- Email: [email protected]
GABA terminal autoreceptors in the pars compacta and in the pars reticulata of the rat substantia nigra are GABAB
European Journal of Pharmacology, 175 (1990) 137-144 Elsevier
137
EJP 51096
GABA terminal autoreceptors in the pars compacta and in the pars reticulata of the rat substantia nigra are GABA 8 M a r i a T e r e s a G i r a l t 1, G i a m b a t t i s t a B o n a n n o a n d M a u r i z i o R a i t e r i Istituto di Farrnacologia e Farmacognosia, Universith degli Studi di Genova, Viale Cembrano 4, 16148 Genova, Italy Received 19 September 1989, accepted 10 October 1989
The depolarization-evoked release of ~,-aminobutyric acid (GABA) and its modulation mediated by autoreceptors were studied in superfused synaptosomes prepared from the pars compacta and from the pars reticulata of the rat substantia nigra. The release of [3H]GABA evoked by 9 mM KC1 was almost totally calcium-dependent in both nigral subregions. In the presence of SK&F 89976A (N-(4,4-diphenyl-3-butenyl)nipecotic acid), a GABA uptake inhibitor added to minimize carrier-mediated homoexchange, GABA (0.3-10 #M) inhibited, in a concentration-dependent way, the K+-evoked overflow of [3H]GABA from both pars compacta and pars reticulata synaptosomes. Similarly to GABA, (-)-baclofen (0.3-10/LM) reduced the [3H]GABA overflow, being roughly equipotent to GABA in both nigral subregions. The ( + ) enantiomer of baclofen was ineffective. The overflow of [3 H]GABA was not consistently affected by muscimol in either the pars compacta or the pars reticulata. The effects of GABA were bicuculline- and picrotoxin-insensitive. However, the inhibition by GABA of the [ 3H]GABA overflow was antagonized by phaclofen. It is concluded that (a) GABA autoreceptors are sited on GABAergic nerve endings in both the pars compacta and pars reticulata of the rat substantia nigra; (b) these autoreceptors belong to the GABA B type. GABA autoreceptors; Substantia nigra (pars compacta); Substantia nigra (pars reticulata); GABA 8 receptor subtypes
1. Introduction In the last few years several reports have focussed on the existence and, particularly, the pharmacological characterization of G A B A autoreceptors in the mammalian CNS. The results of previous publications showing that the release of radioactive G A B A from rat brain cortical slices (Mitchell and Martin, 1978) or synaptosomes (Brennan et al., 1981) or from rat striatal slices (Kuriyama et al., 1984) was inhibited by muscimol
l Present address: Departamento de Farmacologia y Terapeutica, Facultad de Medicina y Odontologia, Universidad del Pais Vasco, 48940 Leioa (Vizcaya), Spain. Correspondence to: M. Raiteri, Istituto di Farmacologia e Farmacognosia, Universith degli Studi di Genova, Viale Cembrano 4, 16148 Genova, Italy.
in a bicuculline-sensitive manner have not been confirmed by recent investigations. In fact, the G A B A A receptor agonist could not inhibit G A B A release from rabbit striatal slices (Limberger et al., 1986), rat cortical synaptosomes (Pittaluga et al., 1987) or slices of rat cerebral cortex, hippocampus and striatum (Waldmeier et al., 1988) or from synaptosomes prepared from human cerebral cortex (Bonanno et al., 1989a). In all these systems the release of G A B A was instead inhibited by the GABAB receptor agonist, baclofen (Bowery et al., 1984), in a stereoselective manner. The findings that ( - ) - b a c l o f e n acted directly on GABAergic nerve terminals (Anderson and Mitchell, 1985; Pittaluga et al., 1987) and, in particular, that its inhibitory action was counteracted by phaclofen (Bonanno et al., 1988; 1989b), a G A B A B receptor antagonist (Kerr et al., 1987; Dutar and Nicoll,
0014-2999/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
138
1988), support strongly the idea that GABA autoreceptors in the mammalian CNS belong to the GABA B type. Of the recent publications, however, the one by Floran et al. (1988) seems to represent an exception. These authors studied the effects of GABA agonists and antagonists on the release of [3HIGABA stimulated by high K + in the pars compacta and in the pars reticulata of the rat substantia nigra. While in the latter nigral subregion the release of [3H]GABA was decreased by ( - ) - b a c l o f e n , muscimol-sensitive GABA A autoreceptors were reported to be present in the pars compacta. It should be noted, however, that the work was carried out with slices in which the inhibitory effects of GABAergic agonists may not reflect activation of autoreceptors sited on GABA terminals. Unfortunately, the sensitivity of the agonist effects to tetrodotoxin was not tested. These considerations, together with the earlier finding that the release of [3H]GABA from slices of rat substantia nigra, although decreased by muscimol, was paradoxically inhibited by bicuculline (Arbilla et al., 1979), prompted us to reinvestigate the problem of GABA autoreceptors in the substantia nigra by using a different experimental approach. In particular: (a) synaptosomes prepared from the pars compacta or the pars reticulata of the rat substantia nigra were used to study [3H]GABA release; (b) the natural agonist GABA was used as an agonist together with muscimol and baclofen; (c) the recently available GABAB receptor antagonist, phaclofen, was also used.
2. Materials and methods
2.1. Preparation of synaptosomes Adult male Sprague-Dawley rats (200-250 g) were used. The animals were killed by decapitation and the brain was removed rapidly. After the substantia nigra had been isolated the pars reticulata was separated from the pars compacta by microdissection. Crude synaptosomes were prepared according to the method of Gray and Whittaker (1962) with some modifications. Briefly, the
tissues pooled from two rats were homogenized with a glass-teflon tissue grinder (clearance 0.25 ram) in 3 ml of 0.32 M sucrose buffered at pH 7.4 with phosphate, The homogenate was centrifuged (5 min, 1 000 × g) to remove nuclei and debris, and synaptosomes were isolated from the supernatant by centrifugation at 12000 x g for 20 min. The synaptosomal pellet was then resuspended in a physiological medium of the following composition (mM): NaC1 125; KC1 3; MgSO4 1.2; CaC12 1.2; N a H 2 P O 4 1.0; N a H C O 3 22; glucose 10 (aeration with 95% 02 and 5% CO 2 at 37°C); pH 7.2-7.4. Protein was measured according to the method of Bradford (1976) with bovine serum albumin as standard.
2.2. Release experiments Synaptosomes obtained from the pars compacta or pars reticulata of the substantia nigra were incubated with 0.04 /~M [3H]GABA, corresponding to 10-12/~Ci/sample, in a rotatory water bath at 3 7 ° C for 15 min. After labeling, the synaptosomes were diluted with tracer-free medium; identical aliquots of the synaptosomal suspensions (ranging between 0.01-0.05 mg protein in the different experiments and for the two different subregions studied) were distributed on 0.65/~m Millipore filters placed at the bottom of a set of parallel superfusion chambers maintained at 3 7 ° C (Raiteri et al., 1974), and layered under moderate vacuum filtration. Superfusion was then started with standard medium, aerated with 95% 02 and 5% CO 2, at a rate of 0.6 m l / m i n and continued for 48 min. The system was allowed to equilibrate, for 36 min and then fractions were collected according to the following scheme: two 3-min samples (basal release) before and after one 6-min sample (evoked release). A 90-s period of depolarization with 9 mM KC1 was applied after the first fraction was collected; KC1 replaced an equimolar concentration of NaC1. GABA was added to the superfusion medium 8 min before depolarization and was present until the end of the experiment. Muscimol or ( - ) - b a c l o f e n was used concomitantly with the depolarizing stimulus. Bicuculline, picrotoxin, SR 95531 (2-(3'-carbethoxy2'-propenyl)-3-amino-6-paramethoxy-phenyl-pyr-
139 idazinium bromide) or phaclofen was added 10 min before GABA. S K & F 89976A (N-(4,4-diphenyl-3-butenyl)-nipecotic acid), a potent GABA uptake inhibitor (Yunger et al., 1984; Bonanno and Raiteri, 1987; Larsson et al., 1988), was present in the superfusion medium (final concentration 30 ~M) to minimize carrier-mediated exchange between intraterminal [3H]GABA and extrasynaptosomal GABA or muscimol. Aminooxyacetic acid (AOAA; final concentration 50 /~M) was present throughout the experiment to prevent GABA metabolism. The low-Ca 2+ medium, when used, was substituted for the standard medium 18 min before depolarization; this medium contained 0.2 mM Ca 2+ and 10 mM Mg 2+. The tritium collected in each fraction and that remaining in the synaptosomes at the end of the experiment was counted.
2.3. Calculation The amount of radioactivity released in each fraction is expressed as a percentage of the total tritium content of the synaptosomes at the start of the respective collection period. The depolarization-evoked overflow was estimated by subtracting the percentage tritium content of the basal release from the release evoked in the 6-min fraction collected during and after the depolarization pulse. The drug effects were evaluated as the ratio of the depolarization-evoked overflow calculated in the presence of the drugs vs. that calculated under control conditions. A two-tailed Student's t-test was used to compare mean values.
2.4. Drugs [3H]GABA (51-59 C i / m m o l ) was obtained from Amersham Radiochemical Centre (Buckinghamshire, U.K.); aminooxyacetic acid, (+)-bicuculline, picrotoxin and muscimol were purchased from Sigma Chemical Co. (S. Louis, MO, U.S.A.); GABA from Serva (Heidelberg, F.R.G.) and phaclofen from Tocris Neuramin (Essex, U.K.). The following drugs were a generous gift of the companies indicated: ( - ) - b a c l o f e n (Ciba Geigy, Basel, Switzerland); S K & F 89976A (Smith, Kline
and French, Welwyn, U.K.); SR 95531 (Sanofi, Bruxelles, Belgium).
3. Results
3.1. Characteristics of the depolarization-evoked release of [3H]GABA from substantia nigra synaptosomes Superfused synaptosomes prepared from the pars compacta and pars reticulata of the rat substantia nigra were exposed to 9 mM KC1. The depolarization-evoked overflow of tritium amounted to 0.86 _+ 0.03% (n = 28) in the pars compacta and to 1.58 + 0.09% (n = 28) in the pars reticulata. Under identical experimental conditions the [3H]GABA overflow from cerebral cortex synaptosomes amounted to about 1.9% of the total tritium content (Bonanno et al., 1989b). When the concentration of Ca 2+ in the superfusion medium was lowered from 1.2 mM to 0.2 mM and that of Mg 2+ was raised to 10 raM, the K+-evoked tritium overflow from both pars compacta and pars reticulata synaptosomes was almost abolished (data not shown). The strict calcium dependence of the depolarization-evoked tritium overflow, the presence in the superfusion medium of 50 /~M of the GABA transaminase inhibitor, aminooxyacetic acid, the characteristics of the superfusion technique used (see Raiteri and Levi, 1978) together with data obtained earlier in various laboratories (see for a review Levi, 1984) make it likely that the radioactivity released upon depolarization consisted largely of authentic [3H]GABA. Therefore, we refer to the K+-evoked tritium overflow as K +evoked [3 H]GABA release in the remainder of the text.
3.2. Effects of GABA receptor agonists on the K +evoked [-¢H]GABA release from nigral synaptosomes In a first set of experiments we examined the effects of exogenous GABA on the release of radiolabeled GABA provided by exposure of synaptosomes prepared from the pars compacta or
140 pars compact a
pars ret icu lat a
6O .o_ J= 50
E_
O
;~ 40 LU O3
<
w ..J w II
q-
30 20
< II1 <
T
"i"
0.3
1
GABA
10
0.3
CONCENTRATION
10 (pM)
Fig. 1. Concentration-dependent inhibition by GABA of the K ÷ (9 mM)-evoked release of [3H]GABA from synaptosomes of the pars compacta and pars reticulata of the rat substantia nigra. Synaptosomes were prelabeled with 0.04/~ M [3H]GABA and superfused (see Methods). After 38 rain of superfusion, the synaptosomes were depolarized with 9 mM KC1. Exogenous GABA was added to the superfusion medium 8 min before depolarization. The [3H]GABA overflow was calculated by subtracting the percentage of [3H]GABA found in the two 3-min fractions corresponding to the basal release from the radioactivity in the 6-min fraction collected during and after the stimulation period. The basal efflux in the fraction preceeding the stimulation amounted to 0.98 5:0.05 nCi/min, corresponding to 0.51 5:0.01% (n = 28), in the pars compacta and to 1.38+0.06 nCi/min, corresponding to 0.545:0.02% (n = 27), in the pars reticulata. Results are means 5: S.E.M. of 3-10 experiments run in triplicate.
the pars reticulata of the rat substantia nigra to high K ÷. Exogenous GABA was added to the superfusion medium in the presence of 30 # M of the GABA uptake inhibitor, S K & F 89976A. This compound effectively prevented GABA homoexchange since the highest concentration of GABA used (10 ~M) increased the basal outflow of [3H]GABA by less than 10%. In the absence of GABA, S K & F 89976A did not affect [3H]GABA release (data not shown). Figure 1 shows that GABA (0.3-10/LM) concentration dependently inhibited the K+-evoked release of [3H]GABA from both pars compacta and pars reticulata synaptosomes.
Figure 2 illustrates the effects of the enantiomers of baclofen and of muscimol on the release of [3H]GABA. In both the pars compacta and reticulata ( - ) - b a c l o f e n (0.3-10 t~M) reduced, in a concentration-dependent manner, the release of [3 H]GABA evoked by depolarization. The GABA receptor agonist was at least as effective as GABA. The effects seen in the pars reticulata appeared to be slightly more pronounced than those seen in the compacta although the differences were not statistically significant. The effect of baclofen was stereoselective since the ( + ) enantiomer was almost inactive at 10/~M (fig. 2). In contrast to ( - ) - b a c l o f e n , the GABA A agonist, muscimol (10-100 /~M), did not inhibit the K+-evoked release of [3H]GABA from pars compacta and pars reticulata synaptosomes. The effects observed were very modest and were not concentration-dependent (fig. 2).
3.3. Antagonism of the GABA-induced inhibition of [3H]GABA release As shown in fig. 3, the inhibition of [3H]GABA release caused by GABA in both nigral subregions was not counteracted by the GABA A receptor antagonists, bicuculline or picrotoxin. Also the more recent GABA A receptor antagonist, SR 95531 (Wermuth and Bizirre, 1986), was ineffective when tested at 10 /~M against 1 ~tM GABA (result not shown). In contrast, the effect of 1 #M GABA was prevented by 100/~M of the GABA B receptor antagonist, phaclofen (fig. 3). In the absence of GABA, the GABA receptor antagonists used had no or only very modest effects on [3H]GABA release (not shown).
4. Discussion
As mentioned in the Introduction, several recent publications have reported that the release of GABA is insensitive to muscimol but can be inhibited by ( - ) - b a c l o f e n . These findings were obtained in slices of rabbit caudate nucleus (Limberger et al., 1986), rat cerebral cortex, hippocampus or striatum (Waldmeier et al., 1988; Raiteri et al., 1989) as well as in synaptosomes of rat or
141 pars compact a
70-
pars
reticulata
60o
r
5C
.E O ae
40
T
uJ u~
3C
7-
...1 213 ii) 0 "r
7"
1£
03
1
10
-
-
0.3
10
-
1
10 10
-
-
10
DRUG
-
100
CONCENTRATION
10
100
i-)baclofen (+)baclofen muscimol
(pM)
Fig. 2. Effects of (-)-baclofen, (+)-baclofen and muscimol on the K+-evoked release of [3H]GABA from nigra compacta and reticulata synaptosomes. The agonists were added to the superfusion medium at the same time as the synaptosomes were depolarized with K +. The experimental details are as in fig. ]. Results are expressed as means+S.E.M, of three to six experiments done in triplicate.
pars corrlpacta
pars retlculata
60 A
.o
T
50
~5 40 a~ LU (13 w .J w cc
T
II
30
20
00
0
10
L
0 1
1 10
1
-
100
10
10 10
10 10
1
1 10
1
CONCENTRATION
10 10
10 10
100 DRUG
10
GABA bicueulline picrotoxin phaclofen
(IJM)
Fig. 3. Antagonism of the GABA-induced inhibition of the [3H]GABA overflow from nigra compacta and reticulata synaptosomes. The experimental details are as in fig. 1. Exogenous G A B A was added to the superfusion m e d i u m 8 rnin before depolarization and the antagonists 10 min before GABA. The results are means -t- S.E.M. of three to six experiments done in triplicate. * P < 0.001 vs. 1 # M G A B A with the two-tailed Student's t-test.
142 human cerebral cortex (Pittaluga et al., 1987; Bonanno et al., 1989a). Thus the present evidence seems to favor the idea that G A B A autoreceptors in various areas of the mammalian CNS are of the G A B A B type. However, the situation appears to be rather peculiar in the substantia nigra. In this area Arbilla et al. (1979) found that muscimol inhibited the K+-evoked release of [3H]GABA; however, bicuculline, instead of antagonizing muscimol, paradoxically inhibited the overflow of the amino acid. ( - ) - B a c l o f e n was not tested. On the other hand, Waldmeier et al. (1988) could not analyze the effects of muscimol or ( - ) - b a c l o f e n on the release of G A B A because G A B A overflow could not be detected when slices of the substantia nigra were exposed to electrical stimuli which induced the release of G A B A from other CNS areas. By using high K + as a depolarizing agent, Floran et al. (1988) did observe a calcium-dependent release of [3H]GABA from slices of rat substantia nigra. Curiously, this release was inhibited by muscimol but not by ( - ) - b a c l o f e n in the pars compacta, while it was inhibited by ( - ) - b a c l o f e n but not by muscimol in the nigra reticulata, suggesting that the two nigral subregions contain pharmacologically different G A B A autoreceptors. The reasons for the above discrepancies may be manifold. For instance, when studying autoreceptors it is important to use the natural transmitter as an agonist. Unfortunately, the use of G A B A as an autoreceptor agonist has been hindered due to the lack of a suitable carrier blocker able to prevent homoexchange between external G A B A and its intraterminal radioactive counterpart. This problem now appears to be solved and G A B A homoexchange can be effectively prevented by S K & F 89976A (Pittaluga et al., 1987), a potent and selective GABA uptake blocker (Yunger et al., 1984). The results of fig. 1 show that, in the presence of S K & F 89976A, G A B A inhibited the K+-evoked release of [3H]GABA from nigral synaptosomes and displayed a similar potency in the pars compacta and pars reticulata. The effect of the natural transmitter on its own release together with the effects of selective receptor antagonists suggest the presence of a receptor-mediated autoregulatory mechanism.
Another critical aspect is that of the exact localization of the receptor involved. Brain slices have often been used in studies of neurotransmitter release because of their anatomical integrity. However, synaptosomes in superfusion represent the preparation of choice when the localization of a release-regulating receptor has to be determined. In fact, modulation of the release of a transmitter by a given agonist could occur through neuronal circuits which remain intact in brain slices so that the receptors involved need not be sited on the releasing nerve terminals. These neuronal circuits are destroyed during the preparation of synaptosomes. Moreover, the release technique used in our laboratory (superfusion of a thin layer of synaptosomes) does not permit indirect interactions to take place. Therefore, when an agonist modulates the release of a transmitter in superfused synaptosomes, it is likely to occur by a direct action on the releasing particles. The results of the present study show that the G A B A A receptor agonist, muscimol, displayed inconsistent effects on the release of [3 H]GABA not only in pars reticulata synaptosomes but also in nerve endings prepared from the pars compacta of the substantia nigra. The technical considerations discussed above together with the finding that the G A B A A receptor antagonists bicuculline~ picrotoxin and SR 95531 did not antagonize the G A B A inhibition of [3H]GABA overflow tend to exclude the possibility that G A B A A receptors are present on GABAergic nerve terminals, as proposed by Floran et al. (1988). Also the possibility that G A B A A receptors desensitize rapidly (Adams and Browm 1975) seems unlikely. In fact, we added muscimol to the superfusion medium concomitantly with the depolarizing stimulus in our experiments. Furthermore, G A B A decreased its own release. In contrast to muscimol, ( - ) - b a c l o f e n stereoselectively inhibited the [3H]GABA overflow in both parts of the substantia nigra. The involvement of G A B A B receptors is strengthened by the finding that the inhibition by G A B A of [3H]GABA overflow was counteracted by phaclofen, a novel G A B A B receptor antagonist (Kerr et al., 1987; Dutar and NicoU, 1988). Thus the present results with synaptosomes are in keeping with those ob-
143
tained with slices by Floran et al. (1988) only when the release of [3H]GABA in the pars reticulata is considered. As to the pars compacta, GABAergic nerve endings appear to possess GABA B instead of GABA A autoreceptors. As a possible explanation one could hypothesize that the muscimol-induced inhibition of [3H]GABA release in slices of the pars compacta observed by Floran et al. (1988) occurs through a more or less complex intra-compacta circuit involving GABAA receptors that are not sited on the GABAergic nerve endings. In other words, in the nigra compacta autoinhibition of GABA release might occur through a dual mechanism, one mediated directly by GABA B autoreceptors sited on GABAergic terminals and the other involving the activation of GABA A receptors present on other structures. The GABAA-mediated mechanism would prevail in the in vitro experimental conditions used by Floran et al. (1988) while the relative importance of the two mechanisms, if present in the living brain, remains to be established. In conclusion, several transmitters are released during stimulation of a brain slice, and several, sometimes complex, interactions may occur. Thus it is difficult to localize receptors involved in the auto- or the heteroregulation of the release of one transmitter. The localization of such a receptor is much easier when using a thin layer of synaptosomes in superfusion, a system in which only release-regulating receptors localized on the particle releasing the transmitter in question can be involved. Our results indicate that GABAergic nerve terminals of both the pars compacta and pars reticulata of the substantia nigra possess autoreceptors. These receptors are activated by GABA itself and, stereoselectively, by ( - ) - b a c l o fen, but not by muscimol. The effect of GABA is phaclofen-sensitive but bicuculline-insensitive. Thus, in analogy to what has been found in several other areas of the rat CNS, the GABA autoreceptors sited on GABAergic terminals in the substantia nigra appear to belong to the G A B A B type.
Acknowledgements This work was supported by grants from the Italian Ministry of Education and from the Italian C.N.R. The authors wish to thank Mrs. Maura Agate for her skilful secretarial assistance.
References Adams, PR. and D.A. Brown, 1975, Actions of y-aminobutyric acid on sympathetic ganglion cells, J. Physiol. (London) 250, 85. Anderson, R.A. and R. Mitchell, 1985, Evidence for GABA B autoreceptors in median eminence, European J. Pharmacol. 118, 355. Arbilla, S., L. Kamal and S.Z. Langer, 1979, Presynaptic GABA autoreceptors on GABAergic nerve endings of the rat substantia nigra, European J. Pharmacol. 57. 211. Bonanno, G., P. Cavazzani, G.C. Andrioli, D. Asaro, G. Pellegrini and M. Raiteri, 1989a, Release-regulating autoreceptors of the GABAB-type in human cerebral cortex, Br. J. Pharmacol. 96, 341. Bonanno, G., G. Fontana and M. Raiteri, 1988, Phaclofen antagonizes GABA at autoreceptors regulating release in rat cerebral cortex, European J. Pharmacol. 154, 223. Bonanno, G., G. Pellegrini, D. Asaro, G. Fontana and M. Raiteri, 1989b, GABAB autoreceptors in rat cortex synaptosomes: response under different depolarizing and ionic conditions, European J. Pharmacol. Mol. Pharmacol. Sect. 172, 41. Bonanno, G. and M. Raiteri, 1987, Coexistence of carriers for dopamine and GABA uptake on a same nerve terminal in the rat brain, Br. J. Pharmacol. 91, 237. Bowery, N.G., G.W. Price, A.L. Hudson, D.R. Hill, G.P. Wilkin and M.J. Turnbull, 1984, GABA receptor multiplicity, Visualization of different receptor types in the mammalian CNS, Neuropharmacotogy 23, 219. Bradford, M.M, 1976, Rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72, 248. Brennan, M.J.W., R.C. Cantrill, M. Oldfield and P. Krogsgaard-Larsen, 1981, Inhibition of "r-aminobutyric acid release by ,¢-aminobutyric acid agonist drugs. Pharmacology of the y-aminobutyric acid autoreceptor, Mol. Pharmacol. 19, 27. Dutar, P. and R.A. Nicoll, 1988, A physiological role for GABA B receptors in the central nervous system, Nature 332, 156. Floran, B., I. Silva, C. Nava and J. Aceves, 1988, Presynaptic modulation of the release of GABA by GABA A receptors
144 in pars compacta and by GABA B receptors in pars reticulata of the rat substantia nigra, European J. Pharmacol. 150, 277. Gray, E.G. and V.P. Whittaker, 1962, The isolation of nerve endings from brain: an electron microscopic study of cell fragments derived by homogenization and centrifugation, J. Anat. 96, 79. Kerr, D.I.B., J. Ong, R.H. Prager, B.D. Gynther and D.R. Curtis, 1987, Phaclofen: a peripheral and central baclofen antagonist, Brain Res. 405, 150. Kuriyama, K., K. Kanmori, J. Taguchi and Y. Yoneda, 1984, Stress-induced enhancement of suppression of [3H]GABA release from striatal slices by presynaptic autoreceptor, J. Neurochem. 42, 943. Larsson, O.M., E. Falch, P. Krogsgaard-Larsen and A. Schousboe, 1988, Kinetic characterization of inhibition of y-aminobutyric acid uptake into cultured neurons and astrocytes by 4,4-diphenyl-3-butenyl derivatives of nipecotic acid and guvacine, J. Neurochem. 50, 818. Levi, G., 1984, Release of putative transmitter amino acids, in: Handbook of Neurochemistry, Vol. 6, ed. A. Lajtha (Plenum Publishing Co., New York) p. 431. Limberger, N., L. Sp~ith and K. Starke, 1986, A search for receptors modulating the release of 3,-[3H]aminobutyric acid in rabbit caudate nucleus slices, J. Neurochem. 46, 1109. Mitchell, P.R. and I.L. Martin, 1978, Is GABA release modulated by presynaptic receptors?, Nature 274, 904.
Pittaluga, A., D. Asaro, G. Pellegrini and M. Raiteri, 1987, Studies on 3H-GABA and endogenous GABA release in rat cerebral cortex suggest the presence of autoreceptors of the GABA 8 type, European J. Pharmacol. 144, 45. Raiteri, M. F. Angelini and G. Levi, 1974, A simple apparatus for studying the release of neurotransmitters from synaptosomes, European J. Pharmacol. 25, 411. Raiteri, M., G. Bonanno and E. Fedele, 1989, Release of [3H]'t-aminobutyric acid from electrically-stimulated rat cortical slices and its modulation by GABA B autoreceptors, J. Pharmacol. Exp. Ther. (in press). Raiteri, M. and G. Levi, 1978, Release mechanisms for catecholamines and serotonin in synaptosomes, in: Reviews of Neuroscience, Vol. 3, eds. S. Ehrenpreis and I. Kopin (Raven Press, New York) p. 77. Waldmeier, P.C., P. Wicki, J.-J. Feldtrauer and P.A. Baumann, 1988, Potential involvement of a baclofen-sensitive autoreceptor in the modulation of the release of endogenous GABA from rat brain slices in vitro, Naunyn-Schmiedeb. Arch. Pharmacol. 337, 289. Wermuth, C.G. and K. Bizi~re, 1986, Pyridazinyl-GABA derivatives: a new class of synthetic GABA A antagonists, Trends Pharmacol. Sci. 7, 421. Yunger, L.M., P.J. Fowler, P. Zarevics and P.E. Setler, 1984, Novel inhibitors of -/-aminobutyric acid (GABA) uptake: anticonvulsant actions in rats and mice, J. Pharmacol. Exp. Ther. 228, 109.
137
EJP 51096
GABA terminal autoreceptors in the pars compacta and in the pars reticulata of the rat substantia nigra are GABA 8 M a r i a T e r e s a G i r a l t 1, G i a m b a t t i s t a B o n a n n o a n d M a u r i z i o R a i t e r i Istituto di Farrnacologia e Farmacognosia, Universith degli Studi di Genova, Viale Cembrano 4, 16148 Genova, Italy Received 19 September 1989, accepted 10 October 1989
The depolarization-evoked release of ~,-aminobutyric acid (GABA) and its modulation mediated by autoreceptors were studied in superfused synaptosomes prepared from the pars compacta and from the pars reticulata of the rat substantia nigra. The release of [3H]GABA evoked by 9 mM KC1 was almost totally calcium-dependent in both nigral subregions. In the presence of SK&F 89976A (N-(4,4-diphenyl-3-butenyl)nipecotic acid), a GABA uptake inhibitor added to minimize carrier-mediated homoexchange, GABA (0.3-10 #M) inhibited, in a concentration-dependent way, the K+-evoked overflow of [3H]GABA from both pars compacta and pars reticulata synaptosomes. Similarly to GABA, (-)-baclofen (0.3-10/LM) reduced the [3H]GABA overflow, being roughly equipotent to GABA in both nigral subregions. The ( + ) enantiomer of baclofen was ineffective. The overflow of [3 H]GABA was not consistently affected by muscimol in either the pars compacta or the pars reticulata. The effects of GABA were bicuculline- and picrotoxin-insensitive. However, the inhibition by GABA of the [ 3H]GABA overflow was antagonized by phaclofen. It is concluded that (a) GABA autoreceptors are sited on GABAergic nerve endings in both the pars compacta and pars reticulata of the rat substantia nigra; (b) these autoreceptors belong to the GABA B type. GABA autoreceptors; Substantia nigra (pars compacta); Substantia nigra (pars reticulata); GABA 8 receptor subtypes
1. Introduction In the last few years several reports have focussed on the existence and, particularly, the pharmacological characterization of G A B A autoreceptors in the mammalian CNS. The results of previous publications showing that the release of radioactive G A B A from rat brain cortical slices (Mitchell and Martin, 1978) or synaptosomes (Brennan et al., 1981) or from rat striatal slices (Kuriyama et al., 1984) was inhibited by muscimol
l Present address: Departamento de Farmacologia y Terapeutica, Facultad de Medicina y Odontologia, Universidad del Pais Vasco, 48940 Leioa (Vizcaya), Spain. Correspondence to: M. Raiteri, Istituto di Farmacologia e Farmacognosia, Universith degli Studi di Genova, Viale Cembrano 4, 16148 Genova, Italy.
in a bicuculline-sensitive manner have not been confirmed by recent investigations. In fact, the G A B A A receptor agonist could not inhibit G A B A release from rabbit striatal slices (Limberger et al., 1986), rat cortical synaptosomes (Pittaluga et al., 1987) or slices of rat cerebral cortex, hippocampus and striatum (Waldmeier et al., 1988) or from synaptosomes prepared from human cerebral cortex (Bonanno et al., 1989a). In all these systems the release of G A B A was instead inhibited by the GABAB receptor agonist, baclofen (Bowery et al., 1984), in a stereoselective manner. The findings that ( - ) - b a c l o f e n acted directly on GABAergic nerve terminals (Anderson and Mitchell, 1985; Pittaluga et al., 1987) and, in particular, that its inhibitory action was counteracted by phaclofen (Bonanno et al., 1988; 1989b), a G A B A B receptor antagonist (Kerr et al., 1987; Dutar and Nicoll,
0014-2999/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
138
1988), support strongly the idea that GABA autoreceptors in the mammalian CNS belong to the GABA B type. Of the recent publications, however, the one by Floran et al. (1988) seems to represent an exception. These authors studied the effects of GABA agonists and antagonists on the release of [3HIGABA stimulated by high K + in the pars compacta and in the pars reticulata of the rat substantia nigra. While in the latter nigral subregion the release of [3H]GABA was decreased by ( - ) - b a c l o f e n , muscimol-sensitive GABA A autoreceptors were reported to be present in the pars compacta. It should be noted, however, that the work was carried out with slices in which the inhibitory effects of GABAergic agonists may not reflect activation of autoreceptors sited on GABA terminals. Unfortunately, the sensitivity of the agonist effects to tetrodotoxin was not tested. These considerations, together with the earlier finding that the release of [3H]GABA from slices of rat substantia nigra, although decreased by muscimol, was paradoxically inhibited by bicuculline (Arbilla et al., 1979), prompted us to reinvestigate the problem of GABA autoreceptors in the substantia nigra by using a different experimental approach. In particular: (a) synaptosomes prepared from the pars compacta or the pars reticulata of the rat substantia nigra were used to study [3H]GABA release; (b) the natural agonist GABA was used as an agonist together with muscimol and baclofen; (c) the recently available GABAB receptor antagonist, phaclofen, was also used.
2. Materials and methods
2.1. Preparation of synaptosomes Adult male Sprague-Dawley rats (200-250 g) were used. The animals were killed by decapitation and the brain was removed rapidly. After the substantia nigra had been isolated the pars reticulata was separated from the pars compacta by microdissection. Crude synaptosomes were prepared according to the method of Gray and Whittaker (1962) with some modifications. Briefly, the
tissues pooled from two rats were homogenized with a glass-teflon tissue grinder (clearance 0.25 ram) in 3 ml of 0.32 M sucrose buffered at pH 7.4 with phosphate, The homogenate was centrifuged (5 min, 1 000 × g) to remove nuclei and debris, and synaptosomes were isolated from the supernatant by centrifugation at 12000 x g for 20 min. The synaptosomal pellet was then resuspended in a physiological medium of the following composition (mM): NaC1 125; KC1 3; MgSO4 1.2; CaC12 1.2; N a H 2 P O 4 1.0; N a H C O 3 22; glucose 10 (aeration with 95% 02 and 5% CO 2 at 37°C); pH 7.2-7.4. Protein was measured according to the method of Bradford (1976) with bovine serum albumin as standard.
2.2. Release experiments Synaptosomes obtained from the pars compacta or pars reticulata of the substantia nigra were incubated with 0.04 /~M [3H]GABA, corresponding to 10-12/~Ci/sample, in a rotatory water bath at 3 7 ° C for 15 min. After labeling, the synaptosomes were diluted with tracer-free medium; identical aliquots of the synaptosomal suspensions (ranging between 0.01-0.05 mg protein in the different experiments and for the two different subregions studied) were distributed on 0.65/~m Millipore filters placed at the bottom of a set of parallel superfusion chambers maintained at 3 7 ° C (Raiteri et al., 1974), and layered under moderate vacuum filtration. Superfusion was then started with standard medium, aerated with 95% 02 and 5% CO 2, at a rate of 0.6 m l / m i n and continued for 48 min. The system was allowed to equilibrate, for 36 min and then fractions were collected according to the following scheme: two 3-min samples (basal release) before and after one 6-min sample (evoked release). A 90-s period of depolarization with 9 mM KC1 was applied after the first fraction was collected; KC1 replaced an equimolar concentration of NaC1. GABA was added to the superfusion medium 8 min before depolarization and was present until the end of the experiment. Muscimol or ( - ) - b a c l o f e n was used concomitantly with the depolarizing stimulus. Bicuculline, picrotoxin, SR 95531 (2-(3'-carbethoxy2'-propenyl)-3-amino-6-paramethoxy-phenyl-pyr-
139 idazinium bromide) or phaclofen was added 10 min before GABA. S K & F 89976A (N-(4,4-diphenyl-3-butenyl)-nipecotic acid), a potent GABA uptake inhibitor (Yunger et al., 1984; Bonanno and Raiteri, 1987; Larsson et al., 1988), was present in the superfusion medium (final concentration 30 ~M) to minimize carrier-mediated exchange between intraterminal [3H]GABA and extrasynaptosomal GABA or muscimol. Aminooxyacetic acid (AOAA; final concentration 50 /~M) was present throughout the experiment to prevent GABA metabolism. The low-Ca 2+ medium, when used, was substituted for the standard medium 18 min before depolarization; this medium contained 0.2 mM Ca 2+ and 10 mM Mg 2+. The tritium collected in each fraction and that remaining in the synaptosomes at the end of the experiment was counted.
2.3. Calculation The amount of radioactivity released in each fraction is expressed as a percentage of the total tritium content of the synaptosomes at the start of the respective collection period. The depolarization-evoked overflow was estimated by subtracting the percentage tritium content of the basal release from the release evoked in the 6-min fraction collected during and after the depolarization pulse. The drug effects were evaluated as the ratio of the depolarization-evoked overflow calculated in the presence of the drugs vs. that calculated under control conditions. A two-tailed Student's t-test was used to compare mean values.
2.4. Drugs [3H]GABA (51-59 C i / m m o l ) was obtained from Amersham Radiochemical Centre (Buckinghamshire, U.K.); aminooxyacetic acid, (+)-bicuculline, picrotoxin and muscimol were purchased from Sigma Chemical Co. (S. Louis, MO, U.S.A.); GABA from Serva (Heidelberg, F.R.G.) and phaclofen from Tocris Neuramin (Essex, U.K.). The following drugs were a generous gift of the companies indicated: ( - ) - b a c l o f e n (Ciba Geigy, Basel, Switzerland); S K & F 89976A (Smith, Kline
and French, Welwyn, U.K.); SR 95531 (Sanofi, Bruxelles, Belgium).
3. Results
3.1. Characteristics of the depolarization-evoked release of [3H]GABA from substantia nigra synaptosomes Superfused synaptosomes prepared from the pars compacta and pars reticulata of the rat substantia nigra were exposed to 9 mM KC1. The depolarization-evoked overflow of tritium amounted to 0.86 _+ 0.03% (n = 28) in the pars compacta and to 1.58 + 0.09% (n = 28) in the pars reticulata. Under identical experimental conditions the [3H]GABA overflow from cerebral cortex synaptosomes amounted to about 1.9% of the total tritium content (Bonanno et al., 1989b). When the concentration of Ca 2+ in the superfusion medium was lowered from 1.2 mM to 0.2 mM and that of Mg 2+ was raised to 10 raM, the K+-evoked tritium overflow from both pars compacta and pars reticulata synaptosomes was almost abolished (data not shown). The strict calcium dependence of the depolarization-evoked tritium overflow, the presence in the superfusion medium of 50 /~M of the GABA transaminase inhibitor, aminooxyacetic acid, the characteristics of the superfusion technique used (see Raiteri and Levi, 1978) together with data obtained earlier in various laboratories (see for a review Levi, 1984) make it likely that the radioactivity released upon depolarization consisted largely of authentic [3H]GABA. Therefore, we refer to the K+-evoked tritium overflow as K +evoked [3 H]GABA release in the remainder of the text.
3.2. Effects of GABA receptor agonists on the K +evoked [-¢H]GABA release from nigral synaptosomes In a first set of experiments we examined the effects of exogenous GABA on the release of radiolabeled GABA provided by exposure of synaptosomes prepared from the pars compacta or
140 pars compact a
pars ret icu lat a
6O .o_ J= 50
E_
O
;~ 40 LU O3
<
w ..J w II
q-
30 20
< II1 <
T
"i"
0.3
1
GABA
10
0.3
CONCENTRATION
10 (pM)
Fig. 1. Concentration-dependent inhibition by GABA of the K ÷ (9 mM)-evoked release of [3H]GABA from synaptosomes of the pars compacta and pars reticulata of the rat substantia nigra. Synaptosomes were prelabeled with 0.04/~ M [3H]GABA and superfused (see Methods). After 38 rain of superfusion, the synaptosomes were depolarized with 9 mM KC1. Exogenous GABA was added to the superfusion medium 8 min before depolarization. The [3H]GABA overflow was calculated by subtracting the percentage of [3H]GABA found in the two 3-min fractions corresponding to the basal release from the radioactivity in the 6-min fraction collected during and after the stimulation period. The basal efflux in the fraction preceeding the stimulation amounted to 0.98 5:0.05 nCi/min, corresponding to 0.51 5:0.01% (n = 28), in the pars compacta and to 1.38+0.06 nCi/min, corresponding to 0.545:0.02% (n = 27), in the pars reticulata. Results are means 5: S.E.M. of 3-10 experiments run in triplicate.
the pars reticulata of the rat substantia nigra to high K ÷. Exogenous GABA was added to the superfusion medium in the presence of 30 # M of the GABA uptake inhibitor, S K & F 89976A. This compound effectively prevented GABA homoexchange since the highest concentration of GABA used (10 ~M) increased the basal outflow of [3H]GABA by less than 10%. In the absence of GABA, S K & F 89976A did not affect [3H]GABA release (data not shown). Figure 1 shows that GABA (0.3-10/LM) concentration dependently inhibited the K+-evoked release of [3H]GABA from both pars compacta and pars reticulata synaptosomes.
Figure 2 illustrates the effects of the enantiomers of baclofen and of muscimol on the release of [3H]GABA. In both the pars compacta and reticulata ( - ) - b a c l o f e n (0.3-10 t~M) reduced, in a concentration-dependent manner, the release of [3 H]GABA evoked by depolarization. The GABA receptor agonist was at least as effective as GABA. The effects seen in the pars reticulata appeared to be slightly more pronounced than those seen in the compacta although the differences were not statistically significant. The effect of baclofen was stereoselective since the ( + ) enantiomer was almost inactive at 10/~M (fig. 2). In contrast to ( - ) - b a c l o f e n , the GABA A agonist, muscimol (10-100 /~M), did not inhibit the K+-evoked release of [3H]GABA from pars compacta and pars reticulata synaptosomes. The effects observed were very modest and were not concentration-dependent (fig. 2).
3.3. Antagonism of the GABA-induced inhibition of [3H]GABA release As shown in fig. 3, the inhibition of [3H]GABA release caused by GABA in both nigral subregions was not counteracted by the GABA A receptor antagonists, bicuculline or picrotoxin. Also the more recent GABA A receptor antagonist, SR 95531 (Wermuth and Bizirre, 1986), was ineffective when tested at 10 /~M against 1 ~tM GABA (result not shown). In contrast, the effect of 1 #M GABA was prevented by 100/~M of the GABA B receptor antagonist, phaclofen (fig. 3). In the absence of GABA, the GABA receptor antagonists used had no or only very modest effects on [3H]GABA release (not shown).
4. Discussion
As mentioned in the Introduction, several recent publications have reported that the release of GABA is insensitive to muscimol but can be inhibited by ( - ) - b a c l o f e n . These findings were obtained in slices of rabbit caudate nucleus (Limberger et al., 1986), rat cerebral cortex, hippocampus or striatum (Waldmeier et al., 1988; Raiteri et al., 1989) as well as in synaptosomes of rat or
141 pars compact a
70-
pars
reticulata
60o
r
5C
.E O ae
40
T
uJ u~
3C
7-
...1 213 ii) 0 "r
7"
1£
03
1
10
-
-
0.3
10
-
1
10 10
-
-
10
DRUG
-
100
CONCENTRATION
10
100
i-)baclofen (+)baclofen muscimol
(pM)
Fig. 2. Effects of (-)-baclofen, (+)-baclofen and muscimol on the K+-evoked release of [3H]GABA from nigra compacta and reticulata synaptosomes. The agonists were added to the superfusion medium at the same time as the synaptosomes were depolarized with K +. The experimental details are as in fig. ]. Results are expressed as means+S.E.M, of three to six experiments done in triplicate.
pars corrlpacta
pars retlculata
60 A
.o
T
50
~5 40 a~ LU (13 w .J w cc
T
II
30
20
00
0
10
L
0 1
1 10
1
-
100
10
10 10
10 10
1
1 10
1
CONCENTRATION
10 10
10 10
100 DRUG
10
GABA bicueulline picrotoxin phaclofen
(IJM)
Fig. 3. Antagonism of the GABA-induced inhibition of the [3H]GABA overflow from nigra compacta and reticulata synaptosomes. The experimental details are as in fig. 1. Exogenous G A B A was added to the superfusion m e d i u m 8 rnin before depolarization and the antagonists 10 min before GABA. The results are means -t- S.E.M. of three to six experiments done in triplicate. * P < 0.001 vs. 1 # M G A B A with the two-tailed Student's t-test.
142 human cerebral cortex (Pittaluga et al., 1987; Bonanno et al., 1989a). Thus the present evidence seems to favor the idea that G A B A autoreceptors in various areas of the mammalian CNS are of the G A B A B type. However, the situation appears to be rather peculiar in the substantia nigra. In this area Arbilla et al. (1979) found that muscimol inhibited the K+-evoked release of [3H]GABA; however, bicuculline, instead of antagonizing muscimol, paradoxically inhibited the overflow of the amino acid. ( - ) - B a c l o f e n was not tested. On the other hand, Waldmeier et al. (1988) could not analyze the effects of muscimol or ( - ) - b a c l o f e n on the release of G A B A because G A B A overflow could not be detected when slices of the substantia nigra were exposed to electrical stimuli which induced the release of G A B A from other CNS areas. By using high K + as a depolarizing agent, Floran et al. (1988) did observe a calcium-dependent release of [3H]GABA from slices of rat substantia nigra. Curiously, this release was inhibited by muscimol but not by ( - ) - b a c l o f e n in the pars compacta, while it was inhibited by ( - ) - b a c l o f e n but not by muscimol in the nigra reticulata, suggesting that the two nigral subregions contain pharmacologically different G A B A autoreceptors. The reasons for the above discrepancies may be manifold. For instance, when studying autoreceptors it is important to use the natural transmitter as an agonist. Unfortunately, the use of G A B A as an autoreceptor agonist has been hindered due to the lack of a suitable carrier blocker able to prevent homoexchange between external G A B A and its intraterminal radioactive counterpart. This problem now appears to be solved and G A B A homoexchange can be effectively prevented by S K & F 89976A (Pittaluga et al., 1987), a potent and selective GABA uptake blocker (Yunger et al., 1984). The results of fig. 1 show that, in the presence of S K & F 89976A, G A B A inhibited the K+-evoked release of [3H]GABA from nigral synaptosomes and displayed a similar potency in the pars compacta and pars reticulata. The effect of the natural transmitter on its own release together with the effects of selective receptor antagonists suggest the presence of a receptor-mediated autoregulatory mechanism.
Another critical aspect is that of the exact localization of the receptor involved. Brain slices have often been used in studies of neurotransmitter release because of their anatomical integrity. However, synaptosomes in superfusion represent the preparation of choice when the localization of a release-regulating receptor has to be determined. In fact, modulation of the release of a transmitter by a given agonist could occur through neuronal circuits which remain intact in brain slices so that the receptors involved need not be sited on the releasing nerve terminals. These neuronal circuits are destroyed during the preparation of synaptosomes. Moreover, the release technique used in our laboratory (superfusion of a thin layer of synaptosomes) does not permit indirect interactions to take place. Therefore, when an agonist modulates the release of a transmitter in superfused synaptosomes, it is likely to occur by a direct action on the releasing particles. The results of the present study show that the G A B A A receptor agonist, muscimol, displayed inconsistent effects on the release of [3 H]GABA not only in pars reticulata synaptosomes but also in nerve endings prepared from the pars compacta of the substantia nigra. The technical considerations discussed above together with the finding that the G A B A A receptor antagonists bicuculline~ picrotoxin and SR 95531 did not antagonize the G A B A inhibition of [3H]GABA overflow tend to exclude the possibility that G A B A A receptors are present on GABAergic nerve terminals, as proposed by Floran et al. (1988). Also the possibility that G A B A A receptors desensitize rapidly (Adams and Browm 1975) seems unlikely. In fact, we added muscimol to the superfusion medium concomitantly with the depolarizing stimulus in our experiments. Furthermore, G A B A decreased its own release. In contrast to muscimol, ( - ) - b a c l o f e n stereoselectively inhibited the [3H]GABA overflow in both parts of the substantia nigra. The involvement of G A B A B receptors is strengthened by the finding that the inhibition by G A B A of [3H]GABA overflow was counteracted by phaclofen, a novel G A B A B receptor antagonist (Kerr et al., 1987; Dutar and NicoU, 1988). Thus the present results with synaptosomes are in keeping with those ob-
143
tained with slices by Floran et al. (1988) only when the release of [3H]GABA in the pars reticulata is considered. As to the pars compacta, GABAergic nerve endings appear to possess GABA B instead of GABA A autoreceptors. As a possible explanation one could hypothesize that the muscimol-induced inhibition of [3H]GABA release in slices of the pars compacta observed by Floran et al. (1988) occurs through a more or less complex intra-compacta circuit involving GABAA receptors that are not sited on the GABAergic nerve endings. In other words, in the nigra compacta autoinhibition of GABA release might occur through a dual mechanism, one mediated directly by GABA B autoreceptors sited on GABAergic terminals and the other involving the activation of GABA A receptors present on other structures. The GABAA-mediated mechanism would prevail in the in vitro experimental conditions used by Floran et al. (1988) while the relative importance of the two mechanisms, if present in the living brain, remains to be established. In conclusion, several transmitters are released during stimulation of a brain slice, and several, sometimes complex, interactions may occur. Thus it is difficult to localize receptors involved in the auto- or the heteroregulation of the release of one transmitter. The localization of such a receptor is much easier when using a thin layer of synaptosomes in superfusion, a system in which only release-regulating receptors localized on the particle releasing the transmitter in question can be involved. Our results indicate that GABAergic nerve terminals of both the pars compacta and pars reticulata of the substantia nigra possess autoreceptors. These receptors are activated by GABA itself and, stereoselectively, by ( - ) - b a c l o fen, but not by muscimol. The effect of GABA is phaclofen-sensitive but bicuculline-insensitive. Thus, in analogy to what has been found in several other areas of the rat CNS, the GABA autoreceptors sited on GABAergic terminals in the substantia nigra appear to belong to the G A B A B type.
Acknowledgements This work was supported by grants from the Italian Ministry of Education and from the Italian C.N.R. The authors wish to thank Mrs. Maura Agate for her skilful secretarial assistance.
References Adams, PR. and D.A. Brown, 1975, Actions of y-aminobutyric acid on sympathetic ganglion cells, J. Physiol. (London) 250, 85. Anderson, R.A. and R. Mitchell, 1985, Evidence for GABA B autoreceptors in median eminence, European J. Pharmacol. 118, 355. Arbilla, S., L. Kamal and S.Z. Langer, 1979, Presynaptic GABA autoreceptors on GABAergic nerve endings of the rat substantia nigra, European J. Pharmacol. 57. 211. Bonanno, G., P. Cavazzani, G.C. Andrioli, D. Asaro, G. Pellegrini and M. Raiteri, 1989a, Release-regulating autoreceptors of the GABAB-type in human cerebral cortex, Br. J. Pharmacol. 96, 341. Bonanno, G., G. Fontana and M. Raiteri, 1988, Phaclofen antagonizes GABA at autoreceptors regulating release in rat cerebral cortex, European J. Pharmacol. 154, 223. Bonanno, G., G. Pellegrini, D. Asaro, G. Fontana and M. Raiteri, 1989b, GABAB autoreceptors in rat cortex synaptosomes: response under different depolarizing and ionic conditions, European J. Pharmacol. Mol. Pharmacol. Sect. 172, 41. Bonanno, G. and M. Raiteri, 1987, Coexistence of carriers for dopamine and GABA uptake on a same nerve terminal in the rat brain, Br. J. Pharmacol. 91, 237. Bowery, N.G., G.W. Price, A.L. Hudson, D.R. Hill, G.P. Wilkin and M.J. Turnbull, 1984, GABA receptor multiplicity, Visualization of different receptor types in the mammalian CNS, Neuropharmacotogy 23, 219. Bradford, M.M, 1976, Rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72, 248. Brennan, M.J.W., R.C. Cantrill, M. Oldfield and P. Krogsgaard-Larsen, 1981, Inhibition of "r-aminobutyric acid release by ,¢-aminobutyric acid agonist drugs. Pharmacology of the y-aminobutyric acid autoreceptor, Mol. Pharmacol. 19, 27. Dutar, P. and R.A. Nicoll, 1988, A physiological role for GABA B receptors in the central nervous system, Nature 332, 156. Floran, B., I. Silva, C. Nava and J. Aceves, 1988, Presynaptic modulation of the release of GABA by GABA A receptors
144 in pars compacta and by GABA B receptors in pars reticulata of the rat substantia nigra, European J. Pharmacol. 150, 277. Gray, E.G. and V.P. Whittaker, 1962, The isolation of nerve endings from brain: an electron microscopic study of cell fragments derived by homogenization and centrifugation, J. Anat. 96, 79. Kerr, D.I.B., J. Ong, R.H. Prager, B.D. Gynther and D.R. Curtis, 1987, Phaclofen: a peripheral and central baclofen antagonist, Brain Res. 405, 150. Kuriyama, K., K. Kanmori, J. Taguchi and Y. Yoneda, 1984, Stress-induced enhancement of suppression of [3H]GABA release from striatal slices by presynaptic autoreceptor, J. Neurochem. 42, 943. Larsson, O.M., E. Falch, P. Krogsgaard-Larsen and A. Schousboe, 1988, Kinetic characterization of inhibition of y-aminobutyric acid uptake into cultured neurons and astrocytes by 4,4-diphenyl-3-butenyl derivatives of nipecotic acid and guvacine, J. Neurochem. 50, 818. Levi, G., 1984, Release of putative transmitter amino acids, in: Handbook of Neurochemistry, Vol. 6, ed. A. Lajtha (Plenum Publishing Co., New York) p. 431. Limberger, N., L. Sp~ith and K. Starke, 1986, A search for receptors modulating the release of 3,-[3H]aminobutyric acid in rabbit caudate nucleus slices, J. Neurochem. 46, 1109. Mitchell, P.R. and I.L. Martin, 1978, Is GABA release modulated by presynaptic receptors?, Nature 274, 904.
Pittaluga, A., D. Asaro, G. Pellegrini and M. Raiteri, 1987, Studies on 3H-GABA and endogenous GABA release in rat cerebral cortex suggest the presence of autoreceptors of the GABA 8 type, European J. Pharmacol. 144, 45. Raiteri, M. F. Angelini and G. Levi, 1974, A simple apparatus for studying the release of neurotransmitters from synaptosomes, European J. Pharmacol. 25, 411. Raiteri, M., G. Bonanno and E. Fedele, 1989, Release of [3H]'t-aminobutyric acid from electrically-stimulated rat cortical slices and its modulation by GABA B autoreceptors, J. Pharmacol. Exp. Ther. (in press). Raiteri, M. and G. Levi, 1978, Release mechanisms for catecholamines and serotonin in synaptosomes, in: Reviews of Neuroscience, Vol. 3, eds. S. Ehrenpreis and I. Kopin (Raven Press, New York) p. 77. Waldmeier, P.C., P. Wicki, J.-J. Feldtrauer and P.A. Baumann, 1988, Potential involvement of a baclofen-sensitive autoreceptor in the modulation of the release of endogenous GABA from rat brain slices in vitro, Naunyn-Schmiedeb. Arch. Pharmacol. 337, 289. Wermuth, C.G. and K. Bizi~re, 1986, Pyridazinyl-GABA derivatives: a new class of synthetic GABA A antagonists, Trends Pharmacol. Sci. 7, 421. Yunger, L.M., P.J. Fowler, P. Zarevics and P.E. Setler, 1984, Novel inhibitors of -/-aminobutyric acid (GABA) uptake: anticonvulsant actions in rats and mice, J. Pharmacol. Exp. Ther. 228, 109.