Distribution of cocaine recognition sites in rat brain: In vitro and ex vivo autoradiography with [125I]RTI-55

~'t~:.:g/)i ELSEVIER

Journal of Chemical Neuroanatomy 7 (1994) 13-23

Distribution of cocaine recognition sites in rat brain: in vitro and ex vivo autoradiography with [125I]RTI-55 Masahiro Fujita *a, Shoichi Shimada b, Kazuki Fukuchi a, Masaya Tohyama b, Tsunehiko Nishimura a aDiviMon of Tracer Kinetics, Biomedical Research Center, bDepartment of Anatomy and Neuroscience, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka. Japan Accepted 23 December 1993

Abstract

The distribution of binding sites of [125I]RTI-55 (3~-(4-iodophenyl)tropan-2/3-carboxylic acid methyl ester), a phenyl tropane analog of cocaine, and the selective labelling of the dopamine transporter (DAT) were studied by in vitro and ex vivo autoradiography in the rat whole brain. Recent evidence has shown that RTI-55 binds to not only DAT but also serotonin transporter (SHTT). In the present study, in vitro autoradiography revealed that [125I]RTI-55 bound to the olfactory tubercle, the caudate putamen, the accumbens nucleus, the midline and lateral geniculate nuclei of the thalamus, the hypothalamic nuclei, the substantia nigra compact part, the subthalamic nucleus, the ventral tegmental area, the superior colliculus, the dorsal raphe nucleus, and the facial nucleus. Further, in the presence of clomipramine, a selective ligand for 5HTT, [125I]RTI-55 binding was remarkably inhibited in the midline and lateral geniculate nuclei of the thalamus, the hypothalamic nuclei, the superior colliculus, the dorsal raphe nucleus, and the facial nucleus, while [125I]RTI-55 binding remained in the olfactory tubercle, the caudate putamen, the accumbens nucleus, the substantia nigra compact part, the subthalamic nucleus, and the ventral tegmental area. These findings suggest that [12SI]RTI-55 binds to 5HTT in the former areas and to DAT in the latter areas. It is therefore concluded that RTI-55 is a suitable ligand for studying the action of cocaine in whole brain regions, including the thalamus, the hypothalamus and the dorsal raphe nucleus, regions in which cocaine is thought to act evoking several neurological effects, e.g., analgesia and elevation of adrenocorticotropic hormone. DAT was also labelled selectively both in vitro and in vivo using [1251]RTI-55 combined with clomipramine. Therefore, radiolabelled RTI-55, combined with unlabelled clomipramine, which displaces its binding to 5HI'F, also appears to be suitable for the selective imaging of DAT in vivo.

Keyworda: Dopamine transporter; RTI-55; Cocaine recognition sites; Serotonin transporter; Brain imaging

1. Introduction

Cocaine, prototype psychomotor stimulant drug of abuse, and related drugs, are thought to act mainly via binding to dopamine transporter (DAT) and by inhibiting dopamine (DA) reuptake (Ritz et al., 1987; Bergman et al., 1989; Spealman et al., 1989). Radioligands for cocaine binding sites have been developed to investigate the psycho-pharmacological effects of cocaine, and the distribution of their binding sites has been demonstrated (Carroll et al., 1992). These ligands include [3H]cocaine (Reith et al., 1986; Calligaro and Eldefrawi, 1988; Madras et al., 1989a), * Corresponding author. 0891-0618/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0891-0618(94)00002-3

[3H]nomifensine (Dubocovich and Zahniser, 1985), [3H]methylphenidate (Janowsky et al., 1985), [3H] l-[2(diphenylmethoxy)-ethyl]-4-(3-phenylpropyl) piperazine (GBR 12935) (Janowsky et al., 1986) and [3H]mazindol (Javitch et al., 1984). Further, a series of potent cocaine analogs has been developed recently; 3/~-(4-fluorophenyl) tropan-2/3carboxylic acid methyl ester (WIN-35428) (Madras et al., 1989b) and its iodinated analog 3B-(4-iodophenyl)tropan-2B-carboxylic acid methyl ester (RTI-55) (Boja et al., 1991) being two of these. Since these ligands have both high- and low-affinity binding sites, and provoke behavioral effects like those of cocaine (Cline et al., 1992a), they are suitable for use in the investigation of cocaine recognition sites and should clarify the

14

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

mechanisms responsible for the psychostimulant action of cocaine. The distribution of the binding sites of [3H]WlN35428 (Kaufman et al., 1991; Kaufman and Madras, 1992) and [125I]RTI-55 (Boja et al., 1992; Cline et al., 1992b; Kaufman and Madras, 1992; Scheffel et al., 1992) has been explored to some extent and it has been shown that these ligands bind not only to DAT but also to serotonin transporter (5HTT). However, to understand complicated psychostimulant actions of cocaine better, it is essential to determine whether the cocaine analogs bind to DAT or 5HTT in each brain region. Accordingly, in the present study, we investigated [t25I]RTI-55 binding sites in vitro throughout the whole brain in the presence and absence of clomipramine, a selective ligand to 5HTT, and we labelled DAT selectively. RTI-55 may also be useful as an imaging agent for labelling DAT, especially for disorders of the dopaminergic system such as Parkinson's disease, which is caused by the extensive loss of dopaminergic neurons in the substantia nigra compact part (SNC). Indeed, RTI-55 began to be used as a radio-pharmaceutical for single-photon-emission computed tomography (SPECT) studies in baboons (Carroll et al., 1991; Innis et al., 1991; Shaya et al., 1992). For diagnostic purposes, it is better to label DAT selectively; we therefore also labelled DAT selectively in vivo, using [125I]RTI-55 combined with unlabelled clomipramine. 2. Materials and methods

2.1. In vitro experiment Tissue preparation Male Wistar rats weighing 200 g were decapitated under pentobarbital anesthesia and the brains were rapidly removed and frozen in powdered dry ice. Tissues were sectioned at a thickness of 20 /zm in a cryostat, thaw-mounted onto silane-coated slides, stored at -80°C, and dried rapidly with a stream of cool air before being used for autoradiographic experiments. A utoradiography Triplicate adjacent slide-mounted tissue sections were preincubated in buffer (50 mM Tris-HCl containing 100 mM NaCI, pH 7.4, at 4°C) for 10 s at 4°C, following which the slides were incubated in buffer containing 0.1 nM [12SI]RTI-55 for 2 h at 4°C to measure total binding, [t25IIRTI-55 plus 1 mM (-)-cocaine to explore cocaine-displaceable specific binding, or [125I]RTI-55 plus 10/~M clomipramine to determine the distribution of clomipramine-displaceable binding sites of [125I]RTI-55. Before being incubated with [125I]RTI-55 plus cocaine, slides were preincubated in buffer containing 1 mM (-)-cocaine for 20 min at 4°C. In addition to this

experiment, one series of whole brain tissue sections was incubated in the buffer containing 0.1 nM [125I]RTI-55 and 1 ~M GBR-12909 to determine the distribution of GBR-120909-displaceable binding sites. Incubation was terminated by two consecutive 1 min washes in fresh icecold buffer. After a final dipping in ice-cold distilled water, the slides were dried rapidly with a stream of cool air. Slide-mounted tissue sections and calibrated standards ([t25I]Microscale, Amersham) were placed in Xray cassettes along with imaging film (X-OMAT AR, Kodak) and exposed for 2-14 days, after which time the films were removed and developed. 2.2. Ex vivo experiment Injection procedures and tissue preparation As it is easier to inject drugs intravenously in larger rats, rats weighing 300 g rather than 200 g were used in ex vivo experiments. Male Wistar rats were anesthetized with pentobarbital and prepared with intravenous (i.v.) catheters (left femoral vein) for drug injections; 814 kBq (10 pmol) [125I]RTI-55, in 0.2 ml saline, was administered i.v., with i.v. injections of clomipramine (1 /~mol, n = 1; 10/zmol, n = 2) dissolved in 0.5 ml saline or 0.5 ml saline (n = 2) 5 min prior to this injection of [125I]RTI-55. Sixty minutes after the i.v. administration of [125I]RTI-55, the rats were sacrificed and the brains removed and sectioned, as described above. Slidemounted sections were dried rapidly with a stream of cool air before being used for autoradiographic experiments. Autoradiography Duplicate tissue sections and calibrated standards ([125IlMicroscale, Amersham) were placed in X-ray cassettes along with [3HlHyperfilm (Amersham) and exposed for 4 weeks; the films were then developed. 2.3. Localization and quantitative analysis of [1251]RTI55 binding sites in vitro and ex vivo Adjacent sections for in vitro experiments and sections used for autoradiographic experiments for ex vivo study were stained with thionin. When thionin-stained sections and autoradiographic films were compared, anatomical areas were identified and named according to the atlas of Paxinos and Watson (1986). 2.4. Quantitative analysis of in vitro and ex vivo autoradiographs Regional binding of radioligands in the brain was measured by utilizing imaging software for microcomputer systems (MCID Image Analysis System, Imaging Research, St Catharines, Ontario). The density of the binding was expressed in fmol/mg tissue according to

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

the optical density of the calibrated standards. Three to five bilateral measurements were made for each nucleus and the mean value was calculated. For in vitro experiments, values for specific binding without clomipramine were obtained by subtracting the amount of binding with cocaine from the total value; specific binding with clomipramine was obtained by subtracting the amount of binding with cocaine from that with clomipramine.

and GBR-12909 were obtained Biochemicals, Inc. (Natick, MA).

15

from

Research

3. Results

3.1. Distribution of [uSI]RTI-55 binding sites without clomipramine in vitro The detailed distribution of [125I]RTI-55 binding sites without clomipramine in vitro is shown in Table 1. Under the conditions used for the autoradiographic procedures, 80-95% of the total binding was cocainedisplaceable specific. We arbitrarily assigned areas in

2.5. Chemicals and radioligands [125I]RTI-55 (81.4 TBq/mmol) was purchased from New England Nuclear (Boston, MA). Clomipramine

Table 1 Densities of [125I]RTI-55 specific binding in vitro in rat brain determined by quantitative autoradiography Brain regions

Without clomipramine (fmol/mg tissue)

With clomipramine (fmol/mg tissue)

Anterior olfactory nucleus Anterior olfactory nucleus, medial part Cerebral cortex Perirhinal cortex Hippocampus CA 1 Hippocampus CA2, 3 Olfactory tubercle Amygdaloid nucleus Amygdalopiriform transition Caudate putamen (whole) Caudate putamen, lateral half Caudate putamen, medial half Accumbens nucleus Globus pallidus Lateral septal nucleus Medial septal nucleus Diagonal band Anteroventral thalamic nucleus Reuniens thalamic nucleus Midline nuclei and mediodorsal nucleus of the thalamus Laterodorsal thalamic nucleus Lateral geniculate nucleus Zona incerta Subthalamic nucleus Lateral hypothalamic nucleus Other hypothalamic nucleus Suprachiasmatic nucleus Substantia nigra (whole) Substantia nigra, compact part Substantia nigra, reticular part Ventral tegmental area Superficial gray layer of superior colliculus Other deep layers of superior colliculus Inferior colliculus Dorsal raphe nucleus Median raphe nucleus Facial nucleus Locus coeruleus Laterodorsal tegmental nucleus Cerebellar cortex

1.514 3.026 0.924 2.016 0.971 1.324 4.426 2.239 2.659 4.945 5.566 4.560 4.845 3.058 1.190 2.020 2.385 1.953 2.761 2.259

4. 0.157 4. 0.268 4- 0.201 4- 0.455 4. 0.186 4. 0.405 4. 0.650 4. 0.297 4- 0.403 4- 0.367 4. 0.419 4- 0.308 4. 0.649 4. 0.380 4- 0.130 4- 0.443 4. 0.326 4. 0.062 4. 0.413 4. 0.302

<0.25 <0.25 <0.25 <0.25 <0.25 <0.25 1.857 4- 0.169 <0.25 <0.25 3.608 4. 0.406 3.613 4- 0.385 2.531 4- 0.394 2.417 4. 0.260 0.485 4. 0.065 <0.25 <0.25 <0.25 <0.25 <0.25 <0.25

1.655 2.567 1.570 2.385 2.565 1.959 2.063 2.267

4. 0.806 4- 0.533 4. 0.364 4. 0.617 4. 0.459 4. 0.288 4. 0.098 4. 0.210

<0.25 <0.25 <0.25 0.650 4. 0.087 <0.25 <0.25 <0.25

3.516 2.695 2.343 1.477 4.539 2.995 2.115 2.805 2.357 0.386

+ 4. 4. 4. 4. 4. 4. 44. 4.

0.340 0.204 0.317 0.161 0.684 0.310 0.353 0.281 0.228 0.066

0.637 4- 0.196 0.121 4- 0.074 0.846 4- 0.206 <0.25 <0.25 <0.25 <0.25 <0.25 <0.25 <0.25 <0.25 <0.25

The data are expressed in the form of mean 4- S.D. and represent the specific [125I]RTI-55 binding at a concentration of 0.1 nM and 10/~M of clomipramine (in the 'with clomipramine' column). Non-specific binding was determined with 1 mM (-)-cocaine.

16

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

which the amount of the ligand was above 4.5 fmol/mg tissue as very high density [125I]RTI-55 binding sites. High density binding sites were defined as those which bound between 3 and 4.5 fmol/mg tissue; moderate density areas were defined as those containing ligand binding of between 1.5 and 3 fmol/mg tissue; and low level binding sites were those below 1.5 fmol/mg tissue. The distribution of [125I]RTI-55 binding sites without clomipramine is shown in the photographic atlas presented in Figs. 1A-C, 2A-C, 3A-C, 4 A - C and described in detail below. 3.2. Forebrain

Markedly high levels of [125I]RTI-55 binding were found in the medial part of the anterior olfactory nucleus (Fig. 1A). In other parts of the anterior olfactory nucleus, a moderate content of [125I]RTI-55 was observed. Moderate levels of [125I]RTI-55 were detected in the perirhinal cortex (Figs. 1C, 2A), while low levels of binding sites were observed in all regions of the cerebral

cortex. Although there were no large differences in the levels of [125IIRTI-55 binding among individual layers of the cerebral cortex, the binding in the fourth layer was somewhat higher than in the other layers. In the hippocampal formation, low levels of [125I]RTI-55 binding sites were detected, while [t25I]RTI-55 levels in the CA2 and CA3 fields were higher than those in the CAI field (Figs. 2B,C, 3A). The medial septum and the medial nuclei of the diagonal band of Broca showed moderate [125I]RTI-55 binding density and the lateral septum contained low levels of binding sites. Moderate levels of [125I]RTI-55 binding were found in the amygdaloid complex and higher levels were observed in the amygdalopiriform transition. No clear differences were detected in the levels of [n25I]RTI-55 binding among individual nuclei of the amygdaloid complex. The striatum displayed the highest levels of [L25I]RTI-55 binding of all brain regions (Fig. IB,C). Of the nuclei in the striatum, the caudate putamen, the accumbens nucleus and the olfactory tubercle showed higher binding levels and the globus pallidus showed

D k H

h,

Fig. 1. In vitro autoradiographicimages showingthe distribution of [1251]RTI-55binding sites in coronal sectionsof the rat brain. Groups of three adjacent serial sections were incubated in the presenceof 0.1 nM [1251]RTI-55(A-C), 0.1 nM [1251]RTI-55plus 10 t~M clomipramine(D-F) or 0. I nM [ t251]RTI-55plus I mM (-)-cocaine (G-I). Dark areas indicate relevantbinding levels.Acb, accumbensnucleus; AOM, anterior olfactory nucleus; medialpart; AV, anteroventralthalamic nucleus; CPu, caudate putamen; GP, globuspallidus; PRh, perirhinalcortex; PV, paraventricular thalamic nucleus; Tu-olfactorytubercle. Bar = 2.5 mm

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

lower levels of binding sites. In the caudate putamen, the lateral half had a greater content of the ligand than the medial half (Fig. IB, Table 1). 3.3. Diencephalon

In the thalamus, several nuclei showed moderate binding levels; the anteroventral (Fig. 1C), reuniens (Fig. 2A), mediodorsal, lateral geniculate (Fig. 3A,B), subthalamic and midline nuclei, e.g., the paraventricular and paratenial nuclei, showing higher binding levels of [125IIRTI-55. A moderate [125I]RTI-55 content was also found in the hypothalamus (Fig. 2B,C), especially in the lateral hypothalamic nucleus (Fig. 2C).

17

the reticular part (SNR) were not discriminated (Fig. 3B,C). A high level of binding sites was found in the ventral tegmental area (VTA) (Fig. 3B,C). Raphe nuclei were rich in [125I]RTI-55 binding sites; the dorsal raphe nucleus contained very high levels of [125I]RTI-55 binding sites, while the median raphe nucleus showed moderate levels (Fig. 4A). The superior colliculus showed moderate levels of [125I]RTI-55 binding, with higher levels in the superficial gray layer than in the other layers located in the ventral areas (Fig. 4A). In other regions in the brainstem, the facial nucleus (Fig. 4C), the locus coeruleus (Fig. 4B) and the laterodorsal tegmental nucleus (Fig. 4B) contained moderate levels of [1251]RTI-55 binding sites, while the inferior colliculus (Fig. 4B) contained low levels.

3.4. Brainstem 3.5. Cerebellum

The substantia nigra contained moderate levels of [125IIRTI-55 binding sites. Differences in [125I]RTI-55 binding density between the compact part (SNC) and

The lowest level of binding sites in the brain was detected in the cerebellar cortex (Fig. 4C).

Fig. 2. In vitro autoradiographic images showing the distribution of [ 125I]RTI-55binding sites. Groups of three adjacent serial sections were incubated in the presence of 0.1 nM [125I]RTI-55 (A-C), 0.1 nM [1251]RTI-55 plus 10 ~M clomipramine (D-F)or 0.1 nM [1251]RTI-55 plus 1 mM (-)-cocaine (G-I). Amy, amygdaloid nucleus; CA 2, 3, CA2 and CA3 sectors of Ammon's horn; CPu, caudate putamen; LD-laterodorsal thalamic nucleus; LH, lateral hypothalamic area; PRh, perirhinal cortex; Re, reuniens thalamic nucleus; SCh, suprachiasmatic nucleus; STh, subthalamic nucleus; ZI, zona incerta. Bar = 2.5 mm.

18

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

13

m

Fig. 3. In vitro autoradiographic image showing the distribution of [1251]RTI-55 binding sites. Groups of three adjacent serial sections were incubated in the presence of 0.1 nM 1125I]RTI-55 (A-C), 0.1 nM [t251]RTI-55 plus 10 #M clomipramine (D-F) or 0.1 nM [125I]RTI-55 plus 1 mM (-)-cocaine (G-l). CA2, 3, CA2 and CA3 sectors of Ammon's horn; LG, lateral geniculate nucleus; SC, superior colliculus; SN, substantia nigra; VTA, ventral tegmental area. Bar = 2.5 mm.

3.6. Distribution of [1251]RTI-55 binding sites with clomipramine in vitro In the presence of clomipramine, a selective ligand that labels 5HTT, [125I]RTI-55 binding was displaced almost completely in some brain regions, e.g., the midline and lateral geniculate nuclei of the thalamus (Fig. 3A,B,D,E), the hypothalamic nuclei (Fig. 2A-F), the superior colliculus (Figs. 3C,F, 4A,D), the dorsal raphe nucleus (Fig. 4A,D) and the facial nucleus (Fig. 4C,F). However, there were also some areas where [t25I]RTI-55 binding was not completely displaced by clomipramine (summarized in Table 1); the olfactory tubercle (Fig. I E), the caudate putamen (Fig. 1E), the accumbens nucleus (Fig. 1E), the globus pallidus (Fig. IF), the subthalamic nucleus (Fig. 2F), the SNC (Fig. 3F) and the ventral tegrnental area (Fig. 3F). In these areas, [125I]RTI-55 binding in the caudate putamen was displaced by clomipramine to an extent of only 27%; in the olfactory tubercle and the accumbens nucleus, the displacement was 58 and 50%, respectively. In addition,

in the SNC, the subthalamic nucleus, the ventral tegmental area and the globus pallidus, 72, 73, 76 and 84%, respectively, of the [125I]RTI-55 binding was displaced by clomipramine. Further, most of RTI-55 binding in all of these regions, namely, the olfactory tubercle, the caudate putamen, the accumbens nucleus, the globus pallidus, the subthalamic nucleus, the SNC and the ventral tegrnental area, were displaced with GBR-12909. The remaining amount of RTI-55 binding in these areas was 20-35%. 3.7. Distribution of [~esI]RTI-55 binding sites by ex vivo autoradiography The distribution of [125I]RTI-55 binding sites after the intravenous injection of [125I]RTI-55 is shown in Figs. 5A-D and 6A-E. The distribution pattern of [125I]RTI-55 binding sites was very similar to that in the in vitro autoradiography (Figs. 1A-C, 2A-C, 3A-C, 4A-C). Namely, the highest levels of [~25I]RTI55 binding were found in the caudate putamen, the ac-

19

M. Fufita et al./J. Chem. Neuroanat. 7 (1994) 13-23

k

k

(

IIE

Fig. 4. In vitro autoradiographic images showing the distribution of [t251]RTI-55 binding sites. Groups of three adjacent serial sections were incubated in the presence of 0.1 nM [125I]RTI-55 (A-C), 0.1 nM ll25I]RTI-55 plus 10 #M clomipramine (D-F) or 0.1 nM [1251lRTI-55plus 1 mM (-)-cocaine (G-I). DR, dorsal raphe nucleus; IC, inferior colliculus; LC, locus coeruleus; LDTg, laterodorsal tegmental nucleus; MnR, median raphe nucleus; SC, superior colliculus; 7, facial nucleus. Bar = 2.5 mm.

cumbens nucleus and the olfactory tubercle (Fig. 5A,B). In addition, moderate binding levels were detected in the anteroventral (Fig. 5C) and lateral geniculate nuclei in the thalamus (Fig. 6A), the hypothalamic nuclei (Fig. 5D), the superficial gray layer of the superior colliculus (Fig. 6B), the dorsal and median raphe nuclei (Fig. 6C) and the facial nucleus (Fig. 6E). A very low amount of [125IIRTI-55 binding was detected in the cerebellar cortex (Fig. 6D,E). When 1 tzmol ciomipramine was injected before injection of [125I]RTI-55, binding in the thalamic nuclei, hypothalamic nuclei and the dorsal raphe nucleus was not completely displaced (data not shown). Therefore, it was necessary to inject 10 #mol clomipramine to displace [125I]RTI-55 binding to 5HTT completely. The distribution pattern of [125I]RTI-55 binding in the ex vivo autoradiography with 10 #mol clomipramine (Figs. 5E-H, 6F-J) was similar to that of the in vitro autoradiography with clomipramine (Figs. 1D-F, 2D-F, 3D-F, 4D-F). That is, [125IIRTI-55 binding was detected only in the caudate putamen, the accumbens nucleus, the olfactory tubercle (Fig. 5E,F), the

subthalamic nucleus (Fig. 5H), the ventral tegrnental area (Fig. 6G) and the substantia nigra (Fig. 6G); in other areas binding was very low in comparison with that in the ex vivo autoradiography without clomipramine. As most of the [125I]RTI-55 binding in the cerebral cortex was displaced by clomipramine, the contrast between the caudate putamen and the cerebral cortex was much clearer in the ex vivo autoradiography with clomipramine than in that without clomipramine. 4. Discussion

4.1. [1251]RTI-55 binding to D A T and 5 H T T in vitro

Recent evidence has shown that RTI-55 binds to DAT and 5HTT (Boja et al., 1992; Kaufman and Madras, 1992; Scheffel et al., 1992). In the present experiment, we distinguished RTI-55 binding to DAT from its binding to 5HTT by displacing with clomipramine (Table 1). In the areas where RTI-55 binding was completely displaced by clomipramine (e.g. the midline and lateral geniculate nuclei of the thalamus,

20

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

Ill at

m

m

B

Fig. 5. Ex vivo autoradiographic images showing the distribution of [1251]RTI-55 binding sites in coronal sections of the rat brain. [1251]RTI-55 (814 kBq [10 pmol]) was intravenously injected (A-D) or intravenously administered 5 min after the intravenous injection of 10 #mol clomipramine (E-H). Dark areas indicate relevant binding levels. Acb, accumbens nucleus; AV, anteroventral thalamic nucleus; CPu, caudate putamen; HyP, hypothalamus; STh, subthalamic nucleus: Tu, olfactory tubercle. Bar = 2.5 mm.

the hypothalamic nuclei, the superior colliculus, the dorsal raphe nucleus, and the facial nucleus), RTI-55 was considered to bind to 5HTT. This distribution of 5HTT coincides well with the distribution of the terminals of 5HT neurons (Steinbusch, 1981). In some regions, [J25I]RTI-55 binding was not completely displaced by clomipramine and most of the binding was displaced with GBR-12909, a selective ligand to DAT, suggesting that the remaining [125I]RTI-55 binding sites were DAT; these regions were the olfactory tubercle, the caudate putamen, the accumbens nucleus, the globus pallidus, the subthalamic nucleus, the SNC and SNR and the ventral tegmental area. However, the proportions of [125I]RTI-55 binding displaced by clomipramine were different in each area. That is, only 27% of [125I]RTI-55 binding was displaced with clomipramine in the caudate putamen, while 84% of the binding was displaced in the globus pallidus. Although clomipramine does displace binding to DAT to a small

extent (Friedman et al., 1977; Kinnier et al., 1984), it can be concluded that the proportion of DAT and 5HTT varied in each region. Namely, the caudate putamen contains large amounts of DAT and small amounts of 5HTT, and the globus pallidus is relatively rich in 5HTT and poor in DAT. 4.2. D A T as cocaine recognition sites

The psychomotor stimulant effects of cocaine are thought to be brought about predominantly by its binding to DAT and by its inhibition of DA reuptake (Ritz et al., 1987; Bergman et al., 1989; Spealman et al., 1989). Cells containing DAT mRNA are distributed in the SNC (A9) and VTA (AI0) and some positive cells are also scattered in the SNR (Shimada et al., 1992; Cerruti et al., 1993). These neurons project mainly to the olfactory tubercle, caudate putamen and the accumbens nucleus, most DAT proteins being present in the plasma

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

21

A

C

E

....

..............

E

IE

Fig. 6. Ex vivoautoradiographicimagesshowingthe distributionof [125I]RTI-55bindingsites, ll251]RTI-55(814 kBq [10 pmoll) was intravenously injected(A-E) or intravenouslyadministered 5 min after the intravenousinjectionof 10 t*molclomipramine(F-J). DR, dorsal raphe nucleus; HyP, hypothalamus;LC, locuscoeruleus;LG, lateralgeniculatenucleus;MnR,medianraphe nucleus;SC, superiorcolliculus;SN, substantianigra; STh, subthalamicnucleus; VTA, ventral tegmentalarea; 7, facial nucleus. Bar = 2.5 mm.

membrane of the presynaptic axon terminals of DA neurons (Kuhar et al., 1990). Thus, the present results showing the dense distribution of binding sites in the olfactory tubercle, caudate putamen and the accumbens nucleus coincide well with the projection areas of the VTA and SNC. In the caudate nucleus and putamen of the monkey brain, medial-to-lateral gradients of [125I]RTI-55 binding were found in ex vivo auto-radiography (Kaufman and Madras, 1992). We also found such gradients in our experiments. [3H]Mazindol binding to DAT was shown to be heightened in the extrastriosomal matrix and reduced in the striosomes, this tendency being prominent in the monkey or cat brain, but not in the rat brain

(Graybiel and MorataUa, 1989). This differential distribution in the striosome and matrix was not prominent in the present results. [1251]RTI-55 binding was partially displaced by clomipramine in the globus pallidus, the subthalamic nucleus, the SNC and the ventral tegmental area, suggesting that these areas contain both DAT and 5HTT. 4.3. 5 H T T as cocaine recognition sites

Cocaine abusers experience both physiologic (e.g., anorexia and hyposomnia) and psychiatric disorders (e.g., anxiety, depression and psychoses) (Lowenstein et al., 1987) in which the 5HT system has been implicated

22

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

(Iversen, 1984; King et al., 1985). Experimental data support this hypothesis; cocaine potently inhibits the spontaneous activity of the dorsal raphe neurons in rats not previously treated with cocaine (Cunningham and Lakoski, 1990) and this inhibitory response is significantly enhanced in cocaine-treated rats (Cunningham et al., 1992). In the present experiment, very high levels of [125I]RTI-55 binding were found in the dorsal raphe nucleus, compatible with the action of cocaine in this region. Cocaine also elevates adrenocorticotrophic hormone and corticosterone levels in plasma (Borowsky and Kuhn, 1991; Levy et al., 1991). This activation of the hypothalamopituitary-adrenal axis is though to be brought about not only by the hypothalamic dopaminergic system (Borowsky and Kuhn, 1991), but also by the serotonergic system (Levy et al., 1991). In fact, in the present study, 1125I]RTI-55 bound to 5HTT in the zona incerta and the nuclei of the hypothalamus. Cocaine abusers experience sleep disturbances (Washton and Tatarsky, 1983). This may be due to cocaine binding to the suprachiasmatic nucleus of the hypothalamus, which may have a key function in maintaining circadian rhythms (Cassone, 1990). Little is known about the action of cocaine in the thalamus. It is possible that [12SI]RTI-55 binding to the midline nuclei of the thalamus may be involved in cocaine-induced analgesia (Shyu et al., 1992). We found moderate levels of RTI-55 binding in the lateral geniculate nucleus; this may be closely related to the visual hallucinations experienced by some patients with cocaine-induced psychosis (Brady et al., 1991). In addition, [125I]RTI-55 binding to the superior colliculus may also contribute to such visual hallucinations. [125I]RT1-55 thus appears to be a suitable ligand for studying the actions of cocaine in the whole brain. 4.4. [1251]RTI-55 binding to DAT and 5HTT in vivo: suitable imaging agent for cocaine recognition sites and DATs

As the in vivo distribution of [125I]RTI-55 binding sites was similar to the distribution pattern of the in vitro autoradiography, it can be concluded that [125I]RTI-55 binds to both DAT and 5HTT in vivo. Indeed, WIN-35428, a non-iodinated analog of RTI-55, stimulates locomotor activity, an effect similar to that of cocaine (Cline et al., 1992a), although WIN-35428 binds less to 5HTT than RTI-55 (Kaufman and Madras, 1992). With clomipramine, a selective ligand for 5HTT, [125I]RTI-55 bound very selectively to areas rich in DATs (Figs. 5E,F, 6G). Thus, to detect changes in DA neurons in conditions such as Parkinson's disease, it is better to image DAT with both radiolabelled RTI-55 and unlabelled clomipramine than with radiolabelled RTI-55 alone. On the other hand, for the detection of

DAT and 5HTT as cocaine recognition sites, imaging with RTI-55 alone is suitable. References Bergman, J., Madras, B.K., Johnson, S.E. and Spealman, R.D. (1989) Effects of cocaine and related drugs in nonhuman primates. Ill. Self-administration by squirrel monkeys. J. Pharmacol. Exp. Ther. 251, 150-155. Boja, J.W., Patel, A., Carroll, F.I., Rahman, M.A., Philip, A., Lewin, A.H., Kopajtic, T.A. and Kuhar, M.J. (1991) [125I]RTI-55: a potent ligand for dopamine transporters. Eur. J. Pharmacol. 194, 133-134. Boja, J.W., Mitchell, W.M., Patel, A., Kopajtic, T.A., Carroll, F.I., Lewin, A.H., Abraham, P. and Kuhar, M.J. (1992) High-affinity binding of [125IIRTI-55 to dopamine and serotonin transporters in rat brain. Synapse 12, 27-36. Borowsky, B. and Kuhn, C.M. (1991) Monoamine mediation of cocaine-induced hypothalamo-pituitary-adrenal activation. J. Pharmacol. Exp. Ther. 256, 204-210. Brady, K.T., Lydiard, R.B., Malcolm, R. and Ballenger, J.C. (1991) Cocaine-induced psychosis. J. Clin. Psychiatry 52, 509-512. Calligaro, D.O. and Eldefrawi, M.E. (1988) High affinity sterospecific binding of [3H]cocaine in striatum and its relationship to the dopamine transporter. Membr. Biochem. 7, 87-106. Carroll, F.I., Rahman, M.A., Abraham, P., Parham, K., Lewin, A.H., Dannals, R.F., Shaya, E., Scheffel, U., Wong, D.F., Boja, J.W. and Kuhar, M.J. (1991) [123113/3-(4-iodophenyl)tropan-2Bcarboxylic acid methyl ester (RTI-55), a unique cocaine receptor ligand for imaging the dopamine and serotonin transporter in vivo. Med. Chem. Res. 1,289-294. Carroll, F.I., Lewis, A.H., Boja, J.W. and Kuhar, M.J. (1992)Cocaine receptor: biochemical characterization and structure-activity relationship of cocaine analogues at the dopamine transporter. J. Med. Chem. 35, 969-981. Cassone, V.M. (1990) Effects of melatonin on vertebrate circadian systems. Trends Neurosci. 13, 457-464. Cerruti, C., Walther, D.M., Kuhar, M.J. and Uhl, G.R. (1993) Dopamine transporter mRNA expression is intense in rat midbrain neurons and modest outside midbrain. Mol. Brain Res. 18, 181-186. Cline, E.J., Scheffel, U., Boja, J.W., Carroll, F.I., Katz, J.L. and Kuhar, M.J. (1992a). Behavioral effects of novel cocaine analogs: a comparison with in vivo receptor binding potency. J. Pharmacol. Exp. Ther. 260, 1174-1179. Cline, E.J., Scheffel, U,, Boja, J.W., Mitchell, W.M., Carroll, F. I., Abraham, P., Lewin, A.H. and Kuhar, M.J. (1992b). In vivo binding of [1251]RTI-55 to dopamine transporters: pharmacology and regional distribution with autoradiography. Synapse 12, 37-46. Cunningham, K.A. and Lakowski, J.M. (1990) The interaction of cocaine with serotonin dorsal raphe neurons, single-unit extracellular recording studies. Neuropsychopharmacology 3, 41-50. Cunningham, K.A., Paris, J.M. and Goeders, N.E. (1992) Chronic cocaine enhances serotonin autoregulation and serotonin uptake binding. Synapse 11, 112-123. Dubocovich, M.L. and Zahniser, N.R. (1985) Binding characteristics of the dopamine uptake inhibitor [3H]nomifensine to striatal membranes. Biochem. Pharmacol. 34, 1137-1144. Friedman, E., Fung, F. and Gershon, S. (1977) Antidepressant drugs and dopamine uptake in different brain regions. Eur. J. PharmacoL 42, 47-51. Graybiel, A.M. and Moratella, R. (1989) Dopamine uptake sites in the striatum are distributed differentially in striosome and matrix compartments. Proc. Natl. Acad. Sci. USA 86, 9020-9024. lnnis, R., Baldwin, R., Sybirska, E., Zea, Y., Laruelle, M., AI-Tikriti, M., Charney, D,, Zoghbi, S., Smith, E., Wisniewski, G., Hoffer, P.,

M. Fujita et al./J. Chem. Neuroanat. 7 (1994) 13-23

Wang, S., Milius, R. and Neumeyer, J. (1991) Single photon emission computed tomography imaging of monoamine reuptake sites in primate brain with [123I]CIT. Fur. J. Pharmacol. 200, 369-370. Iversen, S.D. (1984) Serotonin and anxiety. Neuropharmacology 23, 1553-1560. Janoswky, A., Schweri, M.M., Berger, P., Long, R., Skolnick, P. and Paul, S.M. (1985) The effects of surgical and chemical lesions on striatal [3H]threo-(+/-)-methylphenidate binding: correlation with [3H]dopamine uptake. Fur. J. Pharmacol. 108, 187-191. Janowsky, A., Berger, P., Vocci, F., Labarca, R., Skolnick, P. and Paul, S.M. (1986) Characterization of sodium-dependent [3H]GBR-12935 binding in brain: a radioligand for selective labeling of the dopamine transporter system. J. Neurochem. 46, 1272-1276. Javitch, J.A., Blaustein, R.O. and Snyder, S.H. (1984) [3H]Mazindol binding associated with neuronal dopamine and norepinephrine uptake sites. Mol. Pharmacol. 26, 35-44. Kaufman, M.J., Spealman, R.D. and Madras, B.K. (1991) Distribution of cocaine recognition sites in monkey brain: I. in vitro autoradiography with [3H]CFT. Synapse 9, 177-187. Kaufman, M.J. and Madras, B.K. (1992) Distribution of cocaine recognition sites in monkey brain: Il. ex vivo autoradiography with [3HICFT and [1251]RTI-55. Synapse 12, 99-11 I. King, R., Faull, K.F., Stahl, S.M., Mefford, I.N., Thieman, S., Barchas, J.D. and Berger, P.A. (1985) Serotonin and schizophrenia: correlation between serotonergic activity and schizomotor behavior. Psychiatric Res. 14, 235-240. Kinnier, W.J., Tabor, R.D. and Norrell, L.Y. (1984) Neurochemical properties of AHR-9377: a novel inhibitor of norepinephrine reuptake. Biochem. Pharmacol. 33, 3001-3005. Kuhar, M.J., Sanchez-Roa, P.M., Wong, D.F., Dannals, R.F., Grigoriadis, D.E., Lew, R. and Milberger, M. (1990) Dopamine transporter: biochemistry, pharmacology and imaging. Fur Neurol. 30, 15-20. Levy, A.D., Li, Q., Kerr, J.E., Rittenhouse, P.A., Milonas, G., Cabrera, T.M., Battaglia, G., Alvarez Sanz, M.C. and van de Kar, L.D. (1991) Cocaine-induced elevation of plasma adrenocorticotropin hormone and corticosterone is mediated by serotonergic neurons. J. Pharmacol. Exp. Ther. 259, 495-500. Lowenstein, D.H., Massa, S.M., Rowbotham, M.C., Collins, S.D., McKinney, H.E. and Simon, R.P. (1987) Acute neurologic and psychiatric complications associated with cocaine abuse. Am. J. Med. 83, 841-846.

23

Madras, B.K., Fahey, M.A. Bcrgman, J., Canfield, D.R. and Spealman, R.D. (1989a). Effects of cocaine and related drugs in nonhuman primates. I. [3H]cocaine binding sites in caudateputamen. J. Pharmacol. Exp. Ther. 251, 131-141. Madras, B.K., Spealman, R.D., Fabey, M.A., Neumeyer, J.L., Saha, J.K. and Milius, R.A. (1989b). Cocaine receptors labeled by [3H]2B-carbomethoxy-3B-(4-flurophenyl)tropane. Mol. Pharmacol. 36, 518-524. Paxinos, G. and Watson, C. (1986) The Rat Brain in Stereotaxic Coordinates. Academic Press, New York. Reith, ME.A, Sershen, H. and Lajtha, A. (1986) Binding sites for [3H]cocaine in mouse striatum and cerebral cortex have different kinetics. J. Neurochem. 46, 263-270. Ritz, MC., Lamb, R.J., Goldberg, S.R. and Kuhar, M.J. (1987) Cocaine receptors on dopamine transporters are related to selfadministration of cocaine. Science 237, 1219-1223. Scheffel, U., Dannals, R.F., Cline, E.J., Ricaurte, G.A., Carroll, F.I., Abraham, P., Lewin, A.H. and Kuhar, M.J. (1992) [J23/125I]RTI55, an in vivo label for the serotonin transporter. Synapse I 1, 134-139. Shaya, E.K., Scheffel, U., Dannals, R.F., Ricaurte, G.A., Carroll, F.I., Wagner Jr., H.N., Kuhar, M.J. and Wong, D.F. (1992) In vivo imaging of dopamine reuptake sites in the primate brain using single photon emission computed tomography (SPECT) and iodine-123 labeled RTI-55. Synapse 10, 169-172. Shimada, S., Kitayama, S., Walther, D. and Uhl, G. (1992) Dopamine transporter mRNA: dense expression in ventral midbrain neurons. Mol. Brain Res. 13, 359-362. Shyu, B.C., Kiritsy-Roy, J.A., Morrow, T.J. and Casey, K.L. (1992) Neurophysiological, pharmacological and behavioral evidence for medial thalamic mediation of cocaine-induced dopaminergic analgesia. Brain Res. 572, 216-223. Spealman, R.D., Madras, B.K. and Bergman, J. (1989) Effects of cocaine and related drugs in nonhuman primates. II. Stimulant effects on schedule-controlled behavior. J. Pharmacol. Exp. Ther. 251, 142-149. Steinbusch, H.W.M. (1981) Distribution of serotoninimmunoreactivity in the central nervous system of the rat-cell bodies and terminals. Neuroscience 6, 557-618. Washton, A.M. and Tatarsky, A. (1983) Adverse effects of cocaine abuse. NIDA Res. Monogr. 49, 247-254.