j. Neural Transmission 64, 163-172 (1985) dourna/o? 1 V ~ Trim.s-miss/on �9 by Springer-Verlag 1986

GABA-Noradrenergic Interaction: Evidence for Differential Sites of Action

for GABA-A and GABA-B Receptors

P. D. Suzdak and G. Gianutsos

Section of Pharmacology and Toxicology, School of Pharmacy, The University of Connecticut, Storrs, Connecticut, U.S.A.

Summary

Treatment of mice with DSP4 (a neurotoxin that abolishes the presynap- tic noradrenergic neuron; Doole 22 etal., 1983) resulted in: (A) a decrease in the B ~ for the low affinity GABA-B receptor site in the cerebral cortex and hippocampus, whereas the B .... for the high affinity GABA-B receptor site was unaffected; (B) a greater potentiation of norepinephrine stimulated adenylate cyclase by baclofen in cerebral cortex slices; and (C) a decrease in the B .... for both the high and tow affinity GABA-A receptor sites in the cerebral cortex and hippocampus. These data, coupled with previous work from our laboratory, suggest that the GABA-B receptor may be associated with both the noradrenergic nerve terminal and the post-synaptic neuron receiving noradrenergic input, whereas the GABA-A receptor may be associated with the noradrenergic nerve terminal. These data further suggest a functional coupling between the noradrenergic and GABA-ergic systems.

Introduction

Biochemical evidence (Lloyd et al., 1983; Karbon et aL, 1983, 1984 a; Suzdak and Gianutsos, 1985 a, 1985 b) suggests a link between the noradrenergic and GABA-ergic systems, which may play a role in the mechanism of action of antidepressant drugs and in the under- lying etiology of depression. The presence of two sub-populations of GABA receptors within the CNS (see Bowery et aL, 1980, for review)

164 P. D. Suzdak and G. G~anutsos

raises the question of the function of each of these receptor subtypes in the coupling between the GABA-ergic and noradrenergic systems. Previous evidence suggests that the GABA-A and GABA-B receptors exert differential effects on the noradrenergic system. For example: (1) In-vivo administration of GABA-A receptor agonists has been shown to increase the release of norepinephrine, whereas GABA-B receptor agonists decrease the release of norepinephrine (Anden and Wachtel, 1977; Suzdak and Gianutsos, 1985 b); (2) Chronic administra- tion of either the GABA-A receptor agonist THIP or the GABA-B receptor agonist baclofen decreased Beta-adrenergic receptor binding (Suzdak and Gianutsos, 1985 a). However, pretreatment with the presynaptic noradrenergic neurotoxin DSP4 (Dooley et aL, 1983) prevented the decrease in Beta-adrenergic receptor binding produced by chronic treatment with THIP, but did not alter Beta-adrenergic receptor down regulation produced by baclofen (Suzdak and Gianut- sos, 1985 b); and (3) Baclofen, but not GABA-A receptor agonists, has been shown to potentiate norepinephrine stimulated adenylate cyclase accumulation (Karbon et al., 1984 a).

This purpose of this investigation was to further examine how the GABA-A and GABA-B receptors are linked to the noradrenergic system. The effect of lesioning the noradrenergic neuron with DSP4 on GABA-A and GABA-B receptor binding, as well as on baclofen potentiated adenylate cyclase activity was examined, in order to better understand the mechanism of coupling between the GABA- ergic and noradrenergic systems.

Materials and Methods

Drug Treatments

Male CD-1 mice (Charles River Farms, Wilmington, MA) were used in all experiments. Mice received either DSP4 (N-[2-chloroethyl]-N-ethyl-2- bromobenzylamine hydrochloride, Research Biochemicals Inc., Wayland, MA: 50 mg/kg i.p.) or saline (10 ml/kg i.p.) and were sacrificed 17 days later. The dose of DSP4 used has been previously shown to selectively deplete norepinephrine from mouse CNS (Dooley et al., 1983; Suzdak and Gianutsos, 1985 b).

GABA-A Receptor Binding

GABA-A receptor binding was measured using 3H-GABA as the ligand by a modification of the method of Enna and Snyder (1977). Briefly, the cerebral cortex of hippocampus from 10 mice were pooled and homog- enized in 0.32 M sucrose and centrifuged. The resulting supernatant was recentrifuged, and the pellet was resuspended by the use of a Polytron at a

GABA-Noradrenergic Interaction 16.5

setting of 5, then washed twic? in ice-cold water. The pellet was then resuspended in 50 mM Tris-citral~e buffer (pH 7.1) containing 0.05% Triton- X 100, and incubated for 30 min at 37 ~ The homogenate was then centri- fuged, and the pellet was frozen for at least three days. The pellet was then washed twice more in Tris buffer. After the final wash, the pellet was resuspended in Tris buffer for the binding assay. Specific binding was determined by the amount of 3H-GABA displaced by 100/aM unlabeled GABA. Varying concentrations (10-9 to 10-5 M) of unlabeled GABA and 10 nM 3H-GABA (Amersham; specific activity 54.1 ci/mmole) were added to each tube. The tubes were incubated at 4 ~ for 20 min, then centrifuged. The resulting pellet was quickly rinsed with 2 ml of Tris buffer, and the inside of the binding tube was wiped with a cotton swab. The pellet was dissolved in Protosol, and quantified by liquid scintillation spectrometry. Protein was determined by the method of Bradford (1976). Nonspecific binding was less than 10% of the total counts. GABA-A receptor binding data was analyzed by the method of Rosenthal (1967), assuming the pre- sence of two binding sites.

GABA-B Receptor Binding

GABA-B receptor binding was measured using 3H-GABA as the ligand by a modification of the method of Karbon and Enna (1984 b). Briefly, the cerebral cortex of hippocampus from 10 mice were pooled and homog- enized in 50 mM Tris buffer and centrifuged. The resulting supernatent was recentrifuged and the pellet resuspended by use of a Polytron at speed 5 in Tris buffer containing 0.03% Triton X-100, and incubated at 37~ for 30 minutes. The homogenate was centrifuged and resuspended in 50 mM Tris buffer. After centrifugation, the pellet was again washed with Tris buffer. The pellet was resuspended in Tris buffer for the binding assay. Specific binding was determined by the amount of BH-GABA displaced by 100/aM baclofen. Varying concentrations of unlabeled (10-9 to 10-5 M) GABA and 20 nM 3H-GABA (Amersham; specific activity 54.1 ci/mmole) were added to tubes containing 2.5 mM GAG12 and 100/aM isoguvacine (to inhibit 3H-GABA binding to GABA-A binding sites). The tubes were incubated at 4 ~ for 20 rain, followed by centrifugation. The resulting pellet was quickly rinsed with 2 ml of Tris buffer, and the inside of the binding tube was wiped with a cotton swab. The pellet was dissolved in Protosol, and quantified by liquid scintillation spectrometry. Protein was determined by the method of Bradford (1976). Nonspecific binding was less than 10% of the total counts. GABA-B receptor binding data was analyzed by the method of Rosenthal (1967), assuming the presence of two binding sites. Each point represents the mean + the standard error obtained from three separate experiments run in triplicate.

Cyclic A M P Accumulation

The in-vitro accumulation of cyclic AMP was measured by a modifica- tion of the method of Sulser and Mobley (1981). Briefly, mice were sacrificed

166 P. D. Suzdak and G. Gianutsos

by decapitation and the cerebral cortex was dissected out and placed on a McIlwain tissue chopper and sliced into 0.3 mm sections. The slices were then preincubated in Krebs-Ringer bicarbonate buffer for 30 min under constant aeration with OffCO2 (95/5%). The incubation buffer was changed and the slices were incubated for an additional 10 min followed by the addition of either buffer, 100/aM norepinephrine, 100/aM baclofen or 100/aM norepinephrine and 100/aM baclofen to the system. The slices were then incubated for an additional 15 min followed by freezing with liquid nitrogen and subsequent homogenization in 0.3 N perchloric acid, and were then centrifuged. Cyclic AMP was measured in the resulting supernatant by a commercially available radioimmunoassay kit (New England Nuclear Corp.). Protein was determined by the method of Bradford (1976). Each point represents the mean_the standard error obtained from twelve separate mice.

Statistical Analysis

Comparisons were made between controls and experimental mice treated under the same conditions. Means and standard errors were calcu- lated by averaging the values obtained for the binding parameters from each separate experiment using different groups of mice. Student's t-test or two- way ANOVA was used to analyze the data, with p < 0.05 chosen as the level of significance.

Results

3H-GABA Receptor Binding

Previously, treatment with the presynaptic noradrenergic neuro- toxin DSP4 (Dooley et al., 1983), at a dose of 50 mg/kg i.p. in mice has been shown to result in a selective depletion ofnorepinephrine (80% decrease), without any effect on either dopamine or serotonin levels in the cerebral cortex and hippocampus (Suzdak and Gianutsos, 1985 b), Since GABA-A and GABA-B agonists appear to have differ- ent sites of action on the noradrenergic system, this led us to investi- gate the effect of DSP4 treatment on GABA-A and GABA-B receptor binding. As shown in Table t, treatment with DSP4 resulted in a significant decrease in the B .... of both the high and low affinity GABA-A receptor binding site in the cerebral cortex (79 and 80% of control, respectively), and hippocampus (80% of controi, for both binding sites). There was no apparent change in the Kd values as the result o f DSP4 treatment. In contrast, DSP 4 treatment had no effect on the B .... for the high affinity GABA-B receptor binding site (Table 2) in either the cerebral cortex or hippocampus (93 and 95% of control, respectively). DSP 4 treatment did, however, result in a significant decrease in Bmax for the low affinity GABA-B receptor

GABA-Noradrenergic Interaction 167

Table 1. The effect of DSP41 treatment on GABA-A receptor binding

High Affinity Low Affinity

Treatment Br~.x 2 Kd 3 Bmax 2 Kd 3

Cerebral Cortex: Control 680 • 10 31 -}- 5 2580 • 20 400 • 31 DSP4 540• 12" 30-4-4 2060___ 18" 476-t-43

Hippocampus: Control 430 • 14 20 • 3 2280 __ 31 350 __ 36 DSP4 340_+ 9* 18• 1 8 1 0 - - + 2 5 " 295•

i Mice were txealed with DSP4 (50 mg/kg i.p.), and sacrificed 14 days laler. 2 E,,~ is expressed as fmole/mg protein. 3 Kd is expressed as nM. * Represents values s[gnificantty (p<0.05) different from the corresponding

control group, Each point represents the mean • the standard error obtained from three separate experiments run in triplicate.

Table 2. The effect of DSP41 treatment on GABA-B receptor binding

High Affinity Low Affinity

Treatment Bm~x 2 Kd 3 Bmax 2 Kd 3

Cerebral Cortex: Control 580 4- 12 30 -t- 3 2680 _ 30 476 • 50 DSP4 555 • 18 31 • 5 2006 • 41" 400 • 37

Hippocampus: Control 490 • 15 46 • 7 2880 • 36 490 • 49 DSP4 460• I9 40• 2061 _.39* 444 • 43

1 Mice were treated with DSP4 (50 mg/kg i.p.), and sacrificed 14 days later. 2 Bmax is expressed as fmole/mg protein. 3 Kd is expressed as nM. * Represents va]ues significantly (p<0.05) different from the corresponding

control group, Each point represents the mean + the standard error obtained from three separate experiments run in triplicate.

b i n d i n g site in b o t h the cerebral cor tex a n d h i p p o c a m p u s (75 a n d 72% o f con t ro l , respect ively) . The re was n o a p p a r e n t c h a n g e in Kd values as the ~esutt o f DS?4 t r e a t m e m ,

Cyclic A M P Accumulation

DSP4 t r e a t m e n t p roduces a n increase in be ta -adrenerg ic receptor b i n d i n g (Dooley et aL, 1983; Suzdak a n d Gianutsos, 1985b) , p r e s u m a b l y due to d e p l e t i o n o f n o r e p i n e p h r i n e f rom the no r -

168 P. D. Suzdak and G. Gianutsos

adrenergic nerve terminal (Dooley et al., 1983). Since, the GABA-B receptor agonist baclofen has been shown to potentiate norepine- phrine stimulated cyclic AMP accumulation in cerebral cortex slices (Karbon et al., 1984), and to prevent the Beta-adrenergic receptor supersensitivity induced by DSP 4 (Suzdak and Gianutsos, 1985 b), we investigated the effect of DSP4 lesioning on baclofen potentiated norepinephrine stimulated cyclic AMP accumulation. The addition of 100/~M norepinephrine to control slices resulted in an increase in cyclic AMP accumulation (222% of control). This effect was greater in slices obtained from DSP4 lesioned mice (274% of control). The in-vitro addition of 100/~M baclofen to slices from either the control or DSP4 treated mice resulted in a non-significant increase in cyclic AMP accumulation (124 and 132% of control, respectively). Thus, the effect of baclofen on norepinephrine stimulated cyclic AMP accumulation does not appear to be a direct effect of baclofen on adenylate cyclase, but due to a potentiation of norepinephrine stimulated cyclic AMP accumulation. The addition of 100/~M baclofen resulted in a significant increase (163% of control) in norepinephrine stimulated cyclic AMP accumulation in slices from control animals, and a facilitation of this effect (183 % of control) in cerebral cortex slices from animals lesioned with DSP 4. This represents the amount of cyclic AMP accumulated in addition to that accumulated in the presence of norepinephrine alone. Thus, lesioning with DSP4 resulted in a greater effect of baclofen on nor- epinephrine stimulated cyclic AMP accumulation despite the fact that the lesion itself potentiated the cyclic AMP response. The basal levels of cyclic AMP accumulation were not significantly altered by DSP 4 treatment.

Table 3. The effect of DSP4 ~ treatment on norepinephrine (N.E.) and baclofen stimulated cyclic A M P accumulation

Cyclic AMP (pmole/mg protein/10 minutes)

Treatment Vehicle 100/iM N.E. 100/~M Baclofen N.E. + Baclofen

Control 55.31_+ 8.9 123.0_+12.9 69 .0_+6.9 201.0_+14.0 DSP4 60.03_+10.0 181.0--10.8 ~ 79.3___8.4 332.0 q- 10.W':-

1 Mice were treated with DSP 4 (50 mg/kg i.p.) and sacrificed 14 days later. ~ Represents a significant (p < 0.05) increase in N.E. stimulated cyclic AMP

accumulation as compared to the corresponding control group. *~' Represents a significant (p < 0.05) increase in baclofen potentiated N.E. stim-

ulated cyclic AMP accumulation as compared to the identical treatment in the control group. Each point represents the mean _+ the standard error obtained from twelve separate mice.

GABA-Noradrenergic Interaction 169

Discussion

The GABA-A and GABA-B receptors have been shown to mediate different physiological effects within the CNS (Dunlap, 1981[; Desarmenian et al., 1982; Guidotti et al., 1983; Hill et aL, 1981). This investigation focused on further elucidating the mechanism of coupling of GABA-A and GABA-B receptors to the noradrenergic system. This interaction may play a role in the mechanism of action of antidepressant agents, as well as the underlying etiology of affec- tive disorders, since norepinephrine (Sulser and Mobley, 1981) and GABA (Suzdak and Gianutsos, 1985 a) may be involved in the mechanism of action of antidepressant agents.

DSP4 treatment resulted in a significant decrease in the gmax for the low affinity GABA-B receptor binding site in both the cerebral cortex and hippocampus. These data are in agreement with Karbon et al. (1983) who demonstrated a decrease in the Bmax for the low affinity GABA-B receptor in the cerebral cortex after a 6- hydroxy-dopamine induced lesion of the dorsal noradrenergic bundle. These data suggest that the low affinity component of the GABA-B receptor may be associated with the noradrenergic nerve terminal. Previously, baclofen has been shown to decrease the release of norepinephrine (Suzdak and Gianutsos, 1985 b), an effect which may also be associated with the noradrenergic nerve terminal. On the other hand, chronic administration of baclofen has been shown to decrease noradrenergic receptor binding (Suzdak and Gianutsos, 1985 a), an effect which is not abolished by pretreatment with DSP4, suggesting that GABA-B receptors may also be associated with the post-synaptic neuron receiving noradrenergic innervation. DS]D 4 treatment which resulted in beta-adrenergic receptor supersensitivity (as measured by both norepinephrine stimulated cyclic AMP accumu- lation and beta-adrenergic receptor binding (Suzdak and Gianutsos, 1985 b), also resulted in an increase in the potentiation of nor- epinephrine stimulated cyclic AMP accumulation produced by baclofen. These results suggest that depletion of norepinephrine by DSP4 also results in a greater biochemical effect mediated by the GABA-B receptors linked with adenylate cyclase. This apparent GABA-B mediated supersensitivity of the cyclic AMP generating system may be explained in one of two ways: (1) There may be an increase in the number of GABA-B receptors coupled to adenylate cyclase. Although there was no increase in the Bmax for GABA-B receptor binding following DSP4 treatment, such a potential increase in this subpopulation of GABA-B receptors may be offset by an even larger decrease in the Bmax of the GABA-B receptors associated with

170 P. D. Suzdak and G. Gianutsos

the noradrenergic nerve terminal; or (2) The GABA-B receptors linked with adenylate cyclase may not undergo a change in B . . . . but instead thei r coupling with adenylate cyclase may become facilitated. These data suggest that stimulation of the GABA-B receptor may act by magnifying the hormonal activation of adenylate cyclase. A similar situation is seen after treatment with a low dose of forskolin, which has little effect on cyclic AMP generation by itself, but can greatly magnify the hormonal activation of adenylate cyclase (Seamon and Daly, 1983). Previously, Karbon et al. (1984 a) have demonstrated that the in-vitro addition of baclofen significantly increases cyclic AMP accumulation in cerebral cortex slices, while in our system the in-vitro addition of baclofen resulted in a non-signifi- cant 20 to 30% increase in cyclic AMP accumulation. Thus, the possibiIity exists that stimulation of the GABA-B receptor may affect the cyclic AMP generating system independent of other hormone- receptor stimulation.

The ability of DSP4 to decrease both the high and low affinity component of GABA-A receptor binding suggests that the GABA-A receptor may be associated with the noradrenergic nerve terminal. Stimulation of the GABA-A receptor has been shown to increase the release of norepinephrine (Suzdak and Gianutsos, 1985 b), and chronic treatment with the GABA-A agonist THIP has also been shown to decrease beta-adrenergic receptor binding (Suzdak and Gianutsos, 1985 a), an effect which is abolished by pretreatment with DSP4 (Suzdak and Gianutsos, 1985 b). These results are consistant with the suggestion that GABA-A receptors are on the noradrenergic nerve terminal. The lack of effect of 6-hydroxy-dopamine on GABA-A receptor binding (Karbon et aL, 1983), can not be explained at this time.

In summary, these data suggest that the GABA-B receptor may be associated with both the noradrenergic nerve terminal (mediating the inhibition of norepinephrine release), as well as the post-synaptic neuron receiving noradrenergic input (associated with adenylate cyclase). On the other hand, the GABA-A receptor may be associated with either the noradrenergic nerve terminal (mediating the release of norepinephrine). These data further demonstrate a coupling between the noradrenergic and GABA-ergic systems, which may pIay a role in the mechanism of action of antidepressant agents.

GABA-Noradrenergic Interaction 171

A c k n o w l e d g e m e n t s

P.D.S. is a fellow of the American Foundation For Pharmaceutical Education. We are grateful for the generous gift of baclofen (Ciba-Geigy, Summit, NO. ). This research was partially supported by NIH grant MH39287.

References

Anden, N., Wachtel, H.: Biochemical effects of baclofen on dopamine and noradrenaline in the rat brain. Acta Pharmacol. Toxicol. 40, 310-320 (1977).

Bowery, N. G., Doble, A,, Hill, D. R., Hudson, A. L., Shaw, J., Turnbull, M.J.: Baclofen is a selective ligand for a novel GABA receptor on nerve terminals. Brain Res. Bull. 5, 497-502 (1980).

Bradford, M.M.: A rapid and sensitive method for the quantification of microgram quantities of protein using the principle of protein dye binding. Anal. Biochem. 72, 248-252 (1976).

Desarmenien, M., Feltz, P., koeffler, J. P., Occhipinti, G., Santangelo, F.: Multiple GABA receptors on A8 and C primary afferent neurons in the adult rat. Br. J. Pharmacol. 76, 289-293 (1982).

Dooley, D.J., Bittiger, H., Hauser, K. L., Bischoff, S. F., Waldmeier, P. C.: Altera- tion of central alpha-2 and beta-adrenergic receptors in the rat after DSP4, a selective noradrenergic neurotoxin. Neuroscience 9, 889-894 (1983 a).

Dunlap, K.: Two types of GABA receptor on embryonic sensory neurons. Br. J. Pharmacol. 74, 579-586 (1981).

Enna, S.J., Snyder, S. H.: Properties of GABA binding in rat brain. Brain Res. 124, 185-190 (1977).

Guidotti, A., Corda, M. G., Wise, B. C., Vaccarino, F., Costa, E.: GABA-ergic synapses: supramolecular organization and biochemical regulation. Neuropharmacology 22, 1471-1479 (1983).

Hill, D. R., Bowe~, N. G.: 3H-baclofen and 3H-GABA bind to bicuculline- insensitive GABA-B sites in rat brain. Nature 290, 149-151 0981).

Hill, D.R., Bowery, N. G., Hudson, A.L.: Inhibition of GABA-B receptor binding by guanyl nucleotides. J. Neurochem. 42, 652-657 (1981).

Karbon, E. W., Duman, R. S., Enna, S.J.: Biochemical identification of mul- tiple GABA-B binding sites: association with noradrenergic terminals in rat forebrain. Brain Res. 274, 393-396 (1983).

Karbon, E. W., Duman, R.S., Enna, S.J.: GABA agonists potentiate nor- epinephrine-stimulated adenylate cyclase in rat brain frontal cortex slices. Brain Res. 306, 327-332 (1984 a).

Karbon, E. W., Duman, R.S., Enna, S.J.: GABA-B receptor modulation of catecholamine stimulated cyclic AMP formation in rat brain: Structure- activity requirements and in-vivo interactions. Soc. Neurosci. Abstr. If), 974 (1984 b).

172 P.D. Suzdak and G. Gianutsos: GABA-Noradrenergic Interaction

Krogsgaard-Larsen, P., Hjeds, H., Curtis, D.R., Lodge, D., Johnston, G.A.R.: Dihydromuscimol, thiomuscimol, and related heterocyclic compounds as GABA agonists. J. Neurochem. 32, 1717-1726 (1979).

Lloyd, K. G., Morse{li, P. L., Depoortere, H., Fournier, V., Scatton, B., Broekkamp, C., Worms, P., Bartholini, G.: The potential use of GABA agonists in psychiatric disorders: Evidence from studies with Progabide in animal models and clinical trials. Pharmacol. Biochem. Behav. 18, 959-963 (1983).

Rosenthal, H. E.: A graphic method for the determination and presentation of binding parameters in a complex system. Anal. Biochem. 20, 525-529 (1967).

Seamon, K.B., Daly, J. W.: Forskolin, cyclic AMP and cellular physiology. Trends Pharmacol. Sci. 7, 210-215 (1983).

Sulser, Iv., Mobley, P.L.: Regulation of central noradrenergic receptor func- tion: New vistas on the mode of action of antidepressant treatment. In: Neuroreceptors: Basic and Clinical Aspects (Enna, S.J., ed.), pp. 321-335. New York: Raven Press. 1981.

Suzda/e, P.D., Gianutsos, G.: Parallel changes in GABA-ergic and noradren- ergic receptor sensitivity following chronic administration of anti- depressant and GABA-ergic drugs: A possible role of GABA in affective disorders. Neuropharmacology 24, 217-221 (1985 a).

Suzdak, P. D., Gianutsos, G.: Differential coupling of GABA-A and GABA-B receptors to the noradrenergic system. J. Neural Transm. 62, 77-89 (1985 b).

Authors' address: Dr. G. Gianutsos, The University of Connecticut, Box U-92, Storrs, CT 06268, U.S.A.