Neuroscience Letters, 126 (1991) 9-12 0 1991 Elsevier Scientific Publishers Ireland Ltd. 0304-3940/91/$03.50 ADONIS0304394091OO202W

9

NSL 07720

Fall in intracellular pH mediated by GABAA receptors in cultured rat astrocytes K. Kaila’, P. Panula2, T. Karhunen2

and E. Heinonen3

‘Departmentof Zoology, Division of Physiology, 2Deparfment of Anatomy and ‘Neurobiology Research Unit, University of Helsinki, Helsinki (Finland)

(Received 11 October 1990; Revised version received 14 December 1990; Accepted 25 January 1991) Key worrls:

GABAA receptor; Astrocyte; Muscimol; Intracellular pH; Bicarbonate permeability; BCECF fluorescence

The influence of muscimol (a specific y-aminobutyric acid-A (GABA*) receptor agonist) on intracellular pH (pH,) was studied in cultured rat astrocytes by means of fluorescence spectrophotometry with BCECF as the H+ indicator. In an HCOr--free medium, muscimol had little effect on pHi. In a solution containing 22 mM HCO,-, muscimol produced a reversible, concentration-dependent fall in pH, with a maximum of about 0.14.15 units. The muscimol-induced fall in pH, was antagonized by an increase in the external K+ concentration, which suggests that the acidosis is an immediate consequence of a net efflux of HCOrm through GABAA receptor channels rather than an indirect effect caused by a change in membrane potential. The present results raise the possibility that astrocytes may participate in the regulation of extracellular pH at GABAergic synapses and contribute to activity-induced pH changes in nervous tissue

Recent work on glial cells has revealed the presence of various transmitter-activated channels [2], including anion channels gated by y-aminobutyric acid (GABA* receptor channels) [4, 14, 21, 23, 241. It has been suggested that glial GABA* receptors may play a role in the regulation of extracellular Cl- at inhibitory synapses [14, 231 and in facilitating intracellular potassium accumulation following an activity-induced increase in extracellular K+ [24]. Working on crayfish muscle fibres and neurones, we have recently shown that, in the presence of COz/ HC03-, activation of GABA-gated anion channels induces a passive net efflux of bicarbonate, which leads to a fall in intracellular pH (pHi) and to a rise in the pH close to the external surface of the plasma membrane [17-19, 321. The present study was undertaken to examine whether activation of GABA* receptor channels will lead to similar acid-base shifts in glial cells. If present in glial cells, such a mechanism could contribute to their role in the regulation and modulation of extracellular pH in nervous tissue [6, 81. In order to avoid complications that might arise due to activation of a GABA uptake system [ 11, 12, 201 or due to effects mediated by

Correspondence: K. Kaila, Department of Zoology, Division of Physiology, University of Helsinki, Arkadiankatu 7, SF-00100 Helsinki, Finland.

GABA receptors other than the A-type [l 1, 151,we chose to examine the effects of muscimol, a specific agonist of the GABA* receptor [3, 111. Brains of newborn Wistar rats were removed as described earlier [26]. Pieces of frontal and parietal cortex were suspended in culture medium by forced pipetting and by passing the cells through a polyester filter (pore size 80 pm). The cells were seeded on poly+lysinecoated glass coverslips in Petri dishes. The culture medium was RPM1 (Gibco) supplemented with fetal calf serum (20% for 3 days, thereafter low), L-glutamine (2 mM), penicillin (50 IU/ml) and streptomycin (50 pg/ml). The cultures were maintained for a minimum of two weeks at 37°C in a 95%+5% mixture of air and C02. Judged by morphology and immunofluorescence of glial fibrillar acidic protein, about 90% of the cells were type- 1 astrocytes [l, 9, 131. Coverslips carrying the cells were incubated at 37°C in standard solution containing 3 PM BCECF-AM ester (2’,7’-(bis carboxyethyl)-5,6-carboxyfluorescein acetoxy methyl ester; Molecular Probes, Inc.) for 15-25 min. BCECF-AM was added from a 1 mM stock solution made in dimethylsulfoxide. The standard HCOs--free solution contained (in mM): NaCl 137, KC1 5.4, KH2P04 0.44, CaC12 1.0, MgC12 1.0, TES (N-Tris[hydroxymethyll-methyl-2-aminoethanesulfonic acid), 20 (pH 7.4). In the HCOs--containing solution, 22 mM NaCl was replaced by an equivalent amount of NaHC03

IO

and TES was omitted. The former solution was equilibrated with 100% 02, and the latter with 95% 02 + 5%

co*. A coverslip was placed diagonally into a cuvette with a PVC cap on its top. The solution within the cuvette was continuously stirred. Drugs were applied by means of a Hamilton syringe inserted across a small hole in the cap. In order to avoid pH changes due to a loss of CO:! from the C02/HCOs--containing solution, the atmosphere within the cuvette was stabilized by a stream of 95% O2 + 5% C02, which was prewarmed and humidified. In some experiments (e.g., Fig. 2E), the cuvette was perfused at a rate of 25 ml/min which yielded a nominal exchange time of 2 s. The experimental temperature was 33-34°C. Fluorescence measurements were made by means of a Hitachi F2000 Fluorescence Spectrophotometer. The excitation wavelength was 500 nm and emission was analyzed at 530 nm. In order to minimize bleaching, excitation was intermittent (pulses of l-2 s given at 0.030.1 Hz) and the excitation beam was attenuated with neutral density filters. The final continuous recordings (cf. Fig. 2) were obtained by linearly connecting the sampled data points. It may be worth pointing out that the responses recorded by the present method are generated by a large population of cells, i.e. those exposed to the excitation beam. Each type of experiment was repeated at least three times with separate cultures of astrocytes, and each coverslip was used for one experiment only. At the end of each experiment, the cells were first exposed to gramicidin (0.5 pug/ml) in order to abolish the transmembrane Na+ and K+ gradients and, thereafter, nigericin (1 .OPM) was added to obtain a passive H + dis-

tribution [25], i.e. pHi = pH,. In order to calibrate the fluorescence signal, pH, (measured by a miniature glass pH electrode in the cuvette) was varied by adding HCl or NaOH. Fluorescence was approximately linearly related to pH within the range examined, 6.8-7.7 (Fig. 1). The average resting pHi in the presence of HCOs- was 7.22 + 0.082 ( + S.D., n = 17). In the absence of HC03--, pH, was consistently lower by about 0.2 units. The increase in pHi produced by adding 10 mM trimethylamine chloride (TMA; Sigma) yielded an estimate for the intracellular non-CO2 H+ buffering power [27,29] of about 20 mM (pH unit) ~ I. Fig. 2A,B shows that while muscimol (1 mM) had little effect on pHi in a bicarbonate-free solution, a substantial acidosis was observed in the presence of HCOs-. Application of TMA led invariably to a rise in pHi. On average, the fall in pHi produced by 1 mM muscimol was 0.12 f 0.04 ( f S.D., n = 9) in the presence of HC03-. The acidosis caused by muscimol did not recover in the presence of the drug. This is in agreement with the idea that muscimol induces a maintained acid load, and an acidotic steady-state pHi is attained when the active extrusion of acid equals the load [18, 321. As shown in Fig. 2C, the cells were capable of completely recovering from an acidosis caused by exposure to 20 mM sodium butyrate: the pHi value before the addition of the weak acid was identical to that observed upon recovery. In Fig. 2C, exposure to butyrate has been done

TMA D

M

L

M

6.95 228

i

c

But M

.-_-I

z23 15

’

Fig. 2. Effect of muscimol

1

6.8

7.0

7.2

7.4

lowed by addition addition

Fig. 1. Fluorescence pH (pH,

= pH,)

from measurements

intensity upon

(arbitrary

application

units) plotted

of gramicidin

on two coverslips,

as a function

+ nigericin.

one in the presence

cles) and one in the absence (open circles) of HCOr-.

(I mM) on pH, in the absence (A) and pres-

ence (B-E) of HC03m. In A and B, application

PH

of

Data

(filled cir-

of 10 mM trimethylamine

of 20 mM sodium

butyrate

The initial value of pH, is indicated indicates

decrease

of muscimol chloride

of

is reversible upon washing.

at the start of each trace. Upward

in pH,. All recordings

separate

(M) is fol-

(TMA); in C, by

(But); and in D, by addition

10 mM KC1 (K+). E: the effect of muscimol deflection

O.spHi Pmin. A-D lmin, E

coverslips.

shown

are from

II

in the continuous presence of muscimol. However, an identical regulatory behaviour was seen in the absence of muscimol. It might be argued that the changes in pHi produced by muscimol are secondary to a change in membrane potential. However, it is known that activation of GABA* channels produces a depolarization in astrocytes [21], and a glial depolarization should give rise to an alkaline rather than to an acid load [6, IO]. In fact, we observed that, in the presence of HCOj-, increasing the extracellular K+ concentration by 10 mM invariably produced an increase in pHi. This occurred both in the presence (Fig. 2D; cf. Fig. 3 in ref. 18) and absence of muscimol. Fig. 2E shows that the effect of muscimol on pHi is reversible upon washing with control solution. A clear concentration dependence of the fall in pHi induced by muscimol was evident in the range examined, l&l000 PM (4 experiments using a perfused cuvette). It should be pointed out that results such as those shown in Fig. 3 are not strictly comparable to concentrationconductance responses (cf. ref. 4). First, the glial GABAgated anion channels show rapid desensitization [4] and, therefore, the plateau acidosis in the presence of a given concentration of muscimol (especially at the upper level of the concentration-response curve) must be due to a rather small fraction of the channel population which is initially activated. Furthermore, the shape of the concentration-response curve is strongly affected by the pHidependence of the acid-extruding mechanisms of the cell [ 18, 191.Finally, the pHi response recorded from a population of cells masks all quantitative variation that might exist between and within individual cells. Variation of the latter kind might arise due to a non-uniform distribution of GABAA receptors at the cellular level [2, 41. Nevertheless, it is obvious that the effective concentra-

tion range shown in Fig. 3 is in agreement with the assumption that the fall in pHi caused by muscimol is due to a net efflux of HCOs- mediated by GABA* receptor channels. The present results suggest that, in situ, local neuronal activity does not modulate glial pHi exclusively via changes in extracellular K+ [8], but perhaps also by neuronally released GABA (cf. ref. 24). An additional interesting consequence is related to the fact that a channel-mediated efflux of HC03- can lead to a marked rise in extracellular pH in the vicinity of the plasma membrane 118, 191. This implies that HCOs- movements mediated by GABA* receptors in glial cells may play a role in activity-induced extracellular pH transients [5, 7, 16, 22, 281. A particularly attractive hypothesis is that glial cells might participate in the regulation and modulation of pH, at, and in the vicinity of, GABAergic synapses. Such a novel modulatory role for glial cells in microenvironmental pH regulation in nervous tissue merits attention in future work, especially since there is an increasing amount of evidence pointing to a role for protons in the control of nervous excitability [6, 19, 30, 3 I]. This work was supported by grants from the Academy of Finland and from the Sigrid Juselius Foundation. I Aloisi, F., Agresti, C. and Levi, G., Establishment, and evolution

of cultures

enriched

2 Barres, B.A., Chun, L.L.Y. and Corey, D.P.. Ion channels brate glia, Annu.

Rev. Neurosci.,

4 Bormann,

J. and Kettenmann,

H., Patch-clamp

5 Chen. J.C. and Chesler,

by GABAA

M. and Chan,

transients 948. Chesler,

M. and Kraig,

and modulation

C.Y.,

cultures

from primary

100

10

[muscimol], Fig. 3. Dependence

1000

of the fall in pH, on muscimol

from two experiments

employing

Kleckner, concentration.

a perfused

cuvette.

Data

pH transients

of astrocyte

rat glial cultures. Manual

and oligoden-

of the Nervous

(Ed.), System,

1989, pp. 12ll133.

co-transport

R., Boland,

of mam-

In A. Shahar

J.W. and Schlue, W.R., An inwardly

Dingledine,

uM

pH

27 (1988) 941-

9 (1989) 201 I-2019.

and Tissue Culture

sodium-bicarbonate (1989) 179-194.

extracellular

Neuroscience,

R.P., Intracellular

J. Neurosci.,

Alan R. Liss, New York, Deitmer,

in

of pH in the nervous

Stimulus-induced

Cole, R. and de Vellis, J., Preparation A Dissection

increase

in turtle cerebel-

34 (1990) 401427.

in the in vitro cerebellum,

malian astrocytes, drocyte

receptors

Proc.

Lett., I I6 (1990) l3&135.

M., The regulation

system, Prog. Neurobiol., Chesler,

astrocytes.

M., A bicarbonate-dependent

pH mediated

receptor

study of y-amino-

butyric acid receptor Cll channels in cultured Natl. Acad. Sci. U.S.A., 85 (1988) 93369340.

Chesler,

I

in verte-

I3 (1990) 441474.

3 Bormann, J., Electrophysiology of GABAA and GABAs subtypes, Trends Neurosci., I I (1988) 112-l 16.

lum, Neurosci.

I

J. Neurosci.

Res., 21 (1988) 188-198.

extracellular

OL

characterization

in type-2 astrocytes.

L.M.,

N.W.. Traynelis,

directed

electrogenic

in leech glial cell. J. Physiol., Chamberlin,

N.L.,

S.F. and Verdoorn,

41 I

Kawasaki,

K.,

T.A., Amino

acid

receptors and uptake systems in the mammalian central system, CRC Crit. Rev. Neurobiol., 4 (1988) I-96.

nervous

12 12 Erecinska, M.. The nemotransmitter amino acid transport systems, Biochem. Pharmacol., 36 (1987) 3547-3555. 13 Hansson, E. and Riinnblck, L., Primary cultures of astroglia and neurons from different brain regions. In A. Shahar (Ed.), A Dissection and Tissue Culture Manual of the Nervous System, Alan R. Liss, New York, 1989, pp. 92-104. 14 Hoppe, D. and Kettenmann, H., GABA triggers a Cl- efflux from cultured mouse oligodendrocytes, Neurosci. Lett., 97 (1989) 334 339. 15 H&Ii, E. and H&Ii, L., Evidence for GABAs-receptors on cultured astrocytes of rat CNS: autoradiographic binding studies, Exp. Brain Res., 80 (1990) 621625. 16 Jarolimek, W., Misgeld, U. and Lux, H.D., Activity dependent alkaline and acid transients in guinea pig hippocampal slices, Brain Res., 505 (1989) 225232. 17 Kaila, K., Pastermck, M., Saarikoski, J. and Voipio, J., Influence of GABA-gated bicarbonate conductance on potential, current and intracellular chloride in crayfish muscle fibres, J. Physiol., 416 (1989) 161-181. 18 Kaila, K., Saarikoski, J. and Voipio, J., Mechanism of action of GABA on intracellular pH and on surface pH in crayfish muscle fibres, J. Physiol., 427 (1990) 241-260. 19 Kaila, K. and Voipio, J., GABA-activated bicarbonate conductance: influence on EoAsA and on postsynaptic pH regulation. In F.J. Alvarez-Leefmans and J.M. Russell (Eds.), Chloride Channels and Carriers in Nerve, Muscle and Glial Cells, Plenum, New York, 1990, pp. 331-353. 20 Kanner, B.I. and Schuldiner, S., Mechanism of transport and storage of neurotransmitters, CRC Crit. Rev. Biochem., 22 (1987) I 38. 21 Kettenmann, H., Backus, K.H. and Schachner, M., y-Aminobutyric acid opens Cl-channels in cultured astrocytes, Brain Res., 404 (1987) l-9.

22 Kraig, R.P., Ferreira-Filho, C.S. and Nicholson, C., Alkaline and acid transients in cerebellar miroenvironment, J. Neurophysiol., 49 (1983) 831-850. 23 MacVicar, B.A., Tse, F.W., Crichton, S.A. and Kettenmann, H., GABA-activated Cl- channels in astrocytes of hippocampal slices, J. Neurosci., 9 (1989) 3577-3583. 24 Malchow, R.P.. Qian, H. and Ripps, H., y-Aminobutyric acid (GABA)-induced currents of skate Muller (glial) cells are mediated by neuronal-like GABA* receptors, Proc. Nat]. Acad. Sci. USA, 86 (1989) 43264330. 25 Nachshen, D.A. and Drapeau, P., The regulation of cytosolic pH in isolated presynaptic nerve terminals from rat brain. J. Gen. Physiol., 91 (1988) 289-303. 26 Panula, P., Rechardt, L. and Hirvonen, H., Observations on the morphology and histochemistry of the rat neostriatum in tissue culture, Neuroscience, 4 (1979) 2355248. 27 Roos, A. and Boron, W.F., Intracellular pH, Physiol. Rev., 61 (1981) 296433. 28 Sykova, E., Svoboda, J., Chvatal. A. and Jendelova, P., Extracellular pH and stimulated neurons, Ciba Found. Symp., 139 (1988) 220-235. 29 Szatkowski, M.S. and Thomas, R.C., The intrinsic intracellular H+ buffering power of snail neurones, J. Physiol., 409 (1989) 899101. 30 Tang, C.-M., Dichter, M. and Morad, M., Modulation of the Nmethyl-o-aspartate channel by extracellular H+, Proc. Natl. Acad. Sci. U.S.A., 87 (1990) 64456449. 31 Traynelis, S.F. and Cull-Candy, S.G., Proton inhibition of Nmethyl-o-aspartate receptors in cerebellar neurons, Nature, 345 (1990) 347-350. 32 Voipio. J., Pasternack, M., Rydqvist, B. and Kaila, K., Effects of y-aminobutyric acid (GABA) on intracellular pH in the crayfish stretch-receptor neurone, J. Exp. Biol., in press.