of the 4 cells studied over the membrane potential range at which the

IO-%ENTO J ~ ~

"~

2xl0"4pENTO ~

~

24mY

.--,,_

~ Z U , ~ G PsP + CONTROL X10-4 P~NTO e 2x1D-4FJENTO

~

"-~

HYPEI~ARIZINGPSP o CoN'r~ • 2x10~~NTO

--J " ~ , ~ , .

~/ ~/

\\ \\

I -75

"i,

1/o

~ I T ' '-65 -55 MEMBRANE POTENTIAL(mY)

I

-45

Fig. 2. Pentobarbital selectively depresses the depolarizing phase of a biphasic PSP in a snail neuron. The amplitudes of the de- and hyperp01arizing phases of the biphasic PSP recorded in cell 11 are graphed as a function of the membrane potential for control, 0. i and 0.2 mMpentobarbitat, s ~ m e n records of the PSPs above the potentials at which they were recorded, Calibration: I0 mV, 5 sec.

81 A CONTROl.

C

- 118-110-I00 -90 - 80 -70-60mV

CONTROL

PENTOeARB

-94mV~ 0.2ram PENTC~ARB

-~.v ~!I!!!!!UlI[|[[!]

UJJ,U . L L I m . ~ ~ WASH

-~v~ B ~ 12-~

oCON InAI • 0.2mM PENTO a WASH

,,ifilllI I

I

~- 8

kmV

!,).-. o i

-120

,

,

~ i ~

,

,

,

-80 -40 MEhe~ANE POTENTIAL (mV)

~

0

,

Fig. 3. Pentobarbital reduces a (presumed) cholinergic EPSP in an Aplysia neuron. EPSP amplitude as a function of membrane potential of R15 before, during, and after exposure to 0.2 mM pentobarbital (B). Specimen records above graph in A. Inset in B is I-V plot of membrane properties under same conditions (5 nA and 10 mV marks with origin at -36 mV). C: facilitation of EPSPs recorded at several potential levels when stimulus frequency was doubled under control conditions and in presence of 0.2 mM pentobarbital. Calibration: 5 mV, 5 sec. BPSPs I were recorded. The inversion potentials of depolarizing and hyperpolarizing phases were not appreciably changed by pentobarbital, nor was the duration of the hyperpolarizing phase. Pentobarbital depressed a (presumed) cholinergic EPSP recorded in Rx5 in a dose-dependent, reversible manner (Fig. 3). (The EPSP is presumed to be cholinergic since it is blocked by tubocurarine and hexamethonium and has the same inversion potential as the ACh response in this cell (unpublished observations).) Neither the inversion potential of the EPSP nor the current-voltage relations of the cell over the membrane potential range used to record the EPSPs was altered by the drug. Increasing the frequency of stimulation from 1 to 2/sec resulted in a facilitation of the amplitude of each succeeding EPSP (Fig. 3C). Although the amplitudes of these EPSPs were reduced by pentobarbital, the facilitatory effect was not altered. On occasion spontaneous IPSPs at a regular frequency could be recorded from cell 11. Relatively high concentrations of pentobarbital (0.5-1.0 mM) did not reduce the K+-dependent IPSPs (Fig. 4). Although the amplitude of the IPSPs did not change appreciably relative to control values, the duration of these potentials increased and the regularity of the input was replaced by an irregular rhythm characterized by repeated bursts of IPSPs. The concentrations of pentobarbital used did not alter the

82

A

CON

.5 P

tP

WASH

B

20-

0 CONTROL Q 0.SmM PENIOBARB • LOrnM PENTOBARB • WASH

CCON ACh ~ . .

GO

no

0.2raM PENTO

[3 o

mDO

-73 ~rCV

YIF/

-92 ~

~

-104

~"

~

~f~( ~ ~

~t@

ego

~"~J ~'

I

o~ o~" -120

I

-do

'

-,~

Fig. 4, Pentobarbital facilitates appearance of spontaneous IPSPs recorded in snail neuron. IPSP amplitude as a function of membrane potential of celt 11 before, during and after exposure to high concentrations of pentobarbital (B). Membrane potential traces of IPSPs in A. Note prolonged character of IPSP in 0.5 mM and facilitated increase in second IPSP in 1.0 mM pentobarbital. C: depression of depolarizing response to bath application of 10- ~M ACh (during bar) by 0.2 mM pentobarbital in same cell with spontaneous IPSPs. Resting membrane potential: ....76 mV. Calibration: 30 mV, 12 sec.

membrane properties of the 4 cells studied in this manner, nor was the IPSP inversion potential changed in the nominally K+-free solution. Pentobarbital (0.2 m M ) did, however, markedly reduce the Na+-dependent depolarizing response to bath application of ACh recorded in these cells receiving spontaneous IPSPs (Fig. 4C).

Excitatory efjects on membrane properties The effects of CNS depressants on neuronal membrane properties were studied using identified molluscan neurons. A number of identified cells (9, 10, and 12 in the terminology of Gainer 20) were relatively resistant to the effects of CNS depressants; however, cell 11, a neurosecretory cell with membrane properties correlated with dormancy or activity 20, was quite sensitive. Low to moderate pentobarbital concentrations (0.1 m M ) reversibly increased the excitability of cell 11 and initiated the generation of bursting pacemaker potential (BPP) activity (Fig. 5A). The I - V relations of the cell over the --30 to --60 mV membrane potential range were changed from linear to non-linear. The threshold, overshoot and duration of the spike were relatively unchanged at these concentrations, but the afterpotentials of the spike were considerably reduced. The afterpotential (of the initial spike in the BPP) in pentobarbital was 8 mV compared to 12 mV in control. Inclusion of elevated Mg 2+ in the saline (to 10 times the physiological concentration) did not prevent the BPP-inducing effect of pentobarbital. The interaction of pentobarbital with cell 11 was dependent on drug concentration and species of divalent cation. In the presence of a saline containing normal complements of K ÷ and Ca ~+, pentobarbital initiated BPP activity and altered the I - V relations of the cell at low concentrations (0.02-0.08 m M ) but restored the celt's control membrane properties at higher concentrations (0.2 m M ) (Figs. 5B and 6C). Replacement of Ca 2+ by Mg z+ in the saline bathing the cell illustrated in Fig. 5B in-

83

A1

CON (10 Ca)

0.1 PENTO

WASH

2 • PENTO o WASH

CON (10Ca,30Mg)

B

O.l PENTO

WASH

(~/~' jD/, / ~/~

g/'~//e

mV CON(10Ca) .02PENTO

CON (10Mg)

0.2 P

.02 P

.04 P

.08 P

.04 P

0.2 P

WASH

.08 P

0.4P

Fig. 5. Effects of pentobarbital on membrane properties of a bursting pacemaker cell. A: membrane potential traces of cell 11 from semi-aestivated snail before, during and after exposure to 0.1 mM pentobarbital in physiological saline (10 Ca) and with elevated Mgz+ saline. Induction of bursting pacemaker rhythm in presence of drug is evident. Resting potential: --46 mV in top control trace and --49 mV in bottom control trace. As: I-V plot of cell's membrane properties under same conditions (1 nA, 10 mV markers: origin at --30 mV). Presence of non-linear I-V curve during pentobarbital. B: another example of cell 11 from semi-aestivated snail showing lack of BPP in control (10 Ca) saline, followed by induction of BPPs up to 0.8 mM and then restoration of beating pattern in 0.2 mM drug. Resting membrane potential: - 4 7 inV. Replacement of Ca with Mg induces BPP activity (Con (10 Mg)). Addition of increasing concentrations of pentobarbital (up to 0.4 mM) produce increasingly larger oscillations. Sustained hyperpolarizing current required above 0.04 mM drug to reveal BPP-like activity. Note loss of afterpotentials and eventual depolarization-inactivation of spikes. Calibration: --40 mV, 12 see. (Threshold for spikes (or initial spike in BPP) remains between --39 and --44 mV for both cells under the various conditions.) duced BPP activity, as has been reported previously e. U n d e r these conditions increasing concentrations of pentobarbital (0.02-0.4 r a M ) resulted in an increase in both the number of spikes per burst and the amplitude o f the BPP (defined as the difference between the trough and the crest o f the BPP), a prolongation of spikes during the burst and a decrease in the amplitude of the afterpotentials. (Injection o f hyperpolarizing current under these conditions without pentobarbital could not produce a similar increase in BPP amplitude.) Eventually the m e m b r a n e potential approached 0 mV and the cell no longer oscillated spontaneously. Injection of sustained hyperpolarizing current under these conditions allowed oscillatory membrane potential behavior characterized by a slow depolarization f r o m - - 6 0 to - - 4 0 m V followed by a rapid depolarization f r o m - - 4 0 to 0 inV. The spikes became reduced in size and rapid-

84 Acc~(~oc,) .0~ PENTO

.04P

.08P

.2P

.4P

t!!i!!t i!!1II!it

BCON(10~) rIO0 ETHAN(D(

~[5CHCI3

tl CHLORALOSE

C CON(IOCa) ~

.05 PENTO

.I P

.2 P

WASH

~L

D CON(10Sr)

.3CHCl3

.6CHCt3

J m Fig. 6. Effects of CNS depressants on membrane properties of a bursting pacemaker cell. Membrane potential traces from cell 11. A: pentobarbital increases the BlaP amplitude (difference between trough and crest of BPP) of cell 11 from an active snail by moving the trough and crest of the BPP in

opposite directions. B: ethanol (100 mM), chloroform (0.5 mM) and chloralose (1 mM) all increase the BPP amplitude of cell 11 from an semi-aestivated snail bathed in 10 Mg2+. Movement of both the trough and the crest of the BPP as well as an increase in the number of spikes per BPP is evident. First section of ethanol and middle sections of chloroform and chloralose at one-tenth paper speed of rest of parts. C: biphasic effect of pentobarbital on cell 11 from semi-aestivated snail in 10 Ca2+. Low concentrations induce BPP activity while moderate concentration restores membrane properties to those resembling control state. D: pentobarbital increases amplitude of and prolongs BPP from active snail bathed in 10 Sr~+. Calibration: 30 mV, 12 sec. ly inactivated. These excitatory effects of pentobarbital on membrane properties were

not reversible by prolonged washing in drug-free solution unless Ca ~+ was added to the saline. Similar, irreversible effects on membrane properties were evident when Sr 2+ was the only available divalent (Fig. 6D). Pentobarbital also reversibly altered the membrane properties of cells from active snails, which generated BPPs in normal saline (Fig. 6A). The drug increased the amplitude of the BPP, from 8 mV in control to 16 mV in 0.4 m M pentobarbital by moving the crest of the BPP 5 mV in the depolarizing direction and the trough 3 mV in a hyperpolarizing direction (relative to the potentials of the crest and trough under control conditions). The threshold, overshoot and duration of the spike appeared relatively unchanged, although the afterpotentials were somewhat reduced in size. The number of spikes per BPP increased by 50 ~ . Similar results were obtained with the other agents tested (Fig. 6B). DISCUSSION

Depressant effects of pentobarbitai Many investigators have reported that excitatory transmission ~2 and postsynapti,

85 cally recorded EPSPs are invariably depressed under conditions of general anesthesia 11,14,21,22,26,86,45,47,48,53,55,56,65,67. Inhibitory transmission recorded in a similar manner under the same conditions is preserved 15,34,4°,66. (In addition, at moderate concentrations general anesthetics are known to potentiate excitatory potentials recorded from primary afferent fibersS,X6,17,a8.) These potentials, recorded postsynaptic in relation to primary afferent-motoneuron synapses reflect a depolarization of primary afferent terminals and are considered to be the event underlying presynaptic inhibitiont7.) The generalized depression of EPSPs and preservation of IPSPs has prompted a number of explanations to account for the depression of neuronal excitability present during general anesthesia. They include: (1) blockade of axonal conductiong,27,52,~2,64; (2) decrease in release of excitatory transmitter 25,28,a6' aT,a9,4z,46,48,ns; (3) increase in release of inhibitory transmitter4°; (4) decrease in the chemosensitivity of the postsynaptic receptor for the transmitterZ2,26,46'47'sa'55; (5) stabilization of the postsynaptic membrane to inhibit action potential generation 1~; (6) increase in postsynaptic membrane potential and conductance12,2s,8°'49'5°; and (7) release of intracellular Ca 2+ (leading to the events described in 6)2a,30. The effects of anesthetics on neuronal membranes have been examined using invertebrate preparations as pharmacological models since these preparations allow prolonged intracellular recording and rapid exchange of media. Pentobarbital selectively depressed EPSPs at crustacean neuromuscular junctions without having an observable effect on the muscle's membrane properties or on the IPSPs. The EPSPs in these preparations are considered to be mediated by glutamate coupled primarily to an increase in Na + conductance57a, 60, while the IPSPs are thought to be mediated by GABA coupled to an increase in C1- conductance 1°,24,61. The drug also depressed Na+-dependent EPSPs in an Aplysia neuron (presumed to be mediated by ACh) without affecting the membrane properties of the cell, the inversion potential of the EPSP or its facilitation. The Na+-dependent depolarizing phase of a presumed dopaminergic BPSP a,4,29 was also selectively reduced by the drug without affecting the K÷-dependent hyperpolarizing phase. High concentrations of pentobarbital did not depress spontaneous K+-dependent IPSPs, but rather changed their regular pattern of input into facilitated bursts. The foregoing results suggest that the selective depression of EPSPs by pentobarbital in these preparations is probably not due to a block of axonal conduction 1.1) since all forms of inhibitory synaptic transmission studied remained intact. No evidence for changes in post-synaptic membrane properties which would selectively affect EPSPs was found, reducing the likelihood that proposals (5), (6), and (7) could account for the results. (The changes observed by previous investigators in postsynaptic membrane properties in these preparations are due to the use of anesthetic concentrations 4-100 times those required to affect synaptic events and to produce anesthesia TM 49,50.) Evidence of an increase in the release of inhibitory transmitter (3) was observed at concentrations higher than those necessary to depress excitatory transmission, but the depression of EPSPs could not be related to these changes in inhibitory synaptic activity. A decrease in postsynaptic chemosensitivity (4) remains a possible explanation, but since only the depolarizing phase of the dopaminergic BPSP was affected,

86 the explanation is not probable. Furthermore, postsynaptic responses to glutamate and ACh which are coupled to conductance mechanisms other than Na + are unaltered by pentobarbital~a,sb, 7, suggesting that depression of chemosensitivity (or receptor blcckade) by pentobarbital is unlikely to account for the depression of EPSPs unless the receptors coupled to different conductances are themselves distinct physicochemically. The fact that pentobarbital selectively and reversibly antagonizes postsynaptic excitatory responses coupled primarily to Na ÷ without affecting inhibitory responses coupled to Cl- or K + in these preparations 5~,7 argues for a postsynaptic site of action either on the mechanism coupling the receptor to the Na + conductance or on the conductance itself. A similar, selective depression of post-synaptic excitations has recently been reported for mouse spinal neurones in tissue cuiturO 6a. It is possible that the EPSP depression results from a selective depression of either axonal conduction in excitatory fibers or of the release of transmitters mediating excitatory events (2). This would suggest that either the axons or the presynaptic release mechanisms of fibers involved in excitatory transmission are different from those of fibers involved in inhibitory transmission. This possibility has not been excluded. Evidence of a decrease in the release of excitatory transmitter derives primarily from observations made at the primary afferent-motoneuron synapse which is subject to a presynaptic form of inhibition 65. The decrease in quantal content of EPSPs recorded at this synapse might be due to a direct effect of pentobarbital on primary afferent membranes so as to change their membrane properties and thus decrease evoked transmitter output. In fact, depolarization of primary afferents associated with an increase in conductance has recently been observed (A, Padjen, personal communication). These events would effectively decrease the amplitude of the presynaptic spike and thereby reduce evoked transmitter output. Thus, at this and other synapses in the CNS which are subject to presynaptic inhibition, pentobarbital might produce a tonic form of presynaptic inhibition in addition to a well-described phasic form of presynaptic inhibition 5,16,17,38.

Excitatory effects of CNS depressants All of the agents tested increased the excitability of a neurosecretory cell either by inducing or potentiating BPP activity. These effects would serve to increase the release of neurosecretory product and were observed using either subanesthetic concentrations or concentrations comparable to those present during anesthesia. (The BPP in these cells is due to a time- and voltage-dependent K + conductance (with slow inactivation) coupled to a voltage-dependent Na + conductance (with slow inactivation), the presence of non-linear I-V relations reflecting the voltage-dependent Na + conductance 54. The excitatory effects of the drugs appeared to result primarily from actions on both the slowly inactivating K + conductance associated with the hyperpolarizing phase of the BPP and the K + conductance associated with spikes. (A shift of the voltage dependency of the K + conductance in the depolarizing direction in pentobarbital has been observed under voltage clamp (J. L. Barker and T. G. Smith, unpublished observations).) The induction of slow depolarizing waves in other

87

molluscan nerve cells by barbiturates 58 and halogenated anesthetics41 may reflect similar actions. These effects were especially evident (and not reversible) in Mg 2+ and Sr z+. Neither divalent cation can regulate the appearance of the BPP (and its underlying monovalent conductances) as well as Ca z+ 6,8. By inducing BPP activity in dormant cells which cannot normally generate this activity in Ca 2+ and by potentiating BPP activity in all cells bathed in Mg z+ and Sr 2+, the CNS depressants appear to be antagonizing the regulatory roles of these divalents. The restoration of control membrane properties by moderate concentrations of the anesthetics in ceils from dormant snails bathed in Ca 2+ suggests a biphasic interaction between the drugs and Ca 2+ in the regulation of excitability. The membrane properties of R15 (homologous to cell 11) have recently been proposed as a model of the presynaptic terminal membrane, since a facilitated Ca 2+ influx occurs during the burst gT. If R15 and cell 11 can be considered as models of presynaptic terminal membrane physiology, then CNS depressants would appear to be able to increase the magnitude and duration of depolarizing events at the terminal through effects on the K + conductance of action potentials invading the terminal. Barbiturates reduce excitatory synaptic transmission at peripheral vertebrate synapses, reducing EPSP and miniature endplate potential amplitude, but increasing quantal content 44,63,67. The increase in quantal content is associated with an increase in the duration of the action potential recorded at the presynaptic terminal 6a, an event known to augment transmitter release. These results indicate that evoked transmitter release is increasing while transmitter effect is decreasing. Thus, depression of EPSP amplitude cannot result from a decrease in evoked transmitter release, but more likely is due to the well-established postsynaptic depressant effects of general anesthetics here1,2, 22, 26,35,45,46,62. Barbiturates may disturb the K + conductance of the action potential at the terminal in a manner similar to that observed at the molluscan model cells. The CNS excitation observed with some anesthetics may therefore be due to (1) an increase in excitatory transmitter release without sufficient antagonism of postsynaptic excitation, and (2) effects on neuronal membrane properties to increase their inherent excitability, as observed with these invertebrate cells.

REFERENCES 1 ADAMS, P. R., The mechanism by which amyobarbitone and thiopcntone block the end plate response to nicotinic agonists, J. Physiol. (Lond.), 241 (1975) 62-63P. 2 ADAMS,P. R., CASH, H. C., AND QUILLIAM,J. P., Extrinsic and intrinsic acetylcholine and barbiturate effects on frog skeletal muscle, Brit. J. Pharmacol., 40 (1970) 552-553P. 3 ASCnER, P., Inhibitory and excitatory effects of dopamine on Aplysia neurons, J. Physiol. (Lond.), 225 (1972) 173-209. 4 Ascn~R, P., Excitatory effects of dopamine on molluscan neurones. In Frontiers in Catecholamine Research, J. Neurochem., Suppl. in press. 5 AUSTIN, G. M., AND PASK, E. A., Effect of ether inhalation upon spinal cord and root action potentials, J. Physiol. (Lond.), 118 (1952) 405-411. 5a BARKER, J. L., Activity of CNS depressants related to hydrophobicity, Nature (Lond.), 252 (1974) 52-54. 5b BARKER, J. L., CNS depressants: effects on post-synaptic pharmacology, Brain Research, 92 (1975) 307-323.

88 6 BARKER,J. L., AND GAINER, H., Role of calcium in the seasonal modulation of pacemaker activity in a molluscan neurosecretory cell, Nature (Lond.), 245 (1973) 462-464. 7 BARKER, J. L., AND GAINER, H., Pentobarbital: selective depression of excitatory post-synapfic potentials, Science, 182 (1973) 720-722. 8 BARKER,J. L., AND GAINER, H., Studies on bursting pacemaker potential activity in molluscan neurons. 1. Ionic mechanisms and underlying membrane properties, Brain Research, (1975) in press. 9 BLAUSTEIN,M. P., Barbiturates block sodium and potassium conductance increases in voltageclamped lobster axons, J. gen. Physiol., 51 (1968) 293-307. 10 BOISTEL,J., AND FATT, P., Membrane permeability change during inhibitory transmitter action in crustacean muscle, J. Physiol. (Lond.), 144 (1958) 176-191. l 1 BROOKS,C. McC., AND ECCLES,J. C., A study of the effects of anaesthesia and asphyxia on the monosynaptic pathway through the spinal cord, J. Neurophysiol., l0 (1947) 349-360. 12 CHALAZONIT1S, N., Selective actions of volatile anesthetics on synaptic transmission and autorhythmicity in single identifiable neurons, Anesthesiology, 28 (1967) 111-122. 13 CHENOWETH,M. B., ROBERTSON,D. N., ERLEY, D. S., AND GOLHKE, R., Blood and tissue levels of ether, chloroform, halothane and methoxy flurane in dogs, Anesthesiology, 23 (1962) 101-106. 14 CRAWFORD,J. M., AND CURTIS, D. R., Pharmacological studies on feline Betz cells, J. Physiol. (Lond.), 186 (1966) 121 138. 15 ECCLES, J. C., FABER, D. S., AND T/~Bof~iKOVA,H., The action of a parallel fiber volley on the antidromic invasion of Purkyne cells of cat cerebellum, Brain Research, 25 (1971) 335-356. 16 ECCLES,J. C., AND MALCOLM, J. L., Dorsal root potentials of the spinal cord, J. Neurophysiol., 9 (1946) 139-160. 17 ECCLES, J. C., SCHMIOX, R. F., AND WILmS, W. D., Pharmacological studies on presynaptic inhibition, J. Physiol. (Lond.), 168 (1963) 500-530. 18 EDWARDS,C., AND LARRABEE,M. G., Effects of anaesthetics on metabolism and on transmission in sympathetic ganglia of rats: measurement of glucose in microgram quantities using glucose oxidase, J. PhysioL,(Lond.), 130 (1955) 456-466. 19 FRAZlER, W. T., KANDEL, E. R., KUPEERMANN,I., WAZIRI, F., AND COGGESHALL, R.E., Morphological and functional properties of identified neurones in the abdominal ganglion of Aplysia californica, J. Neurophysiol., 30 (1967) 1288-1351. 20 GAINER, H., Electrophysiological behavior of an endogenously active neurosecretory cell, Brain Research, 39 (1972) 403-418. 21 GALINDO, A., Procaine, pentobarbital and halothane: effects on the mammalian myoneural junction, J. Pharmacol. exp. Ther., 177 (1971) 330-368. 22 GISSEN, A. J., KAR1S, J. H., AND NASTUK, W. L., Effect of halothane on neuromuscular transmission, J. Amer. reed. Ass., 197 (1966) 770-774. 23 GODERAIND,J. M., KAWAMURA,H., KRNJEVlt~,K., AND PUMA1N,R., Actions of dinitrophenol and some other metabolic inhibitors on cortical neurones, J. Physiol. (Lond,), 215 (1971) 199-222. 24 GRUNDFEST, H., REUBEN, J. P., AND RICKLES, W. H., The electrophysiology and pharmacology of lobster neuromuscular synapses, J. gen. Physiol., 42 (1959) 1301-1323. 25 HOLMES, J. C., AND SCHNEIDER, F . H . , Pentobarbitone inhibition of catecholamine secretion, Brit. J. Pharmacol., 49 (1973) 205-213. 26 KARIS, J. H., GIESSEN, A.J., AND NASTUK, W. L., The effect of volatile anesthetic agents on neuromuscular transmission, Anesthesiology, 28 (1967) 128-134. 27 KENDIG, J.J., AND COHEN, W . N . , Pressure reversal of anesthetic effects: conduction versus synaptic transmission, Fed. Proc., 52 (1974) 495. 28 KENNEDY, R. D., AND GAEINDO, A., Comparative site of action of various anesthetics at the mammalian myoneural junction, Fed. Proc., 52 (1974) 579. 29 KERKUT, G. A., HORN, N., AND WALKER, R. J., Long-lasting synaptic inhibition and its transmitter in the snail Helix a~persa, Comp. Biochem. Physiol., 30 (1969) 1061-1074. 30 KRNJEVIt~,K., Is general anesthesia induced by neuronal asphyxia? In B. R. FINK (Ed.), Molecular Mechanisms of Anesthesia, Raven Press, New York, in press, 31 LARRABEE,M. G., AND BRONK, D. W., Metabolic requirements of sympathetic neurons, Cold Spr. Hath. Syrup. quant. Biol., 17 (1952) 245-266. 32 LARRABEE,M. G., AND POSTERNAK,J. M., Selective action of anesthetics on synapses and axons in mammalian sympathetic ganglia, J. Neurophysiol., 15 (1952) 91-114. 33 LARRABEE,M. G., RAMOS,J. G., AND BULBRING,E., Effects of anesthetics on oxygen consumption and on synaptic transmission in sympathetic ganglia, J. cell. comp. Physiol., 40 (1952) 461-493.

89 34 LARSON,M. D., AND MAJOg, M. A., The effect of hexobarbital on the duration of the recurrent IPSP in cat motoneurons, Brain Research, 21 (1970) 309-311. 35 LEE SON, S., WAUD, B. E., AND WAUD, D. R., The effects of barbiturates on depolarization at the neuromuscular junction, Fed. Prec., 33 (1974) 510. 36 LOYNING,Y., OSHIMA,T., ANDYOKOTA,J., Site of action of thiamylal sodium on the monosynaptic spinal reflex pathway in cats, J. Neurophysiol., 27 (1964) 408-427. 37 MATI'HEWS,E. K., AND QUILLIAM, J. P., Effects of central depressant drugs upon acetylcholine release, Brit. J. Pharmacol., 22 (1964) 415-440. 38 MIYAHARA,J. T., ESPLIN, D. W., AND ZABLOCKA,B., Differential effects of depressant drugs on presynaptic inhibition, J. Pharmacol. exp. Ther., 154 (1966) 118-127. 39 MULDOON,S., LORENZ,R. R., AND VAN DYKE, R. A., Effect of halothane on adrenergic neurotransmission in isolated venous smooth muscle, Fed. Prec., 52 (1974) 496. 40 NICOLL, R. A., The effects of anaesthetics on synaptic excitation and inhibition in the olfactory bulb, J. Physiol. (Lend.), 223 (1972) 803-814. 41 PARMENTIER,J. L., AND WILSON, K. M., The effects of halogenated anesthetics on membrane processes of gastropod central neurons, Fed. Prec., 52 (1974) 495. 42 PATON, W. D. M., AND SPEOEN, R. N., Uptake of anaesthetics and their actions on the central nervous system, Brit. med. Bull., 21 (1965) 44 48. 43 QUASTEL,J, H., Biochemical aspects of narcosis, Curr. Res. Anesth., 31 (1952) 151-163. 44 QUASTEL,D. M. J., HACKETT,J. T., AND OKAMOTO,K., Presynaptic action of central depressant drugs: inhibition of depolarization-secretion coupling, Canad. J. Physiol. Pharmacol., 50 (1972) 279-284. 45 QUASTEL,D. M. J., AND LINDER, T. M., Pre- and postsynaptic actions of central depressants at the mammalian neuromuscular junction. In B. R. FINK (Ed.), Molecular Mechanisms of Anesthesia, Raven Press, New York, in press. 46 QUILLIAM,J. P., The action of hypnotic drugs on frog skeletal muscle, Brit. J. Pharmacol., 10 (1955) 133-140. 46a RANSOM,B. R., AND BARKER,J. L., Pentobarbital modulates transmitter effects on mouse spinal neurones grown in tissue culture, Nature (Lend.), 254 (1975) 703-705. 47 RICHARDS, C. D., On the mechanism of barbiturate anesthesia, J. Physiol. (Lend.), 227 (1972) 749-767. 48 RICHARDS,C. D., On the mechanism of halothane anesthesia, J. Physiol. (Lend.), 233 (1973) 439-456. 49 SATO, M., AUSTIN, G. M., AND YAt, H., Increase in permeability of the post-synaptic membrane to potassium produced by 'Nembutal', Nature (Lend.), 215 (1967) 1506-1508. 50 SATO, M., AUSTIN, G. M., YAI, H., AND MARUHASHI,J., The ionic permeability changes during acetylcholine-induced responses of Aplysia ganglion cells, J. gen. Physiol., 51 (1968) 321-345. 51 SAUBERMAN,A. J., GALLAGHER, M. L., AND HEDLEY-WHITE,J., Uptake distribution and anesthetic effect of pentobarbital 2-t4C after intravenous injection into mice, Anesthesiology, 40 (1974) 41-51. 52 SEEMAN,P., The membrane action of anesthetics and tranquillizers, Pharmacol. Rev., 24 (1972) 583-655. 53 SHAPOVALOV,A. I., Intracellular microelectrode investigation of effect of anesthetics on transmission of excitation in the spinal cord, Fed. Prec., Trans. Suppl., 23 (1963) 113-116. 54 SMITH, T. G., BARKER, J. L., AND GAINER, H., Requirements for bursting pacemaker potential activity in molluscan neurons, Nature (Lend.), 235 (1975) 450-452. 55 SOMJEN,G. G., Effects of anesthetics on spinal cord of mammals, Anesthesiology, 28 (1967) 135143. 56 SOMJEN,G. G., ANDGILL, M., The mechanism of blockade of synaptic transmission in the mammalian spinal cord by diethyl ether and by thiopental, J. Pharmacol. exp. Ther., 140 (1963) 19-30. 57 STINNAKRE,J., AND TAUC, L., Calcium influx in active Aplysia neurons detected by injected aequorin, Nature New Biol., 242 (1973) 113-115. 57a TAKEUCHI,A., ANDONDERA,K., Reversal potentials of the excitatory transmitter and L-glutamate at the crayfish neuromuscular junction, Nature New Biol., 242 (1973) 124-126. 58 TAKEUCHI,H., ET CHALAZONITIS,N., Effets du ph6nobarbital sur les neurones autoactifs, C. R. Soc. Biol. (ParisJ, 162 (1968) 491-494. 59 TAKEUCHI,A., ANDTAKEUCHI,N., On the permeability of end-plate membrane during the action of transmitter, J. Physiol. (Lend.), 154 (1960) 52-67. 60 TAKEUCHI, A., AND TAKEUCHI, N., The effect on crayfish muscle of iontophoretically applied glutamate, J. Physiol. (Lend.), 170 (1964) 296-317.

90 61 TAKEUCHI, A., AND TAKEUCHI, N., Localized action of gamma-aminobutyric acid on the craytish muscle, J. Physiol. (Lond.), 177 (1965) 225-238. 62 THESLEFF, S., The effect of anesthetic agents on skeletal muscle membrane, Acta physioL stand., 37 (1956) 335-347. 63 THOMSON, T. D., AND TURKANIS, S. A., Barbiturate-induced transmitter release at a frog neuromuscular junction, Brit. J. Pharmacol., 48 (1973) 48-58. 64 THORNTON, J. A., WHELPTON, D., AND BROWN, B. H., The effect of general anesthetic agents on nerve conduction velocities, Brit. J. Anaesth., 40 (1963) 583-587. 65 WEAKLY, J. N., Effect of barbiturates on 'quantal' synaptic transmission in spinal motoneurones, J. Physiol. (Lond.), 204 (1969) 63-77. 66 WEAKLY, J. N., ESPLIN, D. W., AND ZABLOCKA,B., Criteria for assessing effects of drugs on post-synaptic inhibition, Arch. int. Pharmacodyn., 171 (1968) 385-393. 67 WESTMORELAND,B. F., WARD, D., AND JOHNS, T. R., The effect of methohexital at the neuromuscular junction, Brain Research, 26 (1971) 465-468. 68 WILHJELM, B. J., AND ARNERED, I., Protective action of some anaesthetics against anoxics, Acta pharmacol. (Kbh.), 22 (1965) 93-98.