Characterization of galanin and 5-HT1A receptor coupling to adenylyl cyclase in discrete regions of the rat brain

ejp ELSEVIER

European Journal of Pharmacology Molecular Pharmacology Section 269 (1994) 209-217

molecularpharmacology

Characterization of galanin and 5-HTIAreceptor coupling to adenylyl cyclase in discrete regions of the rat brain Agn~s Billecocq a, Peter B. Hedlund b,,, Francisco Bolafios-Jim~nez a, Gilles Fillion

a

" Unit of Neuroimmunoendocrinological Pharmacology, Institut Pasteur, Paris, France b Department ofNeuroscience, Karolinska Institutet, S-171 77 Stockholm, Sweden Received 13 June 1994; accepted 5 July 1994

Abstract

We have studied the coupling of galanin and 5-HT1A receptors with adenylyl cyclase in the hypothalamus, the entorhinal cortex and the hippocampus of the rat brain. Furthermore, we have evaluated the effects of simultaneous activation of galanin and 5-HTIA receptors on adenylyl cyclase activity. Galanin-(1-29) and galanin-(1-15) showed a dose-dependent inhibitory effect on forskolin-stimulated adenylyl cyclase activity in the hypothalamus and entorhinal cortex. No clear effects were observed in the hippocampus. Neither galanin-(1-29) nor galanin-(1-15) had any effect on the basal activity of adenylyl cyclase in these regions. The selective 5-HT1A receptor agonist 8-OH-2-(di-n-propylamino)-tetralin (8-OH-DPAT) induced a dose-dependent inhibition of forskolin stimulated adenylyl cyclase activity in the hippocampus and the entorhinal cortex. 5-HT induced an inhibition in the hypothalamus. In all regions the effects could be fully counteracted by methiothepin. 5-HT was shown to stimulate the basal activity of adenylyl cyclase in the hippocampus and the entorhinal cortex. The effects could be counteracted by methiothepin. When galanin-(1-29) and 5-HT/8-OH-DPAT were incubated simultaneously additive inhibitory effects, but no synergistic interactions, could be observed on the stimulated adenylyl cyclase activity. In conclusion, galanin and 5-HT1A receptors seem to be linked to different independent pools of G proteins, indicating that the previously demonstrated intramembrane interactions between galanin and 5-HTIA receptors involve a mechanism not directly related to adenylyl cyclase.

Keywords: Galanin; 5-HT (5-hydroxytryptamine, serotonin); 8-OD-DPAT (8-OH-2-(di-n-propylamino)-tetralin); Adenylyl cyclase; Forskolin; Hypothalamus; Hippocampus; Brain; Receptor subtype; Peptide fragment

I. Introduction

Galanin is a 29-amino acid peptide first isolated from porcine small intestine (Tatemoto et al., 1983). It is widely distributed in the central nervous system of the rat (R6kaeus et al., 1984; Skofitsch and Jacobowitz, 1985; Melander et al., 1986a) and of the human (Gentleman et al., 1989). Specific binding sites for [1251] galanin are localized in many of the terminal regions of galanin-containing neurons (Skofitsch et al., 1986; Melander et al., 1988). Galanin appears to be colocalized with cholinergic and monoaminergic neuronal systems (Melander et al., 1986b). It has been shown to modu-

* Corresponding author. Tel.: +46-8 728 7028; Fax: +46-8 33 79 41. 0922-4106/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0 9 2 2 - 4 1 0 6 ( 9 4 ) 0 0 1 1 5 - 4

late the function of these neurotransmitters and in particular, the existence of galanin/5-hydroxytryptamine (5-HT) interactions in the central nervous system has been well documented. Galanin has been shown to affect 5-HT metabolism in some cortical and hippocampal regions (Fuxe et al., 1988b) and to stimulate the release of 5-HT in the hypothalamus (Martire et al., 1991). Galanin has also been reported to exert a presynaptic inhibitory effect on central 5-HT neurons (Sundstr6m and Melander, 1988). Galanin binding sites (Skofitsch et al., 1986; Melander et al., 1988) and 5-HT1A binding sites (Pazos and Palacios, 1985) appear to be codistributed in several regions, notably in the limbic system and the hypothalamus. Furthermore, it has been possible to demonstrate the existence of a reciprocal intramembrane interaction between both receptors since galanin selectively decreases the affinity of 5-HT~A receptors within the rat ventral limbic cortex

210

A. Billecocq et al. / European Journal of Pharmacology - Molecular Pharmacology Section 269 (1994) 209-217

(Fuxe et al., 1988a), and 8-hydroxy-2-(di-n-propylamino)-tetralin (8-OH-DPAT), a specific 5-HT1A receptor agonist, in contrast increases the affinity of galanin binding sites in various tel- and diencephalic regions (Hedlund et al., 1991a). Such interactions have also been observed in functional studies on serum prolactin and thyroid stimulating hormone levels after combined central and systemic administration of galanin and 8-OH-DPAT respectively (Fuxe et al., 1990), as well as in the development of vasodepressor responses after central coinjection of both receptor agonists (Hedlund et al., 1991b). Galanin has been reported to inhibit the muscarinic agonist stimulated breakdown of inositol phospholipids in the rat ventral hippocampus (Palazzi et al., 1988; Fisone et al., 1989) as well as the noradrenaline-induced accumulation of cyclic AMP (cAMP) in rat cerebral cortex (Nishibori et al., 1988). Recently, it has been shown that galanin also inhibits forskolin-stimulated adenylyl cyclase of rat whole brain membranes (Chen et al., 1992) as previously observed in the pancreatic beta cell line Rin m 5F (Amiranoff et al., 1988). The predominant transduction system linked to the 5-HTIA receptors is that of adenylyl cyclase. In numerous experiments, the 5-HTIA receptor has been demonstrated to exert a negative control of the activity of the adenylyl cyclase stimulated by various agents (Weiss et al., 1983; De Vivo and Maayani, 1986; Bockaert et al., 1987; Dumuis et al., 1988; Oksenberg and Peroutka, 1988; Schoeffter and Hoyer, 1988; M6rk and Geisler, 1990). There have also been some reports indicating that there may be a positive coupling between 5-HTIA receptors and adenylyl cyclase (Barbaccia et al., 1983; Shenker et al., 1985; Markstein et al., 1986; Fayolle et al., 1988; M6rk and Geisler, 1990). The aim of the present study has been to investigate both the positive and negative coupling of galanin receptors with adenylyl cyclase in three rat brain regions, the hypothalamus, the entorhinal cortex and the hippocampus, as well as the action of 8-OH-DPAT or 5-HT on adenylyl cyclase activity in the same regions. Furthermore, we have evaluated that the previously observed interactions between galanin and 5-HT~A receptors might occur at the level of second messenger formation by adenylyl cyclase.

2. Materials and methods 2.1. Animals

Male Wistar rats (250 g body weight; Iffa-Credo, Lyon, France) were used in all experiments. They were group-housed and maintained under standard laboratory conditions with a 12 h light-dark cycle and free access to food pellets and tap water.

2. 2. Membrane preparation

The rats were killed by decapitation and the hypothalamus, hippocampi a n d / o r entorhinal cortices were rapidly dissected. In some experiments the hippocampus was divided into a ventral and a dorsal part, since the dorsal part has a very low density of [le5I]galanin(1-29) binding sites but a relatively high density of [125I]galanin-(1-15) binding sites. The tissue was immediately homogenized with a Teflon homogenizer in 2.5 ml of 2 mM ice-cold Tris-maleate buffer (pH 7.4) containing 2 mM EGTA, 0.3 M sucrose and protease inhibitors (0.1 mM phenyl-methane-sulfonyl fluoride and 5 U / m l aprotinin). The homogenates were centrifuged at 1000 × g for 5 min at 4°C. The supernatants were recentrifuged at 20 000 × g for 20 min at 4°C. For the inhibition assays (see below), the pellets were resuspended in 1.5 to 2.5 ml of the same buffer for the hypothalamus or 2 to 3 ml for the other tissues and frozen at - 80°C before use (between 1 h to 2 months). For the activation assays, the pellets were resuspended in 35 ml of 5 mM ice-cold Tris-HC1 (pH 8.2) containing 2 mM EGTA and protease inhibitors and maintained under mild agitation for 1 h at 4°C and then centrifuged at 20 000 × g for 20 min, resuspended in 1 ml of 50 mM ice-cold Tris-HCl (pH 7.4) containing 0.5 mM EGTA, 4 mM MgSO4, 1 mM 3-isobutyl-l-methylxanthine and 1 /zM pargyline and used immediately. Protein content was determined using the BCA protein assay (Pierce, USA). 2.3. Inhibition o f forskolin-stimulated adenylyl cyclase acticity

To study the inhibition of forskolin-stimulated adenylyl cyclase activity the protocol described by De Vivo and Maayani (1986) was used with slight modifications. An assay medium was prepared with the following components and final concentrations: 25 mM Tris-maleate (pH 7.2), 10/xM forskolin, 100 mM NaCI, 5 /zM GTP, 1 mM MgSO 4, 0.1 mM ATP, 10 mM theophylline, 0.25 mg/ml bacitracin, 20 mM phosphocreatine, 0.2 mg/ml creatine phosphokinase, 50 nM [3H]cAMP and 1 /zM [a-32p]ATP. The assay medium also contained the various peptides and drugs tested (see below). The reaction was started by adding 50/~1 of membrane suspension ~o 150 /zl of assay medium. The membranes were incubated for 20 min at 30°C and the incubation was stopped by the addition of 200/zl of 50 mM Tris-HC1 (pH 7.4) containing 5 mM ATP, 5 mM cAMP and 1% sodium dodecyl sulfate. The [32p]cAMP produced was separated from the [32p]ATP as previously described (Salomon, 1979) on Dowex AG50W-X4 columns followed by a further separation on neutral alumina columns. The [3H]cAMP initially

A. Billecocqet al. / European Journal of Pharmacology - MolecularPharmacologySection 269 (1994) 209-217

added was used to determine the degree of recovery of formed [32p]cAMP during the procedure.

2.4. Activation of adenylyl cyclase basal activity

The experiments to study activation of adenylyl cyclase basal activity was performed as described by Fayolle et al. (1988). An assay medium with the following components and final concentrations was prepared: 50 mM Tris-HC1 (pH 7.4), 100 ~M GTP, 0.5 mM EGTA, 4 mM MgSO4, 1 mM ATP, 1 mM 3-isobutyl-1methylxanthine, 1 /zM pargyline, 0.25 mg/ml bacitracin, 20 mM phosphocreatine, 0.2 m g / m l creatine phosphokinase, 50 nM [3H]cAMP and 1 /xM [a32P]ATP. The assay medium also contained the various peptides and drugs tested (see below). The reaction was started by adding 50 /zl of membrane suspension to 150 /zl of assay medium. The membranes were incubated for 10 min at 30°C. The incubation was stopped and the [32p]cAMP was separated from the [32p]ATP as described for the inhibition assay.

211

2.6. Chemicals

Most drugs and chemicals were purchased from Sigma (St Louis, MO, USA). 8-OH-DPAT (8-hydroxy2-(di-n-propylamino)tetralin was obtained from Research Biochemicals (Natick, MA, USA) and porcine galanin from Peninsula Laboratories (Belmont, CA, USA). Methiothepin was a gift from F. Hoffmann-La Roche. Galanin-(1-15) was kindly provided by Prof. N. Yanaihara (University of Shizuoka, Japan) and the galanin receptor antagonists M35 and M40 were generous gifts from Prof. T. Bartfai (Stockholm University, Sweden). [a-32p]ATP (30 Ci/mmol) and [3H]cAMP (31 Ci/mmol) were purchased from Amersham (Buckinghamshire, UK) and New England Nuclear (Boston, MA, USA) respectively. 2. 7. Stat&tics

All experiments were done in triplicate for each point in one given experiment and 2-7 independent experiments were performed in each set. The results were analyzed by a one-factor ANOVA followed by Scheff's F-test.

ft.5. Effects of galanin peptides and 5 - H T / 8-OH-DPA T alone or in various combinations 3. Results

First the concentration-response effects of galanin(1-29), galanin-(1-15) and two chimeric peptides known as M35 (galanin-(1-12)-Pro-bradykinin-(2-9)) and M40 (galanin-(1-13)-(Pro)2-(Ala-Leu)2-Ala-NH2) putatively acting as galanin receptor antagonists in concentrations ranging from 1 nM-1 ~M on forskolin stimulated adenylyl cyclase activity were evaluated. The possible antagonistic properties of M35 (1 I~M) were also tested against galanin-(1-29) (1 /~M) in the hypothalamus and entorhinal cortex. In another set of experiments the effects of galanin-(1-29) and galanin-(115) (0.1-1 /zM) on basal adenylyl cyclase activity were evaluated. Furthermore, the concentration-response effects of 8-OH-DPAT (hippocampus and entorhinal cortex) and 5-HT (hypothalamus, due to the low density of 5-HTIA receptors in this region) in concentrations ranging from 1 nM-1 ~M on the forskolin stimulated adenylyl cyclase activity were evaluated. The effect of the nonselective 5-HT 1 receptor antagonist methiothepin (1 /~M) was also studied. The simultaneous action of galanin and 5-HT/8OH-DPAT in concentrations ranging from 1 nM-1 /~M on both forskolin stimulated and basal adenylyl cyclase activity was also evaluated in all three brain regions studied. In the hippocampus, the simultaneous effect of galanin-(1-15) and 8-OH-DPAT was also evaluated.

3.1. Effects o f galanin on adenylyl cyclase activity

Galanin-(1-29) showed a dose-dependent inhibitory effect on forskolin-stimulated adenylyl cyclase activity in hypothalamic and entorhinal cortical membranes (Fig. 1). In the hypothalamus, the maximal effect at 1 /zM of galanin-(1-29) was an approximately 14% reduction of the enzyme activity, and the corresponding effect in the entorhinal cortex was - 7 % . Galanin-(129) had no clear inhibitory effect on adenylyl cyclase activity in the hippocampus as a whole (Table 1), nor in the ventral or dorsal hippocampus when analyzed separately (data not shown). Galanin-(1-15) showed an inhibitory effect on the adenylyl cyclase in hypothalamic and entorhinal cortical membranes with a similar maximal effect as seen with galanin-(1-29) at 1/zM, but the affinity seemed to be somewhat lower since the concentration-response curve was shifted slightly to the right (Fig. 1). Galanin(1-15) had no clear inhibitory effect on the stimulated adenylyl cyclase activity in the hippocampus, neither when the hippocampus was analyzed as a whole (Table 1) nor when the ventral and dorsal parts were analyzed separately (data not shown). Neither galanin-(1-29) nor galanin-(1-15) had any effect on the basal unstimulated activity of adenylyl cyclase in any of the studied brain regions (Table 2).

A. Billecocq et al. / European Journal of Pharmacology

212

Hypothalamus

Molecular Pharmacology Section 269 (1994) 209-217

110-

Table 1 Effects of galanin-(1-29), galanin-(1-15), M35 and M40 on forskolin stimulated adenylyl cyclase activity in membrane preparations from the hippocampus

I O0 -

(~M)

90-

0.01 0.1 1

tO

"6

-

Concentration

.>

==

GAL-(1-29)

GAL-(1-15)

M35

M40

97.7+-0.7 96.7+-1.2 96.1+-1.2

98.5+-1.3 98.65:1.3 97.4+-1.1

n.d. 95.1+-1.9 96.15:1.4

n.d. 99.0+-1.0 97.3+-1.8

Values are given as mean + S.E.M. in percent of control. The amount of cAMP formed under control conditions was 39.5 +-5.1 pmol/mg protein/min n = 2-5 independent experiments performed in triplicate. N.D. = not determined.

80¢"0

70 -9

-

-

-6

Drug concentration [ log M]

Entorhinal cortex

3.2. Effects of 5-HT or 8-OH-DPA T on adenylyl cyclase activity

110-

8-OH-DPAT induced a dose-dependent inhibition of the forskolin stimulated adenylyl cyclase activity (Fig. 2) with a maximal effect at 1 ~ M o f - 19% and - 1 1 % in the hippocampus and the entorhinal cortex,

eo

-6 100-

*L .>_ ~

90. Table 2 Effects of 5-HT, galanin-(1-29) and galanin-(1-15) on basal adenylyl cyclase activity in membrane preparations from the hippocampus, entorhinal cortex and hypothalamus

u ~

M35 was unable to affect the inhibition of adenylyl cyclase activity induced by galanin-(1-29). Hence, neither counteractive nor additive effects were seen between galanin-(1-29) and M35 (data not shown).

80"

e.

Treatment

.;

.;

I#M

I~M+ METI~M

Hypothalamus 5-HT GAL-(1-29) 5-HT + GAL-(1-29)

102.4+0.8 102.3 5:1.6 105.1+0.7

106.05:1.1a 103.05:1.5 105.05:2.0

102.25:1.7

Entorhinal cortex 5-HT GAL-(1-29) 5-HT + GAL-(1-29)

110.5=1=0.1 a 100.45:1.7 109.85:1.0

112.9+1.5 a 102.1 + 1.5 114.2+-0.8

97.5+_1.1

Hippocampus 5-HT GAL-(1-29) GAL-(1-15) 5-HT + GAL-(1-29) 5-HT+ GAL-(1-15)

112.4+- 1.8 a 101.5 5:2.8 101.45:1.1 110.0+1.0 109.35:1.0

114.35:1.1 b 102.4+ 1.7 98.7+-1.0 115.2+-1.4 114.3+- 1.9

.;

Drug concentration [ log M] Fig. 1. Effect of galanin-(1-29) (O), galanin-(t-15) (o), M35 (zx) and M40 (o) on forskolin (10 /~M) stimulated adenylyl cyclase activity in membrane preparations from the hypothalamus and entorhinal cortex. The amounts of cAMP formed under control conditions were (pmol/mg protein/min) 51.7 + 5.5 and 31.3 + 3.2 for the hypothalamus and entorhinai cortex, respectively, n = 10 (galanin(1-29)) and 2-3 (other peptides) independent experiments performed in triplicate. In the hypothalamus P < 0.001 at 0.1/zM and 1 /zM for galanin-(1-29), M35 and M40, and in the entorhinal cortex P < 0.01 at 0.1 ~M for M35 and P < 0.001 at 0.1/~M for galanin-(129) and M40 and at 1 ~M for galanin-(1-29), M35 and M40 as compared with respective controls (Scheff's F-test).

M35 and M40 both inhibited the forskolin-stimulated adenylyl cyclase in the hypothalamus and entorhinal cortex (Fig. 1). M35 appeared to have the same affinity and potency as galanin-(1-29) with a maximal effect of approximately - 1 3 % and - 1 2 % , respectively, at 1 /zM. M40 was less potent and had a lower affinity. M35 and M40 had no effect on the adenylyl cyclase activity in the hippocampus (Table 1).

Concentration 0.1~M

99.45:1.7

96.9+-3.1 102.4+2.1

103.0+-1.0

The concentrations in the table head refer to both 5-HT and galanin(1-29) or galanin-(1-15). MET = methiothepin, a 5-HT 1 receptor antagonist. Values are given as means+S.E.M, in percent of the control. The amounts of cAMP formed under control conditions were (pmol/mg protein/min) 6875:135, 203 +50 and 180+43 for the hypothalamus, entorhinal cortex and hippocampus, respectively. n = 2-3 independent experiments, each performed in triplicate. a p < 0.01, b p < 0.001 as compared with respective controls (Sheff6's F-test). The effects of 5-HT are similar also in combination with galanin peptides.

A. Billecocq et aL /European Journal of Pharmacology - Molecular Pharmacology Section 269 (1994) 209-217

respectively. The same concentration of 5-HT induced an inhibition of - 1 2 % in the hypothalamus. In all regions the effects could be fully counteracted by 1 / , M of the 5-HT t receptor antagonist methiothepin (Fig. 2). 5-HT at concentrations 0.1 and 1 /zM was shown to stimulate the basal activity of adenylyl cyclase in the hippocampus and the entorhinal cortex by + 14% and + 13% (1 ~M), respectively (Table 2). The effects could be counteracted by 1 /zM methiothepin (Table

Hypothalamus 110o

8 ~0

213

tO0

2). 90 ¸

3.3. Effects of 5-HT / 8-OH-DPAT and galanin in combination on adenylyl cyclase activity

(J

80e-

"0

7O Drug concentration [ log M]

Entorhinal cortex llOe tO

°2 ._~ "~ 90o

80-

g 70

i

!

i

i

-9

-8

-7

-6

When galanin-(1-29) and 5-HT/8-OH-DPAT were incubated simultaneously, additive inhibitory effects could be observed on the stimulated adenylyl cyclase activity, but no synergistic interactions could be detected (Fig. 2). In contrast, in the hypothalamus and entorhinal cortex, at high concentrations of both receptor agonists, a slightly smaller inhibition than the theoretical one was observed, possibly due to a saturation of the transduction system (Fig. 2). In the hippocampus galanin-(1-15) had no additive effect on the one induced by 8-OH-DPAT (Fig. 2). In the combination experiments on stimulated adenylyl cyclase activity methiothepin was only able to block the effects induced by 5-HT or 8-OH-DPAT, whereas the effects of galanin remained unchanged (Fig. 2). When evaluating the basal adenylyl cyclase activity neither galanin-(1-29) nor galanin-(1-15) was able to influence the stimulatory effect of 5-HT, and the observed effects could be completely counteracted by methiothepin (Table 2).

Drug concentration [ log M]

Hippocampus 110 0 0

'.6 100

e-"

70

i

i

T

i

-9

-8

-7

-6

Drug conoehtration [ log M]

Fig. 2. Effects of 5-HT (12; hypothalamus), 8-OH-DPAT ([2; entorhinal cortex and hippocampus), galanin-(1-29) (11; all regions) and galanin-(1-15) (A; hippocampus) on forskolin (10/zM) stimulated adenylyl cyclase activity in membrane preparations. 5-HT/8OH-DPAT was also incubated together with 1 p.M of galanin-(1-29) (©). In the hippocampus 8-OH-DPAT was also incubated with 1/zM of g a l a n i n - ( 1 - 1 5 ) ( - ) . In the hypothalamus and entorhinal cortex galanin-(1-29) was also incubated with 1/zM of 5-HT/8-OH-DPAT (e). The ability of methiothepin (1 /zM) to counteract the effects of 5-HT/8-OH-DPAT (l~l) alone or in combination with 1 /zM of galanin-(1-29) ( [ ] ) is also shown. Values are given as mean + S.E.M. in percent of respective control values. The amounts of cAMP formed under control conditions were (pmoi/mg protein/min) 50.3 +12.8, 30.5+2.7 and 39.5:1:5.1 for the hypothalamus, entorhinal cortex and hippocampus respectively, n = 3-5 independent experiments each performed in triplicate. In the hypothalarnus and the entorhinal cortex P < 0.001 at 0.1 ~ M and 1 /~M for galanin-(1-29) and 5-HT as compared with respective controls. In the hippocampus P < 0.01 at 0.1/~M and P < 0.001 at 1/zM for 8-OH-DPAT (Scheffe's F-test).

214

A. Billecocq et al. / European Journal of Pharmacology - Molecular Pharmacology Section 269 (1994) 209-217

4. Discussion

The amounts of cAMP formed under control conditions are comparable to those previously reported both for the inhibition experiments and for the stimulation experiments (De Vivo and Maayani, 1986; Fayolle and Fillion, 1988) under the respective experimental conditions. However, there are differences in the amounts of cAMP formed under control conditions in the inhibition experiments compared to the stimulation experiments (see Table 1 and 2, and Fig. 1 and 2). These differences may be explained by e. g. differences in pH, Mg 2+ and GTP concentrations in the different assay mediums used. It should, however, be noted that the ability of forskolin to stimulate the cAMP formation was similar in both assay conditions. In this study, we show that galanin receptors may be negatively coupled to the adenylyl cyclase transduction pathway in discrete brain regions as has previously been demonstrated in membranes prepared from whole rat brain (Chen et al., 1992). However, the degree of coupling or the G protein activity level seems to be different in different brain regions, since galanin inhibited forskolin-stimulated adenylyl cyclase activity in membrane preparations from the hypothalamus and entorhinal cortex, whereas no clear effect could be seen in the hippocampus. A previous report has suggested such a regional heterogeneity in G protein coupling or activity level related to galanin receptor occupancy (Lagny-Pourmir and Epelbaum, 1992). Furthermore, a negative coupling is supported by binding studies suggesting a linkage of the galanin receptor with an inhibitory G protein (Fisone et al., 1989). This notion is also in agreement with the finding that no positive coupling between galanin receptors and adenylyl cyclase could be detected in the three brain regions studied. In the hippocampus the absence of a clear galanin effect on adenylyl cyclase activity may suggest that in this region the galanin receptor is not coupled to this transduction pathway. However, this is most likely not due to an absence of galanin receptors, since a high density of galanin binding sites have been observed, at least in the ventral hippocampus (Skofitsch et al., 1986; Melander et al., 1988). This would suggest that in this region the receptor is not coupled to the adenylyl cyclase transduction pathway. In fact, it has already been shown that these receptors may be coupled to another pathway, since galanin has been shown to inhibit the muscarinic stimulation of the phospholipase C (Palazzi et al., 1988), probably via a G i protein (Fisone et al., 1989). The existence of specific binding sites for the Nterminal fragment galanin-(1-15) with a unique distribution in the rat brain has recently been demonstrated (Hedlund et al., 1992). One notable difference is that these binding sites extend also into the dorsal part of

the hippocampus. However, it seems unlikely that this putative receptor subtype is linked to adenylyl cyclase, since we were unable to detect any effect of galanin(1-15) on adenylyl cyclase activity. This was found both in the hippocampus as a whole and when the ventral and dorsal parts were analyzed separately. N-terminal fragments such as galanin-(1-15) have been shown to be ligands of the hypothalamic [~25I]galanin binding site, although with a lower affinity than the full molecule (Land et al., 1991). Also galanin-(1-16) acts as an agonist at the hippocampal galanin receptor (Fisone et al., 1989). In the hypothalamus and the entorhinal cortex galanin-(1-15) is shown to act as a full agonist of the galanin receptor, since it inhibits the forskolin-stimulated adenylyl cyclase with the same efficacy as galanin itself, although with a lower affinity. The two putative galanin receptor antagonists (Bartfai et al., 1992) M35 and M40 in this study appear to be functional agonists of the galanin receptor, since they inhibit the stimulated adenylyl cyclase activity without antagonizing the action of galanin. The efficacy of M35 is comparable to that of galanin-(1-29), and that of M40 is comparable to that of galanin-(1-15), indicating that they are full agonists at the galanin receptor. The finding that when both galanin and M35 or M40, are present at concentrations that give maximal effects, no additive actions could be detected, supports the idea that M35 and M40 act at the galanin receptor and not at some other receptor. Agonistic properties of M35 have been described in other studies (Ogren et al., 1993; Hedlund et al., 1994). These properties may be due to the fact that the N-terminal part of these peptides is identical to galanin-(1-13) (Langel et al., 1992). Furthermore, differential actions of these chimeric peptides have been observed in several studies, possibly indicating the presence of multiple receptor subtypes for galanin (Gu et al., 1993; Wynick et al., 1993; 0gren et al., 1993). The different subclasses of 5-HT 1 receptors have been shown to be G protein coupled and associated with adenylyl cyclase; the subtype previously known as 5-HTlc is now classified as 5-HTzc (Humphrey et al., 1993; for a review, see Zifa and Fillion, 1992). A negative coupling has been well demonstrated for the 5-HT1A, 5-HT m and 5-HTlo receptors, and has been demonstrated in guinea-pig and rat hippocampus, and in rat and calf substantia nigra. Furthermore, in the rat and guinea-pig hippocampus as well as in the rat cortex, a positive coupling has been reported in some studies for a 5-HT receptor showing 5-HT1 characteristics, but where the specific subtype is not well established. Two subtypes might be involved, a 5-HTtA or a 5-HTlnonA, nonB,nonC receptor. It should be noted that in transfected cells the human and dog 5-HT~D receptor has been shown to both activate and inhibit adenylyl cyclase (Van Sande et al., 1993). Molecular biology

A. Billecocq et al. / European Journal of Pharmacology - Molecular Pharmacology Section 269 (1994) 209-217

techniques have also revealed new 5-HT receptor subtypes shown to mainly activate adenylyl cyclase, such as the 5-HT 7 (Bard et al., 1993; Ruat et al., 1993b) and the 5-HT6 receptor (Monsma et al., 1993; Ruat et al., 1993a), the latter with an apparent lower affinity. In the three brain regions analyzed, we have shown the presence of 5-HT~ receptors displaying a positive as well as a negative coupling to adenylyl cyclase. Although no extensive pharmacological characterization was performed all the effects observed could be counteracted by the 5-HT 1 receptor antagonist methiothepin. However, it should be noted that 8-OH-DPAT show a relatively high affinity for 5-HT 7 receptors (Shen et al., 1993) which can be found in the hypothalamus and hippocampus. Using the selective 5-HTIA receptor agonist 8-OHDPAT the 20% inhibition of forskolin stimulated adenylyl cyclase activity observed is comparable to the previously reported inhibition in the hippocampus (De Vivo and Maayani, 1986; Schoeffter and Hoyer, 1988; Oksenberg and Peroutka, 1988). In the entorhinal cortex the 5-HTxA binding sites were also negatively coupled to adenylyl cyclase, although a smaller inhibition was observed with a maximal effect of 10%. In the hypothalamus, the action of 8-OH-DPAT was very weak, but a 12% inhibition could be induced by 5-HT. This is in agreement with autoradiographical studies showing that the predominant 5-HT 1 subtype in the hypothalamus is the 5-HT1B subtype, and that the density of 5-HTIA receptors is very low (Pazos and Palacios, 1985). The 5-HT induced stimulation of adenylyl cyclase observed in hippocampus and the entorhinal cortex is in agreement with the reports of a positive coupling of some 5-HT~ receptors. The amplitude of the stimulation (15%) is close to or slightly lower than thepreviously reported stimulatory effects (Shenker et al., 1985; Fayolle et al., 1988; M6rk and Geisler, 1990). The stimulatory effect observed in the hypothalamus was very weak (6%) but nevertheless significant. One aim of the present study was to evaluate a possible involvement of adenylyl cyclase in the previously demonstrated intramembrane receptor interaction between 5-HTIA and galanin receptors (Fuxe et al., 1988a; Hedlund et al., 1991a; Hedlund et al., 1991b). In a recent study such an interaction has also been demonstrated for the putative galanin fragment receptor subtype and 5-HT~A receptors in the dorsal hippocampus (Hedlund et al., 1994). Hence, all the three brain regions studied have been shown to be involved in galanin/5-HTxA receptor interactions. Since both galanin and 5-HT1A receptors seem to be coupled to adenylyl cyclase, an interaction might take place at this level. However, no synergistic, only additive, effects could be observed when agonists for both receptors were incubated together. Although an earlier prelimi-

215

nary finding indicated that the interaction between galanin and 5-HT1A receptors is G protein independent (Hedlund et al., 1991a) the involvement of G proteins cannot be excluded. It is known that both galanin and 5-HT1A receptors are linked to other G protein coupled mechanisms, such as phospholipase C activity and K ÷ channels (see Zifa and Fillion, 1992; Bartfai et al., 1993). Receptor-receptor interactions can also be explained by other mechanisms than the involvement of membrane associated proteins (such as G proteins), e. g. interactions can take place within the plasma membrane itself or involve intracellular processes (Zoli et al., 1993). Hence, the synergistic interactions which have been observed at the network level (Fuxe et al., 1990; Hedlund et al., 1991b) seem not to directly involve adenylyl cyclase but might be explained by one or more of the other possible mechanisms for receptor interactions. In conclusion, we have demonstrated that in the hypothalamus and the entorhinal cortex, but not clearly in the hippocampus, galanin receptors are negatively linked to the adenylyl cyclase transduction pathway. Furthermore, we have shown that there is no positive coupling in any of these regions. In addition, we have demonstrated an inhibitory role for 5-HT1A or 5HTIA/B/D receptors on stimulated adenylyl cyclase activity and a positive linkage to basal activity in these regions. When incubated together, the galanin and 5-HT~A receptor agonists showed additive inhibitory effects, but no synergistic effects. Hence, galanin and 5-HT~A receptors seem to be linked to different independent pools of G proteins and, furthermore, this indicates that the previously demonstrated intra-membrane interactions between galanin and 5-HT~A receptors involve a mechanism not directly related to adenylyl cyclase.

Acknowledgements This study has been supported by grants from the Swedish Medical Research Council, the Swedish Society for Medicine and Karolinska Institutet Research Funds (PBH).

References Amiranoff, B., A.-M. Lorinet, I. Lagny-Pourmirand M. Laburthe, 1988, Mechanismof galanin-inhibitedinsulinrelease: Occurrence of a pertussis-toxin-sensitiveinhibitionof adenylatecyclase.,Eur. J. Biochem.177, 147. Barbaccia, M.L., N. Brunello, D.M. Chuang and E. Costa, 1983, Serotonin-elicited amplificationof adenylate cyclase activityin hippocampal membranesfrom adult rat, J. Neurochem.40, 1671. Bard, J.A., J. Zgombick,N. Adham, P. Vaysse,T.A. Branchek and R.L. Weinshank, 1993, Cloning of a novel human serotonin

216

A. Billecocq et al. / European Journal of Pharmacology - Molecular Pharmacology Section 269 (1994) 209-217

receptor (5-HT 7) positively linked to adenylate cyclase, J. Biol. Chem. 268, 23422. Bartfai, T., G. Fisone and I]. Langel, 1992, Galanin and galanin antagonists: molecular and biochemical perspectives, Trends Pharmacol. Sci. 13, 312. Bartfai, T., T. H6kfelt and 0 . Langel, 1993, Galanin - a neuroendocrine peptide, Crit. Rev. Neurobiol. 7, 229. Bockaert, J., A. Dumuis, R. Bouhelal, M. Sebben and R.N. Cory, 1987, Piperazine derivatives including the putative anxiolytic drugs, buspirone and ipsapirone, are agonists at 5-HTIA receptors negatively coupled with adenylate cyclase in hippocampal neurons, Naunyn Schmiedebergs Arch. Pharmacol. 335, 588. Chen, Y., M. Laburthe and B. Amiranoff, 1992, Galanin inhibits adenylate cyclase of rat brain membranes, Peptides 13, 339. De Vivo, M. and S. Maayani, 1986, Characterization of the 5-hydroxytryptaminelA receptor-mediated inhibition of forskolinstimulated adenylate cyclase activity in guinea pig and rat hippocampal membranes, J. Pharmacol. Exp. Ther. 238, 248. Dumuis, A., M. Sebben and J. Bockaert, 1988, Pharmacology of 5-HT1A receptors which inhibit cAMP production in hippocampal and cortical neurons in primary cultures, Mol. Pharmacol. 33, 178. Fayolle, C. and G. Fillion, 1988, Reversed-phase liquid chromatographic determination of cyclic adenosine 3', 5'-monophosphate in rat brain cortex, J. Chromatography 426, 177. Fayolle, C., M.P. Fillion, P. Barone, P. Oudar, J.C. Rousselle and G. Fillion, 1988, 5-Hydroxytryptamine stimulates two distinct adenylate cyclase activities in rat brain: high-affinity activation is related to a 5-HT 1 subtype different from 5-HTIA, 5-HTm, and 5-HTtc, Fundam. Clin. Pharmacol. 2, 195. Fisone, G., lJ. Langel, M. Carlquist, T. Bergman, S. Consolo, T. H6kfelt, A. Und6n, S. Andell and T. Bartfai, 1989, Galanin receptor and its ligands in the rat hippocampus, Eur. J. Biochem. 181,269. Fuxe, K., G. von Euler, L.F. Agnati and S.O. tDgren, 1988a, Galanin selectively modulates 5-hydroxytryptamine 1A receptors in the rat ventral limbic cortex, Neurosci. Lett. 85, 163. Fuxe, K., S.-O. Ogren, A. Jansson, A. Cintra, A. H~iffstrand and L.F. Agnati, 1988b, Intraventricular injections of galanin reduces 5-HT metabolism in the ventral limbic cortex, the hippocampal formation and the fronto-parietal cortex of the male rat, Acta Physiol. Scand. 133, 579. Fuxe, K., L.F. Agnati, G. yon Euler, K. Lundgren, M. Zoli, B. Bjelke, P. Eneroth and S.-O. Ogren, 1990, Galanin/5-HT receptor interactions. A new integrative mechanism in the control of 5-HT neurotransmission in the CNS, in: Serotonin from cell biology to pharmacology and therapeutics, eds. R. Paoletti and P.M. Vanhoutte (Kluwer Academic Publ., Dordrecht) p. 169. Gentleman, S.M., P. Falkai, B. Bogerts, M.T. Herrero, J.M. Polak and G.W. Roberts, 1989, Distribution of galanin-like immunoreactivity in the human brain, Brain Res. 505, 311. Gu, Z.F., W.J. Rossowski, D.H. Coy, T.K. Pradhan and R.T. Jensen, 1993, Chimeric galanin analogs that function as antagonists in the CNS are full agonists in gastrointestinal smooth muscle, J. Pharmacol. Exp. Ther. 266, 912. Hedlund, P., G. von Euler and K. Fuxe, 1991a, Activation of 5-hydroxytryptamineiA receptors increases the affinity of galanin receptors in di- and telencephalic areas of the rat, Brain Res. 560, 251. Hedlund, P.B., J.A. Aguirre, J.A. Narvaez and K. Fuxe, 1991b, Centrally coinjected galanin and a 5-HT1A agonist act synergistically to produce vasodepressor responses in the rat, Eur. J. Pharmacol. 204, 87. Hedlund, P.B., N. Yanaihara and K. Fuxe, 1992, Evidence for specific N-terminal galanin fragment binding sites in the rat brain., Eur. J. Pharmacol. 224, 203. Hedlund, P.B., U.-B. Finnman, N. Yanaihara and K. Fuxe, 1994,

Galanin-(1-15), but not galanin-(1-29), modulates 5-HTlA receptors in the dorsal hippocampus of the rat brain: possible existence of galanin receptor subtypes, Brain Res. 634, 163. Humphrey, P.P., P. Hartig and D. Hoyer, 1993, A proposed new nomenclature for 5-HT receptors, Trends Pharmacol. Sci. 14, 233. Lagny-Pourmir, I. and J. Epelbaum, 1992, Regional stimulatory and inhibitory effects of guanine nucleotides on [12SI]galanin binding in rat brain: relationship with the rate of occupancy of galanin receptors by endogenous galanin, Neuroscience 49, 829. Land, T., U. Langel and T. Bartfai, 1991, Hypothalamic degradation of galanin(1-29) and galanin(1-16): identification and characterization of the peptidolytic products, Brain Res. 558, 245. Langel, 13., T. Land and T. Bartfai, 1992, Design of chimeric peptide ligands to galanin receptors and substance P receptors, Int. J. Pept. Protein Res. 39, 516. Markstein, R., D. Hoyer and G. Engel, 1986, 5-HT1A receptors mediate stimulation of adenylate cyclase in rat hippocampus, Naunyn Schmiedebergs Arch. Pharmacol. 333, 335. Martire, M., K. Fuxe, L.F. Agnati, G. Pistritto and P. Preziosi, 1991, Galanin increases potassium evoked release of [3H]5-hydroxytryptamine from rat hypothalamic synaptosomal preparations, Neurosci. Lett. 122, 87. Melander, T., T. H6kfelt and ,~. R6kaeus, 1986a, Distribution of galanin-like immunoreactivity in the rat central nervous system, J. Comp. Neurol. 248, 475. Melander, T., T. H6kfelt, ,~. R6kaeus, A.C. Cuello, W.H. Oertel, A. Verhofstad and M. Goldstein, 1986b, Coexistence of galanin-like immunoreactivity with catecholamines, 5-hydroxytryptamine, GABA and neuropeptides in the rat CNS, J. Neurosci. 6, 3640. Melander, T., C. K6hler, S. Nilsson, T. H6kfelt, E. Brodin, E. Theodorsson and T. Bartfai, 1988, Autoradiographical quantitation and anatomical mapping of lzSI-galanin binding sites in the rat central nervous system, J. Chemical Neuroanatomy 1, 213. Monsma Jr., F.J., Y. Shen, R.P. Ward, M.W. Hamblin and D.R. Sibley, 1993, Cloning and expression of a novel serotonin receptor with high affinity for tricyclic psychotropic drugs, Mol. Pharmacol. 43, 320. M6rk, A. and A. Geisler, 1990, 5-HT receptor agonists influence calcium-stimulated adenylate cyclase activity in the cerebral cortex and hippocampus of the rat, Eur. J. Pharmacol. 175, 237. Nishibori, M., R. Oishi, Y. Itoh and K. Saeki, 1988, Galanin inhibits noradrenaline-induced accumulation of cyclic AMP in the rat cerebral cortex, J. Neurochem. 51, 1953. tDgren, S.O., A. Pramanik, T. Land and r~. Langel, 1993, Differential effects of the putative galanin receptor antagonists M15 and M35 on striatal acetylcholine release, Eur. J. Pharmacol. 242, 59. Oksenberg, D. and S.J. Peroutka, 1988, Antagonism of 5-HT1A receptor mediated modulation of adenylate cyclase activity by pindolol and propranolol isomers, Biochem. Pharmacol. 37, 3429. Palazzi, E., G. Fisone, T. H6kfelt, T. Bartfai and S. Consolo, 1988, Galanin inhibits the muscarinic stimulation of phosphoinositide turnover in rat ventral hippocampus, Eur. J. Pharmacol. 148, 479. Pazos, A. and J.M. Palacios, 1985, Quantitative autoradiographic mapping of serotonin receptors in the rat brain. I. Serotonin-1 receptors, Brain Res. 346, 205. Ruat, M., E. Traiffort, J.M. Arrang, J. Tardivel-Lacombe, J. Diaz, R. Leurs and J.C. Schwartz, 1993a, A novel rat serotonin (5-HT 6) receptor: molecular cloning, localization and stimulation of cAMP accumulation, Biochem. Biophys. Res. Commun. 193, 268. Ruat, M., E. Traiffort, R. Leurs, J. Tardivel-Lacombe, J. Diaz, J.M. Arrang and J.C. Schwartz, 1993b, Molecular cloning, characterization, and localization of a high-affinity serotonin receptor (5-HT 7) activating cAMP formation, Proc. Natl. Acad. Sci. USA 90, 8547. R6kaeus, A., T. Melander, T. H6kfelt, J.M. Lundberg, K. Tatemoto, M. Carlquist and V. Mutt, 1984, A galanin-like peptide in the

A. Billecocq et al. / European Journal of Pharmacology - Molecular Pharmacology Section 269 (1994) 209-217 central nervous system and intestine of the rat, Neurosci. Lett. 47, 161. Salomon, Y., 1979, Adenylate cyclase assay, Adv. Cyclic Nucleotide Res. 10, 35. Schoeffter, P. and D. Hoyer, 1988, Centrally acting hypotensive agents with affinity for 5-HT1A binding sites inhibit forskolinstimulated adenylate cyclase activity in calf hippocampus, Br. J. Pharmacol. 95, 975. Shen, Y., F.J. Monsma Jr., M.A. Metcalf, P.A. Jose, M.W. Hamblin and D.R. Sibley, 1993, Molecular cloning and expression of a 5-hydroxytryptamine 7 serotonin receptor subtype, J. Biol. Chem. 268, 18200. Shenker, A., S. Maayani, H. Weinstein and J.P. Green, 1985, Two 5-HT receptors linked to adenylate cyclase in guinea-pig hippocampus are discriminated by 5-carboxamidotryptamine and spiperone, Eur. J. Pharmacol. 109, 427. Skofitsch, G. and D.M. Jacobowitz, 1985, Immunohistochemical mapping of galanin-like neurons in the rat central nervous system, Peptides 6, 509. Skofitsch, G., M. Sills and D. Jacobowitz, 1986, Autoradiographic distribution of 125I-galanin binding sites in the rat central nervous system, Peptides 7, 1029. Sundstr6m, E. and T. Melander, 1988, Effects of galanin on 5-HT neurons in the rat CNS, Eur. J. Pharmacol. 146, 327.

217

Tatemoto, K., ,~. R6kaeus, H. J6rnvall, T.J. McDonald and V. Mutt, 1983, Galanin - a novel biologically active peptide from porcine intestine, FEBS Lett. 164, 124. Van Sande, J., A. Allgeier, C. Massart, A. Czernilofsky, G. Vassart, J.E. Dumont and C. Maenhaut, 1993, The human and dog 5-HTlo receptor can both activate and inhibit adenylate cyclase in transfected cells, Eur. J. Pharmacol. Mol. Pharmacol. Sect. 247, 177. Weiss, S., M. Sebben, D.E. Kemp and J. Bockaert, 1983, Serotonin 5-HT 1 receptors mediate inhibition of cyclic AMP production in neurons, Eur. J. Pharmacol. 120, 227. Wynick, D., D.M. Smith, M. Ghatei, K. Akinsanya, R. Bhogal, P. Purkiss, P. Byfield, N. Yanaihara and S.R. Bloom, 1993, Characterization of a high-affinity galanin receptor in the rat anterior pituitary: Absence of biological effect and reduced membrane binding of the antagonist M15 differentiate it from the brain/gut receptor, Proc. Natl. Acad. Sci. USA 90, 4231. Zifa, E. and G. Fillion, 1992, 5-hydroxytryptamine receptors, Pharmacol. Rev. 44, 401. Zoli, M., L.F. Agnati, P.B. Hedlund, X.M. Li, S. Ferr~ and K. Fuxe, 1993, Receptor-receptor interactions as an integrative mechanism in nerve cells, Mol. Neurobiol. 7, 293.