GABAB receptors expressed in Xenopus oocytes by guinea pig cerebral mRNA are functionally coupled with Ca2+-dependent Cl− channels and with K+ channels, through GTP-binding proteins

Molecular Brain Research, 8 (1990) 301-309

301

Elsevier BRESM 70237

GABAB receptors expressed in Xenopus oocytes by guinea pig cerebral mRNA are functionally coupled with Ca2÷-dependent C1channels and with K ÷ channels, through GTP-binding proteins Masayuki Sekiguchi, Hidenari Sakuta, Koichi Okamoto and Yutaka Sakai Department of Pharmacology, National Defense Medical College, Saitama (Japan) (Accepted 1 May 1990)

Key words: y-Aminobutyric acid-B receptor; Xenopus oocyte; mRNA; Baclofen; 2-Hydroxysaclofen; Ca2÷-dependent CI- channel; K ÷ channel

Transmembrane currents induced by (-)-baclofen (BAC), a specific agonist of the 7-aminobutyric acid-B (GABAB) receptor, in Xenopus oocytes injected with guinea pig cerebral mRNA were electrophysiologically and pharmacologically characterized under a voltage-clamp condition. The oocytes injected with mRNA acquired responsiveness to BAC and showed two types of currents at a holding potential of -50 inV. One was the slow and smooth inward current which had a short latency and associated with a decrease in membrane conductance, and its amplitude was decreased by hyperpolarization and increased by depolarization. The other was the large fast oscillatory inward current with a long-latency, which was decreased in amplitude by depolarization and reversed at -26 mV. Both currents were not blocked by bicuculline but were depressed by 2-hydroxysaclofen (2-OH-SAC), though the smooth current was less sensitive to 2-OH-SAC; about 40% blockade at the 2-OH-SAC concentration capable of abolishing the oscillatory current. The smooth current was depressed by Ba2+. The intracellular injection of EGTA into oocytes abolished the oscillatory current but did not affect the smooth current. The oscillatory current was time-dependently attenuated and almost abolished by intracellularly injected pertussis toxin (PTX), while the smooth current was not depressed by this toxin even when the oscillatory current was nearly abolished. The intracellular injection of GTP-y-S into oocytes attenuated both oscillatory and smooth currents. These results suggest the possibility that GABAB receptors expressed in Xenopus oocytes by cerebral mRNA are functionally coupled with two signal transduction systems, one is the opening of Ca:+-dependent C1- channels mediated by PTX-sensitive GTP-binding protein(s) and the other is the closure of K ÷ channels through PTX-insensitive GTP-binding protein(s). INTRODUCTION The ~,-aminobutyric acid-B ( G A B A B ) receptor 12 is a metabotropic receptor. (-)-Baclofen ( B A C ) and 3-aminopropylphosphonic acid are regarded as selective agonists of the G A B A B receptor 2'1°'29, and phaclofen ( P H A C ) and 2-hydroxysaclofen (2-OH-SAC) 16A7 are known as selective antagonists at this receptor. Electrophysiologically, the activation of G A B A B receptor has been reported to change a K ÷- or Ca 2÷conductance in central neurons 2'8"37. For example, B A C induced increase of a K + conductance was observed as a slow inhibitory postsynaptic potential in rat hippocampal pyramidal cells 8'37, and BAC-induced reduction of a Ca 2÷ conductance caused the decrease of the action potential duration in the A-6 or C fiber of rat dorsal root ganglia 2. As for intracellular second messengers, the inhibitory GTP-binding protein, Gi, has been suggested to be the main signal transducer coupled with G A B A B receptors in

central neurons 1'13. However, all G A B A B receptors seem not necessarily to be coupled with Gi, since pretreatment of rat hippocampal CA1 pyramidal neurons with pertussis toxin (PTX) did not impair G A B A B receptor-mediated presynaptic events but blocked the postsynaptic events mediated by this receptor 9. In addition to this heterogeneous sensitivity to PTX, G A B A B receptor-mediated pre- and postsynaptic events in the rat hippocampus have also been demonstrated to show different sensitivities to P H A C , and this finding has led to the suggestion that G A B A B receptors might be divided into sub-classes 9. However, as P H A C is a relatively weak antagonist (at most one-tenth as potent as 2 - O H - S A C ) , the existence of G A B A B receptor subclasses has not yet been substantiated 6. For the purpose of further characterizing not only the electrophysiological and pharmacological properties of the G A B A B receptor but also coupled G-proteins and second messengers, we utilized the Xenopus oocyte expression system 4"15'31 and expressed, for the first time,

Correspondence: Masayuki Sekiguchi, Department of Pharmacology, National Defense Medical College, 3-2 Namiki, Tokorozawa, Saitama 359, Japan. 0169-328X/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

302 GABA B receptors

in

the

oocytes

from

guinea

pig

c e r e b r a l m R N A . V o l t a g e c l a m p studies w e r e carried out with e x p r e s s e d G A B A a r e c e p t o r s in the p r e s e n t study. MATERIALS AND METHODS Messenger RNA used was fractionated by oligo-dT cellulose column chromatography from the whole RNA which was extracted from the cerebral microsomes of adult guinea pigs by the CsCI/ guanidinium thiocyanate method 3 as described previously33. Ten cerebra were usually treated in one batch extraction (yield: 70-100 /~g mRNA), and mRNAs obtained by 3 batch extractions were used in this study. Xenopus oocytes were dissected out of mother frogs, dispersed in modified Barth's medium (MBM (in mM): NaCI 88, KC! 1, NaHCO 3 2.4, MgSO 4 0.82, Ca(NO3) 2 0.33, CaC12 0.41, HEPES 10, 1% streptomycin and 0.7% penicillin, pH 7.4), and defolliculated by gentle shaking in the MBM containing collagenase (1 mg/mi) for 3-4 h at 20-24 °C. Messenger RNA solution (50-100 mg mRNA/ml distilled water) was pressure-injected (30 nl/oocyte) into each defolliculated oocyte (stages V-VI, > 1.2 mm in diameter) in Ca2+-free MBM (Ca salts and antibiotics were removed from MBM). mRNA-injected oocytes were incubated for 3-4 days at 19 °C in MBM and supplied to electrophysioiogical recordings which were carried out in the same manner as described previously33. Briefly, each oocyte was voltage-clamped by a conventional twoelectrode method using the voltage clamp amplifier (CEZ-1100, Nihon Kohden, Tokyo). Two glass microelectrodes were filled with 3 M KC1 (1-5 MI2), and one was used for voltage monitoring and

the other was for current injection and command pulse application. One oocyte was placed in the superfusion chamber 33 at a time and superfused with frog-Ringer solution (control medium (in mM): NaCI 120, KCI 2, CaC12 1.8, HEPES 5, pH 7.4) at a constant rate of 5 ml/min throughout recording. BAC, 2-OH-SAC, bicuculline methiodide (BIC) and BaC! 2 were dissolved in frog-Ringer solution and applied by superfusion (5 ml/min). PTX, EGTA, guanosine5"-O-(3-thiotriphosphate) tetralithium salts (GTP-y-S), inositol 1, 4,5-trisphosphate (IP3) and 3",5"-cyclic AMP (cAMP) were dissolved in distilled water to be 0.5 g/I, 1 mM, 100 mM, 2.4 mM and 10 mM, respectively, and each solution was filled into a glassmicropipette and pressure injected into oocytes (2-10 nl/oocyte). BAC ((-)-baclofen) used was kindly donated by Ciba-Geigy Japan (Takarazuka, Japan). The sources of other compounds used were as follows: 2-OH-SAC, Tocris Neuramin (Essex, U.K.); PTX, Funakoshi Chemicals (Tokyo); GTP-y-S, Boehringer MannheimYamanouchi (Tokyo); BIC, Pierce (Rockford, U.S.A.), and IP3, cAMP and EGTA, Sigma Chemicals (St. Louis, U.S.A.). All other chemicals including BaCI 2 were obtained from Wako Pure Chemicals (Tokyo).

RESULTS T h e resting m e m b r a n e p o t e n t i a l s o f m R N A - i n j e c t e d o o c y t e s w e r e c o n s i s t e n t l y in a r a n g e f r o m - 4 0 to - 6 0 m V as r e p o r t e d p r e v i o u s l y 33. T h i r t e e n o o c y t e s w h i c h w e r e

BAC 1 -50mV

BAC 1

~

10

- lOmV

25nA BAC 1 60s

30s 25s

D

BAC 0.1

BAC 0.5

BAG 1

-60mY :

- ' ~

'l ~

I~1~50hA

15s

E

BAC 0.1

BAC 0.5

I

BAC 1

-30mV 38hA

40s

I

Fig. 1. Two types of transmembrane currents induced by (-)-baclofen (BAC) in Xenopus oocytes injected with guinea pig cerebral mRNA. In this and following figures, BAC was applied by superfusion for the period indicated by the bar in the concentration shown in mM and at the clamped membrane potential denoted on the left of each record, and a downward deflection corresponds to an inward current. A: typical responses of a mRNA-injected oocyte to 1 mM BAC at a holding potential of -50 mV, which consist of a small and smooth inward current with a short latency (arrowhead) and a fast and oscillatory inward current with a long latency (arrow). B: BAC-induced currents in a mRNA-injected oocyte (other than that in A) clamped at -10 inV. The smooth current response (arrowhead) was not reversed, while the oscillatory response (arrow) was reversed. C: BAC-induced smooth inward current at -10 mV recorded from an oocyte other than that in A or B. In this oocyte, no oscillatory current was induced by BAC. Brief downward deflections are brief pulses (1 s in duration) shifting the membrane potential from -10 to -35 mV for monitoring the membrane conductance. BAC-induced smooth current was associated with a decrease in membrane conductance. D,E: dose-dependence of the oscillatory inward current at -50 mV and the smooth inward current at-30 mV induced by BAC (0.1-1 raM), respectively. Records D and E were from different oocytes, and increasing concentrations of BAC were applied at 5 min intervals from left to right. The oocyte in E was clamped at -30 mV to mask the oscillatory component.

303 defolliculated with collagenase but were not injected with m R N A did not show any response to BAC at concentrations up to 2 mM in a membrane potential range from 0 to -100 m V (data not shown). On the other hand, 39 out of about 400 oocytes injected with guinea pig cerebral m R N A acquired responsiveness to BAC. As shown in Fig. 1A, two types of inward currents were elicited by bath-applied 1 mM BAC in 31 of 39 oocytes at this holding potential of - 5 0 mV; one was a slowly developing smooth current with a short latency of 3-20 s (arrowhead in Fig. 1A) and the other was a fast rising oscillatory inward current with a long latency of 15-80 s (arrow in Fig. 1A), the latter being superimposed on the first. These two current components may be seen more clearly at a holding potential o f - 1 0 mV as shown in Fig. 1B, in which the oscillatory current (arrow) was reversed, while the smooth component (arrowhead) was not reversed but rather intensified. The record Fig. 1B thus clearly indicates that the smooth and oscillatory currents are generated by different ionic mechanisms. In 3 out of 39 responsive mRNA-injected oocytes, only the oscillatory current was evoked by BAC at a holding potential o f - 5 0 mV, but the smooth current appeared when the holding potential was shifted to more positive than - 4 0 mV as exemplified in Fig. 3A. In the remaining 5 responsive mRNA-injected oocytes, however, only the smooth-inward current was induced by BAC even at a holding potential o f - 5 0 mV, and this current became larger in amplitude by shifting a holding potential to - 1 0 mV as illustrated in Fig. 1C. Fig. 1C also shows the change of the membrane conductance associated with the smooth inward current induced by BAC. The height of command pulses (brief downward deflections shifting the membrane potential from -10 to -35 mV for a duration of 1 s) was decreased during a BAC-induced smooth inward current. The mean % decrease (+ S.E.M.) in membrane conductance during the smooth inward current induced by 1 mM BAC at a holding potential o f - 1 0 mV was 48.5 + 2.4% in 3 oocytes tested (the mean membrane resistance before BAC application was 780 + 164 kI2, n = 5). The dose-dependent properties of BAC-induced oscillatory and smooth inward currents are illustrated in Fig. 1D and 1E, respectively. Records in Fig. 1D and 1E were obtained from different oocytes, and in order to mask the oscillatory component, the oocyte for Fig. 1E was purposely clamped at the potential level (-30 mV) close to the reversal potential (-26 mV) of the oscillatory component (see Fig. 3). As shown in Fig. 1D and 1E, the lowest effective concentration of BAC to evoke either oscillatory or smooth current was about 500 aM. Fig. 2 shows responses to repetitively applied BAC, in which the right record was obtained 5-10 min after the

first application on the left. As we reported previously on dopamine receptors expressed in oocytes 3°, the oscillatory response mediated by metabotropic receptors expressed in oocytes was often desensitized or facilitated by repetitive applications of a relevant agonist. As shown in Fig. 2A, however, BAC-induced smooth inward current was neither desensitized nor facilitated in all oocytes tested. The oocyte illustrated in Fig. 2A was clamped at -26 mV to mask the oscillatory current (see Fig. 3), but BAC-induced smooth currents recorded from several oocytes at -50 mV also showed no desensitization. On the other hand, the oscillatory current was neither desensitized nor facilitated in many cases as shown in Fig. 2B, but was either strongly desensitized rarely (Fig. 2C)

A

BAC 0.5

BAC 0.5

-26mV

~s

B

BAC 0.5

I

BAC 0.5

-50mV

20s

C

BAC 0.5

BAC 0.5

-50mV

D

BAC I

30s

I

30$

I

BAC I

-50mY

Fig. 2. Effects of repetitive applications of BAC. Righthand record in each line was obtained about 5 min after the left. A: a typical BAC-induced smooth current showing neither desensitization nor facilitation at the second application of BAC. This oocyte was clamped at -26 mV to mask the oscillatory current. B: a typical BAC-induced oscillatory current showing neither desensitization nor facilitation. C: a rare case, in which BAC-induced oscillatory current was strongly desensitized at the second application. D: an example of strong facilitation of BAC-induced oscillatory current at the second application, which was observed sometimes. Only such mRNA-injected oocytes as those in A and B were employed for experiments.

304 or strongly facilitated sometimes (Fig. 2D). The oocytes showing either desensitization or facilitation at the second application of BAC were excluded from subsequent experiments, because drug-effects were hardly evaluated. Fig. 3A shows BAC-induced currents obtained at varied holding potentials. The current-voltage relations of oscillatory and smooth currents are illustrated in Fig. 3B and 3C, respectively, in which data from 3 different oocytes including the oocyte for Fig. 3A were plotted (different closed symbols indicate different oocytes). In Fig. 3B, the amplitude of the oscillatory current at each membrane potential was normalized as the percentage of the response at -50 mV (open circle), while in Fig. 3C, it was normalized as the percentage of the response at 0 mV (open circle). As shown in Fig. 3A, the oscillatory inward current (arrows) at -50 mV was decreased by depolarization and reversed to be outward currents at -20 to 0 mV. The mean reversal potential of the oscillatory current estimated from Fig. 3B was -26.0 _+ 1.8 mV ( n - - 3). Since this reversal potential corresponded to the C1- equilibrium potential of oocytes 19, the oscillatory current was indicated to be due to the opening of C1- channels. BAC-induced smooth inward current, on the other hand, was usually small at a holding potential o f - 5 0 mV and became larger with depolarization and smaller with

A

hyperpolarization as shown in Fig. 3A (arrowheads) and 3C. At a more negative potential than -50 mV, the smooth inward current disappeared or became very small (<1 nA), but no reversal could be observed even at a holding potential o f - 1 0 0 mV at which current recording became very noisy (data not shown). As will be discussed later, these behaviors of BAC-induced smooth inward current are most likely to be due to the closure of K + channels. Fig. 4A and 4B show the effects of intracellular injection of E G T A on BAC-induced oscillatory and smooth currents, respectively. As shown in Fig. 4A, BAC-induced oscillatory current was readily abolished by intraceUular injection of E G T A (6 pmol/6 nl/oocyte in this illustrated case). BAC-induced smooth inward current, on the other hand, was little affected by intracellularly injected E G T A (Fig. 4B), in which the oocyte was clamped at -27 mV to mask the oscillatory component and E G T A applied was 10 pmol/10 nl/oocyte. Similar results as Fig. 4A and 4B were also obtained from 2 other oocytes tested for the oscillatory current and from 4 other oocytes tested for the smooth current. These findings suggest that BAC-induced oscillatory current, which is a Cl- current as mentioned above, is mediated by intracellular Ca 2÷ mobilization, while BAC-induced smooth current is not mediated by the Ca 2÷ mobilization. Fig. 4C shows the effect of intracellular injection of

BAC 1

B

C -100

50 -20mV

M.P. (mV) -50 I

-50mV

• : -30 I

-=;o-;o

-30mY

-40mV

-40 I

4"~nA

M.P. (mV) T -0

-50 n

-40 |

-30 J

-20 =

--100

o I .0

|

• -50

! 0

- 10

o

- -50

"~C

-50

• 0

'-100

30s

Fig. 3. Voltage-dependence of BAC-induced currents. A: BAC-induced currents at varied membrane potentials. B,C: current-voltage relations for the oscillatory current (B) and the smooth current (C) obtained from records in A and from 2 other oocytes. Different closed symbols show different oocytes. The current amplitude at each holding potential was normalized as the percentage of the current at -50 mV for B and at 0 mV for C (open circles). The oscillatory current (arrows) was reversed between -20 and -30 mV (A,B), while the smooth current (arrowheads) became larger by depolarization and smaller by hyperpolarization without showing reversal (A,C).

305 inhibitory G-protein (Gi) by ADP-ribosylation of the a-subunit of Gi, upon BAC-induced currents in a m R N A - i n j e c t e d oocyte clamped at - 5 0 mV. IntraceUular injection of PTX (2 ng/4 nl/oocyte) attenuated B A C induced oscillatory currents in a time-dependent manner, namely, the oscillatory currents recorded 3, 30 and 50 min after the injection of P T X became to 83, 43 and 6%, respectively, of the control response before PTX injection. On the other hand, B A C - i n d u c e d smooth inward current became rather clearer 30 and 50 min after the injection of PTX (arrowheads in Fig. 4D) probably as a result of unmasking due to the depression of a superimposed large oscillatory component. In 3 oocytes tested, BAC-induced smooth currents were still observable even 40-50 min after the P T X injection, while BAC-induced oscillatory currents were strongly depressed. The mean % blockade after 40-50 min from P T X injection was 93.8 + 0.4% (n = 3) for the oscillatory current induced by 1

GTP-~-S, a non-hydrolyzable analog of GTP, upon BAC-induced currents in a mRNA-injected oocyte clamped at - 4 0 mV. In agreement with the previous report 7, the injection of GTP-y:S alone (400 pmol/4 nl/oocyte) evoked an inward current accompanied by small fluctuations which lasted for about 7 - 8 min as shown in Fig. 4C. W h e n B A C (1 mM) was applied 1-2 min after GTP-~-S injection, an outward current (double arrowheads) was induced, while the oscillatory component (double arrows) was attenuated. However, 10-15 min after GTP-~-S application, B A C (1 mM) showed practically no response as illustrated in Fig. 4C, rightmost record. GTP-y-S showed similar depressive actions as Fig. 4C in 2 other oocytes, and the mean % blockade at 15 min after GTP-~-S injection was 94.5 + 0.6% (n = 3) for the oscillatory current induced by 1 m M B A C . Fig. 4D shows the effect of PTX, which is known to interfere with the interaction between a receptor and the

A

B

EGTA

EGTA

BAC 0.5

BAC 0.5

BAC 1

BAG 1

-2rmv '

-50mY

~

25nA

1

25s

_3 25s

C

GT~-I~S BA..~CI

_40m~

[v

BAC.. "~ 1

11rn

~

BA.~C 1

"-::-~

° \ j ~" "%

-----

"~ A,~ o ,I

J

20s

D

3 min after PTX-ink

Control

BAC 1 -50mY ~

BAC 1

30 min BAC 1

E

50 min

-- 100nA

20s

IP3

_50mV..~~

BAC 1 I OOnA

/-60s

I

50nA

60s

I

F

cAMP

10OnA -50mY

I 120s

Fig. 4. Effects of intracellularly injected EGTA, GTP-),-S and PTX on BAC-induced currents, and responses to intracellularly applied I P 3 and cAMP. The aqueous solution of EGTA (1 mM), GTP-),-S (100 mM), PTX (0.5 g/l), IP3 (2.4 mM) or cAMP (10 mM) was pressure injected from a glass-micropipette into mRNA-injected oocytes. A,B: the effect of EGTA (6 pmol/6 nl/oocyte for A and 10 pmol/10 nl-oocyte for B) on BAC-induced oscillatory (A) and smooth (B) inward currents. The oocyte in B was clamped at -27 mV to mask the oscillatory current. The oscillatory current was completely blocked by EGTA (A), while the smooth current was not affected at all by EGTA (B). C: the effect of GTP-~,-S (400 pmol/4 nl/oocyte) on BAC-induced currents. Both oscillatory (double arrows) and smooth (arrowhead) inward currents were finally abolished by GTP-y-S (the rightmost record), though GTP-),-S itself induced an inward current associated with small current fluctuations (middle record), and BAC transiently evoked an outward current in the presence of GTP-),-S (double arrowheads in middle record). D: the effect of PTX on BAC-induced currents. BAC was applied before and 3, 30 and 50 min after PTX injection (2 rig/4 nl/oocyte). The oscillatory current (arrows) was time-dependently attenuated after PTX injection, while the smooth current (arrowheads) became rather clearer at 30 min and did not decrease, even at 50 min. E: the oscillatory inward current induced by IP3 (5 pmol/2 nl/oocyte). F: the smooth outward current induced by cAMP (20 pmol/2 nl/oocyte). Records A-F were from different oocytes.

306

A

BAC I

BAC

-50mV ~

"

1 + BIC 0.014

~

~

50nA

25sI B

BAC I

BAC I

BAC 2

BAC

1

÷2-OH--SAC0.014 +2-Of-P-SAC0.014 10OnA

26s

C

BAC I

RAC 1

+2"OH-SAC0.014

BAC I I"BIC0.014 2P'nA

308 D

Ba2~ 10 BAC 1

OmV

~

BAC

1

/\~nA 60s

Fig. 5. Pharmacology of BAC-induced currents. Each substance tested was dissolved in frog Ringer solution and applied by superfusion at denoted concentrations (mM). A: the effect of BIC (14 ~M) on BAC (1 mM)-induced oscillatory current. The right record was obtained 5 rain after the left. BAC showed no effect on the oscillatory current. B: the effect of 2-hydroxysaclofen (2OH-SAC, 14~M) on BAC (1 mM)-induced oscillatorycurrent. The responses were recorded from left to right at 3 rain intervals. 2-OH-SAC (14 ~M) completely and reversibly abolished the oscillatory current induced by 1 mM BAC and partially depressed that induced by 2 mM BAC. C: effects of 2-OH-SAC (14 ~M) and BIC (14 ~M) on BAC (1 mM)-induced smooth inward current. The oscillatory current was not evoked in this oocyte. BAC-induced smooth current was partially antagonized (by about 40%) by 2-OH-SAC, but was little affected by BIC. D: the effect of Ba~÷ (10 raM) on BAC (1 mM)-induced smooth current. The oscillatory current was not observed in this oocyte. Ba2+ itself elicited a smooth inward current, and BAC-induced smooth inward current was depressed to about 20% by Ba2÷. mM BAC. These findings clearly indicate that BACinduced oscillatory current is sensitive to PTX, while BAC-induced smooth current is either insensitive or much less sensitive to this toxin. In Fig. 4E and 4F, typical currents induced by intracellularly injected inositol trisphosphate (IP3) and cAMP are illustrated, respectively. Intracellular injection of IP 3 (4.8 pmol/2 nl/oocyte) consistently induced a fast rising inward current but with a short latency as shown in Fig. 4E in agreement with Nomura et al. 24. Intracellularly injected cAMP (20 pmol/2 nl/oocyte), on the other hand, elicited a slow outward current (Fig. 4F). Fig. 5 shows the pharmacology of BAC-induced

currents in mRNA-injected oocytes. The oscillatory current induced by BAC (1 mM) was not affected by BIC (14/tM), a selective antagonist of the G A B A A receptor, as illustrated in Fig. 5A and also in 2 other oocytes tested. On the other hand, 2-OH-SAC (14 /.tM, a selective antagonist of the G A B A B receptor) completely and reversibly blocked BAC-induced oscillatory current as illustrated in Fig. 5B, though 2 mM BAC partially reversed the blockade (Fig. 5B, B A C 2). The oscillatory current induced by 1 mM BAC was completely blocked by 14 #M 2-OH-SAC in 9 out of 10 oocytes tested and depressed to 30% in the remaining 1 oocyte. Fig. 5C shows the effects of BIC and 2-OH-SAC on BAC-induced smooth inward current in a mRNAinjected oocyte (this oocyte showed only a smooth current response to BAC). The smooth inward current induced by BAC (1 mM) was partially depressed by 14 /~M 2-OH-SAC (middle record), but was not affected at all by 14 #M BIC (rightmost record). Similar results were obtained from 3 other oocytes, and the mean % blockade by 14/~M 2-OH-SAC of the smooth current induced by 1 mM BAC was 40.7 + 3.0% (n = 4). It seems likely from these results that the BAC-induced smooth current is less sensitive to 2-OH-SAC than the oscillatory current. Fig. 5D shows the effect of Ba 2÷, a K + channel blocker TM, on the smooth inward current induced by 1 mM BAC in a mRNA-injected oocyte clamped at 0 mV. Ba 2÷ (10 mM) itself induced a smooth inward current in mRNA-injected oocytes as reported by Parker et al. 27. In the presence of 10 mM Ba 2+, however, the amplitude of the smooth inward current induced by 1 mM BAC was depressed to about 20% of the control in this illustrated case and also in 1 other oocyte. Ba 2÷ did not induce any smooth inward current in mRNA-non-injected oocytes (data not shown). DISCUSSION It was demonstrated by this study that two types of currents were evoked by BAC in Xenopus oocytes injected with guinea pig cerebral mRNA. One is a smooth inward current, and the other is an oscillatory current (Fig. 1). Both smooth and oscillatory currents induced by BAC were likely to be mediated by the activation of G A B A B receptors expressed from guinea pig cerebral mRNA, because the mRNA-non-injected control oocytes never responded to BAC in a membrane potential range from 0 to -100 mV, and because both currents were depressed by 2-OH-SAC, a selective antagonist of the G A B A B receptor, but not by BIC, a selective antagonist of the G A B A A receptor (Fig. 5). BAC-induced oscillatory current was found to be

307 mediated by PTX-sensitive G-protein(s) (Fig. 4D), and this current seems to be evoked by the opening of Clchannels, which are endogenous to oocytes, by the mobilization of intracellular Ca 2+. In agreement with Nomura et al. 24 and Oron et a l Y , injection of IP 3 into the oocytes to which guinea pig cerebral mRNA had been injected induced a fast rising inward current (Fig. 4E), while intracellularly injected cAMP elicited a slow outward current (Fig. 4F). Therefore, it is likely that BAC-induced oscillatory current might mainly be mediated by an increase in IP 3, but apparently not by an increase in cAMP. Activation of CaE÷-dependent Clchannels by phosphatidylinositol (PI) turnover is well documented in oocytes21'24'25'28"35"36. It has been reported, however, in the cerebral cortical slices of the rat 5 and mouse 11 that the activation of G A B A B receptors depresses the serotonin- or histamine-stimulated formation of IP 3. Thus, the G A B A B receptors mediating BAC-induced oscillatory inward current in mRNAinjected oocytes seem to be coupled with the G-proteins which are different from those in central neurons. Such a change in the coupling to G-proteins has been reported for [ArgS]vasopressin (AVP) receptorsE2; AVP receptors expressed in oocytes are coupled with PTX-sensitive G-protein(s), while those in the liver are with PTXinsensitive G-protein(s). Another possible mechanism of generation of BACinduced oscillatory current may be that activation of G A B A B receptors by BAC inhibits phospholipase C (PLC) via PTX-sensitive G-protein(s) as in the case of central neurons TM, then as soon as this BAC action is over, PLC is disinhibited to trigger PI turnover and generate the Ca2÷-dependent Cl- current. This is an attractive possibility, because the long latency of BACinduced oscillatory current may be accounted for. On the other hand, BAC-induced smooth inward current was found to have the following properties: insensitivity to PTX (Fig. 4D), association with a decrease in membrane conductance (Fig. 1C), no reversal (Fig. 3A,C), amplitude increases with depolarization (Fig. 3A,C), and blockade by GTP-7-S (Fig. 4C) and Ba 2+ (Fig. 5D). The decrease of the membrane conductance indicates that certain ion channels were closed by BAC. The insensitivity to PTX and the sensitivity to GTP-~-S suggest that PTX-insensitive G-protein(s) participate in BAC-induced smooth inward current, since it is well established that GTP-F-S binds irreversibly to the a-subunit of G-proteins. The inward current induced by GTP-~-S alone (Fig. 4C, downward arrow) and the outward current induced by BAC in the presence of GTP-~-S (Fig. 4C, double arrowheads), however, were not studied in detail here. What ion channels are closed during a BAC-induced

smooth inward current? In the case of Xenopus oocytes bathed in frog Ringer solution, the equilibrium potentials of Na ÷ and K + are estimated to be at +40 to +50 mV and about -100 mV 19, respectively, and the Ca 2÷ equilibrium potential may be at about + 120 mV based on the intracellular free Ca 2+ level of about 0.17 ktM36. The Cl- equilibrium potential is well known to be at about-25 mV. Thus, the inward current caused by ion channel closure in the membrane potential range from -50 to 0 mV can theoretically be possible only when K + channels are closed. The blocking action of Ba 2+ on BAC-induced smooth current (Fig. 5D) and the increases of the amplitude of the smooth current with depolarizations (Fig. 3A,C) are in support of this conclusion. Parker et al. 27 have recently analyzed the smooth inward current induced by serotonin in oocytes injected with rat cerebral mRNA. This serotonin-induced current is very similar to BAC-induced smooth inward current in many respects, and they have concluded that the closure of K ÷ channels is the underlying mechanism. It is likely that the K ÷ channels which are closed by BAC are not endogenous to oocytes but are expressed from injected cerebral mRNA, because Ba 2÷ did not induce any inward Current in mRNA-non-injected oocytes, while Ba 2+ not only induced an inward current by itself but also depressed BAC-induced smooth inward current in mRNA-injected oocytes (Fig. 5D). At least two possible mechanisms may be considered as to the BAC-induced closure of K + channels. One is the direct closing of K ÷ channels by PTX-insensitive G-protein(s) coupling with expressed G A B A a receptors, and the other is closure of K ÷ channels through the intracellular second messenger system operated by PTXinsensitive G-protein(s). Although the direct opening of G A B A B receptor-mediated K ÷ channels by PTX-sensitive G-protein(s) has been reported in rat hippocampal neurons 23, there is no report demonstrating the direct closure of K + channels through PTX-insensitive Gprotein(s). The latter possibility is similar to the serotonin-induced S-current reported in Aplysia neurons 1s'34, which was evoked by the closure of K ÷ channels through the formation of cAMP mediated by the PTX-insensitive G-protein(s) coupling with serotonin receptors. The S-current, however, is not blocked by Ba ions 18. In 5 out of 39 oocytes, BAC induced only the smooth inward current without the oscillatory component. This suggests the possibility that the smooth current might be mediated by the G A B A B receptor other than that mediating the oscillatory current. As described in Introduction, it has already been reported in rat hippocampal neurons 9 that both PTX and PHAC block the postsynaptic event, but not the presynaptic event, mediated by G A B A B receptors. This led to the suggestion that

308 G A B A B receptors might be able to be classified into sub-classes2'9. The results obtained in the present study, that B A C induced both PTX-sensitive and PTX-insensitive responses and the former was due to an increase in CI- conductance, while the latter was due to a decrease in K ÷ conductance, seem to be in favor of the existence or expression of subclasses of G A B A B receptors. As reviewed by Schofield et al. 32, it is now obvious that several n e u r o t r a n s m i t t e r receptors are heterogeneous at the gene level. W h e t h e r the G A B A B receptor mediating

concentration such as 0.5 m M was required for B A C to show electrophysiological responses in this study, while it has been demonstrated that 0.1-1 ktM serotonin evoked oscillatory inward currents of nearly equivalent sizes to those induced by 0.5-1 m M B A C in oocytes 2°'26, and 0.1 ~ M and 0.7/~M B A C inhibited serotonin-stimulated PI turnover in mouse cerebral cortex 11 and histamineinduced PI turnover in rat cerebral cortex 5, respectively. Therefore, the interaction b e t w e e n B A C and the stimu-

B A C - i n d u c e d smooth and oscillatory currents are different molecular entities seems to be a matter of interest for

lators of PI turnover, such as serotonin and histamine, is also to be investigated electrophysiologically utilizing the X e n o p u s oocyte expression system to further reveal the

further studies. It should be noted finally that a relatively high

physiological role and function of central G A B A B receptors.

REFERENCES

14 Hille, B., Ionic Channels of Excitable Membranes, Sinauer, Sunderland, MA, 1983. 15 Houamed, K.M., Bilbe, G., Smart, T.G., Constanti, A., Brown, D.A., Barnard, E.A. and Richards, B.M., Expression of functional GABA, glycine and glutamate receptors in Xenopus oocytes injected with rat brain mRNA, Nature, 310 (1984) 318-321. 16 Kerr, D.I.B., Ong, J., Johnston, A.G.R., Abbenante, J. and Prager, R.H., 2-Hydroxy-saclofen: an improved antagonist at central and peripheral GABAB receptors, Neurosci. Lett., 92 (1988) 92-96. 17 Kerr, D.I.B., Ong, J., Prager, R.H., Gynther, B.D. and Curtis, D.R., Phaclofen: a peripheral and central baclofen antagonist, Brain Res., 405 (1987) 150-154. 18 Klein, M., Camardo, J. and Kandel, E.R., Serotonin modulates a specific potassium current in the sensory neurons that show presynaptic facilitation in Aplysia, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 5713-5717. 19 Kusano, K., Miledi, R. and Stinnakre, J., Cholinergic and catechotaminergic receptors in Xenopus oocytes membrane, J. Physiol., 328 (1982) 143-170. 20 Lubbert, H.A., Snutch, T.P., Dascal, N., Lester, H.A. and Davidson, N., Rat 5-HT1c receptors are encoded by a 5-6 kbase mRNA size class and are functionally expressed in injected Xenopus oocytes, J. Neurosci., 7 (1987) 1159-1165. 21 Miledi, R. and Parker, I., Chloride current induced by injection of calcium into Xenopus oocytes, J. Physiol., 357 (1984) 173-183. 22 Moriarty, T.M., Sealfon, S.C., Carty, D.J., Roberts, J.L., Iyengar, R. and Landau, E.M., Coupling of exogenous receptors to phospholipase C in Xenopus oocytes through pertussis toxin-sensitive and -insensitive pathways, J. Biol. Chem., 264 (1989) 13524-13530. 23 Nicoll, R.A., The coupling of neurotransmitter receptors to ion channels in the brain, Science, 241 (1988) 545-551. 24 Nomura, Y., Kaneko, S., Kato, K., Yamagishi, S. and Sugiyama, H., Inositol phosphate formation and chloride current responses induced by acetylcholine and 5-HT through GTPbinding proteins in Xenopus oocyte after injection of rat brain messenger RNA, Mol. Brain Res., 2 (1987) 141-143. 25 Oron, Y., Dascal, N., Nadler, E. and Lupu, M., Inositol 1,4,5-trisphosphate mimics muscarinic receptors in Xenopus oocytes, Nature, 313 (1985) 141-143. 26 Parker, I., Gundersen, C.B. and Miledi, R., Intracellular Ca2÷-dependent and Ca2+-independent responses of rat brain 5-HT receptors transplanted to Xenopus oocytes, Neurosci. Res., 2 (1985) 491-496. 27 Parker, I., Panicker, M.M. and Miledi, R., Serotonin receptors expressed in Xenopus oocytes by mRNA from brain mediate a closing of K÷ membrane channels, Mol. Brain Res., 7 (1990)

1 Asano, T., Ui, M. and Ogasawara, N., Prevention of the agonist binding to ?,-aminobutyric acid B receptors by guanine nucleotides and islet-activating protein, pertussis toxin, in bovine cerebral cortex, J. Biol. Chem., 260 (1985) 12653-12658. 2 Bowery, N., GABAB receptors and their significance in mammalian pharmacology, Trends Pharrnacol. Sci., 10 (1989) 401407. 3 Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J., Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease, Biochemistry, 18 (1979) 52945299. 4 Constanti, A., Houamed, K.M., Smart, T.G., Bilbe, G., Brown, D.A. and Barnard, E.A., Functional brain GABA and glycine receptors in mRNA-injected Xenopus oocytes, Neuropharmacology, 23 (1984) 817-818. 5 Crawford, M.L.A. and Young, J.M., GABAB receptor-mediated inhibition of histamine Hi-receptor-induced inositol phosphate formation in slices of rat cerebral cortex, J. Neurochem., 51 (1988) 1441-1447. 6 Curtis, D.R., Gynther, B.D., Beattie, D.J., Kerr, D.I.B. and Prager, R.H., Baclofen antagonism by 2-hydroxy-saclofen in the cat spinal cord, Neurosci. Lett., 92 (1988) 97-101. 7 Dascal, N., Ifune, C., Hopkins, R., Snutch, T.P., Lubbert, H., Davidson, N., Simon, M.I. and Lester, H.A., Involvement of a GTP-binding protein in mediation of 5-HT and acetylcholine responses in Xenopus oocytes injected with rat brain messenger RNA, Mol. Brain Res., 1 (1986) 201-209. 8 Dutar, P. and Nicoll, R.A., A physiological role for GABAB receptors in the central nervous system, Nature, 332 (1988) 156-158. 9 Dutar, P. and Nicoll, R.A., Pre- and postsynaptic GABA B receptors in the hippocampus have different pharmacological properties, Neuron, 1 (1988) 585-591. 10 Falch, E., Hedegaard, A., Nielsen, L., Jensen, B.R., Hjeds, H. and Krogsgaard-Larsen, P., Comparative stereostructure-activity studies on GABA A and GABAa receptor sites and GABA uptake using rat brain membrane preparations, J. Neurochem., 47 (1986) 898-903. 11 Godfrey, P.P., Grahame-Smith, D.G. and Gray, J.A., GABAB receptor activation inhibits 5-hydroxytryptamine-stimulated inositol phospholipid turnover in mouse cerebral cortex, Eur. J. Pharmacol., 152 (1988) 185-188. 12 Hill, D.R. and Bowery, N.G., 3H-Baclofen and 3H-GABA bind to bicuculline-insensitive GABA B sites in rat brain, Nature, 290 (1981) 149-152. 13 Hill, D.R., Bowery, N.G. and Hudson, A.C., Inhibition of GABA B receptor binding by guanyl nucleotides, J. Neurochem., 42 (1984) 652-657.

309 31-38. 28 Parker, I., Sumikawa, K. and Miledi, R., Activation of a common effector system by different brain neurotransmitter receptors in Xenopus oocytes, Proc. Roy. Soc. Lond. Ser. B, 231 (1987) 37-45. 29 Pratt, G.D., Knott, C., Davey, R. and Bowery, N.G., Characterisation of 3-aminopropyl phosphinic acid (3-APPA) as a GABAs agonist in rat brain tissue, Br. J. Pharmacol., 96 (1989) 141P. 30 Sakai, Y., Sekiguchi, M., Okamoto, K., Kuwano, R. and Takahashi, Y., Dopamine receptors expressed in the Xenopus oocytes injected with bovine striatal mRNA, Mol. Brain Res., 7 (1990) 183-187. 31 Schofield, P.R., Darlison, M.G., Fujita, N., Burt, D.R., Stephenson, EA., Rodriguez, H., Rhee, L.M., Ramachandran, J., Reale, V., Glencorse, T.A., Seeburg, P.H. and Barnard, E.A., Sequence and functional expression of the GABAA receptor shows a ligand-gated receptor super-family, Nature, 328 (1987) 221-227. 32 Schofield, P.R., Shivers, B.D. and Seeburg, P.H., The role of receptor subtype diversity in the CNS, Trends Neurosci., 13

(1990) 8-11. 33 Sekiguchi, M., Okamoto, K. and Sakai, Y., Low-concentration N-acetylaspartylglutamate suppresses the climbing fiber response of Purkinje cells in guinea pig cerebellar slices and the responses to excitatory amino acids of Xenopus laevis oocytes injected with cerebellar mRNA, Brain Res., 482 (1989) 87-96. 34 Siegelbaum, S.A., Camardo, J. and Kandel, E.R., Serotonin and cyclic AMP close single K+ channels in Aplysia sensory neurones, Nature, 299 (1982) 413-417. 35 Snutch, T.R, The use of Xenopus oocytes to probe synaptic communication, Trends Neurosci., 11 (1988) 250-256. 36 Takahashi, T., Neher, E. and Sakmann, B., Rat brain 5-HT receptors in Xenopus oocytes are coupled by intracellular calcium to endogenous channels, Proc. Natl. Acad. Sci. U.S.A., 84 (1987) 5063-5067. 37 Thalmann, R.H., Evidence that guanosine triphosphate (GTP)binding proteins control a synaptic response in brain: effect of pertussis toxin and GTP-z-S on the late inhibitory postsynaptic potential of hippocampal CA3 neurons, J. Neurosci., 8 (1988) 4589-4602.