Numbers of neurotensin-immunoreactive neurons selectively increased in rat ventral striatum following acute haloperidol administration

Neuropefhfes 0 Longman

(1988) 11, 1X-132 Group UK Ltd 198X

Numbers of Neurotensin-lmmunoreactive Neurons Selectively Increased in Rat Ventral Striatum Following Acute Haloperidol Administration. K. W. EGGERMAN

and D. S. ZAHM

Department of Anatomy and Neurobiology, St. Louis University Blvd., St. Louis, MO 63104. (reprint requests to DSZ)

School

of Medicine,

7402 S. Grand

Abstract-The effect of haloperidol (HAL) on neurotensin (NT) levels in various structures of the rat brain was evaluated using an immunoperoxidase method. Adult, male SpragueDawley rats were given intraperitoneal injections of either HAL (2mg/kg) or vehicle at twenty-four and four hours prior to sacrifice. The brains were fixed, cut at 50pm on the vibratome, and prepared to demonstrate NT immunoreactivity, or its absence following appropriate control incubations. The distributions of NT-immunoreactive (IR) cell bodies were plotted using the camera lucida, and the numbers of NT-IR neurons in various structures were recorded. The numbers of NT-IR perikarya in striatal and ventral striatal structures of HAL-treated rats greatly exceeded those observed in the same structures of control animals. In other NT-IR rich regions including the bed nucleus of the stria terminalis, central amygdala, hypothalamus and septum, HAL and control values did not differ. Conversely, HAL treatment appeared to effect a decrease in the number of immunoreactive perikarya in the medial amygdala and caudal part of the endopiriform area. It was noted that in brain regions where D-2 receptors are reported to be numerous, the number of NT-stained cells increased following HAL treatment, whereas in regions where D-l receptors predominate, the number remained stable or decreased. Subjective evaluation of axon terminal immunoreactivity revealed a change only in the globus pallidus, where the proportional area of that structure exhibiting NT-immunoreactivity expanded following HAL.

Introduction The regional distribution in rat brain of neurotensin, a basic tridecapeptide originally isolated from bovine hypothalamus and characterized by Carraway and Leeman (3, 4, .5), has been desDate received 8 January 1988 Date accepted 20 January 1988

cribed using radioimmunoassay (e.g. 8) and immunocytochemical techniques (elg-7, 15, 17, 30, 31-33, 37, 38). In the absence of measures taken to enhance perikaryal immunocytochemical reactivity, immunoreactive (IR) cell bodies are frequently present in the substantia gelatinosa of the spinal cord and medullary trigeminal complex, bed nucleus of the stria terminalis (BST),- and central amygdaloid nucleus and can occasionally 125

126 be demonstrated in the midbrain ventral tegmental area of Tsai (VAT). Under the same conditions, however, few if any, cell bodies are immunoreactive in the nucleus accumbens (Acb), olfactory tubercle (Tu), or the caudate-putamen (CPU), structures which display numerous stained perikarya following colchicine pretreatment (17, 30, 37, 38). Moderate to substantial densities of immunoreactive fibers and axon terminals are widely distributed throughout the CNS and are found in various parts of the hypothalamus, septal area, BST, Acb, centromedial amygdaloid complex, globus pallidus (GP) and ventral pallidum (VP). NT immunoreactivity co-localizes with dopamine in some neurons which project to the Tu, Acb, and ventral CPU via the mesolimbic pathway (15), and it can also be demonstrated, following intraventricular injections of colchicine, in neurons of the ventral striatum [Tu, Acb, and ventromedial part of CPU (37, 38)], a striatal region which projects to the mesencephalic ventral tegmentum and substantia nigra (see 12, 24). Moderate to abundant NT receptor levels are found both in the VTA and in ventral striatal regions (26,31,35,36). NT injected into the VTA activates dopaminergic neurons, as indicated by their increased firing rates (l), increases in locomotor activity (18-20)) and increases in DA metabolites in the limbic forebrain (22). NT injected into the Acb has an opposite, neuroleptic-like effect (22, 25), attenuating the increase in locomotor activity induced by injection into the VTA of amphetamine (11) or NT (20, 22, 25). NT, unlike neuroleptics, fails to displace the in vitro binding of dopamine antagonists and dopamine, but some evidence has been reported that NT reduces the in vitro affinity of dopamine receptors for ligand (2). Drugs which block dopamine transmission also perturb the ventral striatal NT system. For example, Herve et al. (14) found that blockade of mesolimbic dopamine transmission by a long acting dopamine antagonist, pipotiazine palmitate, caused an increase in the number of striatal NT-binding sites. Govoni et al. (10) reported increases in the level of the peptide itself in the nucleus accumbens and adjacent caudateputamen after chronic administration of haloperidol, but reported that NT levels in hypothalamus , preoptic area, septum, hippocampus, amygdala, central gray, and frontal cortex remained unaltered, a result confirmed by Frey et al. (9). The detection method used in those

NEUROPEPTIDES

studies, radioimmunoassay of homogenized extracts, however, does not allow differentiation of neurotensin contained in cell bodies from that present in axon terminals, an important distinction since the available evidence suggests that neurotensin synthesized in and transported from neurons in the striatum could have profoundly different physiological effects than neurotensin produced in and transported from neurons located in VTA. The present study was undertaken to evaluate whether perikaryal immunoreactivity is enhanced in a manner concordant with the increased levels of NT measured in ventral striaturn by RIA following blockade of dopamine neurotransmission.

Materials and Methods Male adult Sprague-Dawley rats weighing between 175 and 350 grams were used. In each of eight trials involving two weight-matched rats, one rat was given an IP injection of haloperidol (HAL, 2mg/kg) in vehicle while the pair matched control rat was given only a vehicle (VEH) which consisted of 20% ethanol, 50% propylene glycol and 30% H20. The rats received two injections, at 24 hours and 4 hours prior to sacrifice. The brains of each HAL/VEH pair were processed in parallel through the same solutions utilizing the following protocol. At sacrifice the rats were anesthetized with sodium pentobarbital (50mg/kg). The anterior thoracic wall was removed and a cannula was introduced through the apex of the heart into the left ventricle. The vascular system was then perfused first briefly with sodium phosphate buffered saline (NaPBS) containing 2.5% polyvinylpyrrolidone (PVP-40) and 0.5% procaine-NC1 and then with 0.07M Sorensen’s phosphate buffer containing 4% paraformaldehyde, 0.1% glutaraldehyde and 2.5% PVP-40. Brains were removed after 20-30 minutes and stored for four additional hours in fresh fixative. Vibratome sections were cut at a thickness of 50km in six adjacent series. One series was incubated with anti-NT antiserum (Immunonuclear, 1:3000). Another series, incubated with anti-NT antiserum preabsorbed with excess NT (50kg/ml), served as a control with which to compare the anti-NT staining produced in the first series. A third series was used for Nissl staining, and the other three series were held in reserve. Sections were pretreated, with intervening rinses,

NT/IR NEURONS

INCREASED

FOLLOWING

ACUTE

HALOPERIDOL

with l”k sodium borohydride for 20 minutes to remove unbound aldehydes, with 0.25% Triton X-100 for 3 hours at room temperature, and, finally, with 5% normal goat serum. Primary antibody incubations were carried out for 18 hours at 4°C in NaPBs containing 0.25% carrageenan and 2% normal goat serum. Incubations in biotinylated anti-rabbit IgG raised in goat (1:200) and abidin-biotin peroxidase complex (1:200) in the same carrier followed, again, with intervening rinses. A reaction product was produced using DAB and the coupled glucose oxidase method (16) and the sections were mounted in serial order. The immunostain was intensified with 0.005% osmium tetroxide and 0.1% thiocarbohydrazide, and the sections were coverslipped under Permount. Each section was drawn using a projecting microscope, and stained cells were subsequently plotted on the diagrams with the aid of a camera lucida. The numbers of immunoreactive neurons in each of the several neuroanatomical structures evaluated in HAL rats were recorded and compared directly with the numbers recorded in the same structures at corresponding levels of the VEH brains. Statistical evaluation of the data was accomplished using the Mann-Whitney u-test.

ADMINISTRATION

127

Results

Immunoreactive (IR) cells (Fig 1) were identified based upon criteria established in earlier reports (17.30,32.37,38): 1. granularreactionproduct, 2. proximal portion of dendrites stained, 3. nucleus unstained. Slight neuronal staining not conforming to these criteria occasionally was observed in the polymorph layer of olfactory tubercle and endopyriform area. Such spuriously labelled cells, however, were eliminated from the evaluation in consideration of their coincident presence in sections incubated with preabsorbed antiserum, since this series reflected a degree of non-specific, staining which varied somewhat from case to case. A representation of the mean numbers of stained neurons at corresponding levels of the HAL and VEH sample groups is given in Figure 2A-L. Abscissae depict consecutive coronal levels along the rostrocaudal axis of each structure. The levels, which are identified by consecutive integers, with lower numbers more rostral, are separated by approximately 300 pm. Data generated using sections from the HAL brains and the corresponding sections from the pair-matched VEH brains are displayed at each level. Ordinates depict the numbers of NT-IR neurons, expressed

Fig 1 A. Neurotensin immunoreactive (NT-IR) neurons (arrowheads) in the olfactory tubercle (Tu) and nucleus accumbens (Acb) of a rat given 2mg/kg of haloperidol24 and 4 hours prior to sacrifice. B. An enlargement of the boxed area. exhibits the cytoplasmic staining characteristics, relatively unstained nucleus (arrowheads), and stained proximal dendrites of NT-IR neurons, VP: ventral pallidum. Scale bars = 5OOpm: inset: 1OOpm.

128

NEUROPEPTIDES Tu

B

Figure

Acb

40

1

dmCPu

D

60

2

Axa* Idontltlod In leoand and text

1

VICCU

-s-l--s-‘-s-“iT~~

,*54se7

F

VrnCPU

!1.,,I5cJdAx

10 0

-

-TJi-.--‘-“-

1

H

10 10

0

B8T

1

Il-4

lkf v--r 1

1

10

a

b0

”

J CPUtoll 1

.O-

DO-

¶O-

lo-

o_

I:: I.

L

K

1

60

I

6

@

7

0

0

101*

En

w-1

NVIR

NEURONS

INCREASED

FOLLOWING

ACUTE

HALOPERIDOL

as medians and ranges. Each point reflects sixteen separate counts, generated from evaluation of structures on both sides of the eight HAL and eight VEH brains. For convenience, median values of HAL rats are connected by solid lines, of VEH rats by broken lines. The structures listed in Figure 2 are organized in approximately rostra1 to caudal sequence. In olfactory tubercle (Tu, Fig 2A), nucleus accumbens (Acb, Fig 2B), and the body of the caudateputamen (CPU, Figs 2C-F), which was evaluated as dorsolateral (dl, Fig ZC), dorsomedial (dm, Fig 2D), ventrolateral (VI, Fig 2E), and ventromedial (vm, Fig 2F) quadrants, the numbers of NT-IR ‘neurons in the HAL rats exceeded the numbers in the VEH controls (p < 0.01 using the Mann-Whitney u-test). The amount of increase varied among these areas with the most striking effects recorded in Tu, Acb, and the dorsomedial quadrant of the CPU. Dorsolateral and ventromedial CPU were less affected with few NT-IR neurons, being counted in the HAL or VEH brains. Numbers of NT-IR neurons were moderately elevated in the tail of the CPU (Fig 2J). No differences between HAL and VEH brains were recorded in the lateral septum (LS, Fig 2G), bed nucleus of the stria terminalis (BST, Fig 2H). or central amygdaloid nucleus (CA, Fig 2K). Interestingly, counts of NT-IR neurons were depressed compared to controls (p < 0.05) in caudal parts of the deep piriform cortical layers [endopiriform area, (En)], Figure 21 indicated by asterisk, and medial amygdaloid area (MeA. Fig 2L).

ADMINISTRATION

129

Evaluation of the immunoreactivity displayed by fibers and axon terminals was restricted primarily to known projection fields of the ventral striatum. These structures included GP, VP, BST, the lateral hypothalamic area, zona incerta, substantia nigra, VTA, the retrorubral field and nucleus, pedunculopontine tegmental nucleus, parabrachial region and nucleus, and central grey of the brainstem. Collectively, these contribute to a more or less continuous column of NT-immunoreactivity which extended along the trajectory of the medial forebrain bundle from the striatum to the pontomesencephalic junction. With one exception, indicated below, subjective evaluation revealed no obvious differences between the HAL and VEH groups, either in the numbers of immunostained elements or the intensity of staining along this continuum. Clear and consistent increases, however, were observed in the GP, not necessarily as a supplemented intensity of immunostaining, but rather, and more persuasively, as a conspicuous expansion in the HAL-treated brains of the proportional area of the nucleus which displayed NT immunoreactivity (Fig 3). Untreated (38, 39) or VEH brains exhibit NT immunoreactivity only in a thin, medialmost strip of GP which extends darsoventrally along the lateral edges of the BST and lateral ventricle. In the HAL brains, the NT-IR zone consistently included 60% to 80% of the mediolateral dimension of the GP.

Discussion Fig 2 Numbers of immunostained neurons at corresponding levels in several stuctures of brains receiving 2mg/kg haloperidol (HAL) or vehicle (VEH) 24 and 4 hours prior to sacrifice. Abscissae: coronal levels along the rostrocaudal axis of each structure with lower integers indicating more rostra1 levels. Ordinates: numbers of NT-IR neurons expressed as the medians and ranges of counts done bilaterally in eight HAL and eight VEH brains. Each point reflects sixteen separate counts, generated from evaluation of structures on both sides of the eight HAL and eight VEH brains. HAL medians are connected by solid lines, VEH medians by broken lines. Structures evaluated include olfactory tubercle (Tu, A), nucleus accumbens (Acb, B), dorsolateral (dl, C), dorsomedial (dm, D), ventrolateral (VI. E), and ventromedial (vm, F) quadrants of the caudate-putamen (CPU), lateral spetum (LS, G), bed nucleus of stria terminalis (BST, H), endopiriform area (En, I). tail of the CPU (J). central amygdaloid nucleus (CeA, K), and medial amygdaloid nucleus (MeA, L), HAL values exceed VEH values (Mann-Whitney u-test) in A-F and J (p < 0.01). VEH values exceed HAL vahres in the caudate part of En (* in I) and L (p < 0.05). No differences were observed in G. H. and

IL

The results of this study indicate that increases in NT-immunoreactivity reported to occur in the Acb, adjacent ventromedial CPU and striatal parts of the Tu after neuroleptic treatment (9,lO) can be related, at least in part, to increases in the amount of an NT-immunoreactive substance associated with neuronal perikarya in these regions. It is interesting in this regard that perikaryal staining was absent in midbrain dopaminergic areas in the HAL and VEH groups, although NT-IR perikarya can readily be demonstrated in these regions following colchicine pretreatment (37) or an immunocytochemical protocol that favors cell staining (e.g. 15). This result, which suggests that the dramatic increase of NT staining found in ventral striatal structures following neuroleptic treatment is not seen in midbrain regions, is supported by the RIA data (9, 10).

130

NEUROPEPTIDES

Fig 3 Neurotensin immunoreactivity at corresponding levels of the globus pallidus in rat braina treated with 2mgIkg haloperidol (A) or vehicle (B) 24 and 4 hours prior to sacrifice. Note the expanded territory displaying NT-IR in the HAL globus pal&s (GP) v: lateral ventricle; ac: temporal limb of the anterior commissure. Scale bar = Worn.

Differences were not observed in the numbers of NT-IR neurons in the bed nucleus of stria terminalis, septum, and central amygdala of VEH and HAL groups, and it is interesting that these structures contain low levels of dopamine receptors (28,29). Notably, the numbers of NT-stained perikarya declined following HAL treatment in medial amygdaloid nucleus and caudal parts of the endopyriform area. It may be correlative that regions that showed an increase in the numbers of NT-IR cells following HAL-treatment (CPU, Acb, Tu) are regions rich in both D-l and D-2 receptors, whereas those regions in which the numbers of NT-stained perikarya remained unchanged or declined are regions where levels of D-l receptors are reported to exceed those of D-2 receptors (6, 27-29). D-l receptors are reported to stimulate adenylate cyclase activity. whereas D-2 receptors are not linked or inhibitory of adenylate cyclase activity. If it is supposed that adenylate cyclase activity positively correlates with NT production, blockade of D-l receptors may effectively reduce both adenylate cyclase activity and NT synthesis. Conversely, in ventral striatal regions where both D-l and D-2 receptors are abundant, blockade of the D-2 receptor class by haloperidol may tend to predominate, since haloperidol binds the D-2 receptor with greater affinity than the D-l. The resultant disinhibition of adenylate cyclase might result in facilitated production of NT in these regions. This speculation is supported by the work of Frey et al. (9) who showed that the selective D-l receptor antagonist, SCH-23390, reduced NT in the striatum below the control levels.

As applied to tissue sections, immunocytochemica1 technique is generally considered to allow a purely qualitative evaluation, although it has been observed that the intensity of staining does tend to reflect the quantitative measure of antigen as assayed by radio immunoassay (13,21,23). It was therefore, that no consistent suprising, supplementation of staining intensity was observed in the projection fields of the ventromedial striatal regions where the numbers of NT-IR neurons increased following HAL. It may be that axonal transport of newly synthesized NT from the perikaryon in the normal rat brain is carried out with an efficiency that keeps perikaryal NT at levels undetectable with routine immunohistochemistry and terminal NT levels at near maximal, as demonstrable with the same methods. While HAL facilitates visualization of striatal NT-IR perikarya, limitations of the immunoperoxidase method may interfere with observation of increased NT levels in axon terminals, if they obtain. In the GP, a clear expansion of the proportional area of that structure which displayed NT-IR was observed. This finding also was intriguing because NT-IR following HAL extended into parts of the GP that are topographically related to striatal districts that display few NT-IR perikarya following HAL, Hz02 pretreatment of sections (37) or colchicine administration prior to sacrifice (36). A possible explanation is that HAL increased NT synthesis in striatal neurons projecting to those regions to levels which allowed demonstration of the NT-IR terminal field, but still left NT-IR

NT/IR NEURONS

INCREASED

FOLLOWING

ACUTE

HALOPERIDOL

neurons in the topographcially appropriate striatal districts undetectable with immunoperoxidase methods. This situation is entirely analogous to the condition in untreated rats where ventral pallidal axon terminal immunoreactivity is striking in the absence of detectable cell staining in ventral striatum (38, 39). Whether neurons in the dorsolateral reaches of the adult rat striatum synthesize NT under basal conditions remains unaddressed, although neurons in these areas clearly are NT-IR during the perinatal period of the rat (7) and in the cat (30). A functional consideration. NT transmission from ventral striatal neurons projecting to the midbrain may increase in a compensatory manner as a response to reduction of dopaminergic neurotransmission in the “ascending” mesolimbic projection. For example, neuroleptic blockade in ventral striatum, by reducing the dopaminergic input to cells which project to VTA, based on the present evidence, may stimulate increased production of NT by those neurons and possibly increased release of NT in the VTA. That an increase in NT released in VTA might result in activation of the dopaminergic mesolimbic system gains support from several studies (1, 18-20, 34). The potential involvement of such a feedback mechanism as a confounding factor in therapeutic neuroleptic treatment is intriguing, but the interplay between striatal neurotensin production and striatopetal dopaminergic neurotransmission, as an influence upon normal homeostasis in the ventral basal ganglia, presents an equally interesting problem. Acknowledgements This work was supported by a Summer Research Training Fellowship awarded to K.W.E. by the St. Louis University School of Medicine, and USPHS NIH grant NS-23805 and a grant from the American Parkinson Disease Association, both to D.S.Z. Steve Johnson provided expert technical assistance, and Sharon Hughes is gratefully acknowledged for preparing the manuscript. We thank G. Fred Wooten, M.D.. of the University of Virginia, Department of Neurology, for helpful advice.

References 1. Andrade. R. and Aghajanian, G. K. (1981). Neurotensin selectively activates dopaminergic neurons of the substantia nigra. Sot. Neurosci. Abstr. 7: 753. 2. Agnati, L. F.. Fuxe, K., Benfenati, F. and Battistini, H. (1983). Neurotensin in vitro markedly reduces the affinity in subcortical limbic 3H-N-propylnorapomorphine binding sites. Acta Physiol. Stand. 119: 459-461.

ADMINISTRATION

131

3. Carraway, R. and Leeman. S. E. (1973). The isolation of a new hypotensive peptide, neurotensin, from bovine hypothalami. J. Biol. Chem. 248: 6854-6861. 4. Carraway. R. and Leeman, S. E. (1975). The synthesis of neurotensin. J. Biol. Chem. 250: 1912-1918. 5. Carraway, R. and Leeman, S. E. (1975). The amino acid sequence of a hypothalamic peptide. neurotensin. 250: 1907-1911. 6. Dawson, T. M., Gehlert, D. R. and Walmsley. J. K. (1987). Quantitative autoradioeraohic localization of central dopamine D1 and Dz receitdrs. In: Breese, G. R. and Creese, I. (eds.). Neurobiology of central Dt-dopamine receptors. Plenum, New York and London, p. 93-118. 7. Eggerman. K. W.. Sprung, R. F., Wesche. D. E., Payne, E. and Zahm, D. S. (1987). Distributionsandrelationships to each other of clusters of neurotensin- and substance P-immunoreactive neurons in the caudate-putamen and ventral striatum of the Sprague-Dawley rat in the early postnatal period. Sot. Neurosci. Abstr. 13: 995. 8. Emson. P.C. Goedert, M., Horsefield, P.. Rioux. F. and St. Pierre, S. (1982). The regional distribution and chromatographic characterization of neurotensin-like immunoreactivity in the rat central nervous system. J.Neurochem. 38: 992-999. 9. Frey. P. Fuxe. J.. Eneroth, P. and Agnati. L. F. (1986). Effects of acute and long-term treatment with neuroleptics on regional telencephalic neurotensin levels in the male rat. Neurochem. Int. 8: 429-434. 10 Govoni, S.. Hong. J. S., Yang. H. -Y. T. and Costa, E. (1980). Increase of neurotensin content elicited by neuroleptics in nucleus accumbens. J. Pharmacol. Exp. Therap. 215: 313-417. 11 Gregory, N. E., Birkemo. L. S., Nemeroff, C. B. and Prange. A. J. Jr. (1981). Neurotensin blocks certain amphetamine-induced behaviors. Nature 291: 43-76. 12 Heimer. L.. Alheid, G. F. and Zaborszky, L. (1985). Basal ganglia. In: Paxinos. G. (ed.). The Rat Nervous System. Academic Press, New York, p. 37-86. 13 Hellendall, R. P., Godfrey, D. A.. Ross, C. D., Armstrong, D. M. and Price, J. L. (1986). The distribution of choline acetyltransferase in the rat amygdaloid complex and adjacent cortical areas, as determined by quantitative microassay and immunohistochemistry. J. Comp. Neurol. 249: 486.498. 14. Herve. D., Tassin, J. P.. Studier. J. M., Dana, D., Kitabgi. P., Vincent. J. P., Glowinski, J. and Rostene. W. (1986). Dopaminergic control of ‘ZSI-labeled neurotensin binding site density in corticolimbic structures of the rat brain. Proc. Natl. Acad. Sci. USA 83: 6203-6207. 15 Hokfelt. T.. Everitt. B. J., Theodorsson-Norheim. E. and Goldstein. M. (1984). Occurrence of neurotensinlike immunoreactivity in subpopulations of hypothalamic, mesencephalic. and medullary catecholamine neurons, J Comp. Neurol. 222: 543-559. 16 Itoh. 2.. Akiva. K.. Nomura, S., Miguno, N., Nakamura. Y. and Sugimoto, T. (1979). Application of a coupled oxidation reaction to electron microscopic demonstration of horseradish peroxidase: Cobalt-glucose oxidase method. Brain Res. 175: 341-347. 17 Jennes, L.. Stumpf. W. E. and Kalivas, P. W. (1982). Neurotensin: Topographical distribution in rat brain by immunohistochemistry. J. Comp. Neurol. 210: 211-224. 18 Kalivas, P. W.. Burgress, S. K., Nemeroff, C. B. and Prang. A. J. Jr. (1981). Stimulant effects of neurotensin

132

19.

20.

21.

22.

23.

24.

25. 26.

27.

28.

microinjection into the ventral tegmentai area. Sot. Neuroscil. Abstr. 7: 31. Kalivas, P. W., Nemeroff, C. B. and Prange. A. J. (1981). Increases in spontaneous motor activity following infusion of neurotensin into the ventral tegmental area. Brain Res. 229: 525-529. Kalivas, P. W., Nemeroff, C. B., Prange, A. J. Jr. (1982). Neuroanatomical sites of action of neurotensin. Ann. NY Acad. Aci. 400: 307-318. Kawata, M., Nakao, K., Morrii, N., Kiso, Y.. Yamashita, H.. Imura. H. and Sano. Y. (1985). Atria1 natiuretic polypeptide: Topographical distribution in the rat brain by radioimmunoassay and immunohistochemistry. Neuroscience 16: 521-546. Lipton, M. A., Ervin, G. N., Birkemo, I. S. et al. (1979). Neurotensin-neuroleptic similarities: An example of peptide-catecholamine interaction. In: Usdin, E., Kopin, I. S., Barchas, J. (eds.). Catecholamines. Basic and Clinical Frontiers. Pergammon Press, New York, p. 657-662. McLean, S., Bannon, M. J., Zamir, N. and Pert, C. B. (1985). Comparison of the substance P and dynorphin-containing projections to the substantia nigra: A radioimmunocytochemical and biochemical study. Brain Res. 361: 185-192. Nauta, W. J. H. andDomesick, V. B. (1984). Afferent and efferent relationships of the basal ganglia. In: Evered, D. and O’Connor, M. (eds.). Functions of the basal ganglia. Ciba Foundation Symposium 107. Pitman, London, p. 3-23. Nemeroff, C. B. (1980). Neurotensin: Perchance an endogenous neuroleptic? Biol. Psych. 15: 283-286. Quirion, R., Gaudreau, P.,St. Pierre, S. Rioux, F. and Pert, C. B. (1982). Autoradiographic distribution of 3Hneurotensin receptors in rat brain. Visualization by tritium sensitive films. Peptides 33: 757-763. Richfield, E. K., Young, A. B. and Penney, J. B. (1987). Comparative distribution of dopamine D-l and D-2 receptors in the basal ganglia of turtles, pigeons, rats, cats, and monkeys. J. Comp. Neurol. 262: 446-463. Savasta, M. Dubois, A. and Scatton, B. (1986). Autoradiographic localization of Dl dopamine receptors in the rat

NEUROPEPTIDES

brain with [“H]SCH 23390. Brain Res. 375: 291-301. 29. Seeman. P. (1980). Brain dopamine receptors. Pharm. Rev. 32: 229-313. 30. Sugimoto, T. and Mizuno, N. (1987). Neurotensin in projection neurons of the striatum and nucleus accumbens. with reference to coexistence with enkephalin and GABA: An immunohistochemical study in cat. J. Comp. Neurol. 257: 383-395. 31. Uhl, G. R. (1982). Distribution of neurotensin and its receptor in the central nervous system. Ann. NY Acad. Sci. 400: 132-149. 32. Uhl, C. R., Goodman. R. R. and Snyder, S. H. (1979). Neurotensin containing cell bodies, fibers and nerve terminals in the brain stem of the rat. Immunohistochemical mapping. Brain Res. 167: 77-91. 33. Uhl, G. R., Kahan, M. J. and Snyder, S. H. (1977). Neurotensin: Immunohistochemical localization in rat central nervous system. PNAS 74: 4059-4063. 34. Widerlov. E.. Kilts, C., Mueller, R., Mailman, R., Nemeroff, C.. Prange. A. J. Jr. and Breese, G. R. (1981). Neurotensin increase dopaminemetabolites in rat brain. Pharmacologist 23: 139-142. 35. Young, W. S. III and Kuhar, M. J. (1979). Neurotensin receptors: autoradiographic localization in rat CNS. Eur. J. Pharmacol. 59: 161-163. 36. Young, W. S. III and Kuhar, M. J. (1981). Neurotensin receptor localization by light microscopic autoradiography in rat brain. Brain Res. 206: 273-285. 37. Zahm, D. S. (1987). Concentration of neurotensin immunoreactive neurons in the ventral striatum of the rat: ventromedial caudate-putamen, nucleus accumbens and olfactory tubercle. Neurosci. Lett. 81: 41-47. 38. Zahm, D. S. and Heimer. L. (1988). Ventral striatopallidal parts of the basal ganglia in the rat: I. Neurochemical compartmentation as reflected by the distributions of neurotensin and substance P immunoreactivity. J. Comp. Neurol. (in press). 39. Zahm, D. S. and Heimer, L. (1985). Regional specialization of the basal ganglia; the corticostriatopallidal system in the olfactory tubercle. Sot. Neurosci. Abstr. 11: 199.