THE JOURNAL OF COMPARATIVE NEUROLOGY 421:247–255 (2000) Localization of Dopaminergic Markers in the Human Subthalamic Nucleus SARAH J. AUGOOD,1,2* ZANE R. HOLLINGSWORTH,2 DAVID G. STANDAERT,2 PIERS C. EMSON,1 AND JOHN B. PENNEY, JR.2 1 Department of Neurobiology, Babraham Institute, Cambridge CB2 4AT, United Kingdom 2 Neurology Research, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114 ABSTRACT The potential role for dopamine in the subthalamic nucleus was investigated in human postmortem tissue sections by examining; (1) immunostaining for tyrosine hydroxylase, the rate-limiting enzyme in catecholamine synthesis; (2) binding of [3H]-SCH23390 (D1-like), [3H]-YM-09151-2 (D2-like), and [3H]-mazindol (dopamine uptake); and (3) expression of dopamine D1 and D2 receptor mRNAs. Immunostaining for tyrosine hydroxylase was visualized in Bouin’s-fixed tissue by using a monoclonal antibody and the avidin-biotin-complex method. The cellular localization of the dopamine D1 and D2 receptor mRNAs was visualized by using a cocktail of human specific oligonucleotide probes radiolabeled with 35S-dATP. Inspection of immunostained tissue revealed a fine network of tyrosine hydroxylaseimmunostained fibers traversing the nucleus; no immunopositive cells were detected. Examination of emulsion-coated tissue sections processed for D1 and D2 receptor mRNA revealed, as expected, an abundance of D1 and D2 mRNA-positive cells in the caudate nucleus and putamen. However, no D1 or D2 receptor mRNA-expressing cells were detected in the subthalamic nucleus. Further, semiquantitative analysis of D1-like, D2-like and dopamine uptake ligand binding similarly revealed an enrichment of specific binding in the caudate nucleus and putamen but not within the subthalamic nucleus. However, a weak, albeit specific, signal for [3H]-SCH23390 and [3H]-mazindol was detected in the subthalamic nucleus, suggesting that the human subthalamic nucleus may receive a weak dopaminergic input. As weak D1-like binding is detected in the subthalamic nucleus, and subthalamic neurons do not express dopamine D1 or D2 receptor mRNAs, together these data suggest that the effects of dopaminergic agents on the activity of human subthalamic neurons may be indirect and mediated via interaction with dopamine D1-like receptors. J. Comp. Neurol. 421: 247–255, 2000. © 2000 Wiley-Liss, Inc. Indexing terms: basal ganglia; Parkinson’s disease; brain; dopamine receptor The role of dopamine (DA) within the subthalamic nucleus (STN) has received considerable interest in recent years (see Chesselet and Delfs, 1996; Joel and Weiner, 1997, for review), as numerous studies have suggested that DA and DAergic agents may have a direct effect within this elliptical structure. One of the early pioneering studies suggestive of a role for DA in the rat STN was that of Brown and Wolfson (1978) who reported a marked increase in glucose utilization in the STN of rats treated with apomorphine, a potent DA agonist (Brown and Wolfson, 1978). Furthermore, the increase in 2-deoxyglucose signal was blocked by pretreating the animals with haloperidol (Brown and Wolfson, 1978), an antagonist at D2like receptors. Consequently, the nature of the DAergic input to the rodent STN has been the focus of biochemical © 2000 WILEY-LISS, INC. (Rosales et al., 1997), histofluorescence (Meibach and Katzman, 1979), anatomical tracing (Campbell et al., 1985; Hassani et al., 1997), and more recently, electrophysiological studies (Mintz et al., 1986; Kreiss et al., 1996, 1997). Together, these studies have demonstrated This paper is dedicated to the memory of Dr. Jack Penney for his friendship and guidance of this work. Grant sponsor: U.K. Parkinson’s Disease Society; Grant number: 3061; Grant sponsor: USPHS; Grant number: NS31579. *Correspondence to: Dr. Sarah Augood, Neurology Research, Warren 408, Massachusetts General Hospital, 32 Fruit Street, Boston, MA 02114. E-mail: [email protected] Received 7 October 1999; Revised 19 January 2000; Accepted 21 January 2000 248 S.J. AUGOOD ET AL. unequivocally a direct DAergic input to the rat STN and that administration of DAergic agents can markedly affect the firing pattern of these tonically active fast-firing neurons in vivo. The precise nature of this DAergic response is unclear as binding studies (Savasta et al., 1986; Mansour et al., 1990, 1992; Johnson et al., 1994; Kreiss et al., 1996) suggest that the rodent STN is enriched in either D1- or D2-like receptors, however there is no clear consensus. The subcellular localization of these DA receptors, that is, whether they are encoded by STN neurons and/or are present on STN afferents, has been addressed recently by using sensitive reverse transcriptase-polymerase chain reaction (RT-PCR) coupled with receptor autoradiography: the rat STN is enriched in D1-like, D2-like, and D4-like binding sites, and intrinsic STN neurons express DA D1, D2, and D3 but not D4 receptor mRNAs (Flores et al., 1999). To date, no studies have investigated in detail the DAergic nature of the human STN. Recent immunochemical studies have reported the presence of fine tyrosine hydroxylase-immunopositive (TH) fibers traversing this structure (Cossette et al., 1999; Hedreen, 1999), in line with rodent (Hassani et al., 1997) and primate (Francois et al., 1998) data. To our knowledge, no studies have investigated the nature of the DAergic interaction within the human STN. Thus, in this study we investigated the immunostaining for TH coupled with DA D1-like and D2like receptor autoradiography, whilst the DA receptor phenotype of these STN neurons was addressed by using in situ hybridization. This molecular approach for localizing the DA receptor subtypes was favored over immunohistochemistry primarily because in situ hybridization techniques allow the cellular origin of a mRNA to be identified in addition to the subtype specificity provided by examining the cellular expression of the individual genes. Finally, the presence of DA uptake sites, a marker of functional DA terminals, was assessed by 3H-mazindol autoradiography. MATERIALS AND METHODS Human tissue Bouin’s (0.9% picric acid, 9% formaldehyde, 5% glacial acetic acid: Sigma, St. Louis, MO)-fixed tissue blocks (n ⫽ 3) containing the medial-lateral extent of the human STN were provided by the Harvard Brain Tissue Resource Center (Belmont, MA). Frozen human tissue was provided as fresh-frozen cryostat sections (16 m) thawmounted onto poly-L-lysine-coated microscope slides by the U.K. Parkinson’s Disease Society Tissue Brain Bank (London, U.K.). Tissue was stored at ⫺80°C until used. All the Bouin’sfixed and fresh-frozen tissue was from control cases with no history of neurological disease and no overt signs of neuropathology. Brain postmortem intervals did not exceed 24 hours for the in situ hybridization studies or 56 hours for the ligand binding studies. All cases used in this study had brain pH values in the range of 6.1– 6.6 (Kingsbury et al., 1995). TH immunohistochemistry Bouin’s-fixed tissue blocks containing the medial-lateral extent of the STN were cryoprotected in 30% sucrose (w/v), rinsed briefly with 95% ethanol, and frozen rapidly on aluminium plates as described previously (Vonsattel et al., 1995). Fifty-micron-thick sections were cut on a freezing microtome, collected into 0.1 M phosphate-buffered saline (PBS) pH 7.4/50% glycerol and either processed immediately for TH immunostaining or stored at ⫺20°C. For immunostaining, the free-floating sections were first washed extensively in 0.1 M PBS, treated with 30% methanol ⫹ 1.5% H2O2 to deactivate endogenous peroxidase activity, rinsed again in 0.1 M PBS, blocked with 10% normal goat serum (NGS)/0.3% Triton X-100 and then incubated overnight at 4°C in mouse anti-TH antiserum (1:1,000; Sigma). Sections were then washed extensively in PBS, incubated in biotinylated-goat anti-mouse IgG (1:200; Jackson Labs, Bar Harbor, ME) for 1 hour, rinsed, then incubated in avidin-biotin-complex (Vectastain Elite Kit, Vector Laboratories, Burlingame, CA). Finally, THimmunopositive structures were visualized by using 3,3diaminobenzidine (0.5 mg/ml) as chromogen activated with H2O2 (0.0045%). DA D1 and D2 receptor in situ hybridization Hemicoronal tissue sections containing the posterior putamen, globus pallidus internus and externus, and the medial-lateral aspect of the STN were processed for in situ hybridization by using a standardized protocol as reported previously (Harrington et al., 1995). In brief, sections were fixed, rinsed in 0.1 M phosphate-buffered saline, dehydrated, and then hybridized at 37°C with human-specific DA D1 and D2 receptor 35S-oligonucleotide probes as described previously (Harrington et al., 1995; Augood et al., 1997). Following an overnight hybridization, tissue sections were washed stringently in 1 ⫻ saline sodium citrate at 55°C and then processed for emulsion autoradiography (Ilford K5, Polysciences, Warrington, PA) for 8 –12 weeks. Finally, tissue sections were counterstained with methylene-blue to visualize all brain nuclei and examined with light microscopy by using a Leica microscope. Some tissue sections were also processed with the addition of an excess of unlabeled oligonucleotide to the hybridization buffer (pH 7.0) to demonstrate that the binding of the 35 S-probes was displaceable. Dopamine D1 and D2 receptor binding and dopamine uptake sites Hemicoronal tissue sections (n ⫽ 4) containing the posterior putamen, globus pallidus internus and externus, the medial-lateral aspect of the STN and the rostral substantia nigra pars reticulata were processed for [3H]SCH23390 (DA D1-like), [3H]-YM-08151-2 (DA D2-like) and [3H]-mazindol (DA uptake) ligand binding by using standardized protocols as reported previously (Cha et al., 1998). All [3H]-ligands were obtained from New England Nuclear (Boston, MA). In brief, assay buffer contained 25 mM Tris-HCl (pH 7.5), 100 mM NaCl, 1 mM MgCl2, 1 M pargyline, and 0.001% ascorbate. For D1-like binding, slides were incubated in the dark with 1.54 nM [3H]-SCH23390 (s.a. 70.3 Ci/mmol) for 2.5 hours at room temperature. Nonspecific binding was defined in the presence of 1 M cis-flupentixol. For D2-like binding, slides were incubated in the dark with 187 pM [3H]-YM-09151-2 (s.a. 85.5 Ci/mmol) for 3 hours at room temperature. Nonspecific binding was defined in the presence of 100 M DA. For DA uptake, slides were prewashed in ice-cold binding buffer (50 mM Tris-HCl, 5 mM KCl, and 300 mM NaCl, pH 7.9) for 5 minutes, then incubated with 5.5 nM [3H]-mazindol (s.a. 24 Ci/mmol) in the presence of 300 nM desipramine TH-LI AND DA RECEPTORS IN THE HUMAN STN for 1 hour at 4°C. Nonspecific binding was defined in the presence of 10 M nomifensine. After incubation with [3H]-ligand, slides were rinsed in cold assay buffer for 10 minutes, rinsed quickly in cold distilled water, dried under a stream of cool air, and apposed to tritium-sensitive film (Hyperfilm 3H, Amersham, Arlington Heights, IL) with calibrated 14C-standards (ARC, Inc, St. Louis, MO) for 2– 4 weeks. Films were analyzed by using computerassisted image analysis (M1, Imaging Research, St. Catharine’s, ON, Canada) and the density of the image converted to Ci/g by automated extrapolation from the calibrated standards. Specific binding was determined by subtracting the nonspecific value from the total binding value. RESULTS TH immunohistochemistry A robust TH-immunopositive signal was detected in all the Bouin’s-fixed tissue sections processed with the anti-TH antibody. In a pilot study comparing different fixatives on the quality of TH-like immunoreactivity, we found Bouin’s-fixed tissue blocks to give the most robust staining by using this TH monoclonal antibody. Of the three Bouin’s-fixed cases processed, b-3932 displayed the most robust immunostaining. A dense network of intensely labeled TH-immunopositive neurons and fibers was observed in the lateral tip of the substantia nigra (Fig. 1A,B). On the same tissue sections, a heterogeneous pattern of TH immunoreactivity was observed in the STN: in addition to weak neuropil staining, a fine network of TH-immunopositive fibers was observed traversing the nucleus (Fig. 1C,E,F). Many of these fibers have the appearance of long smooth axons with very occasional bifurcated endings. TH-positive fibers were more numerous medially than laterally. No TH-positive somata were detected. By contrast, on adjacent tissue sections processed in parallel but with the omission of the anti-TH antibody, no specific immunostaining was observed (Fig. 1B): The presence of darkly stained punctate dots in Figure 1D represents an intensification of the endogenous lipofuscin pigment contained within the cytoplasm of these human STN neurons. DA D1 and D2 receptor mRNA On film autoradiograms, intense DA D1 and D2 receptor hybridization signals were observed in the caudate nucleus and putamen of all tissue sections. On sections from cases containing the lateral tip of the substantia nigra pars compacta, an intense D2 receptor hybridization signal was observed; no D1 signal was detected in this nucleus. Within the STN, no hybridization signal was detected for either DA receptor transcript. At the cellular level, DA receptor mRNA-positive cells were identified by the accumulation of silver grains overlying methylene blue-counterstained neurons. In the caudate-putamen, numerous DA receptor mRNA-positive cells were detected throughout the entire extent of the nucleus. In addition to the extensive labeling of medium-sized D1 and D2 receptor (Fig. 2A) mRNA-positive cells, D2 mRNA-positive large cells were also observed. In addition to the striatal labeling, numerous D2 receptor mRNA-positive neurons were observed in the lateral tip of the substantia nigra; these corresponded to the pigmented neurons in the sub- 249 stantia nigra pars compacta (Fig. 2B). No D1 receptor mRNA-positive cells were observed in the substantia nigra. Neurons of the human STN were readily identified by the presence of endogenous yellow pigment (lipofuscin) contained within the soma of numerous cells, a characteristic of these cells in mature adults. In contrast to the caudate-putamen, no DA receptor mRNA-positive cells were detected for either D1 or D2 (Fig. 2C); an occasional D2 receptor mRNA-positive cell was detected on the ventrolateral edge of the nucleus, although it was not clear if this was a displaced nigral neuron (Meibach and Katzman, 1979). DA D1 and D2 receptor binding Quantitative receptor autoradiography was used to reveal specific sites of DA D1-like, D2-like (Fig. 3) and DA uptake binding sites (Fig. 4). Sites of specific binding were determined by subtracting nonspecific binding (Figs. 3B,D; 4B) from total binding (Figs. 3A,C; 4A). Of the four control brains examined, a robust and reproducible signal was measured in all brain areas. For D1-like binding sites, an intense signal was detected within the head of the caudate and putamen: a less intense signal was observed in the globus pallidus internus (GPi), substantia nigra pars reticulata, and insular cortex. A very weak signal was detected in the STN (Fig. 3A,B). The intensity of D1-like binding was similar in the head of the caudate nucleus and putamen. For D2-like binding, an intense signal was detected within the head of the caudate and putamen, the intensity of the signal being greater in the putamen than in the head of the caudate. A less intense signal was observed within the globus pallidus externus (GPe). Similarly, for 3H-mazindol binding, a marker of DA uptake sites, an intense signal was detected within the head of the caudate and putamen. As for D2-like binding, the intensity of the signal was greater in the putamen than in the head of the caudate. A faint specific 3Hmazindol binding signal was detected within the substantia nigra pars reticulata, insular cortex and STN (see Table 1). The quantitative data are summarized in Table 1. DISCUSSION The present study provides a comprehensive examination of markers of DA transmission in the normal human STN. We demonstrate that the normal human STN contains a sparse network of fine TH-immunopositive fibers, a finding reported recently in formalin-fixed tissue (Cossette et al., 1999; Hedreen, 1999). Additionally, we demonstrate that intrinsic STN neurons are not enriched in DA D1 or D2 receptor mRNA or DA D1-like or D2-like binding sites. However, a specific, albeit weak, D1-like binding signal is detected, as is a weak specific DA uptake signal ([3H]-mazindol). Together these data suggest that the human STN receives a sparse DAergic innervation. Unlike the rat, intrinsic human STN neurons do not express D1 or D2 receptors. Although low levels of D1-like (D1/D5) binding are present, this may be associated with STN afferents and/or may reflect expression of DA D5 receptors. Whether the TH-immunopositive fibers in the human STN represent nigrostriatal fibers of passage or a direct DAergic innervation is unclear and requires further investigation. In the rat (Hassani et al., 1997), tracing studies 250 S.J. AUGOOD ET AL. Figure 1 TH-LI AND DA RECEPTORS IN THE HUMAN STN 251 Fig. 2. Darkfield images of tissue sections hybridized with the D2 receptor probe and then processed for emulsion autoradiography. Tissue sections were viewed under darkfield microscopy. Dopamine D2 receptor mRNA-positive cells (arrows) are visualized by the accu- mulation of silver grains (white dots). Examples of D2 receptor mRNA-positive cells are shown in the (A) dorsal striatum and (B) substantia nigra pars compacta. In marked contrast, no D2 mRNApositive cells are detected in the STN (C). Scale bar ⫽ 50 m. coupled with immunohistochemistry for TH have provided convincing evidence for a DAergic innervation of this nucleus, although ultrastructural studies are required to clarify this point conclusively. In the African green monkey, sparse TH-immunopositive axonal endings (bifurcations and varicosities) are also found traversing the STN (Francois et al., 1998), suggestive of a direct DAergic innervation. Consistent with these findings, we report here a specific, albeit weak, 3H-mazindol binding signal in the human STN, indicating a low level complement of DA uptake sites and thus supporting the idea of functional DA terminals within this structure. As the specific [3H]mazindol signal in the STN is weak (approximately 20fold less than the putamen), several important points must be considered when interpreting this data: (1) the impact of postmortem interval on DA uptake, (2) the specificity of 3H-mazindol (in the presence of desipramine to block noradrenaline uptake sites) for DA uptake sites, and (3) the small number (n ⫽ 4) of human cases studied. Considering each point in turn, the impact of postmortem interval on DA uptake is unknown, although our data in the caudate-putamen (this study) coupled with the recent studies of Piggott et al., (1999) suggest that the membrane protein is stable for up to 80 hours (see case 20 in Piggott et al., 1999). The selectivity of 3H-mazindol for labeling DA uptake sites (in the presence of desipramine) has been shown elsewhere and is demonstrated here by the greater binding in the putamen versus the caudate nucleus, consistent with the respective DA enrichment (Hornykiewicz, 1973). Further, the distribution of other DA uptake ligands such as 125RTI-121 (in the presence of clomipramine to block serotonin uptake sites), 3H-WIN-35,428, and 3 H-CFT suggests that 3H-mazindol does indeed bind to sites of DA uptake. However, differences in 125I-RTI-55 and 3H-WIN 35,428 binding density have been noted in the rat STN and have been attributed to variants of the dopamine transporter and/or posttranslational modifications of the membrane protein (Coulter et al., 1995). To date, despite the presence of DA uptake binding sites (Coulter et al., 1995; Flores et al., 1999; Fujita et al., 1994), no DA transporter-like immunoreactivity has been detected within this structure (Ciliax et al., 1995), further suggestive of DA transporter variants. The weak specific [3H]-mazindol binding measured in the human STN is comparable to the binding in the insular cortex on the same tissue sections (see Table 1). Within the insular cortex, scattered TH- and DA transporterimmunopositive axons, both varicose and nonvaricose, have been reported, which are particularly enriched within layer 1 (Ciliax et al., 1999; Gaspar et al., 1989). These immunohistochemical findings coupled with our DA uptake autoradiography data suggest that the low level of [3H]-mazindol binding detected in this structure is indeed indicative of DA terminals. Similarly, the binding of [3H]SCH23390 in the human insular cortex (this study) is consistent with the immunohistochemical localization of D1 and D5 (D1 family) receptors to interneurons (Muly et al., 1998) and dendritic spines (D1) and dendritic shafts (D5) of intrinsic pyramidal neurons in the primate prefrontal cortex (Bergson et al., 1995). The lack of D1 receptor mRNA by in situ hybridization analysis (this study) or Northern analysis (Mrzljak et al., 1996) suggests that intrinsic STN neurons do not express this DA receptor subtype. However, as the expression of D5 receptors was not examined in this study, it is possible that these intrinsic neurons express D5 receptors. Thus, our data showing a low-level complement of D1-like binding in the human STN may reflect the local expression of D5 receptors coupled with the presence of D1/D5 receptors on STN affer- Fig. 1. Montage of tyrosine hydroxylase (TH) immunostaining in the human subthalamic nucleus (STN) and lateral SN. THimmunopositive neurons and nerve fibers are highly enriched within the SN (A,B). Within the STN (C), a fine network of THimmunopositive fibers are detected traversing the nucleus. This fine network of TH-positive fibers (E,F) is more pronounced in the medial aspect of the nucleus. The specificity of the TH immunostaining is demonstrated by the absence of specific staining on a control section processed without any primary antibody. Scale bars ⫽ 200 m in D (also applies to A,C); 100 m in E (also applies to B); 50 m in F. 252 S.J. AUGOOD ET AL. Fig. 3. Total (A,C) and nonspecific (B,D) binding of 3H-SCH-23390 (A,B) and 3H-YM-09151-2 (C,D) to human basal ganglia tissue at the level of the subthalamic nucleus (STN). A: Total binding. Note the enrichment of D1-like binding sites in the head of the caudate nucleus (CN), the putamen (Put), the globus pallidus internus (GPi), and the substantia nigra pars reticulata (SNr) when compared with nonspecific binding in B. C: Total binding. Note the enrichment of D2-like binding sites in the head of the caudate nucleus (CN), the putamen (Put), and the globus pallidus externus (GPe) when compared with nonspecific binding in D. Scale bar ⫽ 1 cm. ents. Ultrastructural studies using subtype specific antisera will be required to clarify this point. Functionally, the DA receptor phenotype of STN neurons in mammals is of interest, as this structure has been identified as a potential therapeutic target in the treatment of Parkinson’s disease. Pharmacological studies have shown that direct infusion of apomorphine, a DA receptor agonist, into the STN elicits oral dyskinesias in rats (Parry et al., 1994), suggestive of functional DA receptors within this structure. Furthermore, systemic administration of ephedrine, an indirect catecholamine agonist (Zarrindast, 1981), results in a massive induction of Fos protein in the striatum and STN (Blandini et al., 1997). However, the precise mechanism of this Fos induc- TH-LI AND DA RECEPTORS IN THE HUMAN STN 253 Fig. 4. Total (A) and nonspecific (B) binding of 3H-mazindol to human basal ganglia tissue at the level of the subthalamic nucleus (STN). Note the enrichment of dopamine (DA) uptake sites in the head of the caudate nucleus (CN) and the putamen (Put) when compared with nonspecific binding in (B). SNr, substantia nigra pars reticulata. Scale bar ⫽ 1 cm. TABLE 1. Specific Binding of Dopamine (DA) D1-Like, D2-Like and DA Uptake Sites in the Human Basal Ganglia From Four Individual Cases1 3 3 [ H]SCH23390 [3H]YM-09151-24 [3H]mazindol CN Put STN SNr Insular ctx 1.56 ⫾ 0.17 1.59 ⫾ 0.12 0.70 ⫾ 0.18 1.43 ⫾ 0.22 1.94 ⫾ 0.16 0.96 ⫾ 0.24 0.06 ⫾ 0.02 0.002 0.05 ⫾ 0.03 0.44 ⫾ 0.00 0.07 ⫾ 0.06 0.02 ⫾ 0.02 0.32 ⫾ 0.06 0.08 ⫾ 0.03 0.06 ⫾ 0.04 1 Values ⫽ mean ⫾ S.E.M. (Ci/g ⫻ 10). N ⫽ 2. D1-like. 4 D2-like. 2 3 tion is not clear and may relate, in part, to the vasoactive properties of this compound; the STN is one of the most densely vascularized structures in the human central nervous system. By contrast, no such induction of Fos immunoreactivity is observed following DA depletion (Cooper and Mitchell, 1995) or systemic haloperidol (Suzuki et al., 1998). Together these data invite the question of whether DA exerts a direct effect on intrinsic STN neurons and, if so, what is the DA receptor phenotype of STN neurons? Extracellular recordings in anesthetized rats have revealed a predominantly excitatory effect of DA on the basal activity of STN neurons (Mintz et al., 1986; Kreiss et al., 1996), although earlier conflicting data exist (Campbell et al., 1985). Similarly, an increase in STN firing rate is observed following systemic administration of apomorphine and SKF-82958 (a D1 receptor agonist) and following DA depletion (Hassani et al., 1996; Kreiss et al., 1997) but not after administration of quinpirole, a D2 receptor agonist (Kreiss et al., 1997). Further, the increase in tonic firing rate of these STN neurons in the DA-depleted rat model of parkinsonism can be “normalized” by systemic administration of the nonselective DA agonists, apomorphine and I-DOPA (Kreiss et al., 1997), underscoring the importance of DAergic interactions within this structure. To dissect the DA receptor phenotype of the rat STN, ligand binding studies have been carried out by using a variety of radiolabeled ligands. In general, studies using tritiated ligands favor a preferential enrichment of D1like over D2-like binding sites (Savasta et al., 1986; Mansour et al., 1990, 1992; Kreiss et al., 1996; Flores et al., 1999), although conflicting data exist using iodinated ligands (Johnson et al., 1994). Further as D1, D2, and D3 receptor mRNAs are expressed by intrinsic STN neurons in the rat (Flores et al., 1999), yet no D1-like or D2-like immunopositive cells have been observed (Yung et al., 1995), these data suggest that the DA receptor mRNAs expressed by rat STN neurons are translated and the receptor proteins exported outside of the STN. With regard to the cellular localization of DA D1 and/or D2 receptors within the human STN, our study demonstrates clearly that DA D1 and D2 receptor mRNAs are differentially expressed within the human basal ganglia, as might be predicted from in situ mapping studies in rodents (Mansour et al., 1990, 1992). Of particular note is the apparent lack of D1 or D2 receptor mRNAs within the STN despite prominent expression within other basal ganglia nuclei on the same tissue section. These findings agree with Northern analysis data (Matsumoto et al., 254 S.J. AUGOOD ET AL. 1996) which similarly report a lack of expression of D1 and D2 receptor mRNAs within the human STN, although an enrichment of D4 receptor mRNA has been observed. These mRNA studies are exactly contrary to rat data which report an enrichment of D1, D2, and D3 receptor mRNA but not D4 mRNA (Flores et al., 1999). Collectively these findings add to the known species differences in dopaminergic systems between rodents and human (Berger et al., 1991; Betarbet et al., 1997) and underscore the importance of human brain studies. In summary, we have demonstrated that THimmunopositive processes traverse the human STN and that these processes may constitute functional DA terminals. By using in situ hybridization and DA receptor receptor autoradiography, we were unable to detect a robust DA D1-like, D2-like, or DA uptake signal within the human STN of any of the normal brains examined. However, weak and specific 3H-mazindol and D1-like binding was observed in this structure (see Table 1), suggestive of a low-level complement of DA terminals and DA D1 receptor binding sites. It will be important to validate these findings by using specific DA receptor subtype-specific antisera coupled with sensitive ultrastructural techniques. Understanding the DA receptor complement of the human STN is of clinical importance, as this structure plays a central role in motor (Lee and Marsden, 1994; Wichmann et al., 1994; Joel and Weiner, 1997) and possibly executive functioning (Baunez et al., 1995; Ardouin et al., 1999). ACKNOWLEDGMENTS Bouin’s-fixed human tissue was provided by the Harvard Brain Tissue Resource Center, which is supported, in part, by USPHS grant number NS 31579. K. Westmore and A. Kingsbury are thanked for their expert assistance. Dr. S. Daniel is thanked for her neuropathological evaluation of the frozen tissue. Dr. John Hedreen is thanked for his comments on this manuscript. LITERATURE CITED Ardouin C, Pillon B, Peiffer E, Bejjani P, Limousin P, Damier P, Arnulf I, Benabid AL, Agid Y, Pollak P. 1999. Bilateral subthalamic or pallidal stimulation for Parkinson’s disease affects neither memory nor executive functions: a consecutive series of 62 patients. Ann Neurol 46:217– 223. Augood S, Faull R, Emson P. 1997. Dopamine D1 and D2 receptor gene expression in the striatum in Huntington’s disease. Ann Neurol 42: 215–221. Baunez C, Nieoullon A, Amalric M. 1995. In a rat model of parkinsonism, lesions of the subthalamic nucleus reverse increases of reaction time but induce a dramatic premature responding deficit. J Neurosci 15: 6531– 6541. Berger B, Gaspar P, Verney C. 1991. Dopaminergic innervation of the cerebral cortex: unexpected differences between rodents and primates. Trends Neurosci 14:21–27. Bergson C, Mrzljak L, Smiley JF, Pappy M, Levenson R, Goldman-Rakic PS. 1995. Regional, cellular, and subcellular variations in the distribution of D1 and D5 dopamine receptors in primate brain. J Neurosci 15:7821–7836. Betarbet R, Turner R, Chockkan V, DeLong MR, Allers KA, Walters J, Levey AI, Greenamyre JT. 1997. Dopaminergic neurons intrinsic to the primate striatum. J Neurosci 17:6761– 6768. Blandini F, Joseph S, Tassorelli C. 1997. Systemic administration of ephedrine induces Fos protein expression in caudate putamen and subthalamic nucleus of rats. Functional Neurol 12:293–296. Brown L, Wolfson L. 1978. Apomorphine increases glucose utilization in the substantia nigra, subthalamic nucleus and corpus striatum of rat. Brain Res 140:188 –193. Campbell GA, Eckardt MJ, Weight FF. 1985. Dopaminergic mechanisms in subthalamic nucleus of rat: analysis using horseradish peroxidase and microiontophoresis. Brain Res 333:261–270. Cha J-H, Kosinski C, Kerner J, Alsdorf S, Mangiarini L, Davies S, Penney J, Bates G, Young A. 1998. Altered brain neurotransmitter receptors in transgenic mice expressing a portion of an abnormal human Huntington disease gene. Proc Natl Acad Sci 95:6480 – 6485. Chesselet M-F, Delfs JM. 1996. Basal ganglia and movement disorders: an update. TINS 19:417– 422. Ciliax BJ, Heilman C, Demchyshyn LL, Pristupa ZB, Ince E, Hersch SM, Niznik HB, Levey AI. 1995. The dopamine transporter: immunochemical characterization and localization in brain. J Neurosci 15:1714 – 1723. Ciliax BJ, Drash GW, Staley JK, Haber S, Mobley CJ, Miller GW, Mufson EJ, Mash DC, Levey AI. 1999. Immunocytochemical localization of the dopamine transporter in human brain. J Comp Neurol 409:38 –56. Cooper AJ, Mitchell IJ. 1995. Fos immunopositive neurons in the subthalamic nucleus following reversal of parkinsonian symptoms by antagonism of excitatory amino acid transmission in the entopeduncular nucleus of the monoamine depleted rat. Neurosci Lett 201:251–254. Cossette M, Levesque M, Parent A. 1999. Extrastriatal dopaminergic innervation of human basal ganglia. Neurosci Res 34:51–54. Coulter CL, Happe HK, Bergman DA, Murrin LC. 1995. Localization and quantification of the dopamine transporter: comparison of [3H]WIN 35,428 and [125I]RTI-55. Brain Res 690:217–224. Flores G, Liang JJ, Sierra A, Martinez-Fong D, Quirion R, Aceves J, Srivastava LK. 1999. Expression of dopamine receptors in the subthalamic nucleus of the rat: characterization using reverse transcriptasepolymerase chain reaction and autoradiography. Neuroscience 91:549 – 556. Francois C, Savy C, Tande D, Yelnik J. 1998. Is there a dopaminergic innervation of the subthalamic nucleus in the primate? Mov Disord 13:P2.006. Fujita M, Shimada S, Fukuchi K, Tohyama M, Nishimura T. 1994. Distribution of cocaine recognition sites in rat brain: in vitro and ex vivo autoradiography with RTI-55. J Chem Neuroanat 7:13–23. Gaspar P, Berger B, Febvret A, Vigny A, Henry JP. 1989. Catecholamine innervation of the human cerebral cortex as revealed by comparative immunohistochemistry of tyrosine hydroxylase and dopamine-betahydroxylase. J Comp Neurol 279:249 –271. Harrington KA, Augood SJ, Faull RL, McKenna PJ, Emson PC. 1995. Dopamine D1 receptor, D2 receptor, proenkephalin A and substance P gene expression in the caudate nucleus of control and schizophrenic tissue: a quantitative cellular in situ hybridisation study. Brain Res Mol Brain Res 33:333–342. Hassani OK, Mouroux M, Feger J. 1996. Increased subthalamic neuronal activity after nigral dopaminergic lesion independent of disinhibition via the globus pallidus. Neuroscience 72:105–115. Hassani OK, Francois C, Yelnik J, Feger J. 1997. Evidence for a dopaminergic innervation of the subthalamic nucleus in the rat. Brain Res 749:88 –94. Hedreen J. 1999. Tyrosine hydroxylase-immunoreactive elements in the human globus pallidus and subthalamic nucleus. J Comp Neurol 409: 400 – 410. Hornykiewicz O. 1973. Dopamine in the basal ganglia: its role and therapeutic implications (including the clinical use of L-DOPA). Brit Med Bull 29:1972–1978. Joel D, Weiner I. 1997. The connections of the primate subthalamic nucleus: indirect pathways and the open-interconnected scheme of basal ganglia-thalamocortical circuitry. Brain Res Rev 23:62–78. Johnson AE, Coirini H, Kallstrom L, Wiesel F-A. 1994. Characterization of dopamine receptor binding sites in the subthalamic nucleus. NeuroReport 5:1836 –1838. Kingsbury AE, Foster OJF, Nisbet AP, Cairns N, Bray L, Eve DJ, Lees AJ, Marsden CD. 1995. Tissue pH as an indicator of mRNA preservation in human post-mortem brain. Brain Res Mol Brain Res 28:311–318. Kreiss DS, Anderson LA, Walters JR. 1996. Apomorphine and dopamine D1 receptor agonists increase the firing rates of subthalamic nucleus neurons. Neuroscience 72:863– 876. Kreiss DS, Mastropietro CW, Rawji SS, Walters JR. 1997. The response of subthalamic nucleus neurons to dopamine receptor stimulation in a rodent model of Parkinson’s disease. J Neurosci 17:6807– 6819. Lee MS, Marsden CD. 1994. Movement disorders following lesions of the thalamus or subthalamic region. Mov Disord 9:493–507. Mansour A, Meador-Woodruff J, Bunzow J, Civelli O, Akil H, Watson S. TH-LI AND DA RECEPTORS IN THE HUMAN STN 1990. Localization of dopamine D2 receptor mRNA and D1 and D2 receptor binding in the rat brain and pituitary: an in situ hybridization-receptor autoradiographic analysis. J Neurosci 10:2587– 2600. Mansour A, Meador WJ, Zhou Q, Civelli O, Akil H, Watson SJ. 1992. A comparison of D1 receptor binding and mRNA in rat brain using receptor autoradiographic and in situ hybridization techniques. Neuroscience 46:959 –971. Matsumoto M, Hidaka K, Tada S, Tasaki Y, Yamaguchi T. 1996. Low levels of mRNA for dopamine D4 receptor in human cerebral cortex and striatum. J Neurochem 66:915–919. Meibach R, Katzman R. 1979. Catecholaminergic innervation of the subthalamic nucleus: evidence for a rostral continuation of the A9 (substantia nigra) dopaminergic cell group. Brain Res 173:364 –368. Mintz I, Hammond C, Feger J. 1986. Excitatory effect of iontophoretically applied dopamine on identified neurons of the rat subthalamic nucleus. Brain Res 375:172–175. Mrzljak L, Bergson C, Pappy M, Huff R, Levenson R, Goldman-Rakic PS. 1996. Localization of dopamine D4 receptors in GABAergic neurons of the primate brain. Nature 381:245–248. Muly EC, Szigeti K, Goldman-Rakic PS. 1998. D1 receptor in interneurons of macaque prefrontal cortex: distribution and subcellular localization. J Neurosci 18:10553–10565. Parry T, Eberle-Wang K, Lucki I, Chesselet M-F. 1994. Dopaminergic stimulation of subthalamic nucleus elicits oral dyskinesia in rats. Exp Neurol 128:181–190. Piggott MA, Marshall EF, Thomas N, Lloyd S, Court JA, Jaros E, Costa 255 D, Perry RH, Perry EK. 1999. Dopaminergic activities in the human striatum: rostrocaudal gradients of uptake sites and of D1 and D2 but not of D3 receptor binding or dopamine. Neuroscience 90:433– 445. Rosales M, Martinez-Fong D, Morales R, Nunez A, Flores G, GongoraAlfaro J, Floran B, Aceves J. 1997. Reciprocal interaction between glutamate and dopamine in the pars reticulata of the rat substantia nigra: a microdialysis study. Neuroscience 80:803– 810. Savasta M, Dubois A, Scatton B. 1986. Autoradiographic localization of D1 dopamine receptors in the rat brain with [3H]SCH 23390. Brain Res 375:291–301. Suzuki M, Sun YJ, Murata M, Kurachi M. 1998. Widespread expression of Fos protein induced by acute haloperidol administration in the rat brain. Psychiatry Clin Neurosci 52:353–359. Vonsattel JP, Aizawa H, Ge P, DiFiglia M, McKee AC, MacDonald M, Gusella JF, Landwehrmeyer GB, Bird ED, Richardson EP, HedleyWhyte ET. 1995. An improved approach to prepare human brains for research. J Neuropathol Exp Neurol 54:42–56. Wichmann T, Bergman H, DeLong MR. 1994. The primate subthalamic nucleus. I. Functional properties in intact animals. J Neurophysiol 72:494 –506. Yung KKL, Bolam JP, Smith AD, Hersch SM, Ciliax BJ, Levey AI. 1995. Immunocytochemical localization of D1 and D2 dopamine receptors in the basal ganglia of the rat: light and electron microscopy. Neuroscience 65:709 –730. Zarrindast M. 1981. Dopamine-like properties of ephedrine in rat brain. Br J Pharmacol 74:119 –122.