- Email: [email protected]
Modulation of GABA flux across rat brain membranes resolved by a rapid quenched incubation technique
ELSEVIER
Neuroscience Letters 182 (1994) 73-76
HEUROSCIINCE lETTERS
Modulation of GABA flux across rat brain membranes resolved by a rapid quenched incubation technique Julianna Kardos a'*, Ilona Kovfics a, Tamers Blandl a, Derek J. Cash b aDepartment of Molecular Pharmacology, Central Research Institute for Chemistry, The Hungarian Academy of Sciences, Pusztaszeri (tt 59-67, H-1025 Budapest, Hungary bDepartment of Biochemistry, School of Medicine, University of MissourL Columbia, MO, USA Received 5 August 1994; Revised version received 12 September 1994; Accepted 28 September 1994
Abstract The progress and inhibition of [3H]GABA influx in native plasma membrane vesicles from the rat cerebral cortex was studied on a subsecond to minute time scale under different conditions by applying a rapid quenched incubation technique. In the absence of Ca 2+ ([Ca2+]free = 10-s M), the progress of influx followed by the addition of 10 nM [3H]GABA to the membrane vesicle suspension with time (500 ms to 15 min) can be described by a first-order rate equation giving an overall rate constant, k, of 3.93 + 0.48 x 10 -3 s-1 and equilibrium influx value, INFe, of 8.84 + 0.41 pmol [3H]GABA/mg protein. In the presence of Ca 2÷ ([Ca2+]free = 2.4 × 10 -3 M) a significant increase in the INFe value was observed (k = 4.64 + 0.41 x 10 -3 s -l and INFe = 13.9 + 0.40 pmol [3H]GABA/mg protein). Multiplicity of GABA transporters was indicated in the time-dependent inhibition of [3H]GABA influx by different uptake blockers. In the absence of Ca 2÷, depolarization (75 mM KCI) inhibited the influx of [3H]GABA into the vesicles by ~70% and initiated the efflux from vesicles loaded with [3H]GABA. Different uptake blockers inhibited the Ca2+-independent translocation of [3H]GABA in both directions with similar specificities.
Key words: [3H]GABA translocation; Rapid quenched incubation technique; Membrane vesicle; Rat cerebral cortex
The pre-synaptic release of G A B A [13] and its postsynaptic action [3] occurs on a time scale of subsecond to seconds. To shape synaptic events the translocation of G A B A from the synaptic cleft needs to be fast and regulated. Examination of the effect of Ca 2+ on G A B A transport [10] showed that Ca 2+ strongly stimulated G A B A transport at low N a ÷ concentrations ([Na+]) but had only slight effects at high [Na+]. Since it is known that activators of protein kinase C and phosphatases stimulated G A B A influx via the cloned rat brain G A B A transporter (GAT1) expressed in Xenopus oocytes [4] it was of interest to see if Ca 2+ does have an effect on the transporterregulated G A B A influx on a physiologically relevant time scale. The release of cytoplasmic G A B A by reversal of the Na+-coupled transporters was considered artifactual consequence of unphysiologically prolonged de-polarization [11]. However, in the absence of Ca 2+, a tran-
*Corresponding author. 0304-3940194/$7.00 © 1994 Elsevier Science Ireland Ltd. All rights reserved SSD1 0 3 0 4 - 3 9 4 0 ( 9 4 ) 0 0 7 5 8 - 6
sient (< 1 s) nipecotic acid-sensitive component of high [K+]-induced G A B A release was observed by rapid superfusion of synaptosomes, suggesting that the transporter-regulated release of cytoplasmic G A B A m a y occur physiologically [13]. To address these issues, we have measured transmembrane G A B A influx in relevant time intervals with the use of a rapid quenched incubation technique which can resolve fast processes on a time scale of tens of milli-seconds to minute [2,3,12]. We have thus investigated (a) the progress of [3H]GABA influx in the presence and absence of Ca2+; (b) the progress of the inhibition of [3H]GABA influx by uptake blockers [1,6] in the absence of Ca2+; and (c) the effect of 75 m M KCI on [3H]GABA influx and efflux in the absence of Ca 2+. As -- 80% of synaptosomes were disrupted by fast displacements in the rapid kinetic apparatus native plasma m e m b r a n e vesicles [3,12] were applied. In order to characterise native plasma membrane vesicles, the Km and Vmax parameters of [3H]GABA influx and the inhibitory potencies of several G A B A uptake blockers [6] in mem-
74
J. Kardos et al./Neuroscienee Letters 182 (1994) 73 ~76
brane vesicles were determined in the presence of Ca 2+ ion using a conventional mixing-incubation technique. The cerebral cortex of a 4-6 weeks old male Wistar rat was homogenized by Ultraturrax (at 20,500 rpm for 5 s) in a HEPES-buffered sucrose solution containing protease inhibitors, antioxidant (phenyl-methanesulfonyl fluoride, 1 mM; aprotinin, 0.01 mg/ml; antipain, 0.005 mg/ml; leupeptin, 0.005 mg/ml; pepstatin A, 0.005 mg/ ml; butylated hydroxy-toluene, 0.02 m M [3]) and 1 mM ATE Native membrane vesicle fraction was obtained by the short, differential centrifugation procedure [3] with slight modifications [12]. Final pellet was suspended in either the buffer A [13] (in mM: NaC1, 145; KC1, 5; MgC12, 1; NaHCO3, 2; ATP, 1; D-glucose, 10; aminooxyacetate, 0.1; EGTA-Tris, 1; pH 7.5, [Ca2+]r~ = 10-8 M, pCa 8) or in buffer B (buffer A containing in addition 3.4 mM CaC12 ([Ca2+]f~ee= 2.4 mM, [13]). Native plasma membrane vesicles kept on ice for 3 hours have retained the initial transport activity. Membrane vesicle suspension (final protein concentration: 0.4-0.5 mg/ml, [8]) was usually incubated with 10 nM [3H]GABA (94 Ci/mmol, Amersham) in the presence of increasing concentration of GABA (0.1-10/IM) for 1, 5 and 20 rain at 30°C, pH 7.5. At the end of the incubation, the suspension was filtered (Whatman GF/B filters), washed with 2 x 3 ml ice-cold buffer and the radioactivity on the filter was counted by standard liquid scintillation techniques. Nonspecific influx was determined in the presence of 1 mM guvacine. The non-specific to total ratio was below 4% for the membrane vesicle preparation. The K m values were unchanged during the different periods of incubation: 4.3 + 0.6 y M (n = 4-5) and were similar to the Km of 4 / I M reported for synaptosomes [9]. The Vm~xvalues
were decreased with increasing incubation time: 900 + 280 ~ 380 + 125 pmol/min x mg protein). The inhibitory effect of 0.3 ¢tM to 1 mM GABA, 0.3-100 ~tM guvacine 0.2 ¢tM to 1 mM L-diaminobutyric acid (LDABA) and fl-alanine was measured in 1 rain of incubation using 100 nM [3H]GABA (9.1 Ci/mmol). The ICs0 values obtained in membrane vesicles were as follows (in /IM): GABA, 6.3 + 0.1; guvacine, 8.7 + 0.2; L-DABA, 28 + 6; fl-ALA, 2700 + 100. Aliquots of the membrane vesicle suspension in buffer A (B), kept at 4°C, were incubated at 30°C for 1 min, centrifuged (3 rain, 11,340 x g, SIGMA 112 microfuge); the resulting pellet was rinsed twice with 1 ml buffer A (B) and resuspended either in buffer A (B) or in buffer A (B) containing the different uptake blockers and kept at 30°C for 10 min before use. Fast kinetic experiments were performed with the use of the apparatus described previously [2]. The pulsed-incubation, quenched-flow technique was applied and the apparatus was thermostatted at 30°C [3,12]. Aliquot (0.34 ml) of the membrane vesicle suspension (0.8-1.0 mg protein/ml), p r e incubated with the uptake blocker was mixed with equal amount of buffer A (B) containing 20 nM [3H]GABA with and without the uptake blocker applied in the same concentration as in the p r e i n c u b a t i o n . After different time intervals of incubation, the suspension was mixed with equal amount of buffer A containing 2 mM guvacine (quench-solution) and filtered on Whatman GF/B filter; the filter was washed with 2 x 3 ml ice-cold buffer A and radioactivity was counted by standard liquid scintillation technique; to give specific [3H]GABA influx (100%), the [3H]GABA content of the filtered membrane vesicles measured in the presence of 1 mM guvacine was
8J o o_ 10
!
i
C~ L_ ~IL
o:
0.8
"° l
....
" /
4
o.6
4
./// . . . . . .
(]-
O
LI ........
~"
0.4
J
/
U
O
....
i
~JJ ............
!
f
o.2
:/i ..... ]
5
0
2S0
500
7S0
1@00
0
%
10
1%
Fig. 1. The progress of specific[3]GABAinfluxinto native plasma membranevesiclesfrom the rat cerebralcortex at 30°C, pH 7.5. o, [Ca2+]rr~= 10 8 M; [], [Ca2+]fr,o= 2.4 x 10-3 M. Lineswere calculated(Scientist,MicroMath Version2.03) accordingto a first-orderkineticswith the followingoverall rate constants (k) and equilibrium influx values (INFe) (mean +_S.D.): ©, k = 3.93 _+0.48 x 10-3 sq and INF,. = 8.84 -+0.41 pmol [3H]GABA/mg protein; [], k -- 4.64 + 0.41 x 10_3s-~ and INFe -- 13.9_+0.40 pmol [3H]GABA/mgprotein. Data are from a representativeexperiment,with measurements performed in duplicate. Other experimentsgave similar results. A: specificinfluxof 10 nM [3H]GABAfollowednear to completion;B: initial phase of [3H]GABAinflux.
J. Kardos et al./Neuroscience Letters 182 (1994) 73 76
subtracted from the [3H]GABA content of the filtered membrane vesicles measured in buffer A (B). Progress of [3H]GABA influx was followed in the 0.5 s to 15 min range of incubation in the presence or absence of Ca ~+ at 10 nM [3H]GABA concentration (Fig. 1). The first order rate equation, M = INFe[1-exp(-kt)], predicts the influx data in the whole range of equilibration with the overall rate constant, k of 3.93 + 0.48 x 1 0 -3 s -1 and the equilibrium influx value (INFe) of 8.84 + 0.41 pmol [3H]GABA/mg protein with 10 nM [3H]GABA at pCa 8 (buffer A). The presence of Ca 2÷ (buffer B) altered the influx of 10 nM [3H]GABA, characterized by k = 4.64 + 0.41 x 1 0 -3 S -1 and INFe = 13.9 + 0.40 pmol [3H]GABA/mg protein. Overall rate constant of 8.9 + 0.58 s -l and equilibrium influx value of 81.1 + 3.4 pmol [3H]GABA/mg protein were obtained when influx was initiated by 100 nM [3H]GABA. The progress of [3H]GABA influx (10 nM) in vesicles preincubated with the uptake blocker was followed in the 1 s to 60 s range of incubation in the absence of Ca 2+ (Table 1). A significant decrease of inhibition by fl-alanine and L-DABA was observed with increasing incubation time. The effect of depolarization on the influx of [3H]GABA by applying high concentration of K ÷ ion was studied when an aliquot of membrane vesicles in buffer A was mixed with equal amount of buffer C (in Table 1 Inhibition of [3H]GABA translocation by different GABA uptake inhibitors in native membrane vesiclesfrom the rat cerebral cortex at 30°c in the absence of calcium ion Uptake blockers (pM)
Inhibition (%) Influxa
Nipecotic acid (10) OH-nipecote acid (10) Guvacine (10) L-DABA (100) fl-Alanine (100)
Efflux b
1s
6s
60s
1s
82 + 7 60 + 7 60 + 9 85 + 5 46 + 3
84 + 5 65 + 4 70 + 5 85 + 6 39 + 7
86 + 1 70 + 2 64 + 2 39 + 6¢ 0 + 2c
80 + 17 58 + 18 62 + 13 67 + 16 54 + 22
Data are given as mean _ S.D. from 3 5 preparations with measurements performed in duplicate or triplicate (n = 610). aPre-incubation experiment: pelleted membrane vesicleswerepre-incubated in buffer A in the presence or absence of the uptake blocker in the following concentration (in pM): nipecotic acid, 10; OH-nipecotic acid, 10; guvacine, 10; L-DABA, 100;fl-alanine, 100; and kept at 30°C for 10 min before use; in the quech-flow experiment the pre-incubated vesicleswere mixed with an equal amount of buffer A containing 20 nM [3H]GABA with and without the uptake blocker applied in the same concentration. bCo-incubation experiment: aliquots of membrane vesiclesin buffer A, kept at 4°C, were incubated at 30°C; after 1 min equal amount of [3H]GABA in buffer A was added (40 nM) and the suspension was incubated at 30°C for 13 min (loading); in the quench-flowexperiment loaded vesicles were mixed with equal amount of buffer C with and without the uptake blocker applied in the following concentration (in /~M): nipecotic acid, 20; OH-nipecotic acid, 20; guvacine, 20; L-DABA, 200; fl-alanine, 200. cp < 0.001, when [3H]GABAinflux in 1 s is compared (Student's t-test).
75
mM: NaC1, 5; KC1, 145; MgC12, 1; NaHCO3, 2; ATE 1; D-glucose, 10; amino-oxyacetate, 0.1; EGTA-Tris, 1; p H 7.5) containing 40 and 80 n M of [3H]GABA. In the absence of Ca 2+, substitution of 75 m M KC1 for NaC1 (reportedly [7], this decrease of N a + ion concentration, final [Na +] = 75 m M , has no effect on G A B A influx) inhibited [3H]GABA influx in 1 s ( 6 4 _ 8% at 20 nM [3H]GABA and 80 _+ 9% at 40 nM [3H]GABA, n = 6). Prolonged incubation (6 s) did not affect the high [K+] induced inhibition significantly ( 6 9 _ 7% at 20 nM [3H]GABA and 72 _+ 6% at 40 n M [3H]GABA). These data suggest that we can disregard the relatively slow process of swelling as well as the isotopic dilution of the label by the release of endogenous G A B A as possible sources [9] of the high K+-induced inhibition of [3H]GABA influx. To compare CaE+-insensitive influx with that of efflux, an aliquot of the membrane vesicle suspension in buffer A, kept at 4°C, was incubated at 30°C. After 1 rain, an equal volume of [3H]GABA in buffer A was added (40 nM) and the suspension was incubated at 30°C for 13 rain (loading); 0.34 ml of the loaded membrane vesicle suspension containing 20 nM [3H]GABA was mixed with equal amount of buffer C containing the different uptake blockers. To give specific [3H]GABA efflux (100%), the [3H]GABA content of the filtered membrane vesicles measured in the absence of uptake blockers (100000_+ 5000 dprn) was subtracted from the [3H]GABA content of the filtered membrane vesicles measured in the presence of 1 m M guvacine (130000 + 7000 dpm). When the inhibition of the efflux of [3H]GABA from membrane vesicles in 1 s was compared with that of [3H]GABA influx in 1 s (Table 1), the different G A B A uptake blockers showed similar inhibitory specificities suggesting an identity of the transporters moving G A B A in cis---~trans and trans--)cis directions. Slow kinetic properties of the transporter(s) we observed suggest that the concentration of G A B A in the synaptic cleft could be critical to remove G A B A effectively in 'synaptic' time. We can estimate the half-response concentration for the faster and the slower desensitizing G A B A g receptors, a concentration of ~ 100 p M G A B A [3] in the synaptic cleft during inhibition. Under the condition, the G A B A transporter(s) would remove G A B A more than 100 times faster, which is on the timescale as the slower desensitizing GABAA receptor with 0.5 s half-time. On the basis of the known pharmacological specificity of the cloned G A B A transporters, GAT1, GAT2 and GAT3 [1], our data indicate that [3H]GABA influx into native membrane vesicles from the rat cerebral cortex is regulated by GAT1- and GAT2-type transporters. Both transporters are to be active in short times, whereas [3H]GABA influx near to completion is regulated by the G A T l - t y p e transporter. In addition to the requirements for N a + and CI ions [6] the pharmacological specificity of G A B A influx was found to be similar
76
Z Kardos et aL / Neuroscience Letters 182 (1994; 73-76
to its efflux suggesting that transporter-mediated GABA translocation may change from cis--*trans to trans---~cis direction when an action potential reaches the presynaptic terminal. GABA uptake in GATl-expressing oocytes by protein kinase C activators has been demonstrated to be mediated by a regulated subcellular redistribution of the transporter showing the onset 20 min after injection [4]. The onset we observed for the Ca 2÷induced stimulation of GABA uptake in native plasma membrane vesicles occurs on a subsecond time scale suggesting that the modulation may involve a fast Ca 2+induced conversion of transporter between active and inactive states. In conclusion, our fast kinetic experiments performed on native plasma membrane vesicles from the rat cerebral cortex indicated that the rate of GABA influx is sufficient to shape slow synaptic events. It is proposed now that Ca 2÷ ion and depolarization may play a role in regulating transporter-mediated GABA translocation in neurotransmission. J.K. was supported by OTKA T 4030 and US-Hungarian JF No. 277 grants. The skillful assistance of Mrs. Erzs6bet Fekete-K6ti is gratefully acknowledged. [1] Borden, L.A., Smith, K.E., Hartig, RR., Branchek, T.A. and Weinshank, R.L., Molecular heterogeneity of the y-amino-butyric acid (GABA) transport system, J. Biol. Chem., 267 (1992) 21098 21104. [2] Cash, D.J. and Hess, G.P., Quenched-flow technique with plasma membrane vesicles: acetylcholine-receptor-mediated transmembrane ion-flux, Anal. Biochem., 112 (1981) 39-51.
[3] Cash, D.J. and Subbarao, K., Desensitization of y-amino-butyric acid receptor from rat brain: two distinguishable receptors on the same membrane, Biochemistry, 26 (1987) 7556-7562. [4] Corey, J.L., Davidson, N., Lester, H.A., Brecha, N. and Quick, W., Protein kinase C modulates the activity of a cloned y-aminobutyric acid transporter expressed in Xenopus Oocytes via regulated subcellular redistribution of the transporter, J. Biol. Chem., 269 (1994) 14759-14767. [5] Kanner, B.I. and Kifer, L., Effiux of y-aminobutyric acid by synaptic plasma membrane vesicles isolated from rat brain, Biochemistry, 20 (1981) 3355-3358. [6] Krogsgaard-Larsen, P., Jacobsen, P., Brehm, L., Larsen, J. and Schaumburg, K., GABA agonists and uptake inhibitors designed as agents with irreversible actions, Eur. J. Med. Chem., 15 (1980) 529-535. [7] Liron, Z., Wong, E. and Roberts, E., Studies on uptake of yaminobutyric acid by mouse brain particles; toward the development of a model, Brain Res., 444 (1988) 119 132. [8] Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J., Protein measurements with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265 275. [9] Martin, D.L., Kinetics of the sodium-dependent transport of gamma-aminobutyric acid by synaptosomes, J. Neurochem., 21 (1973) 345 356. [10] Martin, D.L. and Smith III, A.A., Ions and the transport of gamma-aminobutyric acid by synaptosomes, J. Neurochem., 19 (1972) 841 855. [11] Nicholls, D.G., Release of glutamate, aspartate and y-aminobutyric acid from isolated nerve terminals, J. Neurochem., 52 (1989) 331-341. [12] Serfdz6, P. and Cash D.J., Effect of a benzodiazepine (chlordiazepoxide) on a GABA receptor from rat brain: requirement of only one bound GABA molecule for channel opening, FEBS Lett., 310 (1992) 55~59. [13] Turner, T.J. and Goldin, S.M., Multiple components of synaptosomal [3H]-?'-aminobutyric acid release resolved by a rapid superfusion system, Biochemistry, 28 (1989) 586-593.
Neuroscience Letters 182 (1994) 73-76
HEUROSCIINCE lETTERS
Modulation of GABA flux across rat brain membranes resolved by a rapid quenched incubation technique Julianna Kardos a'*, Ilona Kovfics a, Tamers Blandl a, Derek J. Cash b aDepartment of Molecular Pharmacology, Central Research Institute for Chemistry, The Hungarian Academy of Sciences, Pusztaszeri (tt 59-67, H-1025 Budapest, Hungary bDepartment of Biochemistry, School of Medicine, University of MissourL Columbia, MO, USA Received 5 August 1994; Revised version received 12 September 1994; Accepted 28 September 1994
Abstract The progress and inhibition of [3H]GABA influx in native plasma membrane vesicles from the rat cerebral cortex was studied on a subsecond to minute time scale under different conditions by applying a rapid quenched incubation technique. In the absence of Ca 2+ ([Ca2+]free = 10-s M), the progress of influx followed by the addition of 10 nM [3H]GABA to the membrane vesicle suspension with time (500 ms to 15 min) can be described by a first-order rate equation giving an overall rate constant, k, of 3.93 + 0.48 x 10 -3 s-1 and equilibrium influx value, INFe, of 8.84 + 0.41 pmol [3H]GABA/mg protein. In the presence of Ca 2÷ ([Ca2+]free = 2.4 × 10 -3 M) a significant increase in the INFe value was observed (k = 4.64 + 0.41 x 10 -3 s -l and INFe = 13.9 + 0.40 pmol [3H]GABA/mg protein). Multiplicity of GABA transporters was indicated in the time-dependent inhibition of [3H]GABA influx by different uptake blockers. In the absence of Ca 2÷, depolarization (75 mM KCI) inhibited the influx of [3H]GABA into the vesicles by ~70% and initiated the efflux from vesicles loaded with [3H]GABA. Different uptake blockers inhibited the Ca2+-independent translocation of [3H]GABA in both directions with similar specificities.
Key words: [3H]GABA translocation; Rapid quenched incubation technique; Membrane vesicle; Rat cerebral cortex
The pre-synaptic release of G A B A [13] and its postsynaptic action [3] occurs on a time scale of subsecond to seconds. To shape synaptic events the translocation of G A B A from the synaptic cleft needs to be fast and regulated. Examination of the effect of Ca 2+ on G A B A transport [10] showed that Ca 2+ strongly stimulated G A B A transport at low N a ÷ concentrations ([Na+]) but had only slight effects at high [Na+]. Since it is known that activators of protein kinase C and phosphatases stimulated G A B A influx via the cloned rat brain G A B A transporter (GAT1) expressed in Xenopus oocytes [4] it was of interest to see if Ca 2+ does have an effect on the transporterregulated G A B A influx on a physiologically relevant time scale. The release of cytoplasmic G A B A by reversal of the Na+-coupled transporters was considered artifactual consequence of unphysiologically prolonged de-polarization [11]. However, in the absence of Ca 2+, a tran-
*Corresponding author. 0304-3940194/$7.00 © 1994 Elsevier Science Ireland Ltd. All rights reserved SSD1 0 3 0 4 - 3 9 4 0 ( 9 4 ) 0 0 7 5 8 - 6
sient (< 1 s) nipecotic acid-sensitive component of high [K+]-induced G A B A release was observed by rapid superfusion of synaptosomes, suggesting that the transporter-regulated release of cytoplasmic G A B A m a y occur physiologically [13]. To address these issues, we have measured transmembrane G A B A influx in relevant time intervals with the use of a rapid quenched incubation technique which can resolve fast processes on a time scale of tens of milli-seconds to minute [2,3,12]. We have thus investigated (a) the progress of [3H]GABA influx in the presence and absence of Ca2+; (b) the progress of the inhibition of [3H]GABA influx by uptake blockers [1,6] in the absence of Ca2+; and (c) the effect of 75 m M KCI on [3H]GABA influx and efflux in the absence of Ca 2+. As -- 80% of synaptosomes were disrupted by fast displacements in the rapid kinetic apparatus native plasma m e m b r a n e vesicles [3,12] were applied. In order to characterise native plasma membrane vesicles, the Km and Vmax parameters of [3H]GABA influx and the inhibitory potencies of several G A B A uptake blockers [6] in mem-
74
J. Kardos et al./Neuroscienee Letters 182 (1994) 73 ~76
brane vesicles were determined in the presence of Ca 2+ ion using a conventional mixing-incubation technique. The cerebral cortex of a 4-6 weeks old male Wistar rat was homogenized by Ultraturrax (at 20,500 rpm for 5 s) in a HEPES-buffered sucrose solution containing protease inhibitors, antioxidant (phenyl-methanesulfonyl fluoride, 1 mM; aprotinin, 0.01 mg/ml; antipain, 0.005 mg/ml; leupeptin, 0.005 mg/ml; pepstatin A, 0.005 mg/ ml; butylated hydroxy-toluene, 0.02 m M [3]) and 1 mM ATE Native membrane vesicle fraction was obtained by the short, differential centrifugation procedure [3] with slight modifications [12]. Final pellet was suspended in either the buffer A [13] (in mM: NaC1, 145; KC1, 5; MgC12, 1; NaHCO3, 2; ATP, 1; D-glucose, 10; aminooxyacetate, 0.1; EGTA-Tris, 1; pH 7.5, [Ca2+]r~ = 10-8 M, pCa 8) or in buffer B (buffer A containing in addition 3.4 mM CaC12 ([Ca2+]f~ee= 2.4 mM, [13]). Native plasma membrane vesicles kept on ice for 3 hours have retained the initial transport activity. Membrane vesicle suspension (final protein concentration: 0.4-0.5 mg/ml, [8]) was usually incubated with 10 nM [3H]GABA (94 Ci/mmol, Amersham) in the presence of increasing concentration of GABA (0.1-10/IM) for 1, 5 and 20 rain at 30°C, pH 7.5. At the end of the incubation, the suspension was filtered (Whatman GF/B filters), washed with 2 x 3 ml ice-cold buffer and the radioactivity on the filter was counted by standard liquid scintillation techniques. Nonspecific influx was determined in the presence of 1 mM guvacine. The non-specific to total ratio was below 4% for the membrane vesicle preparation. The K m values were unchanged during the different periods of incubation: 4.3 + 0.6 y M (n = 4-5) and were similar to the Km of 4 / I M reported for synaptosomes [9]. The Vm~xvalues
were decreased with increasing incubation time: 900 + 280 ~ 380 + 125 pmol/min x mg protein). The inhibitory effect of 0.3 ¢tM to 1 mM GABA, 0.3-100 ~tM guvacine 0.2 ¢tM to 1 mM L-diaminobutyric acid (LDABA) and fl-alanine was measured in 1 rain of incubation using 100 nM [3H]GABA (9.1 Ci/mmol). The ICs0 values obtained in membrane vesicles were as follows (in /IM): GABA, 6.3 + 0.1; guvacine, 8.7 + 0.2; L-DABA, 28 + 6; fl-ALA, 2700 + 100. Aliquots of the membrane vesicle suspension in buffer A (B), kept at 4°C, were incubated at 30°C for 1 min, centrifuged (3 rain, 11,340 x g, SIGMA 112 microfuge); the resulting pellet was rinsed twice with 1 ml buffer A (B) and resuspended either in buffer A (B) or in buffer A (B) containing the different uptake blockers and kept at 30°C for 10 min before use. Fast kinetic experiments were performed with the use of the apparatus described previously [2]. The pulsed-incubation, quenched-flow technique was applied and the apparatus was thermostatted at 30°C [3,12]. Aliquot (0.34 ml) of the membrane vesicle suspension (0.8-1.0 mg protein/ml), p r e incubated with the uptake blocker was mixed with equal amount of buffer A (B) containing 20 nM [3H]GABA with and without the uptake blocker applied in the same concentration as in the p r e i n c u b a t i o n . After different time intervals of incubation, the suspension was mixed with equal amount of buffer A containing 2 mM guvacine (quench-solution) and filtered on Whatman GF/B filter; the filter was washed with 2 x 3 ml ice-cold buffer A and radioactivity was counted by standard liquid scintillation technique; to give specific [3H]GABA influx (100%), the [3H]GABA content of the filtered membrane vesicles measured in the presence of 1 mM guvacine was
8J o o_ 10
!
i
C~ L_ ~IL
o:
0.8
"° l
....
" /
4
o.6
4
./// . . . . . .
(]-
O
LI ........
~"
0.4
J
/
U
O
....
i
~JJ ............
!
f
o.2
:/i ..... ]
5
0
2S0
500
7S0
1@00
0
%
10
1%
Fig. 1. The progress of specific[3]GABAinfluxinto native plasma membranevesiclesfrom the rat cerebralcortex at 30°C, pH 7.5. o, [Ca2+]rr~= 10 8 M; [], [Ca2+]fr,o= 2.4 x 10-3 M. Lineswere calculated(Scientist,MicroMath Version2.03) accordingto a first-orderkineticswith the followingoverall rate constants (k) and equilibrium influx values (INFe) (mean +_S.D.): ©, k = 3.93 _+0.48 x 10-3 sq and INF,. = 8.84 -+0.41 pmol [3H]GABA/mg protein; [], k -- 4.64 + 0.41 x 10_3s-~ and INFe -- 13.9_+0.40 pmol [3H]GABA/mgprotein. Data are from a representativeexperiment,with measurements performed in duplicate. Other experimentsgave similar results. A: specificinfluxof 10 nM [3H]GABAfollowednear to completion;B: initial phase of [3H]GABAinflux.
J. Kardos et al./Neuroscience Letters 182 (1994) 73 76
subtracted from the [3H]GABA content of the filtered membrane vesicles measured in buffer A (B). Progress of [3H]GABA influx was followed in the 0.5 s to 15 min range of incubation in the presence or absence of Ca ~+ at 10 nM [3H]GABA concentration (Fig. 1). The first order rate equation, M = INFe[1-exp(-kt)], predicts the influx data in the whole range of equilibration with the overall rate constant, k of 3.93 + 0.48 x 1 0 -3 s -1 and the equilibrium influx value (INFe) of 8.84 + 0.41 pmol [3H]GABA/mg protein with 10 nM [3H]GABA at pCa 8 (buffer A). The presence of Ca 2÷ (buffer B) altered the influx of 10 nM [3H]GABA, characterized by k = 4.64 + 0.41 x 1 0 -3 S -1 and INFe = 13.9 + 0.40 pmol [3H]GABA/mg protein. Overall rate constant of 8.9 + 0.58 s -l and equilibrium influx value of 81.1 + 3.4 pmol [3H]GABA/mg protein were obtained when influx was initiated by 100 nM [3H]GABA. The progress of [3H]GABA influx (10 nM) in vesicles preincubated with the uptake blocker was followed in the 1 s to 60 s range of incubation in the absence of Ca 2+ (Table 1). A significant decrease of inhibition by fl-alanine and L-DABA was observed with increasing incubation time. The effect of depolarization on the influx of [3H]GABA by applying high concentration of K ÷ ion was studied when an aliquot of membrane vesicles in buffer A was mixed with equal amount of buffer C (in Table 1 Inhibition of [3H]GABA translocation by different GABA uptake inhibitors in native membrane vesiclesfrom the rat cerebral cortex at 30°c in the absence of calcium ion Uptake blockers (pM)
Inhibition (%) Influxa
Nipecotic acid (10) OH-nipecote acid (10) Guvacine (10) L-DABA (100) fl-Alanine (100)
Efflux b
1s
6s
60s
1s
82 + 7 60 + 7 60 + 9 85 + 5 46 + 3
84 + 5 65 + 4 70 + 5 85 + 6 39 + 7
86 + 1 70 + 2 64 + 2 39 + 6¢ 0 + 2c
80 + 17 58 + 18 62 + 13 67 + 16 54 + 22
Data are given as mean _ S.D. from 3 5 preparations with measurements performed in duplicate or triplicate (n = 610). aPre-incubation experiment: pelleted membrane vesicleswerepre-incubated in buffer A in the presence or absence of the uptake blocker in the following concentration (in pM): nipecotic acid, 10; OH-nipecotic acid, 10; guvacine, 10; L-DABA, 100;fl-alanine, 100; and kept at 30°C for 10 min before use; in the quech-flow experiment the pre-incubated vesicleswere mixed with an equal amount of buffer A containing 20 nM [3H]GABA with and without the uptake blocker applied in the same concentration. bCo-incubation experiment: aliquots of membrane vesiclesin buffer A, kept at 4°C, were incubated at 30°C; after 1 min equal amount of [3H]GABA in buffer A was added (40 nM) and the suspension was incubated at 30°C for 13 min (loading); in the quench-flowexperiment loaded vesicles were mixed with equal amount of buffer C with and without the uptake blocker applied in the following concentration (in /~M): nipecotic acid, 20; OH-nipecotic acid, 20; guvacine, 20; L-DABA, 200; fl-alanine, 200. cp < 0.001, when [3H]GABAinflux in 1 s is compared (Student's t-test).
75
mM: NaC1, 5; KC1, 145; MgC12, 1; NaHCO3, 2; ATE 1; D-glucose, 10; amino-oxyacetate, 0.1; EGTA-Tris, 1; p H 7.5) containing 40 and 80 n M of [3H]GABA. In the absence of Ca 2+, substitution of 75 m M KC1 for NaC1 (reportedly [7], this decrease of N a + ion concentration, final [Na +] = 75 m M , has no effect on G A B A influx) inhibited [3H]GABA influx in 1 s ( 6 4 _ 8% at 20 nM [3H]GABA and 80 _+ 9% at 40 nM [3H]GABA, n = 6). Prolonged incubation (6 s) did not affect the high [K+] induced inhibition significantly ( 6 9 _ 7% at 20 nM [3H]GABA and 72 _+ 6% at 40 n M [3H]GABA). These data suggest that we can disregard the relatively slow process of swelling as well as the isotopic dilution of the label by the release of endogenous G A B A as possible sources [9] of the high K+-induced inhibition of [3H]GABA influx. To compare CaE+-insensitive influx with that of efflux, an aliquot of the membrane vesicle suspension in buffer A, kept at 4°C, was incubated at 30°C. After 1 rain, an equal volume of [3H]GABA in buffer A was added (40 nM) and the suspension was incubated at 30°C for 13 rain (loading); 0.34 ml of the loaded membrane vesicle suspension containing 20 nM [3H]GABA was mixed with equal amount of buffer C containing the different uptake blockers. To give specific [3H]GABA efflux (100%), the [3H]GABA content of the filtered membrane vesicles measured in the absence of uptake blockers (100000_+ 5000 dprn) was subtracted from the [3H]GABA content of the filtered membrane vesicles measured in the presence of 1 m M guvacine (130000 + 7000 dpm). When the inhibition of the efflux of [3H]GABA from membrane vesicles in 1 s was compared with that of [3H]GABA influx in 1 s (Table 1), the different G A B A uptake blockers showed similar inhibitory specificities suggesting an identity of the transporters moving G A B A in cis---~trans and trans--)cis directions. Slow kinetic properties of the transporter(s) we observed suggest that the concentration of G A B A in the synaptic cleft could be critical to remove G A B A effectively in 'synaptic' time. We can estimate the half-response concentration for the faster and the slower desensitizing G A B A g receptors, a concentration of ~ 100 p M G A B A [3] in the synaptic cleft during inhibition. Under the condition, the G A B A transporter(s) would remove G A B A more than 100 times faster, which is on the timescale as the slower desensitizing GABAA receptor with 0.5 s half-time. On the basis of the known pharmacological specificity of the cloned G A B A transporters, GAT1, GAT2 and GAT3 [1], our data indicate that [3H]GABA influx into native membrane vesicles from the rat cerebral cortex is regulated by GAT1- and GAT2-type transporters. Both transporters are to be active in short times, whereas [3H]GABA influx near to completion is regulated by the G A T l - t y p e transporter. In addition to the requirements for N a + and CI ions [6] the pharmacological specificity of G A B A influx was found to be similar
76
Z Kardos et aL / Neuroscience Letters 182 (1994; 73-76
to its efflux suggesting that transporter-mediated GABA translocation may change from cis--*trans to trans---~cis direction when an action potential reaches the presynaptic terminal. GABA uptake in GATl-expressing oocytes by protein kinase C activators has been demonstrated to be mediated by a regulated subcellular redistribution of the transporter showing the onset 20 min after injection [4]. The onset we observed for the Ca 2÷induced stimulation of GABA uptake in native plasma membrane vesicles occurs on a subsecond time scale suggesting that the modulation may involve a fast Ca 2+induced conversion of transporter between active and inactive states. In conclusion, our fast kinetic experiments performed on native plasma membrane vesicles from the rat cerebral cortex indicated that the rate of GABA influx is sufficient to shape slow synaptic events. It is proposed now that Ca 2÷ ion and depolarization may play a role in regulating transporter-mediated GABA translocation in neurotransmission. J.K. was supported by OTKA T 4030 and US-Hungarian JF No. 277 grants. The skillful assistance of Mrs. Erzs6bet Fekete-K6ti is gratefully acknowledged. [1] Borden, L.A., Smith, K.E., Hartig, RR., Branchek, T.A. and Weinshank, R.L., Molecular heterogeneity of the y-amino-butyric acid (GABA) transport system, J. Biol. Chem., 267 (1992) 21098 21104. [2] Cash, D.J. and Hess, G.P., Quenched-flow technique with plasma membrane vesicles: acetylcholine-receptor-mediated transmembrane ion-flux, Anal. Biochem., 112 (1981) 39-51.
[3] Cash, D.J. and Subbarao, K., Desensitization of y-amino-butyric acid receptor from rat brain: two distinguishable receptors on the same membrane, Biochemistry, 26 (1987) 7556-7562. [4] Corey, J.L., Davidson, N., Lester, H.A., Brecha, N. and Quick, W., Protein kinase C modulates the activity of a cloned y-aminobutyric acid transporter expressed in Xenopus Oocytes via regulated subcellular redistribution of the transporter, J. Biol. Chem., 269 (1994) 14759-14767. [5] Kanner, B.I. and Kifer, L., Effiux of y-aminobutyric acid by synaptic plasma membrane vesicles isolated from rat brain, Biochemistry, 20 (1981) 3355-3358. [6] Krogsgaard-Larsen, P., Jacobsen, P., Brehm, L., Larsen, J. and Schaumburg, K., GABA agonists and uptake inhibitors designed as agents with irreversible actions, Eur. J. Med. Chem., 15 (1980) 529-535. [7] Liron, Z., Wong, E. and Roberts, E., Studies on uptake of yaminobutyric acid by mouse brain particles; toward the development of a model, Brain Res., 444 (1988) 119 132. [8] Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J., Protein measurements with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265 275. [9] Martin, D.L., Kinetics of the sodium-dependent transport of gamma-aminobutyric acid by synaptosomes, J. Neurochem., 21 (1973) 345 356. [10] Martin, D.L. and Smith III, A.A., Ions and the transport of gamma-aminobutyric acid by synaptosomes, J. Neurochem., 19 (1972) 841 855. [11] Nicholls, D.G., Release of glutamate, aspartate and y-aminobutyric acid from isolated nerve terminals, J. Neurochem., 52 (1989) 331-341. [12] Serfdz6, P. and Cash D.J., Effect of a benzodiazepine (chlordiazepoxide) on a GABA receptor from rat brain: requirement of only one bound GABA molecule for channel opening, FEBS Lett., 310 (1992) 55~59. [13] Turner, T.J. and Goldin, S.M., Multiple components of synaptosomal [3H]-?'-aminobutyric acid release resolved by a rapid superfusion system, Biochemistry, 28 (1989) 586-593.