SR 48692, a non-peptide neurotensin receptor antagonist differentially affects neurotensin-induced behaviour and changes in dopaminergic transmission

Neuroscience Vol. 59, No. 4, pp. 921-929, 1994

0306-4522(93jE0067-Z \ ,

Elsevier ScienceLtd Copyright 0 1994IBRO Printed in Great Britain. All rightsreserved 0306-4522/94-$6.00 + 0.00

SR 48692, A NON-PEPTIDE NEUROTENSIN RECEPTOR ANTAGONIST DIFFERENTIALLY AFFECTS NEUROTENSIN-INDUCED BEHAVIOUR AND CHANGES IN DOPAMINERGIC TRANSMISSION R. STEINBERG,*P. BRUN, M. FOURNIER, J. SOUILHAC, D. RODIER, G. G. LE FUR and P. SOUBRIB

MONS, J. P. TERRANOVA,

Sanofi Recherche, Neuropsychiatry Research Department, 371 rue du Prof. J. Blayac, 34184 Montpellier Cedex 04, France Abstract-Unilateral microinjection of neurotensin in the ventral tegmental area of the rat (2.5 fig/O.5 ~1) produced behavioural excitation illustrated by contralateral circling. Given orally, SR 48692, a selective and potent non-peptide neurotensin receptor antagonist, significantly reduced these rotations with a triphasic dose-effect relationship. Inhibition occurred at 0.12 mg/kg; further increases in dose up to 2.5 mg/kg produced no significant antagonism, then at doses > 5 mg/kg, a second phase of antagonism was observed. Bilateral injection of neurotensin (0.5 pg each side) into the nucleus accumbens antagonized the increase in locomotor activity following intraperitoneal injection of amphetamine. Given orally, SR 48692 reduced dose-dependently (0.1-l mg/kg) these intra-accumbens neurotensin effects. Using high pressure liquid chromatography with electrochemical detection, we showed that microgram amounts of neurotensin injected into the ventral tegmental area increased dihydroxyphenylacetate/dopamine ratios in the nucleus accumbens. Using in viva voltammetry techniques, we found that the injection of nanogram and picogram amounts of neurotensin in the ventral tegmental area stimulated dopamine efflux in the nucleus accumbens. None of these biochemical changes were affected by SR 48692 (0. I-10 mg/kg). These results indicate complex interactions between neurotensin and the mesolimbic dopamine system. More particularly, the differential ability of SR 48692 to affect neurotensin-evoked behavioural versus biochemical changes supports the concept of neurotensin receptor heterogeneity.

It is now well established that the tridecapeptide neurotensin is closely associated with dopamine (DA) transmission. Neurotensin-like immunoreactivity and neurotensin receptors have been detected in all brain structures containing DA cell bodies and terminals such as the substantia nigra (pars compacta), the ventral tegmental area (VTA), the striatum, the nucleus accumbens and the prefrontal cortex.‘6.20It has been suggested that neurotensin decreases the affinity of DA agonists for D, autoreceptors via intramembrane interactions.‘0,28,32,33 Furthermore, administration of neurotensin into the substantia nigra (pars compacta) or VTA increases the metabolism and release of DA in the striatum or the nucleus accumbens, respectively, with a greater effect in the mesolimbic system. 22,27 Moreover, microiontophoretic applications of neurotensin in the VTA present variable effects on the basal activity of DA *To whom correspondence should be addressed. DA, dopamine; DNPV, differential normal

Abbreviations:

pulse voltammetry;. DOPAC, dihydroxyphenylacetate; DPA, differential pulse amoerometrv: EDTA. ethvlenediaminetetra-acetate; PBS: phosphaie-buffered saline; SR 48692, {2-[(l-(7-chloro-4-quinolinyl)-5-(2, 6-dimethoxyphenyl) pyrazol - 3 - yl )carbonylamino ]tricyclo(3.3.1.1 .3.7)decan-2-carboxylic acid; VTA, ventral tegmental area.

cell~,~~but produced a consistent antagonistic effect on DA-induced inhibiton. At the level of areas of DA projections, the modulatory effect of neurotensin on DA responsiveness depends on the brain structure considered, namely inhibition in the cortex and nucleus accumbens, and facilitation in the striatum.2 Behavioural studies showed that the injection of neurotensin into the VTA produced a facilitatory effect on behavioural responses associated with an activation of the mesolimbic DA system.‘“,” Conversely, neurotensin injection in the nucleus accumbens blocked the behavioural excitation obtained following injection of drugs known to release mesolimbic DA.‘,” SR 48692 has recently been described as a selective non-peptide antagonist of neurotensin receptors. This compound presents a high affinity for [‘251]neurotensin binding on rat, guinea-pig and human brain, and blocks many effects of neurotensin, including the stimulation of the K+-evoked striatal release of [3H]DA in guinea-pig,” but does not display any significant affinity for most neurotransmitter binding sites studied so far, including receptors for neuropeptides. The first aim of this study was to investigate the effects of SR 48692 on two main behavioural changes induced by central administration of neurotensin:

921

R. STEINBERGem al.

922

excitation or inhibition depending on the injected brain area. When neurotensin is applied onto DA perikarya in the VTA, behavioural activation occurs,” and in the case of unilateral injection, rotational behaviour can be identified.” When injected in the nucleus accumbens, neurotensin decreases enhanced locomotor activity such as that produced by amphetamine.’ In addition, we investigated the extent to which SR 48692 may affect the enhanced DA transmission induced by application of neurotensin in the VTA. As already reported, microgram amounts of neurotensin were administered in the VTA to increase the 3,4_dihydroxyphenylacetate (DOPAC)/DA ratio from the mesolimbic terminal areas.” However, in microdialysis studies, a dissociation between changes in metabolite levels and DA release after intracerebral peptide injection has been described.22 Therefore, using differential normal pulse voltammetry (DNPV)“,” or differential pulse amperometry (DPA),“.3” we examined the effect of nanogram and picogram amounts of neurotensin in the VTA on the DA efflux in the nucleus accumbens.

EXPERIMENTAL

PROCEDURES

Behavioural studies

Sprague-Dawley rats (Charles River, France) weighing 260-330 g were anaesthetized with chloral hydrate (400 mg/kg, i.p.) and mounted in a stereotaxic frame accordMale

ing to the Paxinos and Watson atlas.25 Rats were implanted one week before behavioural testing either with bilateral guide cannulae situated 0.5 mm over the site of interest in the medial nucleus accumbens (1.7mm anterior to bregma, 1.S mm lateral to the midline and 0.5 mm below the cortical surface) or with unilateral guide cannulae placed 0.5 mm over the VTA (5.5 mm posterior to bregma, 0.6 mm lateral to the midline and 7.5 mm below the cortical surface). Guide cannulae (0.6 mm in diameter, Europhore) were secured to the skull with stainless steel screws and dental cement. An obturator was placed in each guide cannula. For VTA administration of neurotensin, and for studies of rotational behaviour, the animals were placed into separate glass cylindrical cages (40 cm in height and 20 cm in diameter) and habituated for 10 min. The rats were then removed and microinjected with neurotensin (2.5 pg/O.S ~1 per 1 min) dissolved in phosphate-buffered saline solution (PBS: 8 g/l NaCl, 0.2g/l KCI, 1.44g/l Na2HP0,.2H20 0.2 g/l KH,PO,, pH 7.4, added with 132 mail CaC1,‘2H,O delivered by a microinfusion pump (CMA’lOO, Bibanaiytical Systems). Two minutes after the end of the injection, the cannulae were removed and the obturator replaced. The rats were returned to the cylindrical cages and the number of contralateral rotations were recorded over 30 min. Rats were administered SR 48692 at 0.06, 0.12, 0.25, 1.25, 2.5, 5 and lOmg/kg p.o. or vehicle, 60 min before VTA injection of neurotensin. For 85 rats, a second trial identical to the first one was conducted one week later, except that the doses of SR 48692 were restricted to 0.06, 0.12, 1.25 and lOmg/kg p.o. (rats being randomly assigned to a given dose). In an additional experiment, rats (n = 41) were pretreated with thiorphan (7.5 pg/O.5 pl), a 24-11 endopeptidase inhibitor, 10 min before neurotensin application in the VTA. SR 48692 was given at 0.12,2.5, 5 and 10 mg/kg p.o. For nucleus accumbens administration of neurotensin, rats were habituated individually for a period of 1 h in polycarbonate cages (30 cm x 40 cm x 18 cm). Bilateral

nucleus accumbens administration of neurotensin (0.5 pg/l PI/side) was conducted identically to that into the VTA. Immediately after intra-accumbens injection, rats were given amphetamine (2 mg/kg i.p.) and were returned to the test cages for 60min. Motility was automatically recorded with a digital image analysis system (Videotrack 5 12, View Point) and the last 40 min were used for statistical analysis. Rats were administered SR 48692 at 0.01, 0.1 and I mg/kg p.o. or vehicle, 60min before nucleus accumbens injection of neurotensin. Measurement vivo

of dopamine and dihydroxyphenylacetarr

ex

Sixty-four rats of the first trial of behavioural studies were decapitated 60 min after neurotensin injection. The brains were rapidly removed and the nucleus accumbens and olfactory tubercles dissected on an ice-cooled glass plate. Tissue samples were homogenized in 500 ~1 of 0.1 M HCIO, containing 4 mM sodium metabisulphite and 1 mM EDTA, and centrifuged. The pellet was assayed for protein with the biorad assay technique and 100 ~1 aliquots of supernatant were stored at -80°C until analysed for DA and DOPAC levels. DA and DOPAC were measured using high pressure liquid chromatography with electrochemical detection, as described in detail elsewhere.i A Bondapack phenyl column (Waters Ass.) was used for the separation. The mobile phase consisted of 3% methanol in 0.1 M sodium phosphate buffer (pH 2.5) and I mM l-octane sulphonic acid. Measurement of dopamine in vivo In pargyline-treated rats, voltamperometric techniques (DNPV and DPA) enable the measurement of an oxidation current corresponding to the extracellular DA concentration.“~‘*~30 EIeclrochemical techniques. Male Sprague-Dawley rats (250-300 g, Charles River, France) were treated with pargyline (7Smg/kg, i.p.) and 30 min later anaesthetized with urethane (1.15 g/kg, i.p.). Rats were mounted in a stereotaxic frame according to Paxinos and Watson’s atlas.‘5 Their body temperature was monitored by a rectal probe and adjusted to 37.5 + O.S”C by a homeothermic blanket. The skull and the dura mater were opened at the level of the nucleus accumbens and the VTA, and the pia mater was punctured by means of a used carbon fibre electrode. The freshly treated carbon fibre electrode was implanted in the nucleus accumbens at the following coordinates: 2.2mm anterior to bregma, 1.5 mm lateral to the midline and 6.5 mm below the cortical surface. A conventional threeelectrode system was connected to a voltamperometric SOLEA Tacussel, France). The apparatus (“Biopulse”, auxiliary electrode (a platinum wire) and the Ag/AgCl reference electrode (a silver wire coated with AgCl, length of the active part = 10 mm) were kept in contact with the skull by means of a semi-liquid junction (sponge moistened with a 0.9% NaCl solution). Carbon fibre electrodes were produced as described previouslyn and electrochemically treated in PBS. For DNPV studies, carbon fibre electrodes (500 pm long and 12 nm in diameter) were electrochemically treated by first applying an anodic potential of a triangular waveform (between 0 and + 2.9 V for 5 s/2.8 for 15 s, at 70 Hz), and then two successive continuous potentials (-0.8 V and + I .5 V) for 5 s each. For DPA, the active part of the carbon fibre electrodes was reduced (250 pm long, 8 pm in diameter) and the electrochemical treatment was modified (first triangular treatment: between 0 and 2.45 V at 70 Hz for 20 s; secondary treatment: -0.75 V and + 1.5 V for 5 s each). The catechol oxidation current was monitored either every minute with DNPV or every second with DPA (final potential: + 85 mV), as described previously.“,” Pneumatic ejection of neurotensin. A calibrated glass tubing (internal diameter 0.3 mm, 15 mm/PI. AssistentE

923

SR 48692, a non-peptide neurotensin receptor ref. 55515) was pulled and broken back to an external tip diameter of 50 pm. Ejection pipettes were filled either with neurotensin or its inactive fragment neurotensin (l-6) dissolved in PBS solution (same composition as mentioned above with the addition of 132 mg/l CaCl,,2H,O) through the tapered tip, by applying a negative pressure to the opposite tip. They were implanted in the central part of the VTA with the following coordinates: 5.5-5.8 mm anterior to bregma, 0.7 mm lateral to the midline and 8.0 mm below the cortical surface. Neurotensin ejection (65 nl) was performed by applying air pressure with a 1 ml syringe connected to the non-tapered side of the pipette by Tygon tubing. The ejected volume was determined under microscope by the movement of the meniscus of the solution in the pipette. Drugs SR 48692 {2-[(I-(7-chloro-4-quinolinyl)-5-(2,6pyrazol-3-yl)carbonylamino]tricyclodimethoxyphenyl) (3.3.1.1 .3‘)decan-2-carboxylic acid; Sanofi Recherche, Montpellier France} was suspended with Tween 80 in distilled water. The injection volume for the intraperitoneal route (i.p.) or oral route (p.0.) was 0.5 ml/l00 g body weight. Pargyline (Sigma) and chloral hydrate (Fluka) were dissolved in saline solution (0.9% NaCl). and urethane (Fluka) was dissolved in water. All these drugs were injected intraperitoneally. Neurotensin (Sigma) and its fragment neurotensin (16) (Sigma) were dissolved in PBS solution and aliquots were stored at -20°C. Thiorphan (Sigma) was dissolved using dimethyl sulphoxide (2.5%). Statistical analyses were performed using Kruskal-Wallis test. I,

.

,

RESULTS Behavioural studies After 10min habituating to the cylindrical cages, animals treated with vehicle followed by an intraVTA PBS injection generally rested or often slept through much of the 30-min testing period. Some rotations did occur but equal rates were observed toward and away from the injected side. In animals pretreated with vehicle, neurotensin injected in the VTA induced circling in a direction contralateral to the side injected (Table 1). As already observed by

Holmes and Wise,15 rotations induced by neurotensin were interrupted by pauses to rear or groom, but the animals maintained the rotative direction. The rate of rotation was slightly lower in the second trial than that observed in the first one (9.5 f 0.9, n = 28 and 12.5 f 1, n = 45, respectively). Oral treatment with SR 48692 1 h before ejection of neurotensin in the VTA significantly reduced neurotensin-induced controlateral rotations with a triphasic dose-effect relationship (Table 1). Inhibition occurred at O.l2mg/kg (-42%, n = 14, of controls P < 0.05 in the first trial, -4l%, n = 17, of controls P <: 0.01, second trial), intermediate doses (0.25-2.5 mg/kg) producing no significant antagonism. At higher doses (5 and 10 mg/kg), a second phase of antagonism occurred (first trial, -34%, II = 7, P < 0.05 and -43%, n = 8, of controls, P < 0.01, respectively; second trial, - 56%, n = 10, of controls, P < 0.01 for the 10 mg/kg dose). In rats pretreated with thiorphan (7.5 pg/O.5 pl), rotations were more vigorous and their number was enhanced (2 1.2 + 2.3; P < 0.05). Under these conditions, SR 48692 showed an identical profile of action, the most marked antagonistic effects being observed at 5 and 10 mg/kg (number of rotations: 13.2 f 1.8, P < 0.05, and 9.1 f 2.0, P < 0.01, respectively). As shown in Table 2, neurotensin (0.5 pg/pl/ structure) injected in the nucleus accumbens significantly antagonized amphetamine-induced hypermotility in rats with 49% of inhibition (P < 0.05). SR 48692 (0.01, 0.1 and 1 mg/kg) dose-dependently reduced the antagonistic effect of neurotensin (-27%, -54% and -69%, respectively), the antagonism being statistically significant only at 1 mg/kg. SR 48692 at 10 mg/kg exerted deleterious effects on behaviour in implanted animals pretreated with amphetamine, and lost its antagonistic effect against neurotensin. A part from this situation, no gross behavioural modifications were observed in rats treated with SR 48692 (0.1. 1 and 10 mg/kg).

Table 1. Reversal by SR 48692 of the induction of contralateral rotations by unilateral ejection of neurotensin (2.5 pg/O.5 ~1) in the ventral tegmental area Drugs Neurotensin @g) Vehicle 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5

Number of rotations

SR 48692 (mg/kg p.o.) 0.06 0.12 0.25 1.25 2.5 5 10

First trial (mean k S.E.M.) 3.1 k 6 12.5 + 1 8.4 f 1.1 7.2 k 0.9; 9.2 + 1.7 9.7 f 1.6 9.9 f 0.7 8.3 * 0.5* 7.1 + 1.2**

n 18 45t IO 14 11 15 7 7 8

Second trial (mean + S.E.M.)

n

3.1 kO.2 9.5 + 0.9 5.9& 1.1* 5.6 f 0.6**

5 28 9 17

9.2 f 1.1

16

4.2 + 0.6**

10

In the first trial rats received vehicle or SR 48692 (0.0610 mg/kg p.o.) 60 min before neurotensin. One week later a second trial was performed on rats randomly assigned to receive vehicle or SR 48692 (0.06, 0.12, 1.25 and 10 mg/kg p.o., 60min before neurotensin). The results are expressed as the mean number f S.E.M. of contralateral rotations recorded over 30min. TPooled values of two independent experiments, Statistical comparisons were made between SR 48692- and corresponding vehicle-treated animals with the Kruskal-Wallis test. *P < 0.05; **P < 0.01.

924

It. STEINBERG er ul Table 2. Reversal by SR 48692 of the antagonism of amphetamine-induced locomotion produced by neurotensin applied into the nucleus accumbens Drugs Neurotensin (tig)

Amphetamine (mg/kg i.p.)

Vehicle (PBS) Vehicle (PBS) 0.5 0.5 OS 0.5

SR 48692 (mgikr P.o.)

2 2 2 2 2

Locomotion Travelled distance (cm) mean + S.E.M.

~1

151 * 106 5870 4 1348 3065 rt 474* 3844 i 767 4315 1463 4698 + 408+

8 8 II 7 I2 12

0.0 I 0.1

I

Animals were given SR 48692 (p.0.) I h before bilateral injection of neurotensin (0.5 pg). Amphetamine was administered immediately after neurotensin injections. Motility was automatically recorded for 40min. n = number of animals. Statistical comparisons were made with the Kruskal-Wallis test: *P < 0.05 compared to PBS-injected rats: iP < 0.05 compared to rats treated with amphetamine + neurotensin. Effects of neurotensirr ejection on the dihydroxyphenyiacetateldopamine ratio

Table 3 shows that 60min after the injection of 2..5pg of neurotensin into the VTA, a significant increase in the DOPAQDA ratio in the nucleus accumbens (+77%, P < 0.01 of controls) and the

olfactory tubercle (+ 120%, P < 0.05 of controls) was observed, which was due both to an increase in DOPAC and a decrease of DA concentrations, as already described by Kalivas et al.” Thirty minutes after neurotensin injection. these changes were present, though iess marked (+53%, P < 0.01 of controls), in the nucleus accumbens and (t 104%, P < 0.01 of controls) the olfactory tubercle. The pretreatment with SR 48692 (0.1, 1 and 10 mg/kg) did not significantly alter this increase in DOPAC/DA ratios measured 30 min (data not shown) and 60 min (Table 3) after neurotensin injection. When injected alone, SR 48692 (0.1-10 mg/kg p.0.) did not significantly affect DOPAC/DA ratios (data not shown). Effects of neurotensin ejection on the extracellular ~iopamine c~nceritratio~

When recorded from the nucleus accumbens of a pargyline-treated rat, voltammograms obtained with

DNPV showed an oxidation peak appearing at the oxidation potential of DA (t85 mV).‘O This peak is generally considered to correspond to the extracellular DA concentration.“.‘? The ejection of neurotensin in VTA (Fig. 1) was concentration-dependently followed by an increase in the amplitude of the DA peak recorded from the nucleus accumbens. This increase reached a maximum (+ 140% and i-230% above baseline for 4.2 and 42 ng in 65 nl, respectively) after 6min and subsequently returned to baseline levels within 35 min. The 420 ng dose (65 nl) of neurotensin produced a very marked effect (+280%), which persisted for 50 min. In contrast, the amino-terminal fragment neurotensin (l-6) (42 ng in 65 nl) did not affect the DA extracellular concentration measured in the nucleus accumbens. In animals injected with saline between neurotensin ejections (Fig. 2), DA responses to two successive ejections (4.2 ng in 65 nl), 60min apart, were very similar (+ 119% vs + 143%, first vs second ejection, )I = 5). When animals received SR 48692 (0.1, 1 and lOmg/kg, i.p.) 30min before the second ejection,

there was no significant difference between the response to the first and the second stimulations. By itself, SR 48692 did not alter basal DA peak height.

Table 3. The effect of SR 48692 on the ability of neurotensin applied in the VTA to increase the dihydroxyphenylacetate/dopamine ratio in the nucleus accumbens and the olfactory tubercle Doses (m&/kg) PBS + vehicle Neurotensin + SR 48692

2.5 &OS 0 0.1

1 10

DOPACIDA Nucleus accumbens Olfactory tubercle 0.31 kO.03 (11) 0.55 i: 0.05** (18) 0.47 * 0.09 ns (9) 0.47 * 0.04 ns (9) 0.54 * 0.07 ns (17)

0.10 10.01 (9) .0.22 + 0.02* (9) 0.22 & 0.03 ns (9) -. 0.24 & 0.02 ns (9)

Sixty minutes after unilateral injection of neurotensin (2.5 pg) or PBS, the nucleus accumbens and the olfactory tubercie were assayed for changes in DOPAC/DA ratio using reverse-phase high pressure liquid chromatography with electrochemical detection. SR 48692 was given orally 1 h before neurotensin application. (n) = number of animals. Statistical comparison between values were made with the Kruskal-Wallis test. *P < 0.05 and **P < 0.01 compared to PBS-injected rats; ns, non-significant, compared to rats treated with neurotensin + vehicle.

925

SR 48692, a non-peptide neurotensin receptor

3300 8 ii 80 200 5

a

-50

0

50

100

TIME

150 200 250

(min)

Fig. I. Effect of neurotensin ejection in the VTA on the extracellular DA concentration recorded by DNPV from the nucleus accumbens of pargyline-pretreated rats. Neurotensin or the fragment neurotensin (l-6) was ejected (m) in the VTA through a glass pipette by pressure (65 nl of neurotensin at various doses). The 100% value was the mean of peak heights measured from the four successive voltammograms recorded just before the first ejection. NT, neurotensin.

The kinetics of the DA release evoked by neurotensin ejections were observed with a better timeresolution by means of DPA. In fact, the differential pulse oxidation current measured at +85 mV and corresponding to the DA oxidation peak”“O was recorded every second. Ejection either of the aminoterminal fragment neurotensin (l-6) (500 pg in 65 nl, ejection duration = 25 s) or the vehicle solution (PBS alone, 65 nl) induced no significant effects (Fig. 3). Neurotensin concentrations up to lo-* M (1 pg in 65 nl) evoked immediate increase in DA efflux (Fig. 3). When applied every 15 min, ejections of neurotensin (IO--* M, 65 nl) evoked highly reproducible increases in DA efflux which lasted more than 90 min. SR 48692 (0.1, 1 and 10 mg/kg i.p.) injected immediateIy after the third control ejection did not signi~cantly alter DA et&x evoked by any subsequent neurotensin ejection (Fig. 3).

DISCUSSION

As reported previously,- this study found that the injection of neurotensin into the rat VTA elicited both biochemical and behavioural changes, including

300 ‘1

300

250

250

200

200

150

150

loo

100

50 1

-20 0

20 40 60 80 100120

100 . -20 0

20 40 60 80 100120

ul SRld

(n-5)

50 ~

-20 0

20 40 60 80 100 120

TtME (min) Fig. 2. Effect of SR 48692 on the increase in extracellular DA concentration recorded by DNPV from the nucleus accumbens, evoked by neurotensin ejection (4.2 ng in 65 nl) in the VTA. Pargyline-treated rats were injected (arrow) either with vehicle (V) or SR 48692 (SR; 0.1, 1 and 10 mg/kg i.p.) 30 min before the second neurotensin ejection, performed 1 h after the first (control) ejection of neurotensin. Results are expressed as in Fig. I. Each point represents the mean f S. E. M. (vertical bars) of values obtained from (n) different animals.

I

100 PA

30 mtn AFTER St3 48682 01 mg/kg

-45 -30

-15

30 TINGE(rni~~

15

SR48692

0

t

45

60

o 0.1 mg/kg v 1 .O mg/kg o 10 mglkg

75

Fig. 3. Effect of SR 48692 (0.1, I and 10 mg/kg i.p.) on the amplitude of the neurotensin-evoked DA elllux recorded by DPA in the nucleus accumbens of pargyline-treated rats after application of the tridecapeptide in the VTA. The differential pulse (DP) oxidation current appearing at i-85 mV was monitored every 1s. Neurotensin was ejected (II) through a glass pipette by pressure (65 nl during 25s). (A) Effect of neurotensin (l-6) (0.5 ng) and neurotensin (1 and IOpg) ejection in the VTA on the extracellular DA concentration. (B)Typical recording with SR 48692. (C) Effect of SR 48692. Ejections of neurotensin (lo-* M, 65 nl) were repeated every 15 min in the VTA. The effect of SR 48692 is expressed as percentage of the mean of the three absolute effects recorded during the control period (before drug injection). Each point represents the mean I S. E. M. of four (SR, 0.1 mgfkg) ar three (SR, I and lOmg/kg) animals. NT. neurotensin.

l-4

lmln

SR 48692, a non-peptide neurotensin receptor increases in DA release and/or metabolism in terminal areas (nucleus accumbens and olfactory tubercles) and a behavioural excitation. SR 48692 was found to counteract one of the main behavioural signs of excitation following unilateral application of neurotensin in the VTA: contralateral rotations.” This antagonism was observed whether the rats were pretreated or not with thiorphan, a protector of neurotensin degradation. In addition, another classical behavioural effect of neurotensin when applied in the nucleus accumbens, blockade of amphetamine-induced locomotion,’ was found to be reduced by SR 48692. In these two behavioural models, the first doses of SR 48692 producing a marked inhibition of neurotensin-induced effects were very close to those already reported to reduce neurotensin-induced turning in mice.‘4,26 Reversal by SR 48692 of the effects of neurotensin applied in the VTA or the nucleus accumbens cannot be accounted for by non-specific effects (motor incapacitation or excitation), since by itself the compound (O.l-lOmg/kg) did not affect spontaneous locomotion The effect of SR 48692 could not be explained by a direct interaction of SR 48692 with DA D, and D, receptors since, in vitro, this compound is devoid of affinity for these receptors (Gully D., unpublished observations) and, in vim, SR 48692 (10 mg/kg) did not induce catalepsy nor did it modify orofacial stereotypy or climbing behaviour induced by apomorphine. Concerning the different doseresponse relationships obtained in the two behavioural models, 5 and lOmg/kg SR 48692 are required to inhibit neurotensin-induced rotations, whereas 1 mg/kg was sufficient to similarly antagonize the effects of neurotensin on the amphetamineinduced hy~rlocomotion. This difference may be due to the doses of agonist injected, i.e. 2.5 pg into the VTG vs 0.5 pg in each nucleus accumbens. A more critical element may be that, in the VTA model, the antagonism produced by SR 48692 followed a complex dose-response relationship. The triphasic pattern of effect resembled that already observed when SR 48692 was tested against turning behaviour induced by intrastriatal injection of neurotensin.26 It may relate to an altered functional equilibrium between D, and D, receptors without any increase in DA release. Indeed, these receptor interactions, which modulate the expression of DA-dependent behaviour,” are known to be affected markedly by neurotensin.‘* The injection of neurotensin in the VTA induced biochemical changes observed both ex vivo using high pressure liquid chromatography techniques, and in V~VO using two electrochemical detection procedures (DNPV and DPA). Ex viva, as already reported, microgramme amounts of neurotensin were required to enhance the DOPAC/DA ratios in both the nucleus accumbens and the olfactory tubercle, one of the main parameters found to be altered by neurotensin.” For in viva studies, we used DNPV, which

921

allows the recording of a basal oxidation signal due to extracellular DA concentration in animals pretreated with pargyline, i2 a monoamine oxidase inhibitor. We found that nanogramme amounts of neurotensin injected locally in the VTA were sufficient to produce a sustained DA efflux in the nucleus accumbens similar to that described in microdialysis studies.22 A new approach has been described recently which combined DPA and treated carbon fibre electrodes, with the main advantages of very fast response (time resolution of 1 s) and sensitivity of the electrodes to DA.‘O Using DPA, we observed that picogramme amounts of neurotensin evoked an immediate DA efflux from the corresponding terminals. In both DNPV and DPA experiments, the DA-releasing effect of neurotensin was likely to be mediated by neurotensin receptors since the injection in the VTA of vehicle or the fragment neurotensin(l-6) lacking affinity for neurotensin receptors did not induce DA efflux. Moreover, the changes in DA levels were maximal only when the injection cannula was strictly placed in the VTA. It is worth adding that in the in viva VTA paradigm, no desensiti~tion occurred, in contrast to that observed in in vitro models, e.g. striatal DA release.839 The present study indicated that SR 48692 was not able to inhibit the neurotensin-induced changes in DA release/metabolism, even in the most sensitive conditions (DPA). This was unexpected in light of the in vitro and in uivo neurotensin receptor antagonist profile of the compound, which includes antagonism of neurotensin-induced Ca*+ mobilization, striatal DA release and turning behaviour, and the high bioavailability of the compound after intraperitoneal or oral administration.14 This inefficacy of SR 48692 cannot be accounted for by any direct (stimulant) effect on DA function, as the compound (0.1-10 mg/kg) was found on its own in vivo (voltammetry) and ex vim (high pressure liquid chromatography) not to affect extracellular DA levels or tissue DA and DOPAC contents measured in the nucleus accumbens. This apparent inefficacy of SR 48692 could relate to the still high concentrations of neurotensin used in the present biochemical models. However, SR 48692 prevents rotations elicited by the intastriatal injection of 10 pg of neurotensini4 and inhibits the behavioural effects induced by microgram amounts of the t~de~~ptide injected either in the VTA or the nucleus accumbens (present study). A clear-cut dissociation thus emerges between behavioural excitation and enhanced mesolimbic DA transmission, two phenomena thought to be strongly associated, especially following intra-VTA application of neurotensin. i6 However, changes in DA transmission generally occurred at lower doses of neurotensin than the behavioural excitation.” Moreover, there was no clear proportionality in the ability of intra-VTA application of neurotensin and neuromedin N to stimulate DA transmission and to

928

R. STEINBERG el al

produce behavioural excitation.” It has also been noticed that whereas intranigrally applied neurotensin caused dramatic alterations in DA metabolism, no behavioural correlate could be detected.‘4 In addition, after intracerebroventricular injection of neurotensin, an absence of any relationship between behavioural changes and alterations in DA activity has been suggested in view of the dissociation in the time courses of these behavioural and neurochemical phenomena.6 The heterogeneity of neurotensin receptors has been proposed in the light of in vitro and in sivo studies using neurotensin and neurotensin-like agonists.’ These possible neurotensin receptor subtypes may include neuromedin binding sites, since when injected in the VTA, neuromedin N has been shown to reproduce the effects of neurotensin.“,23 It can be tentatively suggested that SR 48692 might discriminate between neurotensin receptor subtypes differentially involved in neurotensin-induced DA facilitation, as compared to neurotensin-induced behavioural activation. Experimental evidence indicates that these neurotensin receptor subtypes should be located on distinct DA neuronal populations. The vast majority of neurotensin receptors present in the VTA disappear after local 6-hydroxydopamine injection,5 and double-labelling immunohistochemical experiments have confirmed the predominant association of neurotensin receptors with DA neurons.3’ However,

in the VTA, synapses identified as neurotensin in nature have been found to be associated with neurons lacking detectable tyrosine hydroxylase immunoreactivity. j4 In addition, destruction of the mesolimbic DA systems by 6-hydroxydopamine abolished the ability of neurotensin applied in the VTA to cause hyperlocomotion. ” However, in this study no data were provided on the possible effect of 6-hydroxydopamine on spontaneous locomotor activity. Concerning DA neurons, several studies converge to suggest the existence of neuronal subpopulations within the VTA according to the presence or absence of neurotensin as co-transmitter,16 their sensitivity to neurotoxins such as 6-hydroxydopamine3’ or other pharmacological

agents

and

stress.4

CONCLUSIONS

This study confirms that SR 48692 acts as a neurotensin receptor antagonist able to reverse neurotensin-incuced behavioural manifestations associated with an altered function of the mesolimbic system, but not the changes in DA transmission induced by the tridecapeptide. This dissociation strengthens the possibility of heterogeneous neurotensin receptors. Acknowledgemenfs-We gratefully thank P. Keane for critical reading of the manuscript and D. Gully for supplying details of her unpublished work.

REFERENCES I.

2. 3. 4. 5. 6. 7. 8. 9. 10.

11. 12. 13.

14.

Al-Rodhan N. R. F., Richelson E., Gilbert J. A., McCormick D. J., Kanba K. S., Pfenning M. A., Nelson A., Larson E. W. and Yaksh T. L. (1991) Structure-antinociceptive activity of neurotensin and some novel analogues in the periaqueductal gray region of the brainstem. Brain Res. 557, 227-235. Beauregard M., Ferron A. and Descarries L. (1992) Opposite effects of neurotensin on dopamine inhibition in different regions of the rat brain: an iontophoretic study. Neuroscience 47. 613-619. Cabib S., Oliverio A. and Puglisi-Allegra S. (1989) Stress-induced decrease of 3-methoxytyramine in the nucleus accumbens of the mouse is prevented by naltrexone pretreatment. Life Sci. 45, 1031-1037. Deutch A. Y. and Cameron D. S. (1992) Pharmacological characterization of dopamine systems in the nucleus accumbens core and shell. Neuroscience 46, 49-56. Dilts R. P. and Kalivas P. W. (1989) Auoradiographic localization of p-opioid and neurotensin receptors within the mesolimbic dopamine system. Brnin Res. 488, 311-327. Drumheller A. D., GagnC M. A., St-Pierre S. and Jolicoeur F. B. (1990) Effects of neurotensin on regional brain concentrations of douamine, serotonin and their main metabolites. Neuropeptides 15, 169-178. Ervin G. N., BirkemdL. S., Nemeroff C. B. and Prange A. J. Jr (1981) Neurotensin blocks certain amphetamine-induced behaviours. Nature 291, 73-16. Faggin B. M. and Cubeddu L. X. (1990) Rapid desensitization of dopamine release induced by neurotensin and neurotensin fragments. J. Pharmac. exp. Ther. 253. 812-818. Faggin B. M., Zubieta J. K., Rezvani A. H. and Cubeddu L. X. (1990) Neurotensin-induced dopamine release in viva and in vitro from substantia nigra and nucleus caudate. J. Pharmac. exp. Ther. 252, 817-825. Fuxe K., OConnor W. T., Antonelli T., Osborne P. G., Tanganelli S., Agnati L. F. and Ungerstedt U. (1992) Evidence for a substrate of neuronal plasticity based on pre- and postsynaptic neurotensin-dopamine receptor interactions in the neostriatum. Proc. natn. Acad. Sci. U.S.A. 89, 5591-5595. Gonon F. G. (1988) Non-linear relationship between impulse flow and dopamine released by rat midbrain dopaminergic neurons as studied by in vivo electrochemistry. Neuroscience 24, 19-28. Gonon F. G. and Buda M. (1985) Regulation of dopamine release by impulse flow and by autoreceptors as studied by in vivo voltammetry in the rat striatum. Neuroscience 14, 765-774. Gonon F., Buda M. and Pujol J. F. (1984) In viva voltammetric detection of catecholamine derivatives by means of treated carbon fiber electrodes. In Measurement qf Neurotransmitter Release (ed. Mardsen CA.), pp. 153-171. John Wiley, Chichester. Gully D., Canton M., Boigegrain R., Jeanjean F., Molimard J. C., Poncelet M., Gueudet C., Heaulme M., Leyris R., Brouard A., Pelaprat D., Labb&Julit C., Masella J., Soubrik P., Maffrand J. P., Rost&ne W., Kitabgi P. and Le Fur G. (1993) Biochemical and pharmacological profile of a potent and selective nonpeptide antagonist of the neurotensin receptor. Proc. natn. Acad. Sci. U.S.A. 90, 65-69.

SR 48692, a non-peptide

neurotensin

receptor

929

15. Holmes L. J. and Wise R. A. (1985) Dopamine-dependent contralateral circling induced by neurotensin applied unilaterally to the ventral tegmental area in rats. Brain Res. Bull. 15, 537-538. 16. Kahvas P. W. (1993) Neurotransmitter regulation of dopamine neurons in the ventral tegmental area. Brain Res. Rev. 18, 75-l 13. 17. Kalivas P. W., Burgess P. W., Nemeroff C. B. and Prange A. J. Jr (1983) Behavioral and neurochemical effects of neurotensin microinjection into the ventral tegmental area. Neuroscience 8, 495-505. 18. Kalivas P. W., Nemeroff C. B. and Prange A. J. Jr (1984) Neurotensin microinjection into the nucleus accumbens antagonizes dopamine-induced increase in locomotion and rearing. Neuroscience 11, 919-930. 19. Kalivas P. W., RichardsonCarBon R. and Duffy P. (1986) Neuromedin N mimics the actions of neurotensin in the ventral tegmental area but not in the nucleus accumbens. J. Pharmac. exp. Ther. 238, 11261131. 20. Kitabgi P. (1989) Neurotensin modulates dopamine neurotransmission at several levels along brain dopaminergic pathways. Neurochem. Int. 14, 1 II-1 19. 21. LaHoste G. J. and Marshall J. F. (1993) The role of dopamine in the maintenance and breakdown of D,/D, synergism. Brain Res. 611, 108-116. 22. Laitinen K., Crawley J. N., Mefford I. N. and De Witte Ph. (1990) Neurotensin and cholecystokinin microinjected into the ventral tegmental area modulate microdialysate concentrations of dopamine and metabolites in the posterior nucleus accumbens. Brain Res. 523, 342-346. 23. Li X. M., Finnman U. B., Von Euler G., Hedlund P. B. and Fuxe K. (1993) Neuromedin N is a potent modulator of dopamine D, receptor agonist binding in rat neostriatal membranes. Neurosci. Letf. 155, 121-124. 24. Napier T. C., Gay D. A., Hulebak K. L. and Breese G. R. (1985) Behavioral and biochemical assessment of time-related changes in globus pallidus and striatal dopamine induced by intranigrally administered neurotensin. Pepiides 6, 1057-1068. 25. Paxinos G. and Watson C. (1982) The Rat Brain in Srereotaxic Coordinates. Academic Press, Sydney. 26. Poncelet M., Gueudet C., Soubrie P. and LeFur G. (1994) Turning behavior induced by intrastriatal injection of neurotensin in mice: sensitivity to non-peptide neurotensin antagonists. Naunyn-Schmiedeberg’s Arch. Pharmac. 39, 57-60. 27. Rivest R., Johcoeur F. B. and Marsden C. A. (1991) Neurotensin causes a greater increase in the metabolism of dopamine in the accumbens than in the striatum in viuo. Neuropharmacology 30, 25-33. 28. Shi W. X. and Bunney B. S. (1991) Neurotensin modulates autoreceptor mediated dopamine effects on midbrain dopamine cell activity. Brain Res. 543, 315-321, 29. Stowe Z. N. and Nemeroff C. B. (1991) The electrophysiological actions of neurotensin in the central nervous system. Lt$ Sci. 49, 98771002. 30. Suaud-Chagny M. F., Chergui K., Chouvet G. and Gonon F. (1992) Relationship between dopamine release in the rat nucleus accumbens and the discharge activity of dopaminergic neurons during local in uioo application of amino acids in the ventral tegmental area. Neuroscience 49, 63-72. 3 1. Szigethy E. and Beaudet A. (1989) Correspondence between high affinity ‘251-neutotensin binding sites and dopaminergic neurons in the rat substantia nigra and ventral tegmental area: a combined radioautographic and immunohistochemical light microscopic study. J camp. Neural. 279, 128-137. 32. Tanganelli S., Li X. M., Ferraro L., Von Euler G., OConnor W. T., Bianchi C., Beani L. and Fuxe K. (1993) Neurotensin and cholecystokinin octapeptide control synergistically dopamine release and dopamine D2 receptor affinity in rat neostriatum. Eur. J. Pharmac. 230, 159-166. 33. Tanganelli S., Von Euler G., Fuxe K., Agnati L. F. and Ungerstedt U. (1989) Neurotensin counteracts apomorphineinduced inhibition of dopamine release as studied by microdialysis in rat neostriatum. Brain Res. 502, 319-324. 34. Woulfe J. and Beaudet A. (1992) Neurotensin terminals form synapses primarily with neurons lacking detectable tyrosine hydroxylase immunoreactivity in the rat substantia nigra and ventral tegmental area. J. camp. Neural. 321, 163-176. 35. Zahm D. S. (1992) An electron microscopic morphometric comparison of tyrosine hydroxylase immunoreactive innervation in the neostriatum and the nucleus accumbens core and shell. Brain Res. 575. 341-346. (Accepted 22 November 1993)