Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Ascending serotoninergic pathways from the DRN innervate about 50% of dopaminergic neurones in the VTA (Hervé et al., 1987) and provide a major dopaminergic input to the (SNPC) (Dray et al., 1978; Vertes, 1991). Further, serotoninergic terminals form excitatory synapses with VTA-localized dopaminergic neurons (Van Bockstaele et al., 1994) and 5-HT_1A receptors have been implicated in the modulatory influence of serotoninergic pathways on dopaminergic transmission. However, 5-HT_1A receptors are virtually absent from the SNPC (Miquel et al., 1991) and their concentration in the VTA itself is modest (Pazos and Palacios, 1985). Accordingly, the ability of systemic administration of the 5-HT_1A agonist, 8-OH-DPAT, to increase DA turnover in the VTA (Ahlenius et al., 1990; Chen and Reith, 1995) and to facilitate DA release in the striatum (Benloucif et al., 1993) may reflect activation of inhibitory 5-HT_1A autoreceptors on serotoninergic perikarya localized in the DRN. Indeed, electrolytic destruction of raphe nuclei accelerates DA turnover in both the accumbens (Hervé et al., 1987) and the striatum (Giambalo and Snodgrass, 1978) whereas stimulation of raphe neurones inhibits the activity of SNPC-localized dopaminergic neurones (Dray et al., 1978). However, De Simoni et al. (1987) suggested that DRN stimulation enhanced the activity of dopaminergic pathways in the striatum. Indeed, there are also other contradictory data concerning the influence of 8-OH-DPAT on dopaminergic transmission. For example, it has both inhibitory and stimulatory influences upon VTA dopaminergic neurones in biochemical (Arborelius et al., 1993a; Ahlenius et al., 1990; Ichikawa et al., 1995) and electrophysiological studies (Arborelius et al., 1993b; Prisco et al., 1994). Further, microinjection of 8-OH-DPAT (or 5-HT) into the DRN decreases extracellular levels of DA in the nucleus accumbens (Yoshimoto and McBride, 1992). Finally, as concerns the influence of local administration of 8-OH-DPAT into the VTA on the activity of dopaminergic neurones, there are reports of both an increase in electrical activity (Arborelius et al., 1993a) and of a lack of effect (Zhang and Freeman, 1993).

Interestingly [with two exceptions (Arborelius et al., 1993b; Ichikawa et al., 1995)] the above studies were performed with racemic 8-OH-DPAT, which may be resolved into (+)- and (-)-isomers. Although these do not present a marked difference in terms of their affinity at rat postsynaptic 5-HT_1A sites, the (+)-isomer possesses greater intrinsic activity (Cornfield et al., 1991). Currently, no information is available concerning their relative activities at presynaptic 5-HT_1A receptors. In addition, racemic 8-OH-DPAT exerts significant (partial agonist) activity at dopamine (D)₂ receptors (Gobert et al., 1995a; Smith and Cutts, 1989), which control the activity of ascending dopaminergic projections (see Gobert et al., 1995b). To date, the activity of both (+)- and (-)-8-OH-DPAT at hD₂ sites remains undefined. Further, the closely related dopamine hD₃ (auto)receptor may also be involved in the modulation of the activity of mesolimbic dopaminergic neurones (Bergstrom et al., 1994; Gobert et al., 1995b; Tang et al., 1994) and the putative activity of 8-OH-DPAT at these sites is unknown. It is, thus, intriguing that the aminotetralin, 7-OH-DPAT, which is closely related to 8-OH-DPAT, has been extensively used as a "selective agonist" at hD₃ receptors (Sokoloff et al., 1992; Sokoloff and Schwartz, 1995), although its selectivity for hD₃ vs. hD₂ sites may be less than originally claimed (Burris et al., 1995; Chio et al., 1993; Gonzalez and Sibley, 1995). D₂ as well as D₃ (auto)receptors may contribute to the ability of 7-OH-DPAT to inhibit the electrical activity of VTA dopaminergic neurones and to suppress DA release and turnover in the nucleus accumbens, striatum and cortex (Gobert et al., 1995b and 1996; Lejeune and Millan, 1995). Concerning the stereospecificity of its actions, the (+)-isomer possesses markedly higher affinity for hD₃ sites than the (-)-isomer (Damsma et al., 1993; Malmberg et al., 1994; Newman-Tancredi et al., 1995) and more potently reduces mesolimbic dopamine release (Rivet et al., 1994).

In view of the striking structural homology between 7- and 8-OH-DPAT, an important question that arises is the selectivity of (+)- and (-)-7-OH-DPAT for hD₃ vs. 5-HT_1A sites. Indeed, the influence of 7-OH-DPAT on the activity of DRN neurones has not been documented. This question is also of interest in the light of scarce information concerning a possible reciprocal influence of dopaminergic pathways upon serotoninergic perykarya themselves. Despite the apparent dopaminergic innervation of the DRN by projections from the VTA (Kalén et al., 1988) and the substantia nigra (Beckstead et al., 1979; Stern et al., 1981), and the presence therein of D₂-like receptors (Bouthenet et al., 1987; Palacios and Pazos, 1987), only few and contradictory data are available as concerns a putative dopaminergic modulation of the DRN (Ferré and Artigas, 1993; Ferré et al., 1994; Lee and Geyer, 1984). Our study was carried out to clarify the comparative influence of (+)- and (-)-isomers of 8- and 7-OH-DPAT on the activity of DRN-localized serotoninergic as compared to VTA-localized dopaminergic neurones. First, we determined the in vitro affinities of (+)- and (-)-7- and 8-OH-DPAT at cloned, human (h)5-HT_1A vs. hD₃ and hD₂ receptors. Second, their in vitro efficacies for stimulation of h5-HT_1A receptor-mediated G-protein activation was determined by [³⁵S]GTPS binding (Newman-Tancredi et al., 1996). Third, their in vivo ability to modulate the electrical activity of DRN-localized serotoninergic neurones as compared with VTA-localized dopaminergic cells was determined. Fourth, we examined the influence of the selective 5-HT_1A antagonist, (-)-tertatolol (Jolas et al., 1993; Lejeune et al., 1994) which possesses negligible (K_i > 10 µM) affinity at hD₂ and hD₃ receptors (see "Results"), and of the selective hD₂/hD₃ antagonist, haloperidol, respectively, on their actions (Lejeune et al., 1994; Lejeune and Millan, 1995; Millan et al., 1994, 1995).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Binding at cloned, human 5-HT_1A receptors. Binding was carried out as previously described (Newman-Tancredi et al., 1992) using [³H]8-OH-DPAT (212 Ci/mmol, Amersham corp., Arlington Heights, IL) as a radioligand. Membranes (20 µg protein) prepared from CHO cells stably transfected with the cloned human 5-HT_1A receptor (CHO-h5-HT_1A) were incubated (2.5 hr, 22°C) in triplicate with competing ligands in a buffer containing HEPES 20 mM (pH 7.5) and MgSO₄ 5 mM. Nonspecific binding was defined with 10 µM 5-HT.

Binding at cloned, human dopamine hD₂ and hD₃ receptors. Binding was carried out as previously described (Millan et al., 1994; Sokoloff et al., 1992). Membranes prepared from CHO cells stably transfected with cDNA coding for human hD₂ and hD₃ receptors were incubated with [¹²⁵I]-iodosulpride at 0.1 and 0.2 nM for hD₂ and hD₃ sites, respectively. Nonspecific binding was defined with raclopride (10 µM) and specific binding was >90% for both hD₂ and hD₃ sites. Inhibitory concentration₅₀ (IC₅₀) were calculated by nonlinear regression analysis and K_i were computed according to K_i = IC₅₀/(1+L/K_D) where L is the concentration of radioligand and K_D is its apparent dissociation constant.

Electrophysiological analysis. Male Wistar rats (Iffa Credo, Illskirchen, France), weighing 275 to 325 g, were anesthetized with chloral hydrate (400 mg/kg, i.p.) and mounted in a stereotaxic apparatus after femoral vein cannulation. Additional doses of chloral hydrate were administered i.p. to maintain surgical anesthesia throughout the experiment. Rectal temperature was maintained at 37 ± 1°C using a homeothermic heating pad. A burr hole was made over the VTA or the DRN. A tungsten microelectrode was lowered, according to the atlas of Paxinos and Watson (1986), into the VTA (AP: -5.5 from bregma, L: 0.7, H: -7/-8.5 from dura) or the DRN (AP: -7.8 from bregma, L: 0, H: -5/-6.5 from dura) for recording of extracellular unit activity. Neurones were identified as described elsewhere [Wang (1981) for dopaminergic and Aghajanian et al. (1978) for serotoninergic neurones[. Only one cell was studied in each animal. After base-line recording (5 min) and a first injection of vehicle, drugs were administered by i.v. route in increasing cumulative doses for evaluation of dose-response relationships. For agonist-antagonist interactions, one dose of antagonist was given after a single dose of agonist. The interval between injections was 2 to 5 min. The peak drug action was attained within 1 to 2 min after injection for all drugs excepted haloperidol that had a delay time to maximal efficacy between 3 to 5 min. Drugs were dissolved in sterile water and injected i.v. in a volume of 0.5 ml/kg, followed by 0.1 ml of saline to flush the catheter. Data acquisition was accomplished by using Spike2 software (C.E.D., Cambridge, England). Interspike time interval histograms were calculated over 500 consecutive spikes as previously described (Arborelius et al., 1993b). Burst firing was the percentage ration of spikes in bursts to the total number of spikes. Results were expressed as firing rate (60-sec bins at time of peak drug action) in percent change from baseline (= mean of predrug values) and presented as means ± S.E.M. (3  n  8). Data were analyzed by two-way analysis of variance followed by Newman-Keuls test (for paired data) or Student's t test (for unpaired data). Inhibitory dose (ID)₅₀ and 95% confidence limits were calculated.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

In vitro binding affinities of agonists and antagonists at cloned h5-HT_1A, hD₂ and hD₃ receptors. The affinities of (±)-8-OH-DPAT, (+)-8-OH-DPAT and (-)-8-OH-DPAT at cloned h5-HT_1A receptors were almost identical (table 1). Racemic 8-OH-DPAT, as well as its (+)- and (-)-isomers, all showed markedly lower affinity at hD₂ and hD₃ receptors. However, at both these sites, the (+)-isomer showed about 2-fold higher affinity than the racemate and about 10-fold higher affinity than the (-)-isomer. (+)-7-OH-DPAT displayed higher affinity than (-)-7-OH-DPAT at both hD₃ and hD₂ receptors. Notably, the affinity of each of these isomers was even higher at h5-HT_1A than at hD₂ receptors. The separation was only 5-fold for (-)-7-OH-DPAT at hD₃ vs. h5-HT_1A sites. Haloperidol displayed high affinity for hD₂ receptors and about 5-fold lower affinity for hD₃ sites with its affinity at h5-HT_1A receptors being very low. In contrast, (-)-tertatolol manifested high affinity at h5-HT_1A receptors and negligible affinity at hD₂ and hD₃ receptors.

Influence of (+)- and (-)-7- and 8-OH-DPAT on G-protein activation. (+)-8-OH-DPAT potently and markedly activated h5-HT_1A receptor-mediated stimulation of [³⁵S]GTPS binding with a maximal effect of 90 ± 2.3% relative to that of 5-HT (fig. 1) and an EC₅₀ of 11.7 ± 2.6 nM. (-)-8-OH-DPAT similarly stimulated [³⁵S]GTPS binding with a potency similar (EC₅₀ = 10.3 ± 3.2 nM) to that of its (+)-counterpart, though with significantly (P < .05) less efficacy (57.3 ± 6.2%). (±)-8-OH-DPAT had intermediate efficacy (77.8 ± 5.8%). As concerns 7-OH-DPAT, in line with its higher affinity at h5-HT_1A receptors, the (+)-isomer more potently stimulated [³⁵S]GTPS binding than the (-)-isomer (EC₅₀ were 1,250 ± 160 nM and 10,300 ± 4,400 nM, respectively). It also showed somewhat higher efficacy (76.2 ± 5.7% vs. 53.1 ± 6.6%), although this difference was not significant (fig. 1).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Stimulation of h5-HT1A receptor-activated [35S]-GTPS binding by (+)- and (-)-7- and 8-OH-DPAT. Agonist efficacy is expressed relative to 5-HT-induced stimulation (=100%).

Influence of (±), (+) and (-) 8-OH-DPAT on the electrical activity of DRN serotoninergic neurones. 8-OH-DPAT potently and markedly inhibited the firing rate of DRN cells with the (-)-isomer displaying about 3-fold lower potency and the racemate showing intermediate activity as compared with the (+)-isomer (table 2; fig. 2). Indeed, the inhibitory actions of both (+)- and (-)-8-OH-DPAT were completely antagonized (figs. 4 and 5) by administration of the 5-HT_1A antagonist, (-)-tertatolol (2 mg/kg) that did not itself modify the firing rate of these DRN neurones (see also Lejeune et al., 1994). Haloperidol did not affect their actions and did not modify DRN firing rate alone (fig. 5).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Dose-dependent effects of 8-OH-DPAT (upper graph) and 7-OH-DPAT (lower graph) isomers on firing rates of serotoninergic neurones of DRN (open symbols) and dopaminergic neurones of VTA (filled symbols). Data are means ± S.E.M. (n  4 recorded cells in each condition). Analysis of variance with repeated measures as follows: DRN, (±)-8-OH-DPAT, F (5,30) = 112.67, P < .001; DRN, (+)-8-OH-DPAT, F (5,20) = 49.8, P < .001; DRN, (-)-8-OH-DPAT, F (4,20) = 45.9, P < .001; VTA, (±)-8-OH-DPAT, F (9,27) = 7.7, P < .001; VTA, (+)-8-OH-DPAT, F (8,24) = 3.5, P < .02; VTA, (-)-8-OH-DPAT, F (9,27) = 2.0, P > .05; DRN, (+)-7-OH-DPAT, F (5,20) = 76.4, P < .001; DRN, (-)-7-OH-DPAT, F (5,15) = 93.6, P < .001; VTA, (+)-7-OH-DPAT, F (5,20) = 141.4, P < .001; VTA, (-)-7-OH-DPAT, F (6,18) = 51.4, P < .001. Asterisks indicate significance (P < .05) of differences to vehicle in Newman-Keuls test (paired data). For clarity, asterisks are omitted for the 8-OH-DPAT-DRN curves for which all values, except the first, were significantly (P < .05) different from vehicle.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Modulation of spontaneous unit activity of DRN serotoninergic and VTA dopamergic neurones by (+)- and (-)-isomers of 8- and 7-OH-DPAT. 1, Antagonism of (+)-8-OH-DPAT, 5 µg/kg ((+)-8)-induced inhibition of DRN firing by (-)-tertatolol, 2 mg/kg (TER) and not haloperidol (HAL), 31 µg/kg. 2, Antagonism of (+)-8-OH-DPAT, 40 µg/kg-induced stimulation of VTA firing by (-)-tertatolol, 2 mg/kg and not haloperidol 16 µg/kg. 3, Antagonism of (-)-8-OH-DPAT, 2.5 mg/kg-induced inhibition of VTA firing by haloperidol, 16 µg/kg and not (-)-tertatolol, 2 mg/kg. 4, antagonism of (+)-7-OH-DPAT, 5 µg/kg-induced inhibition of VTA firing by haloperidol, 16 µg/kg but not (-)-tertatolol, 2 mg/kg. 5, Antagonism of (+)-7-OH-DPAT, 125 µg/kg-induced inhibition of DRN firing by (-)-tertatolol, 2 mg/kg, not haloperidol, 31 µg/kg.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Antagonism of a single, maximally effective dose of (+)- and (-)-isomers of 8- and 7-OH-DPAT. VEH, vehicle; TER, (-)- tertatolol and HAL, haloperidol. Doses are the same as in figure 3 plus (-)-7-OH-DPAT (160 µg/kg for DRN and 1000 µg/kg for VTA). Data are means ± S.E.M. (n  3 recorded cells in each condition). Analysis of variance was as follows: DRN, (+)-8-OH-DPAT, F (2,12) = 22.6, P < .001; DRN, (-)-8-OH-DPAT, F (2,12) = 21.2, P < .001; VTA, (+)-8-OH-DPAT, F (2,13) = 10.8, P < .002; VTA, (-)-8-OH-DPAT, F (2,14) = 17.8, P < .001; DRN, (+)-7-OH-DPAT, F (2,13) = 22.3, P < .001; DRN, (-)-7-OH-DPAT, F (2,6) = 618.4, P < .001; VTA, (+)-7-OH-DPAT, F (2,16) = 74.3, P < .001; VTA, (-)-7-OH-DPAT, F (2,13) = 42.4, P < .001. Asterisks indicate significance (P < .01) of differences in Dunnett's test as compared with respective vehicle values.

Influence of (±)-, (+)- and (-)-8-OH-DPAT on the electrical activity of VTA dopaminergic neurones. At a higher dose range (>10 µg/kg), racemic 8-OH-DPAT elicited a dose-dependent and biphasic influence on dopaminergic VTA cell activity with an increase in firing at modest doses and a decrease at high doses (fig. 2). At these modest doses, (+)-8-OH-DPAT increased VTA firing rate and transformed the regular firing pattern into an irregular burst mode (fig. 3); the percentage of bursts increased after (+)-8-OH-DPAT treatment from 20.72 ± 8.62 to 40.45 ± 10.81 (P = .026, paired Student's t test). However, (-)-8-OH-DPAT only inhibited the activity of VTA dopaminergic neurones (fig. 2) at high doses (>0.25 mg/kg) and did not induce a bursting pattern in regular firing cells (not shown). The stimulation of VTA activity by (+)-8-OH-DPAT (40 µg/kg), as well as the induction of a bursting pattern, were blocked by (-)-tertatolol (2 mg/kg; % bursts = 27.26 ± 9.34, P = .014 vs. (+)-8-OH-DPAT and P = .113 vs. baseline, paired Student's t test) but not influenced by haloperidol (16 µg/kg) (figs. 4 and 5). In contrast, (-)-tertatolol (2 mg/kg) did not affect the inhibition of VTA firing induced by (-)-8-OH-DPAT (2.5 mg/kg) whereas this action was antagonized by haloperidol (16 µg/kg) (figs. 4 and 5). Although haloperidol slightly stimulated VTA firing rate alone, (-)-tertatolol was without effect (fig. 5).

View larger version (31K): [in this window] [in a new window]   Fig. 3.   (+)-8-OH-DPAT augments burst firing activity in dopaminergic neurones of the VTA. Upper panel, Regular firing pattern before treatment. Middle panel, 2 min after (+)-8-OH-DPAT (40 µg/kg, i.v.), the interspike interval histogram reflects the increased proportion of burst firing. Lower panel, 3 min after (-)-tertatolol (2000 µg/kg, i.v.), the firing mode has returned to a regular pattern. The three traces were from the same neurone; (-)-tertatolol was injected 2 min after (+)-8-OH-DPAT

Influence of (+)- and (-)-7-OH-DPAT on the electrical activity of VTA dopaminergic neurones. Both (+)-7-OH-DPAT and (-)-7-OH-DPAT induced a dose-dependent and complete inhibition of the firing rate of DA cells in the VTA (fig. 2), the (+)-isomer displaying about 20-fold higher potency than its (-) counterpart (table 2). Haloperidol abolished this action of (+)- and (-)-7-OH-DPAT whereas (-)-tertatolol was inactive (figs. 4 and 5).

Influence of (+)- and (-)-7-OH-DPAT on the electrical activity of DRN serotoninergic neurones. Over a higher dose range, both (+)-7-OH-DPAT and (-)-7-OH-DPAT also dose-dependently and completely inhibited the firing rate of serotoninergic cells in the DRN (table 2; fig. 2) with the (+)- isomer being only 7-fold more potent than the (-). This inhibition was antagonized by (-)-tertatolol and unaffected by haloperidol (figs. 4 and 5).

Correlation analysis between in vitro affinities and in vivo potencies. Across all ligands tested, there was a highly significant correlation between 5-HT_1A affinity and ID_50s for inhibition of DRN serotoninergic cell firing and between hD₃ affinity and ID_50s for inhibition of VTA neuronal activity (table 3). No other significant correlations were observed.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Actions of (±)-, (+)- and (-)-8-OH-DPAT at h5-HT_1A receptors. The affinities of the (+)- and (-)-isomers of 8-OH-DPAT for cloned h5-HT_1A receptors did not differ markedly, in agreement with studies in rat hippocampus (Cornfield et al., 1991; Foreman et al., 1995). Nevertheless, (+)-8-OH-DPAT displayed higher efficacy than (-)-8-OH-DPAT in activating h5-HT_1A receptor-mediated [³⁵S]GTPS binding, an in vitro measure of G-protein activation. This observation is analogous to the greater efficacy of (+)- as compared to (-)-8-OH-DPAT in inhibiting forskolin-stimulated adenylyl cyclase activity in rat hippocampal membranes (Cornfield et al., 1991; Foreman et al., 1995). Such isomeric differences may be an intrinsic property of 5-HT_1A receptors (Björk et al., 1989), although the G-protein coupling of recombinant human 5-HT_1A receptors in CHO cells may differ from that of native rat 5-HT_1A autoreceptors, with respect to G-protein subtypes and subcellular localisation. Further, due to their high receptor reserve in the DRN, even weak partial agonists suppress serotoninergic transmission (Gobert et al., 1995a; Meller et al., 1990). Thus, the higher efficacy, but not potency, of (+)- vs. (-)-8-OH-DPAT at h5-HT_1A receptor, may underlie its more potent inhibition of the firing of DRN serotoninergic neurones (see below). Similar findings at 5-HT_1A autoreceptors controlling 5-HT synthesis were noted, using other stereoresolved ligands, by Foreman et al. (1995) and Malmberg et al. (1994).

Actions of (±)-, (+)- and (-)-8-OH-DPAT at hD₂ and hD₃ receptors. Racemic 8-OH-DPAT possesses modest affinity at native rat D₂ receptors (Gobert et al., 1995a; Van Wijngaarden et al., 1990), and we demonstrate that it displays low affinity at cloned hD₂ receptors. Further, its hD₃ receptor affinity was 4-fold higher than at hD₂ sites suggesting that certain effects of 8-OH-DPAT previously attributed to D₂ sites (Gobert et al., 1995b; Smith and Cutts, 1989) might in fact be mediated by D₃ receptors. In terms of affinity, (-)-8-OH-DPAT was more selective than (+)-8-OH-DPAT for h5-HT_1A vs. hD₂/hD₃ receptors. The greater separation between (+)- and (-)-8-OH-DPAT as concerns their affinities at hD₃ as compared to h5-HT_1A receptors indicates that stereospecificity may be a more pronounced feature of hD₃ than h5-HT_1A receptors.

Actions of (+)- and (-)-7-OH-DPAT at hD₂ and hD₃ receptors. The (+)-isomer of 7-OH-DPAT showed 20-fold higher affinity than its (-) counterpart at cloned hD₃ (and hD₂) receptors suggesting that pharmacological activity at hD₃ receptors resides primarily in the former. These data corroborate the findings of Damsma et al., (1993) who used a different radioligand, [³H]-spiperone. Further, Baldessarini et al. (1993) found that (+)-7-OH-DPAT was 2-fold more potent than racemic 7-OH-DPAT in binding to hD₃ receptors transfected into fibroblasts. In addition, in an extensive study of substituted 2-aminotetralins, Malmberg et al. (1994) revealed that, although hydroxy-substitued ligands (at position 7 and elsewhere) behave as agonists at hD₃ and hD₂ receptors, the (+)- and (-)-isomers may display different binding modes with the latter consistently presenting lower affinities. It must be mentioned that the hD₂ affinities given here may be underestimated owing to the existence of multiple affinity states (Burris et al., 1995; Chio et al., 1993; Gonzalez and Sibley, 1995; Millan et al., 1995; Newman-Tancredi et al., 1995) and a role of hD₂ receptors in the actions of (+)- and (-)-7-OH-DPAT cannot be excluded (Gobert et al., 1995b; Lejeune et al., 1995).

Actions of (+)- and (-)-7-OH-DPAT at h5-HT_1A receptors. This study extends to cloned h5-HT_1A receptors our previous observation (Millan et al., 1995) that (+)-7-OH-DPAT possesses significant affinity for rat 5-HT_1A receptors and demonstrates that the (-)-isomer displays even less (5-fold) selectivity for hD₃ vs. h5-HT_1A receptors. In line with its higher affinity, (+)-7-OH-DPAT was 8-fold more potent than the (-)-isomer in enhancing 5-HT_1A receptor-stimulated [³⁵S]GTPS binding. In analogy to the differential efficacy of (+)- vs. (-)-8-OH-DPAT at h5-HT_1A receptors, the efficacy of (+)-7-OH-DPAT was higher than that of (-)-7-OH-DPAT in [³⁵S]GTPS binding. Nevertheless, both isomers of 7-OH-DPAT completely inhibited the firing of DRN serotoninergic neurones (see below).

Actions of (±)-, (+)- and (-)-8-OH-DPAT upon dopaminergic vs. serotoninergic transmission in vivo. In our study, the robust influence of (±)-8-OH-DPAT on DRN firing rate (fig. 2) (Blier et al., 1988; Lejeune et al., 1994; Lum and Piercey, 1988; Sinton and Fallon, 1988; Sprouse and Aghajanian, 1986) was confirmed and extended to its isomers. The involvement of 5-HT_1A autoreceptors was revealed by the antagonist activity of (-)-tertatolol and by the lack of effect of haloperidol. In contrast, the modulation by 8-OH-DPAT of the electrical activity of dopaminergic cells has proven variable (Lum and Piercey, 1988; Sinton and Fallon, 1988): an inhibitory (Gervais and Rouillard, 1993; Yoshimoto and McBride, 1992) or excitatory (Kelland et al., 1990; Prisco et al., 1994; Zhang and Freeman, 1993) action has been observed. In all these cases racemic 8-OH-DPAT was used but the study of Arborelius et al. (1993b) reported a biphasic modulation of DA cell firing rate after systemic administration of (+)-8-OH-DPAT. Our results corroborate this biphasic influence of systemically administered (+)-8-OH-DPAT on VTA firing (fig. 2). Further, (-)-8-OH-DPAT had no excitatory effect but only inhibited VTA cell firing. This dissociation between the stimulation ((+)-isomer) and the inhibition ((-)-isomer) of VTA electrical activity provides an explanation for conflicting results previously reported for racemic 8-OH-DPAT and emphasizes the importance of examining both stereoisomers. In addition, we show that the stimulation of VTA firing by (+)-8-OH-DPAT is blocked by (-)-tertatolol, but unaffected by haloperidol, implicating the involvement of 5-HT_1A receptors in the activation of dopaminergic neurones by (+)-8-OH-DPAT. Conversely, the inhibition of VTA firing by (-)-8-OH-DPAT was abolished by haloperidol, consistent with a direct interaction at inhibitory D₃ and/or D₂ (auto)receptors. These findings support the hypothesis that stimulation of 5-HT_1A receptors increases the activity of VTA-localized dopaminergic neurones. Our data do not address the issue of whether pre- or postsynaptic 5-HT_1A receptors are involved; however, (+)-8-OH-DPAT was more potent than (-)-8-OH-DPAT in stimulating VTA cells and inhibiting DRN firing. Thus, the excitatory effect of (+) vs. (-)-8-OH-DPAT on dopaminergic neurones more likely reflects its greater potency in inhibiting the firing of DRN neurones via stimulation of 5-HT_1A presynaptic autoreceptors. Further, we have recently observed that the selective 5-HT_1A ligand, S 15535, which behaves as an agonist and antagonist at pre and postsynaptic 5-HT_1A receptors, respectively (Millan et al., 1994), also activates dopaminergic neurones (Gobert et al., 1995c; Lejeune et al., 1996; Millan et al., 1996, submitted for publication). The inhibitory effect of (-)-8-OH-DPAT on VTA firing was reversed by haloperidol, indicating the involvement of dopaminergic autoreceptors, in agreement with the dopaminergic agonist effects of (±)-8-OH-DPAT previously observed by Smith and Cutts (1989) in vitro.

Actions of (+)- and (-)-7-OH-DPAT on dopaminergic vs. serotoninergic transmission in vivo. (+)- and (-)-7-OH-DPAT haloperidol-reversibly inhibited VTA-localized dopaminergic neurones; the (+)-isomer was 20-fold more potent than its (-)-counterpart. These observations are paralleled by previous reports of the more potent influence of (+)- vs. (-)-7-OH-DPAT on release of DA from mesolimbic dopaminergic terminals in vivo, an action reflecting activation of D₃ and, possibly, D₂ autoreceptors (Rivet et al., 1994; Gobert et al., 1995b and 1996; Patel et al., 1995; Tang et al., 1994). (+)- and (-)-7-OH-DPAT (-)-tertatolol-reversibly inhibited the firing rate of serotoninergic neurones in the DRN, suggesting that they activate 5-HT_1A autoreceptors. This is in line with the activation of h5-HT_1A receptor-coupled [³⁵S]GTPS binding. Further, (+)-7-OH-DPAT showed 100-fold lower potency than (+)-8-OH-DPAT in this binding model, a ratio identical to that for their respective potencies in inhibiting DRN firing. To date, there is no evidence for the occurrence of dopamine D₃ receptors in the DRN (Ariano and Sibley, 1994; Herroelen et al., 1994; Sokoloff and Schwartz, 1995) and the contribution of dopaminergic neurones to projections from the VTA (and SNPC) to the DRN is minor (Geffard et al., 1987; Kalén et al., 1988; Swanson, 1982). These observations are consistent with our study in suggesting that dopamine D₃ receptors are unlikely to play a major role in the effect of 7-OH-DPAT on DRN firing rate. This conclusion is supported by recent studies with CGS-15855, a high affinity agonist at hD₃ (K_i = 5 nM) vs. h5-HT_1A (K_i = 620 nM) receptors (Millan et al., 1995). CGS-15855 fails to modify the activity of DRN-localized serotoninergic neurones (unpublished observation) even at high doses relative to those suppressing dopaminergic transmission (Gobert et al., 1995b).

General Discussion. We did not observe systematic differences between "slow" and "fast" VTA-localized dopaminergic neurones in terms of the excitatory influence of (+)-8-OH-DPAT, an observation consistent with the remark of Arborelius et al. (1993a) that the stimulatory influence of 5-HT_1A agonists on dopaminergic neurones is independent of basal firing rate. In contrast, Kelland et al. (1990) suggested that slow SNPC-localized dopaminergic neurones were more susceptible to the excitatory actions of systemic 8-OH-DPAT. A further observation, in common with the study of Arborelius et al. (1993b), is that (+)-8-OH-DPAT transformed regularly firing cells into a bursting pattern of firing, a change associated with an increase in DA release and synaptic transmission (Svensson et al., 1995). Further, the reinforcement by 5-HT_1A agonists of cortical dopaminergic transmission via preferential activation of this subset of VTA-localized dopaminergic neurones is likely relevant to their ability to relieve the cortical "hypofrontality" common to both depressive states and the negative symptomatology of schizophrenia (Svensson et al., 1995; Tanda et al., 1994).

Conclusions. Our data demonstrate that 7-OH-DPAT shows marked stereospecificity in its interactions at D₃, D₂ and 5-HT_1A receptors. Previous authors (e.g., Burris et al., 1995) have pointed out that the selectivity of 7-OH-DPAT for hD₃ vs. hD₂ sites is not as marked as originally claimed (Sokoloff et al., 1992), and our study emphasizes that potential actions at h5-HT_1A receptors should not be ignored. Indeed, the (-)-isomer shows only 5-fold selectivity for hD₃ vs. h5-HT_1A sites suggesting that studies of this ligand must be restricted to the (+)-isomer. As concerns 8-OH-DPAT, the data also reveal potential actions at high doses at D₃ receptors. Nevertheless, (+)-8-OH-DPAT shows marked in vivo selectivity for 5-HT_1A receptors and the activation of 5-HT_1A autoreceptors underlies its disinhibition of VTA-localized dopaminergic neurones. As concerns stereospecificity of actions, this appears to be a more marked feature of D₃ and D₂ vs. 5-HT_1A receptors and of 7- as compared to 8-OH-DPAT. Our data help resolve several discrepancies in the literature. Finally, they underscore the importance of 5-HT_1A (auto)receptor-mediated modulation of the activity of VTA-localized dopaminergic neurones, an action of potential importance to the therapeutic profiles of novel antidepressant and antipsychotic agents (Broekkamp et al., 1995; Deutch et al., 1991; Meltzer, 1992).