Role of K+ in GABA (γ-aminobutyric acid)-evoked depolarization of peripheral nerve

Brain Research, 278 (1983) 127-135

127

Elsevier

Role of K + in GABA (7-Aminobutyric Acid)-Evoked Depolarization of Peripheral Nerve M. E. MORRIS, G. A. DI COSTANZO, A. BAROLET and P. J. SHERIDAN

Departments of Anaesthesia and Pharmacology, University of Toronto, 1 King's College Circle, Toronto, Ont. M5S 1A8 (Canada) (Accepted March 1st, 1983)

Key words: GABA - - 7-aminobutyric acid - - receptors - - potassium ions - - peripheral nerve - - myelinated fibers - - dorsal root - ventral root - - depolarization - - localization

Isolated, desheathed sciatic nerves of the leopard frog or bull frog were used in studies to determine different sources/components of the depolarizing effect of GABA (y-aminobutyric acid) on myelinated fibers. During the depolarization induced by 1 mM GABA - - which was reflected by an increase of 38.3% (S.E. + 2.2) in the amplitude of the evoked half-maximal A-fiber compound action potential - - the level of extracellular potassium ([K+]o) measured at depths ~< 200 /~m in the nerve with ion-selective microelectrodes, increased by 0.096 mM (S.E. + 0.007). Changes in excitability preceded K÷]o, and there was a significant difference between their peak latencies. Artificially raised levels of [K+]o, similar to those induced by GABA, caused extremely small changes (< 10%) in the size of the evoked action potential. From the magnitude and time course of the GABA-evoked augmentation of levels of [K÷]o, it can be concluded that potassium ions probably arise indirectly and play a secondary role in what appears to be a mainly receptor-mediated depolarization of axons. A much greater sensitivity to GABA was found for fibers of the dorsal roots in comparison with those of the ventral roots (maximal changes in excitability of 50% and 6% respectively). This suggests that the depolarization of ventral root fibers could be caused by [K+]o accumulation, and that there may be a preferentially localized distribution of receptors for GABA on the sensory axons of peripheral nerve. INTRODUCTION

this respect it has been reported that there is differential sensitivity of dorsal and ventral roots - - with

Demonstrations

of the depolarizing effects of

G A B A producing a conduction block of tetanically

G A B A (y-aminobutyric acid) on various peripheral nerves *-7.33,34.36 have led to the presumptive conclu-

well, a selective sensitivity to G A B A ' s depolarizing

e v o k e d responses of only the sensory fibers 46. As

sion that these are m e d i a t e d by specific receptors.

action has been r e p o r t e d for C fibers in studies car-

Th e exact location and nature of the extrasynaptic re-

ried out on vagus, cervical sympathetic and phrenic

ceptors, h o w e v e r , remain to be determined. It is

nerves 6.

kn o wn

that

there

is differential distribution

of

G A B A in the spinal cord and roots - - with greater concentrations of the transmitter and enzymes inv o l v e d in its m e t a b o l i s m being found in dorsally located structures 1.15.32. D e p o l a r i z i n g effects of G A B A on the dorsal root ganglia and primary afferent terminals have long b e e n recognized 3.9,10,11,27,28.38,47 and

Th e observation that at sites where G A B A

has

been shown to depolarize neuronal tissue - - the dorsal root ganglion, primary afferent terminals of the cuneate and the spinal cord - - an accompanying accumulation of extracellular K ÷ has been recorded t1.26,33,5°, raises various possibilities - - that the increased [K+]o reflects an increased K + conductance

contrast with the variable responses of m o t o n e u tons 16,37,38 and the hyperpolarizing responses found

of neuronal m e m b r a n e ; that G A B A - e v o k e d depolarization is an event secondary to or not involving re-

for most postsynaptic structures of the central nervous systeml9. A differential distribution for G A B A

ceptor activation (e.g. as a result of electrogenic neuronal and/or non-neuronal uptake17,38); or that the de-

receptors on the different fiber c o m p o n e n t s of the peripheral nerves might t h e r e f o r e be predicted. In

polarization may have multiple components.

0006-8993/83/$03.00 (~) 1983 Elsevier Science Publishers B.V.

In o r d er to further define and describe the recep-

128 tors for G A B A which have been localized to peripheral nerve, the following experiments were carried out to assess (1) differential distributions of depolarizing responses and receptors for G A B A on different fiber types of the sciatic nerve, and (2) the extent and contribution of G A B A - e v o k e d increases in extracellular [K+]o to the observed depolarization. The results of these experiments have previously been reported only partly and in a brief and preliminary form35. MATERIALS AND METHODS Desheathed sciatic nerves of either Rana pipiens (leopard frog) or Rana catesbeiana (bullfrog) were mounted in a five-chambered bath and perfused with Ringer solutions, as previously described 34. Except where specifically mentioned, most experiments were made using the leopard frog preparation; in experiments with nerves from the bullfrog an identical but larger chamber was used. Electrical stimuli were applied to evoke the half-maximal response of the A-fiber compound action potential; these evoked potentials were recorded from either the whole nerve or from one or the other of the attached dorsal or ventral roots. K+-sensitive ion-exchanger micro-electrodes were

prepared from theta-capillary tubing according to previously described techniques 24.29,30. The membrane of the ion-sensitive channel was either the Corning 47717 exchanger or a valinomycin-based mixture 39. The final tip diameter of electrodes was 2-4 Hm. The potential difference between the ionsensing and reference channels of the electrode was recorded by differential amplification with a Keithley Model 604 Electrometer and a Model 7 D A Grass Polygraph. When electrodes were calibrated in solutions between 2.5 and 25 mM [K+]/Ringer the slope of their voltage response was 49.3 mV (S.E. + 0.8, n = 9). Their response time for a 10-fold change in concentration was less than 0.5 s 24.

12Hzl

t tOHz t

_/',...._ t50Hz I

A

50

40

~

B

[ WHOLE NERV~

20

L ROOT

I/-

,0

OLI

o~

J tO - 5

~

I0 4

i

i

10-3

J

I0 2

t ]OOHz t

14"5

~

1.0

L ROOT

LIo

,'04

[oABd :M

io3

b,

Fig. 1. Effect of GABA to increase excitability of peripheral nerve: changes in amplitude of A-fiber half-maximal compound action potentials of bullfrog sciatic nerve. A: recorded separately from dorsal and ventral roots of seven nerves (mean _+ S.E.. n = 3-7 at each concentration). B: recorded from whole nerve (n = 4). Note scale of abscissa is logarithmic: curves fitted using equations from Parker and Waud 4j.

3"5

-

t2ooHzt

1

30

I

I

40s

Fig. 2. Changes in [K+]o evoked in sciatic nerve by electrical stimulation. Frequency was varied between 2 and 200 Hz (0.5 ms duration; 40 s tetanus; intensity = 3 × maximal for A-fiber compound action potential). Polygraph records of voltage changes measured with K+-sensitive microelectrode (depth = 230/~m).

129 RESULTS

GABA-evoked changes in excitability of dorsal and ventral root fibers Studies to assess effects of G A B A on different fiber groups were facilitated by using sciatic nerves of the bullfrog, which are approximately 3-4 times the length of those of the leopard frog. Fig. 1B demonstrates the depolarizing action of G A B A on whole nerves of the bullfrog over the range of 10-5 to 10-2 M, with an apparent threshold in the micromolar range, EDsc~of 0.12 raM, Hill coefficient of 0.70 and maximal increase in excitability of 30% - - values which correspond closely with those already described for the leopard frog 34. A marked difference in the response to G A B A for the two different groups of fibers which form the dorsal and ventral roots can be seen in Fig. 1A. Dorsal root fibers showed a lower threshold for change in the response, as well as a much greater sensitivity (with a computed maximal increase in excitability of 52% as compared to 6.5% for the motor fibers). Their EDs0 value was 0.8 m M and the Hill coefficient was 0.86.

exchanger for K +; valinomycin for ACh)24, 39 this suggested that any changes in [K+]o which might be observed were unlikely to be due to an accumulation of ACh.

Changes in [K+]o evoked by GABA Simultaneous measurements of the change in extracellular potassium (A[K+]o) and change in amplitude of the half-maximal action potential were made in 7 individual nerves during 31 applications of 6 different concentrations of G A B A (0.1-2.5 mM). The resting level of [K+]o measured in these nerves with 9 A

1.o 0"8

~E 0"6 ,~, 0-4 .
0.2 0

Changes in [K+]o evoked by stimulation Graded and reversible changes in extracellular K + levels (A[K+]o) which were detectable with ion-selective electrodes in response to repetitive electrical stimulation of the sciatic nerve with frequencies > 1 Hz, became marked above 20 Hz (see Fig. 2). The graphs of Fig. 3 show the dependence of these increases on changes in the frequency and intensity of stimulation observed in one preparation. Increasing the voltage - - from that which produced a half-maximal response for A fibers to the level giving a maximal response (approximately 3 times threshold strength), and to the intensity (10 times maximal for A fibers) which activates unmyelinated fibers - could, at the frequency of 100 Hz, evoke an increase in [K+]o (see Fig. 3B). However, with the much lower frequencies of 0.5 and 1 Hz used in excitability test were no discernible increases in [K+]o. In tests in which separate Corning and valinomycin-containing electrodes were used, the graded changes which they measured in response to the variation of stimulus frequencies w e r e identical. In view of the differential sensitivity of the two different membranes (Coming

I

I

2

5

I

I

10 20 Hz

I

I

100 200

B

°8F

~. ~6

I"~ 0-2 0L I

I

0"05 0"1

I

I

1"0

10

VOLTS Fig. 3. Increases in [K+]o evoked in sciatic nerve by electrical stimulation: A: graph of frequency dependence between 1 and 200 Hz (intensity of 1 V = 10 × threshold for A-fiber compound action potential; 0.5 ms duration; 25 s tetanus). B: intensity varied from 0.07 to 7 V (0.5 ms pulses at 100 Hz for 25 s); threshold for A-fiber compound action potential was 0.1 Wand maximal response was evoked by 2.1 V; data from same experiment as shown in A.

130 A )

31

U

o~ 120

305

~ Ho[

o J3o '~''

,00 l'0mM

GABA

0

[K~lo: m M

~

'

~

. . I0. .

12

Fig. 6. Different time course for changes in excitability and [K+]o evoked by GABA (application of 1 mM for 2 min) in one nerve and recorded simultaneously. Open circles, amplitude of compound action potential; closed circles, [K+]o level recorded at depth of 200/~m

3-32[

3.20 t- ii 5min

Fig. 4. Increases in [K÷]o evoked in peripheral nerve by 3 different concentrations of GABA (arrows mark period of application; also note flow artefacts on records). Measured at depth of 230 um with ion-selective microelectrode. Note scale for [K +] is logarithmic.

different electrodes at depths between 70 and 300/~m was 3.26 mM (S.E. + 0.08, n = 9). Dose-dependent effects of G A B A to increase [K+]o can be seen in Figs. 4 and 5. Examples of the graded, prolonged increases in [K+]o which 0.5, 1.0 and 2.5 mM G A B A produced ve are demonstrated in the polygraph tracings of Fig. 4. In this and other experiments, control tests - - carried out with the electrode outside the

o c~ Z~

Z~

,at

% % <1

I 10-4

I 5~10 4

I 163

sponse, other than flow-induced transients, to changes in either Ringer or G A B A - R i n g e r solutions. Fig. 5 plots the A[K+]o (peak values) for 5 different nerves (each represented by a different symbol) with the application of 6 different concentrations of G A B A . It can be seen that the changes in [K+]o evoked by doses of G A B A /> 1 mM - - which caused a maximal or near-maximal increase in the excitability of each preparation (cf. Fig. 1B and also Fig. 4 of the companion article 34) - - were in the order of 0.06-0.18 mM and relatively small. The mean increase in [K+]o which was observed with the application of 1 mM G A B A to 7 nerves was 0.096 mM (S.E.

Comparison of time courses of changes in [K+]o and excitability

Z~

E

5~ 10-5

nerve at a similar depth in the bath - - showed no re-

+ 0.007, n = 22); this was associated with an increase in excitability of 38.3% (S.E. __+2.2, n = 20).

02

-I 5~10

[ G A BA'1 : M

Fig. 5. increase in extracellular potassium ([K+],,) of peripheral nerve evoked by various concentrations of GABA. Observations from 5 nerves, each represented by a different symbol. Recorded with ion-sensitive microelectrodes (depths 70-23(I ltm).

. . 2. . . 4

TIME:mlo

C

The potassium accumulation measured in the nerve during its exposure to G A B A lagged behind the evoked increase in its excitability. The data plotted in Fig. 6 show the distinct separation of the onset and peak for the maximal changes in the simultaneously recorded responses from one experiment. When a comparison was made with data available from 14 applications of G A B A (0.2-2.5 mM) to 4 nerves, in which [K+],, was recorded at depths 70-20(I urn, there was a significant difference (P < 0.001) between the time to reach the peak of maximal increase for excitability (2.55 rain (S.E. + 0.19)) and that for [K+],, (4.23 rain (S.E. + 0.24)). The response time of the K+-sensitive electrode makes only a very small contribution to this difference.

131

Changes in excitability evoked by A[K+]o

3-2

3

'

~

"

[K']o:mM

~.~,, ................" ~ - - ~ ~ .......,~, r,-.., .......

' .....

t

30

TIME:rain

Fig. 7. Increases in [K+]o evoked by superfusion of KCl-enriched solution. Applications for 2 min (shown by arrows): A: 3.2 mM KC1-Ringer; B: 3.5 mM KCI-Ringer. During control and recovery superfusion of Ringer contained 3.0 mM KCI. (Data of Fig. 9 show excitability changes measured at same time as A[K+]~for B.)

A

B

Kc, t

t'

T

160

/x

,_

Z

i o

@

i

u ~o

i ,

J

i 8

i

~ ,2

i

~,,,E ~ N 140

//I

/ /

1

S

u.i a

120

x <

100 I

I

0.1

0 2 A~K']

I

I

0"5

1

I 2

: mM

Fig. 8. Increases in excitability of peripheral nerve evoked by artificial alteration of K ÷ levels. A: increase in half-maximal A-fiber compound action potential (relative to control) during applications of 3.1-5.0 mM KCl-enriched Ringer for 5-10 min. Filled circles, early maximal values; open circles, later plateau level; bars mark S.E. (n = 6-7); curves fitted by eye. B: plot of changes in compound action potential of one experiment during superfusion of 3.5 mM KCI-Ringer for 5 min.

~Z120

3 5 ~ M ~Cl

.

.

.

.

~r !

•

r

TIME:rain

Fig. 9. Comparison of increases in excitability evoked by GABA and artificially raising [K+],,. Superimposed plots of changes in amplitude of compound action potential (as % of control) for same nerve during separate applications of l mM GABA and 3.5 mM KCl-Ringer. (Arrows mark period of application; Fig. 7B shows A[K+]o recorded at same time during application of 3.5 mM KCI in this experiment.)

In order to determine the contribution and source of the augmentation of potassium evoked by G A B A , extracellular levels of K ÷ were artificially altered by superfusion of the nerve with K÷-enriched Ringer solutions. The magnitude and time course of the evoked A[K+]o can be seen in the examples in Fig. 7 of measurements.in one nerve during superfusion of 3.2 and 3.5 mM KC1-Ringer. The peak change was reached at ~< 20 s following the change back to the control Ringer. Recovery from A[K+]o was faster than that for changes induced by G A B A (compare Fig. 7 with Figs. 4 and 6); undershoots were small and prolonged, or not noticeable. Fig. 8A shows the graded effects of different concentrations of raised [K+]o on the excitability of 7 different nerves. Although responses were sensitive to small increments in [K+], only when the concentration was raised by 0.6 mM or more were increases similar to those produced by 0.5 and 1.0 mM G A B A observed. In addition, the time course of the excitability change evoked by the addition of K ÷ (see Figs. 8B and 9) consistently showed an early peak during the period of application, followed by a rapid decline and an undershoot during recovery - - features which are absent during G A B A - e v o k e d changes in excitability (see Fig. 9 and Fig. 3C of preceding paper). The return of the excitability level is even more rapid than that of [K+]o (compare Figs. 8B and 9 with Fig. 7) and falls below the control baseline at a time when [K÷]o is still elevated. The maximum change in the amplitude of the action potential preceded the peak [K+]o level by 0.69 min (S.E. + 0.08, n = 9,5 nerves), Differences in changes in excitability and in [K+]o - - evoked with separate applications of G A B A (0.1-1 mM) and of potassium-enriched Ringer (containing 3.1-3.5 mM KC1) - - w e r e measured and compared in each of 5 different preparations. The superimposed records of Fig. 9 show that the initial rates of rise and magnitude of the increases in peak excitability evoked in one nerve by 1 mM G A B A and 3.5 mM KC1-Ringer are nearly comparable; although there is considerable dissimilarity for the time course of the later components. The graph of Fig. 10 shows the marked difference for correlations of changes in excitability with increases in [K+]o levels, evoked in each nerve (each represented by a different symbol) - - by G A B A (filled symbols) and by addition of K ÷

132 reflecting depolarization, which were significantly

160

Z 0 U 140

o~

Y

"~ 120 :E

100 0

0 2

I

0'05

I

0.1

I

0.2

0'~

t~ [K*]o '.mM

Fig. 10. Comparison of increases in excitability evoked by G A B A and KCl-enriched solutions. Changes in amplitude of compound action potential and [K+]o evoked in five nerves by separate applications of 0.1-1.0 mM GABA (filled symbols) and 3.1-3.5 mM KCI-Ringer (open symbols). For each symbol measurements of [K+]o were made at identical position in a single preparation. Regression equation and correlation coefficient for GABA: y = 173 + 32 log x; r = 0.706 (P < 0.05); for addition of K*: y = 138 + 30 log x, r = 0.907 (P < 0.01).

(open symbols). A l t h o u g h slopes for regressions are similar, at c o m p a r a b l e levels of measured [K+]o the associated excitability changes with G A B A are significantly g r e a t e r than those when K + levels were artificially raised (covariance analysis, F-test, P < 0.01). DISCUSSION The changes in [K+]o (~< 0.2 mM) which were m e a s u r e d in the present experiments following the application of 1 and 2.5 m M G A B A may be comp a r e d to the increases of approximately 0.5 mM in the dorsal root ganglion of the rat 8 and during perfusion of the frog spinal cord with 10-2 M GABA26.50; and of < 1.5 m M in the cuneate nucleus of the cat 33. The fact that they are lower may reflect differences in (1) geometry and structure - - e.g. the absence of sheath, synapses, cell bodies and fiber terminals, (2) the ratio of myelinated/unmyelinated fibers, and (3) activity of K + and/or G A B A uptake mechanisms. The [K+]o increases evoked by G A B A were associated with increases in peripheral nerve excitability,

greater than those at the c o m p a r a b l e K + levels induced by adding KCI to superfusates (32-55% for G A B A as compared to 6-15% for KCl). Excitability increases similar to those p r o d u c e d by 1 m M G A B A were evoked only by a tenfold greater change in the amount of added K +. Even if G A B A - e v o k e d changes in [K+]o were to be underestimated (as calculations of the time for diffusion equilibrium suggestS,34) the differences in the magnitude and time course of the changes in excitability and [K+]o evoked by K + and G A B A provide a convincing comparison. They support the conclusion that (1) the observed change in the response of the nerve reflects a depolarization which is mainly m e d i a t e d by G A B A receptor activation, and (2) the associated K + accumulation plays at most an ancillary role. The d e m o n stration that there is a considerable lag of the G A B A evoked A[K+]o behind the depolarization of the nerve further points to a secondary role for the accumulation of potassium ions. The raising of extracellular K + - - either in the superfusate, or as a result of electrical stimulation - - evokes in this and o t h e r preparations a decline in excitability after the p e a k is reached, as well as a subsequent undershoot (see Figs. 8 and 9) - - features which presumably reflect the stimulation of a Na+, K + - A T P a s e d e p e n d e n t m e m b r a n e pump 23.43. The fact that desensitization of the G A B A - e v o k e d excitability increase is not readily, if at all apparent, in the peripheral nerve therefore suggests that the observed depolarization cannot be attributed solely to the effect of potassium ions. Possible sources of the concomitant rise in [K+]o during the depolarization of A fibers which G A B A produces are numerous: (1) most importantly, outward movement of K + accompanying that of CI-, so maintaining electroneutrality, as well as closing the circuit (inward current at the Cl ionophore and outward K + current elsewhere); (2) a direct receptor-mediated activation of m e m b r a n e conductance to K + is a mechanism not generally favouredtg.2°: (3) some activity-evoked change in m e m b r a n e conductance to K ÷ will occur22.Sl - - however, this is clearly negligible, as shown with the recruitment of all fibres at the low frequency of stimulation used for excitability testing; (4) neuronal/non-neuronal uptake of G A B A , which has been shown to be d e p e n d e n t on Na + and associated with changes in m e m b r a n e po-

133 tential of both glia and neurones17,25,30,37,48 may provide a contribution. It is also conceivable that G A B A evokes a release of K ÷ from the large surface area of unmyelinated axons of the nerve bundle. There is a large stimulusevoked accumulation of K ÷ with activation of unmyelinated fibers as compared to myelinated fibers 13 (see also Fig. 3B). The evidence for receptor-activation of C fibers of the sciatic nerve from our studies (Barolet, unpublished observations) is minimal, although the small C-fiber response in this preparation should probably be investigated more thoroughly using averaging. However, it has recently been reported that G A B A evokes no change in the excitability of small-diameter parallel fibers or in resting [K+]o31. Brown and Marsh 6 have demonstrated the depolarization by G A B A of the vagus and cervical sympathetic nerves, which contain much larger numbers of C fibers. Since they were unable to demonstrate a significant response for the myelinated fibers of phrenic nerves, they concluded that G A B A selectively depolarizes the unmyelinated fibers of peripheral nerves. Our results show that this is indeed not the case. Furthermore, since we show that the response of sensory myelinated fibers to G A B A is 8 times that of the motor fibers, the insensitivity of the phrenic nerve could well be explained by its unusually large population of motor fibers and a scarceness of afterents12,52. The likelihood that the accumulation of K +, whatever its cause, plays only a secondary role, receives additional support from the demonstration of differential responses of dorsal and ventral root fibers to G A B A (which also provides an improved estimate for GABA-receptor stoichiometry) - - especially since motor fibers are known to be more sensitive than sensory fibers to actions of potassium ions 49. It is of interest that the magnitude of the small GABA-induced depolarization of ventral root fibers is nearly identical to that produced in the whole nerve by superfusions of K + concentrations similar to those recorded during the application of G A B A . And, although we cannot exclude the possibility of contribution from hyperpolarizing responses to G A B A 42.44, it can be

noted that the responses to G A B A recorded intracellularly in amphibian dorsal root fibers 40 and crayfish motor axons 4.5 are mainly depolarizing. It would seem that there may be either very few receptors for G A B A on fibers of the ventral root as compared to those of the dorsal root, or even more likely that there are no receptors on fibers from the ventral root their depolarization being entirely due to external K ÷ accumulation. In studies in the immature rat a ten-fold difference in threshold for GABA-evoked depolarization of isolated dorsal and ventral spinal roots has also been observed2; however, because of closely corresponding patterns of relative potencies of agonists, these workers concluded that there were similar receptors. Their report that in the mature animal dorsal roots are sensitive to G A B A , while ventral roots are not would to some extent seem to be in agreement with our findings; but they also state that sciatic branches of immature rats are depolarized, while those of mature animals are not. These differences may possibly be due to differences in (1) degree of myelination, (2) the proportion of motor fibers in branches tested, and (3) sensitivity of the preparation and technique. From the evidence of the present studies, as well as our pharmacological characterization of the effect of G A B A 34, it is concluded that the depolarizing action of G A B A on the myelinated fibers of the sciatic nerve: (1) is mainly a receptor-mediated one involving the dorsal root component, and (2) receives a relatively minor contribution from a concomitant rise in [K+]o •

-

-

ACKNOWLEDGEMENTS This work was supported by the Medical Research Council of Canada. The authors are grateful to David Conner for expert technical assistance and to Professor W. Simon for the donation of the valinomycin neutral carrier ligand. In particular we wish to thank Professor Robert Werman for his suggestion and advice concerning the studies of the dorsal and ventral roots, and for reading the manuscript.

134 REFERENCES 1 Albers, R. W. and Brady, R. O., The distribution of glutamic decarboxylase in the nervous system of the Rhesus monkey, J. biol. Chem., 234 (1959) 926-928. 2 Allan, R. D., Evans, R. H. and Johnston, G, A. R., ?,-Aminobutyric acid agonists: an in vitro comparison between depression of spinal synaptic activity and depolarization of spinal root fibers in the rat, Brit. J. Pharmacol., 70 (1980) 609-615. 3 Barker, J. L. and Nicoll, R. A., The pharmacology and ionic dependency of amino acid responses in the frog spinal cord, J. Physiol. (Lond.), 228 (1973) 259-277. 4 Baxter, D. A., Synaptic Plasticity at Crustacean Motor Terminals, Ph.D. Thesis, University of Texas at Austin. 1981. 5 Baxter, D. A. and Bittner, G. D., Mechanism for presynaptic inhibition in crayfish claw opener muscle, Soc. Nearosci. Abstr., 7 (1981) 437. 6 Brown, D. A. and Marsh, S., Axonal GABA-receptors in mammalian peripheral nerve trunks, Brain Research, 158 (1978) 187-191. 7 Brown, D. A., Adams, P. R., Higgins, A. J. and Marsh, S., Distribution of GABA-receptors and GABA-carriers in the mammalian nervous system, J. Physiol. (Lond.), 75 (1979) 667-671. 8 Carslaw, H. S. and Jaeger, J. C., Conduction of Heat in Solids, Clarendon Press, Oxford, 1959, pp. 20(1-201. 9 Davidoff, R. A., The effects of bicuculline on the isolated spinal cord of the frog, Exp. Neurol., 35 (1972) 179-193. 10 Deschenes, M. and Feltz, P., GABA-induced rise of extracellular potassium in rat dorsal root ganglia: an electrophysiological study in vivo, Brain Research, 118 (1976) 494-499. 11 Deschenes, M., Feltz, P. and Lamour, Y., A model for an estimate in vivo of the ionic basis of presynaptic inhibition: an intracellular analysis of the GABA-induced depolarization in rat dorsal root ganglia, Brain Research, 118 (1976) 486-493. 12 Euler, C. von, Fusimotor activity in spindle control and natural movements, with special reference to respiration. In P. Andersen and J. Jansen (Eds.), Excitatory Synaptic Mechanisms, Universitetsforlaget, Oslo, 1970, pp. 341-349. 13 Galvan, M., Forstl, J. and ten Bruggencate, G., Extracellular K + activity during electrical stimulation of rat sympathetic ganglia, vagus and optic nerves in vitro. In E. Sykovfi, P. Hnik and L. Vyklick~ (Eds.), Ion-Selective Microelectrodes and Their Use in Excitable Tissues, Plenum Press, New York, 1981, pp. 201-204. 14 Glusman, S., Correlation between the topographical distribution of [3H]GABA uptake and primary afferent depolarization in the frog spinal cord, Brain Research, 88 (1975) 10%114. 15 Graham, L. T., Jr., Shank, R. P., Werman, R. and Aprison, M. H., Distribution of some transmitter suspects in cat spinal cord: glutamic acid, aspartic acid, y-aminobutyric acid, glycine and glutamine, J. Neurochem., 14 (1967) 465-472. 16 Hackman, J. C., Ross, D. B. and Davidoff, R. A., Some actions of catecbol on synaptic transmission in the isolated spinal cord of the frog, Brain Research, 157 (1978) 47-61. 17 Kehoe. J., Electrogenic effects of neutral amino acids on neurons of Aplysia californica, ("old Spring Harb. Syrup. quant. Biol., 40 (1975) 145-155. 18 K£~, N., Sykov~i, E., Ujec, E. and Vyklick3% L., Changes of extracellular potassium induced by neuronal activity in

19 20

21

22

23

24

25

26

27

28 29

30 31

32

33

34

35

36

37

the spinal cord of the cat, J. Physiol. (Lond.), 238 11974) 1-15. Krnjevid, K., Chemical nature of synaptic transmission in vertebrates, Physiol. Rev.. 54 11974) 418-540. Krnjevid, K., Inhibitory action of GABA and GABA-mimetics on vertebrate neurons. In E. Roberts, T. N, Chase and D. B. Tower (Eds.), GABA in Nervous System Function, Raven Press, New York, 1976, pp. 26%286. Krnjevid, K. and Morris, M, E.. Extracellular K + activity and slow potential changes in spinal cord and medulla, Canad. J. Physiol. Pharmacol.. 50 (1972) 1214-1217. Krnjevi~, K. and Morris, M. E., Correlation between cxtracellular focal potentials and K + potentials evoked by primary afferent activity, Canad. J. Physiol. Pharmacol., 53 (1975) 912-922. Krnjevid, K. and Morris, M. E., Factors determining the decays of K + potentials and focal potentials in the central nervous system, Canad. J. Physiol. Pharmacol,. 53 11975) 923-934. Krnjevid, K., Morris, M. E. and Reiffenstein, R. J., Stimulation-evoked changes in extracellular K+ and Ca 2+ in pyramidal layers of the rat's hippocampus, Canad. J. Physiol. Pharmacol., 60 (1982) 16&'k-1657. Krnjevid, K. and Schwartz, S.. Some properties of unresponsive cells in the cerebral cortex, Exp. Brain Res. 3 (1967) 306-31t9. Kudo, Y. and Fukuda, H., Alteration of extracellular K ~activity induced by amino acids in the frog spinal cord, Jap. J. Pharmacol., 26 (1976) 385-387. Levy, R. A., GABA: a direct depolarizing action at the mammalian primary afferent terminal, Brain Research. 76 (1974) 155-160. Levy, R. A., The role of GABA in primary afferent depolarization, Prog. Neurobiol., 9 (1977) 211-267. Lux, H. D., Fast recording ion specific microelectrodcs: their use in pharmacological studies in the CNS, Neuropharmacology, 13 (1974) 50%517. Lux, H. D. and Ncher, E., The equilibration time course ol [K+ ]o in the cat cortex, Exp. Brain Res.. 17 (1973) 1911-2115. Malenka, R. C. and Kocsis, J. I)., Effects of GABA on stimulus-ew)ked changes in [K ~]<, and parallel fiber excitability, J. Neurophysiol.. 48 (1982) 608-621. Miyata, Y. and Otsuka, M., Quantitative histochemistry of y-aminobutyric acid in cat spinal cord with special reference to presynaptic inhibition, J. Neurochem., 25 (1975) 23%244. Morris, M. E., Di Costanzo, G., Fox, S., MacDonald, J. F. and Werman, R., Actions of GABA, glycine and baclofen on afferent fibers and their central synapses, Soc. Neurosci. Abstr., 4 (1978) 582. Morris, M. E., Di Costanzo, G. A., Fox, S. and Werman, R., Depolarizing action of GABA (y-aminobutyric acid) on myelinated fibers of peripheral nerve, Brain Research, 278 11983) 117-126 Morris, M. E., Di Costanzo, G. A. and Sheridan, P. J., Ionic changes accompanying depolarizing actions of GABA on peripheral nerves, Proc. canad. Fed. biol. Soc., 24 (1981) 241. Morris. M. E.. Di Costanzo, G. and Werman, R., Depolarizing actions of GABA on peripheral nerves. Physiologist. 23 (1980) 68. Nicoll, R. A., Padjen. A. and Barker, ,I. 1~., Analysis ol amino acid responses on frog motoneurones, Neuropharmacology. 15 (1976) 45-53.

135 38 Nistri, A., Constanti, A. and Krnjevi6, K., Electrophysiological studies of the mode of action of GABA on vertebrate central neurons. In G. Pepeu, M. J. Kuhar and S. J. Enna (Eds.), Receptors for Neurotransmitters and Peptide Hormones. Adv. Biochem. Psychopharmacol., Vol. 21, Raven Press, New York, 1980, pp. 81-90. 39 Oehme, M. and Simon, W., Microelectrode for potassium ion based on a neutral carrier and comparison of its characteristics with a cation exchanger sensor, Anal. chim. Acta, 87 (1976) 21-25. 40 Padjen, A. L. and Hashiguchi, T., Intracellular analysis of GABA-evoked responses of primary afferent axons, Soc. Neurosci. Abstr., 8 (1982) 577. 41 Parker, R. B. and Waud, D. R., Pharmacological estimation of drug-receptor dissociation constants. Statistical evaluation. I. Agonists. J. Pharmacol. exp. Ther., 177 (1971) 1-12. 42 Randi6, M., Semba, K. and Murasa, K., Excitability changes induced by iontophoretic GABA at intraspinal single cutaneous afferent C- and Ab-fibres, Proc. Int. Union Physiol. Sci., 14 (1980) 656. 43 Rang, H. P. and Ritchie, J. M., On the electrogenic sodium pump in mammalian non-myelinated nerve fibers and its activation by various external cations, J. Physiol. (Lond.), 196 (1968) 183-221. 44 Rudomin, P., Engberg, I., Jankowska, E. and Jimenez, I., Evidence of two different mechanisms involved in the generation of presynaptic depolarization of afferent and rubrospinal fibers in the cat spinal cord, Brain Research, 189

(1980) 256--261. 45 Rudomin, P., Stefani, E. and Werman, R., Voltage sensitivity of small focal, transient potassium depolarizations in snail neurons: relevance for diagnosis of chemical synaptic activity, J. Neurophysiol., 42 (1979) 912-924. 46 SabeUi, H. C., DeFoe-May, J. and Bulet, M., Selective drug effects on motor sensory nerves: a 'Dale's principle' for receptors, Soc. Neurosci. Abstr., 3 (1977) 593. 47 Schmidt, R. F., Pharmacological studies on the primary afferent depolarization of the toad spinal cord, Pfliigers Arch. ges. Physiol., 277 (1963) 325-346. 48 Sch6n, F. and Kelly, J. S., The characterization of [3H]GABA uptake into the satellite glial cells of rat sensory ganglia, Brain Research, 66 (1974) 289-300. 49 Stiimpfli, R. and Hille, B., Electrophysiology of the peripheral myelinated nerve. In R. Llin~is and W. Precht (Eds.), Frog Neurobiology: A Handbook, Springer, Berlin, 1976, pp. 9-10. 50 Sykov~i, E., GABA-induced changes of extracellular K ÷activity in the frog spinal cord, Physiol. Bohemoslov., 28 (1979) 189-192. 51 Vyklick~', L., Sykov~i, E., Ki'~, N. and Ujec, E., Post-stimulation changes of extracellular potassium concentration in the spinal cord of the rat, Brain Research, 45 (1972) 608-611. 52 Ya§argil, G. M., Systematische Untersuchung der motorischen Innervation des Zwerchfells beim Kaninchen, Helv. Physiol. Pharmacol. Acta, Suppl. 18 (1966) 1-63.