Brain Research, 305 (1984) 343-352 343 Elsevier

BRE 10178

Inhibitory GABAergic Influence on Striatal Serotonergic Transmission Exerted in the Dorsal Raphe as Revealed by In Vivo Voltammetry

B. SCATTON I, A. SERRANO t, J. P. RIVOT 2 and T. NISHIKAWA l

1Biochemical Pharmacology Unit, Synth~labo-LERS, 31 avenue Paul Vaillant Couturier, 92220 Bagneux, and 2Unitd de Recherches de Neurophysiologie Pharmacologique de I'INSERM (Unit(161), 2 rue d'Aldsia, 75014 Paris (France)

(Accepted December 20th, 1983)

Key words: differential pulse voltammetry - - striatum - - dorsal raph6 - - habenula - - GABA

In vivo differential pulse voltammetry with electrochemically treated carbon fiber microelectrodes has been used to investigate the anatomical nature of the GABAergic influence on striatal serotonergic transmission in the rat. Lesion studies and pharmacological treatments demonstrated that the electrochemical signal recorded at 300 mV in the striatum probably corresponds to the oxidation of extracellularly released 5-hydroxyindoleacetic acid. Thus, dorsal raph6 lesions or systemic administration of a-propyldopacetamide, NSD 1015, pargyline and MK212 decreased, whereas reserpine injection increased the amplitude of the signal. Moreover, L-5-hydroxy- tryptophan administration caused an increase in the signal which was almost completely prevented by pargyline pretreatment. Acute administration of dipropylacetamide (150 mg/kg i.p.) reduced the amplitude of the signal from the striatum, while injection of Y- acetylenic GABA (200 mg/kg i.p.) was without effect. Repeated (but not acute) treatment with the GABA receptor agonist, proga- bide (400 mg/kg i.p.b.i.d, for 14 days), led to a pronounced decrease in the amplitude of the signal from the striatum. A similar effect was observed after intradorsal raph6 infusion of GABA (10 and 100/~g), v-vinyl GABA (100/~g) and SL 75102 (10/~g), a principal me- tabolite of progabide. In contrast, local injection of the GABA receptor antagonists, bicuculline (1 and 10/~g) or R5135 (0.05/~g), failed to affect the peak amplitude in the striatum. When infused into the dorsal raph~, R5135 (0.05-0.1/~g) antagonized the diminu- tion of the signal induced by intradorsal raph6 infusion of GABA (100~g) or SL 75102 (10ktg). Finally, electrolyt!c lesion of the habe- nular nuclei completely blocked the diminution of the signal from striatum induced by an intradorsal raph6 infusion of GABA (100 ~g). These results indicate that the inhibitory GABAergic control of striatal serotonergic transmission is exerted at the level of the dor- sal raph6 cells and depends upon the integrity of the habenulo-dorsal pathway.

INTRODUCTION

Evidence suggests that G A B A pathways exert an

overall inhibitory influence on striatal serotonergic

transmission. Thus, enhancement of G A B A synaptic

activity via systemic administration of G A B A recep-

tor agonists, or drugs enhancing G A B A levels in the

synaptic gap, causes a reduction in the synthesis 19,22

and utilization22 of serotonin (5-HT) in the rat stria-

tum. The neuroanatomical site(s), however, at which

G A B A inhibits striatal serotonergic transmission has

not yet been clearly established. We have recently

demonstrated19 that the GABAergic influence on

striatal serotonergic neurons does not involve

G A B A receptors located on striatal serotonergic ter-

minals, but is most probably exerted at the level of

the dorsal raph6 cells, which are the source of the

striatal serotonergic afferents 2. These data are con-

sistent with anatomical and electrophysiological data

pointing to the existence of a G A B A input on dorsal raph6 cells 4,s,9,27.

Recently, differential pulse voltammetry with car-

bon fiber electrodes has been successfully applied to

the in vivo measurement of extracellularly released

5-hydroxyindoleacetic acid (5-HIAA) in rat brainS, 6.

In the present study, this technique has been used to

explore further the anatomical nature of the GA-

BAergic influence on striatal serotonergic transmis-

sion in the rat. To this end, we have investigated the

effect of G A B A agonist drugs on the release of extra-

cellular 5 -HIAA from the striatum after systemic ad-

ministration and after local injection into the dorsal

Correspondence: B. Scatton, Biochemical Pharmacology Unit, Synth61abo-LERS, 31 avenue Paul Vaillant Couturier, 92220 Bag- neux, France.

0006-8993/84/$03.00 © 1984 Elsevier Science Publishers B.V.

344

raph6. Moreover, since the habenulo-dorsal raph6 tract appears to play a part in the regulation of stria- tal serotonergic transmission 21.24.25,27, we have stud-

ied the possible involvement of this pathway in the

GABAergic influence exerted on dorsal raph6 cells.

MATERIALS AND METHODS

Animals and electrochemical measurements Male Sprague-Dawley rats (COBS CD strain from

Charles River, France) weighing 250 g were used. The rats were housed at 22 _+ 0.5 °C in a humidity controlled room under a 12-h light/dark cycle and had free access to food and water. Animals were tra- cheotomized and the femoral vein was cannulated under light ether anesthesia. Muscular relaxation was achieved by i.p. injection of D-tubocurarine (10 mg/kg). Rats were artificially ventilated with a mix- ture of N20/O 2 (3:1) containing 0.5% fluothane. Ar- terial pH, pO 2 and pCO 2 were measured by means of a gas analyser (Corning model 175) and tidal volume was adjusted so as to maintain normocapnia. Heart rate was continuously monitored and body tempera- ture was kept constant (37 °C).

Rats were placed in a stereotaxic frame for differ- ential pulse voltammetry recordings which were per- formed in the striatum at the following coordinates: 2 mm anterior to bregma, 2 mm lateral to the midline and 4 mm below the skull surface. Electrochemical measurements were performed as described pre- viouslyS, 6 using a pyrrolytic carbon fiber electrode (diameter 8/~m, length 500 ~m), an auxiliary plati-

num electrode, a reference calomel electrode (Ta- cussel) and a polarograph (model PRG5, Tacussel, France). Carbon fiber electrodes were treated elec- trochemically before use according to Cespuglio et al. 5,6. Auxiliary and reference electrodes were posi- tioned in the neck muscles. For in vitro or in vivo measurements, the ramp potential ranged from --50 mV to +450 mV. The electrode was calibrated be- fore and after each experiment in a 10 #M 5-H1AA solution in 0.1 M phosphate buffer at pH 7.4. Electro- trochemical signals were quantitated by measuring the height of the oxidation peak recorded at 300 mV.

Lesions Electrolytic lesions of the dorsal raph6 or habenula

were performed under chloral hydrate (400 mg/kg

i.p.) anesthesia, 2 and 1 week, respectively before the experiment. For the dorsal raph6 lesion, a hi- chrome wire coaxial insulated electrode (0.5 mm in diameter) was set at an angle of 45 ° and introduced successively at the following coordinates: A 5, V 5, L 0 and A 6, V 4.9, L 0 (atlas of K6nig and Klippel~5).

A current of 3 mA was applied for 4 s. Bilateral le- sions of the habenula (lateral and medial parts) were performed using a monopolar stainless steel elec- trode (0.35 mm in diameter). The electrode was in- troduced at the following coordinates: A 4, V 5.1, L + 0.6 and a current of 3.5 mA was applied for 15 s. The location and extent of the lesion were verified histologically. The extent of habenula lesion was also checked biochemically by measuring choline acetyl- transferase (CAT) activity in the interpeduncular nu- cleus. The interpeduncular nucleus was punched out from 200/~m thick coronal brain slices cut serially from level A 2970 to 1020 of the atlas of K6nig and KlippeP 5. CAT activity was assayed by the radio- chemical method of Fonnum 7. Protein was assayed according to Lowry et al. 16.

Local injections When drugs were infused into the dorsal raph6,

guide cannulae (external diameter 0.8 mm) were im- planted stereotaxically under chloral hydrate (400 mg/kg i.p.) anesthesia, 5-7 days before the experi- ments. On the day of the experiment, the mandrel was replaced by an injection cannula (external diam- eter 0.45 mm) 1 mm longer than the guide cannula. The following coordinates (atlas of K6nig and Klip- pep 5) were used: A 4.5, L 0, V 5 (implantation at a posterior angle of 45 ° to the vertical axis). The injec- tion cannula was connected to a motor-driven 10-/A Hamilton syringe. Drugs were infused in a volume of 1 ul for 3 min. The injection cannula was left in place for 1 min before removal. The location of the injec-

tion site was in all cases verified histologically.

Measurement of tissue levels of 5-hydroxyindoles Animals were sacrificed by decapitation. The stri-

atum was dissected out in a cold environment, frozen on dry ice and kept at - -80 °C until analysis. 5-HT and 5-HIAA levels were measured by reversed- phase high-performance liquid chromatography with electrochemical detection23.

Chemicals Progabide (4- {[(4-chlorophenyl)(5-fluoro-2-hy-

droxyphenyl)methylene]amino}butanamide), SL 75102 (4- { [(4-chlorophenyl) (5-fluoro-2-hydroxyphe- nyl)methylene]amino}butyric acid) and dipropylacet- amide were synthesized in the Chemistry Depart- ment, LERS. MK212 (6-chloro-2-[1-piperazinyl]- pyrazine), )'-vinyl GABA, )'-acetylenic GABA and

R5135 (3a-hydroxy-16-imino-5fl-17-aza-androstan- l l -one) were gifts from Merck, Sharp and Dohme, Merrell International and Roussel Uclaf, respective- ly. The other drugs were obtained from commercial sources. Progabide, )'-acetylenic G A B A and dipro- pylacetamide were injected intraperitoneally in 0.1% Tween 80 solution. For chronic experiments, animals were treated twice daily with progabide (400 mg/kg). Control rats received the vehicle. NSD 1015

(m-hydroxybenzylhydrazine), L-5-hydroxytrypto- phan (L-5-HTP) and R5135 were dissolved in a few drops of 1 M hydrochloric acid and the pH was ad- justed to neutrality. Pargyline-HCl, a-propyldopacet- amide, MK212 and reserpine phosphate were dis- solved in distilled water. For local injection studies, SL 75102, GABA, ),-vinyl G A B A and bicuculline were dissolved in saline.

Statistics and expression of results For each individual animal, the mean of the

heights of the electrochemical signal measured dur- ing the period preceding drug treatment (usually 5 or 6 measurements performed every 5 or 10 min) was used as a control value (= 100), and individual data were expressed as percentages of the pre-injection period. The means + S.E.M. of results obtained on 4-6 animals were calculated using corresponding pe- riods. Statistical analysis of the differences between post- and preinjection mean peak height was per- formed using the non-paired two-tailed Student's t-test.

RESULTS

In vitro and in vivo voltammograms In vitro, 5-HIAA, 5-HT and L-5-HTP (10HM) pro-

duced an oxidation peak at around 300 mV (Fig. 1 and data not shown), which was concentration-de- pendent over the range 0.1-20/zM. In vivo, a major and stable electrochemical signal was recorded at 300

t , -

IN V ITRO

345

c

IN VlVO

0 LESIONED

I 1 I I J I I I J I I -50 0 100 200 300 400 mV

Fig. 1. Typical differential pulse voltammograms recorded in vitro and in vivo in the striatum of control and dorsal raph6-1e- sioned rats. In vitro recordings were performed using a 10-HM 5-HIAA solution in 0.1 M sodium phosphate buffer at pH 7.4. For both in vitro and in vivo recordings, the following parame- ters were used: potential range --50 to +450 mV; scanning speed 20 mV/s; pulses of 50 mV in amplitude and 28 ms dura- tion at a frequency of 5 Hz.

mV in the striatum (Fig. 1). Electrolytic lesion of the dorsal raph6 caused a marked decrease in the ampli- tude of the striatal signal (Fig. 1). In a series of ex- periments, the mean oxidation current was reduced by 74% (P < 0.001) in lesioned (0.95 + 0.04 nA) as compared to sham-operated (3.60 + 0.27 nA) ani- mals. In parallel experiments, electrolytic lesion of the dorsal raph6 reduced the endogenous tissue lev- els of striatal 5-HT and 5-HIAA, by 68 and 63%, re- spectively (5-HT (ng/g): sham, 640 + 34; lesioned, 203 + 27*; 5-HIAA (ng/g): sham, 467 + 23; lesioned, 174 + 33*; * P < 0.001 vs respective controls).

Pharmacological identification of the nature of the striatal signal recorded at 300 m V

As shown in Figs. 2 and 3, i.p. injection of the aro- matic L-amino acid decarboxylase inhibitor, NSD 1015, of the tryptophan hydroxylase inhibitor, a- propyldopacetamide, of the monoamine oxidase in-

346

NSD 1015

'°°F!

I00

50

r

01

2

,, PROPYLDOPACETAMIDE

Y 150~

RESERPINE . * [ *

100 i

l 50L

PARGYLINE

1 I00 r = ;

50i " " . . . .

0 0 30 60 120 IRD 240 0 60 120 180 2 4 0

TIME Imini TiME Immi

Fig. 2. Changes in the striatal electrochemical signal after administration of drugs affecting 5-HT metabolism. Rats received an i.p. in- jection of NSD 1015 (100 mg/kg), a-propyldopacetamide (500 mg/kg), reserpine (5 mg/kg) or pargyline (75 mg/kg). Results are means + S.E.M. of data obtained on 5 animals and are expressed as percentages of the pre-injection period. * P < 0.05 vs control period.

hibitor, pargyline, or of the 5-HT receptor agonist, MK212, caused a time-dependent decrease in the striatal signal. In contrast, the monoamine-depleting agent, reserpine, increased the peak height. Admin- istration of the 5-HT precursor, L-5-HTP, induced a rapid increase in the signal which reached its apex (+390%) 30 min later. This effect was markedly re-

duced by pargyline pretreatment (Fig. 3). The endogenous tissue levels of 5-HIAA were re-

duced by 61, 43 and 73% and increased by 67% 3 h after injection of NSD 1015, a-propyldopacetamide, pargyline and reserpine, respectively (Table I). Striatal 5-HT levels were decreased by 42 and 77% and were increased by 118% after a-propyldopaceta- mide, reserpine and pargyline injection, respective-

ly, but remained unaffected after NSD 1015 adminis- tration (Table 1).

Effect of systemic and intradorsal raph# injection of GA BA ergic agents on the striatal signal

Acute systemic injection of dipropylacetamide, an indirectly acting GABA mimetic drug, caused a slight diminution of the amplitude of the striatal sig- nal (Fig. 4). In contrast, acute administration of V- acetylenic-GABA, a specific GABA transaminase

inhibitor, or of progabide, a GABA receptor agon- ist-L was without effect (Fig. 4). After repeated treat- ment with progabide for 2 weeks, however, re-injec- tion of the drug 16 h after the last administration caused a pronounced decrease in the striatal signal (Fig. 4).

A decrease in the striatal signal was also observed after intradorsal raphe infusion of GABA (10 and 100 Hg), of the specific GABA transaminase inhibi- tor y-vinyl-GABA (100 #g) and of SL 75102 (10/~g), a soluble metabolite of progabide possessing potent GABA receptor agonist properties 3 (Fig. 5). lntra- dorsal raph6 infusion of the GABA receptor antago- nist, bicuculline (1 and 10Hg), or of R5135 (0.05-0. l Mg), a novel steroid derivative possessing potent GABA receptor antagonist properties 14, failed to af- fect the peak amplitude in the striatum (results not shown). When infused into the dorsal raphd, howev- er, R5135 (0.05 ,ug) prevented the diminution of the striatal signal induced by intradorsal raphe infusion of GABA (100 Hg) (Fig. 5). Moreover, intradorsal raphe infusion of R5135 (0.1 ug) 2 h after injection of SL 75102 (10 Hg) reversed the decrease in the striatal signal elicited by the GABA agonist (Fig. 5).

Finally, while intradorsal raphd infusion of GABA

TABLE I

Changes in the endogenous levels of 5-HT and 5-HIAA in the striatum after administration of drugs affecting 5-HT metabolism

Drugs were injected 3 h before sacrifice. Results are means + S.E.M. of data obtained from 6 rats per group.

347

Drug Dose 5-HT 5-HIAA (mg/kg i.p. ) ng/g % ng/g %

Saline - - 403 + 33 100 375 _+ 29 100 NSD 1015 100 410 + 14 98 146 + 13"** 39 ct-Propyldopacetamide 500 234 + 28** 58 213 + 19"** 57 Reserpine 5 93 + 8*** 23 627 + 37*** 167 Pargyline 75 881 + 30*** 218 99 + 8*** 27

** P < 0.01 vs controls. *** P < 0.001 vs controls.

400

300

200

100

0 I

100

50

L-5HTP

3'o 6'0 9'o TIME (rnin)

MK 212

i , t I I

0 60 120 180

TIME I m i n )

Fig. 3. Changes in the striatal signal elicited by L-5-HTP (alone or after pretreatment with pargyline) and by MK212. Pargyline (75 mg/kg i.p.) was injected 1 h before L-5-HTP (50 mg/kg i.p.). MK212 was injected in a dose of 10 mg/kg i.p. Results are means + S.E.M. of data obtained on 5 rats per group and are expressed as percentages of respective pre-injection periods (for the L-5-HTP + pargyline group, results were compared to pargyline alone). *P < 0.05 vs respective control period.

100

1 0 0 [

PROGABIDE

-- - ACUTE

CHRONIC

I 6h0 I I 0 120 180

"y • acetyienic • GABA

1

t I I I I

DIPROPYLACETAMIDE

1 1oo 50

L / I I t 0 60 120 180 240

TIME Imin.)

Fig. 4. Alteration in the release of extracellular 5-HIAA from the striatum induced by acute and chronic administration of G A B A agonist drugs. Dipropylacetamide (150 mg/kg) and ~,- acetylenic G A B A (200 mg/kg) were administered acutely via the i.p. route. Progabide (400 mg/kg i.p.) or vehicle were ad- ministered twice daily for 14 days. Sixteen hours after the last injection, both groups of rats received a single dose of proga- bide (400 mg/kg i.p.). Results are means +_ S.E.M. of data ob- tained on 5 animals and are expressed as percentages of respec- tive preinjection periods. *P < 0.05 vs respective control peri- od.

348

100

50

GABM (I0 UQI GABA II00 #g}

4 I . . . . . . . . . . . . . . . . .

= I I * * * , , * , ~*z?* I00

O L i I I ~ 0

~ Vinyl GIBAIlO0 .g) J:'

100 , - - - - - - - - - - - - - - - - - - - - - - - - - - - - ....... 100

50

SL 75102 II0 ~gl

" . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

, R5135 lolpg) ,

* , , , * * , * *

R5135 ,005 ,,q GABA 100 , , ] ,

¢ - ~ r - - v v - v -

I 0 ~ ' 1 J

0 o 6'o 1~o ,;0 240 0 i0 'i0 ~8o 240 TIME tmtn) TIME 4mini

Fig. 5. Decrease in the release of extracellular 5-HIAA from the striatum after intradorsal raph~ infusion of GABA, F-vinyl GABA and SL 75102: reversal by intradorsal raph~ injection of R5135. Drugs were infused into the dorsal raph6 in a volume of 1 kd for 3 min by means of cannulae implanted 5-7 days before the experiment. Results are means with S.E.M. of data obtained on 4 rats per group and are expressed as percentages of the control period. *P < 0.05 vs control period.

(100 ktg) caused a marked diminution of the striatai

signal in sham-operated rats, local injection of

G A B A into rats with an electrolytic lesion of the ha-

benular nuclei was without effect (Fig. 6). The effec-

tiveness of the habenula lesion was indicated by a

93% reduction in CAT activity in the interpeduncu-

lar nucleus of lesioned (31 + 2 nmol/h/mg protein) as

compared to sham-operated (432 + 14 nmol/h/mg

protein) animals.

DISCUSSION

Nature o f the electrochemical signal detected in the

striatum at 300 m V

Cespuglio et al.5,6 have previously demonstrated

that the in vivo electrochemical signal recorded in the

striatum under the experimental conditions presently

used, corresponds to the oxidation of 5-hydroxyin-

dole species and more specifically to that of 5 -HIAA.

The present results are in complete agreement with

this view. Thus, the in vivo signal peaked at the same

potential (+300 mV) as that detected in vitro with 5-

hydroxyindole compounds. Moreover, electrolytic

lesion of the dorsal raph6, which gives rise to the se-

rotonergic innervation of the striatum 2, markedly re-

duced the amplitude of the striatal signal. Pharmaco-

GABA 4100 ~g~ i _ ~ _ HABENULA

NORMAL

L_ . . . . . . . . . . L .............. J__ i i

0 60 120 180

TIME Imm~

Fig. 6. Failure of an intradorsal raph6 infusion of GABA to di- minish the release of extraeellular 5-HIAA from the striatum after habenula lesion. Electrolytic lesion of the habenula was performed 6 days before the experiment. GABA was infused into the dorsal raph~ (in a volume of 1 ~1 for 3 min) by means of cannulae implanted 7 days before the experiment. Results are means + S.E.M. of data obtained on 6 animals per group and are expressed as percentages of respective preinjeetion peri- ods. *P < 0.05 vs respective control period.

logical manipulation of the in vivo oxidation peak with agents known to affect 5-HT metabolism also supports the view that extracellular 5-hydroxyindoles mainly contribute to the striatal signal. Thus, inhibi- tion of 5-HT synthesis by a-propyldopacetamide, a tryptophan hydroxylase inhibitor, or by NSD 1015, an inhibitor of aromatic L-amino acid decarboxylase, leads to a marked diminution of the amplitude of the 300 mV peak in the striatum. Moreover, administra- tion of L-5-HTP markedly increased the amplitude of the signal. That 5-HIAA rather than 5-HT is the chemical species measured in vivo in the striatum is indicated by the fact that (1) administration of pargy- line, an MAO inhibitor, decreases the 300 mV peak height; (2) reserpine injection caused an increase in the signal; (3) pretreatment with pargyline almost completely prevented the augmentation of the signal amplitude induced by administration of a high dose of L-5-HTP; and (4) the alterations in the oxidation peak height observed after drug treatment were in parallel with the changes in the endogenous levels of 5-HIAA but not of 5-HT in the striatum.

These data indicate, therefore, that the technique of in vivo differential pulse voltammetry used can monitor changes in extracellular 5-HIAA levels in the striatum. The measurement of extracellular 5- HIAA may serve as an index of 5-HT neuronal activ- ity. Thus, the signal detected in the striatum almost totally disappears after electrolytic lesion of the dor- sal raph6 (present study) or of the medial forebrain bundle 5 and increases after electrical stimulation of the raph6 dorsalis 6.

GABAergic influence on striatal serotonergic trans-

mission Previous studies have provided evidence that

GABA exerts an inhibitory influence on striatal sero- tonergic transmission (see Introduction). The pres- ent study confirms this viewpoint by showing that acute administration of dipropylacetamide or pro- longed treatment with progabide leads to a decrease in the amplitude of the striatal signal. This is in agreement with our recent data demonstrating that subacute administration of progabide reduces the en- dogenous levels of 5-HIAA as well as the synthesis of 5-HT in the rat striatum 28. Acute administration of progabide or ~-acetylenic GABA failed, however, to affect the electrochemical signal. This suggests that a

349

sustained stimulation of GABA receptors is needed

to produce a maximal reduction in striatal serotoner- gic transmission.

The inhibitory GABAergic control of striatal sero- tonergic transmission appears to be exerted at the level of the dorsal raph6 cells. Thus, intradorsal raph6 infusion of GABA, ),-vinyl-GABA or SL 75102 caused a diminution of the striatal signal. This

effect is probably due to a GABA receptor-mediated mechanism as it was antagonized by intradorsal raph6 injection of the GABA receptor antagonists R5135 (Fig. 5) or bicuculline (unpublished data). A diffusion of the drugs to structures adjacent to the dorsal raph6 is unlikely since (1) no significant change in the striatal signal was observed when mi- croinfusions were made 1 mm lateral to the dorsal raph6 (unpublished observations) and (2) we have previously shown in similar experimental conditions that after local infusion of [3H]GABA into the dorsal raph6, the radioactivity remained confined to this area 19.

The view that serotonergic neurons projecting to the striatum are subject to an inhibitory influence by GABA which is exerted within the dorsal raph6 is also supported by our recent finding that intradorsal raph6 application of SL 75102, ),-vinyl-GABA and GABA itself decreases 5-HTP accumulation in the rat striatum 19. Moreover, infusion of GABA into the dorsal raph6 diminishes the release of newly synthe- sized [3H]5-HT from the cat caudate nucleus 13. The inhibitory action of GABA on dorsal raph6 5-HT- containing cells is also compatible with anatomical data showing the presence of high levels of glutamic acid decarboxylase4, ix and of GABA-accumulating neurons 4,10 in the dorsal raph6. Moreover, iontopho- retic application of GABA into the dorsal raph6 sup- presses the firing of 5-HT sensitive raph6 cells in a picrotoxin-reversible manner8.9.

The present results also indicate that bicuculline and R5135 when infused into the dorsal raph6 failed to af- fect the striatal signal. This is in line with our previous studies which showed that intradorsal raph6 infusion of high amounts of bicuculline or picrotoxin did not influence striatal 5-HT synthesis in the rat 19. More- over, in the cat, application of picrotoxin in the dor- sal raph6 did not modify the release of [3H]5-HT in the substantia nigra 21, a brain area also receiving its serotonergic innervation from the dorsal raph6:. Fi-

350

nally, when administered systemically in subconvul-

sant doses, or when iontophoretically applied to the dorsal raph6, picrotoxin failed to affect raph6 cell fir- ing 9. The lack of effect of GABA antagonists on raph6 cell firing and striatal serotonergic transmis- sion suggests that the inhibitory influence of GABA on dorsal raph6 cells is not tonic in nature. This could also indicate that GABA receptors located in the dorsal raph6 do not play a physiological role in the regulation of striatal serotonergic transmission, but may be of pharmacological importance since they

could be acted upon by exogenous G A B A receptor agonists. Finally, it cannot be ruled out that the fail- ure of GABA antagonists to increase striatal seroton- ergic transmission is related to the fact that these

cells are maintained under a tonic excitatory influ- ence by the numerous afferents projecting to these neurons and may, therefore, show a lower suscepti-

bility to further excitation by GABA antagonists. The exact nature of the GABAergic influence

exerted on dorsal raph6 cells is still a matter for spec- ulation. The GABAergic influence might be exerted by afferents arising outside the raph627 or else be me- diated via the GABAergic interneurons or the puta- tive GABAergic neurons co-existing with serotoner- gic neurons 17 which appear to be present in the dorsal raph6. Anatomical data indicate that one of the main sources of afferents to the dorsal raph6 nucleus is the habenulaH2. This neuronal input appears to play a major role in controlling the activity of serotonergic dorsal raph6 neurons projecting to the caudate nucle- us 21,24,25,27. In the present study, electrolytic lesion of

the habenular nuclei blocked the ability of GABA, applied to the dorsal raph6, to diminish the striatal signal. This suggests that the inhibitory GABAergic influence on striatal serotonergic neurons arising from the dorsal raph6 cells depends upon the habenu- lo-dorsal raph6 pathway. This view is further sub- stantiated by our recent data showing that electrolyt- ic lesion of the habenular nuclei or cessation of im- pulse flow in the habenulo-raph6 pathway by local in- jection of tetrodotoxin into the fasciculus retroflexus blocked the ability of GABA agonist agents (given systemically or applied locally into the dorsal raph6) to diminish striatal 5-HT synthesis 2°.

Based on the fact that iontophoretic application of picrotoxin to the dorsal raph6 blocked the inhibition of dorsal raph6 cells elicited by electrical stimulation

of the lateral habenula, it has been proposed that

GABA could be a transmitter of some of the habenu- lar afferents to the dorsal raph6 27. The existence, however, of a monosynaptic habenulo-dorsai raph6 GABAergic tract remains controversial, since lateral habenula lesions failed to alter glutamic acid decar- boxylase activity in the dorsal raph6 Lm6. Interrup- tion of this hypothetical habenulo-dorsal raph6 GA- BAergic pathway does not appear to account for the loss of effectiveness of GABA agonists on dorsal

raph6-striatal serotonergic neurons after habenula lesion, since lesion of this pathway would not be ex-

pected to influence the ability of direct acting GABA receptor agonists to inhibit the activity of the dorsal raph6-striatal serotonergic pathway. Substance P has also been proposed as a transmitter of the habenulo- dorsal raph6 pathway, as a large depletion of the pep- tide has been found in the dorsal raph6 following le- sions of the lateral habenula 1s.26. Thus, a possible ex- planation for the abolition of the GABA-induced de- crease in striatal serotonergic transmission after ha- benula lesion may be that GABA could act upon 5- HT-sensitive dorsal raph6 cells by blocking the action of a facilitatory influence on dorsal raph6 neurons. This may be explained by different neuronal set-ups. GABA receptors may be localized on a habenulo- dorsal raph6 excitatory (possibly substance Pergic) tract which innervates dorsal raph6 serotonergic neu- rons. Increase in GABAergic transmission might thus lead to a reduction of the putative excitatory in- fluence on dorsal raph6 cells resulting in decreased striatal serotonergic transmission. Alternatively, GABA interneurons may impinge on 5-HT-sensitive dorsal raph6 cells, these being inhibited by GABA only in the presence of a tonic excitatory drive exerted by a habenulo-raph6 input. Finally, adaptive mechanisms occurring in serotonergic neurons as a consequence of the degeneration of the habenulo- dorsal raph6 pathway may render them refractory to GABAergic inhibition. This possibility, however, seems unlikely, since habenula lesion failed to affect the ability of the 5-HT receptor agonist MK212, given systemically, to diminish striatal 5-HT synthe- sis 20.

In conclusion, the present results indicate that the inhibitory GABAergic influence on striatal seroton- ergic transmission is mediated via stimulation of

GABA receptors located in the dorsal raph6 and is

351

d e p e n d e n t upon the in tegr i ty of the habenu lo -do r sa l

raph6 pa thway.

ACKNOWLEDGEMENTS

The au thors wish to thank Dr . A . D e g u e u r c e for

pe r fo rming e lec t ro ly t ic les ion of the dorsa l raph6 , T.

Denn i s and R. L~Heureux for H P L C d e t e r m i n a t i o n s

of tissue levels of 5 - H T and 5 - H I A A , D. Fage for

C A T m e a s u r e m e n t s and V. N e v e u x for typing the

manuscr ip t . W e also thank Dr . P. H u n t (Rousse l

Uclaf , Roma inv i l l e , F rance ) for the gift o f R5135.

REFERENCES

1 Aghajanian, G. K. and Wang, R. Y., Habenular and other midbrain raph6 afferents demonstrated by a modified ret- rograde tracing technique, Brain Research, 122 (1977) 22%242.

2 Azmitia, E. C. and Segal, M., An autoradiographic analy- sis of the differential ascending projections of the dorsal and median raphe nuclei in the rat, J. comp. Neurol., 179 (1978) 641-667.

3 Bartholini, G., Scatton, B., Zivkovic, B. and Lloyd, K. G., On the mode of action of SL 76002, a new GABA receptor agonist. In P. Krogsgaard-Larsen, J. Scheel-Kriiger and H. Kofod (Eds.), GABA-Neurotransmitter, Munksgaard, Co- penhagen, 1979, pp. 326--339.

4 Belin, M. F., Aguera, M., Tappaz, M., McRae-De- gueurce, A., Bobillier, P. and Pujol, J. F., GABA-accumu- lating neurons in the nucleus raph6 dorsalis and periaque- ductal gray in the rat: a biochemical and radioautographic study, Brain Research, 170 (1979) 279-297.

5 Cespuglio, R., Faradji, H., Ponchon, J. L., Buda, M., Riou, F., Gonon, F., Pujol, J. F. and Jouvet, M., Differen- tial pulse voltammetry in brain tissue. I. Detection of 5-hy- droxyindoles in the rat striatum, Brain Research, 223 (1981) 287-298.

6 Cespuglio, R., Faradji, H., Riou, F., Buda, M., Gonon, F., Pujol, J. F. and Jouvet, M., Differential pulse voltammetry in brain tissue. II. Detection of 5-hydroxyindoleacetic acid in the rat striatum, Brain Research, 223 (1981) 299-311.

7 Fonnum, F., Radiochemical assays for choline acetyltrans- ferase and acetylcholinesterase. In N. Marks and R. Rod- night (Eds.), Research Methods in Neurochemistry, Vol. 3, Plenum Press, New York, 1975, pp. 253-275.

8 Gallager, D. W., Benzodiazepines: potentiation of a GABA inhibitory response in the dorsal raph6 nucleus, Eu- top. J. Pharmacol., 49 (1978) 133-143.

9 Gallager, D. W. and Aghajanian, G. K., Effect of antipsy- chotic drugs on the firing of dorsal raph6 cells. II. Reversal by picrotoxin, Europ. J. Pharmacol., 39 (1976) 357-364.

10 Gamrani, H., Calas, A., Belin, M. F., Aguera, M. and Pu- jol, J. F., High resolution radioautographic identification of [3H]GABA labeled neurons in the rat nucleus raph6 dorsa- lis, Neurosci. Lett., 15 (1979) 43-48.

11 Gonesfeld, Z., Hoover, D. B., Muth, E. A. and Jacobo- witz, D. M., Lack of biochemical evidence for a direct ha- benulo-raph6 GABAergic pathway, Brain Research, 14 (1978) 353--356.

12 Herkenham, M. and Nauta, W. J. H., Efferent connections of the habenular nuclei in the rat, J. cornp. Neurol., 187 (1979) 19-48.

13 H6ry, F. and Ternaux, J. P., Regulation of release process- es in central serotoninergic neurons, J. Physiol. (Paris), 77 (1981) 287-301.

14 Hunt, P. and Clements-Jewery, S., A steroid derivative, R5135, antagonizes the GABA/benzodiazepine receptor interaction, Neuropharmacology, 20 (1981) 357-361.

15 K6nig, J. F. R. and Klippel, R. A., The Rat Brain, Williams and Wilkins, Baltimore, MD, 1973.

16 Lowry, O. H., Rosebrough, M. J., Farr, A. L. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-268.

17 Nanopoulos, D., Belin, M. F., Maitre, M., Vincendon, G. and Pujol, J. F., Immunocytochemical evidence for the ex- istence of GABAergic neurons in the nucleus raph6 dorsa- lis. Possible existence of neurons containing serotonin and GABA, Brain Research, 232 (1982) 375-389.

18 Neckers, L. M., Schwartz, J. P., Wyatt, R. J. and Speciale, S. G., Substance P afferents from the habenula innervate the dorsal raph6 nucleus, Exp. Brain Res., 37 (1979) 61%623.

19 Nishikawa, T. and Scatton, B., Evidence for a GABAergic inhibitory influence on serotonergic neurons originating from the dorsal raph& Brain Research, 279 (1983) 325-329.

20 Nishikawa, T., Scatton, B. and Serrano, A., GABAergic inhibition of striatal serotonergic transmission exerted in the dorsal raph6 requires the integrity of the habenula, Brit. J. Pharmacol., (1984) in press.

21 Reisine, T. D., Soubri& P., Artaud, F. and Glowinski, J., Involvement of lateral habenula-dorsal raph6 neurons in the differential regulation of striatal and nigral serotonergic transmission in cats, J. Neurosci., 2 (1982) 1062-1071.

22 Scatton, B., Zivkovic, B., Dedek, J., Lloyd, K. G., Con- stantinidis, J., Tissot, R. and Bartholini, G., 7-Aminobu- tyric acid (GABA) receptor stimulation. III. Effect of pro- gabide (SL 76002) on norepinephrine, dopamine and 5-hy- droxytryptamine turnover in rat brain areas, J. Pharmacol. exp. Ther., 220 (1982) 678-688.

23 Semerdjian-Rouquier, L., Bossi, L. and Scatton, B., Con- current determination of 5-hydroxytryptophan, serotonin and 5-hydroxyindole acetic acid in rat and human brain and biological fluids by reversed-phase high performance liquid chromatography with electrochemical detection, J. Chro- matog., 218 (1981) 663-670.

24 Speciale, S. G., Neckers, L. M. and Wyatt, R. J., Habenu- lar modulation of raph6 indoleamine metabolism, Life Sci., 27 (1980) 2367-2372.

25 Stern, W. C., Johnson, A., Bronzino, J. D. and Morgane, P. J., Neuropharmacology of the afferent projections from the lateral habenula and substantia nigra to the anterior raph6 in the rat, Neuropharmacology, 20 (1981) 97%989.

26 Vincent, S. R., Staines, W. A., McGeer, E. G. and Fibiger, H. C., Transmitters contained in the efferents of the habe- nula, Brain Research, 195 (1980) 479-484.

27 Wang, R. Y. and Aghajanian, G. K., Physiological evi- dence for habenula as major link between forebrain and midbrain raph6, Science, 197 (1977) 8%91.

352

28 Zivkovic, B., Scatton, B., Dedek, J. and Bartholini, G., G A B A influence on noradrenergic and serotonergic trans- missions: implications in mood regulation. In S. Z. Langer,

R. Takahashi, T. Segawa and M. Briley (Eds.), Advances in the Biosciences, Vol. 40, New Vistas in Depression, Per- gamon Press, Oxford and New York, 1982, pp. 195-201.