Distribution and morphology of the catecholaminergic neural elements in the human hypothalamus

Neuroscience 171 (2010) 187–195

DISTRIBUTION AND MORPHOLOGY OF THE CATECHOLAMINERGIC NEURAL ELEMENTS IN THE HUMAN HYPOTHALAMUS B. DUDAS,a* M. BAKER,a G. ROTOLI,a G. GRIGNOL,a M. C. BOHNb AND I. MERCHENTHALERc

The human catecholaminergic system is composed of dopaminergic, noradrenergic, and adrenergic perikarya located mostly in the rhombencephalon, mesencephalon, and diencephalon with fibers projecting into widespread areas of the brain and spinal cord. Since the catecholaminergic neurons appear to control numerous physiological functions associated with the hypothalamo-hypophyseal system, the distribution of the catecholaminergic elements in the hypothalamus has been extensively studied in numerous species including human, using primarily immunohistochemistry and in situ hybridization histochemistry to detect tyrosine hydroxylase (TH), the key and rate-limiting enzyme of the catecholaminergic synthesis present in all classes of catecholaminergic neurons (Hokfelt et al., 1976; Kitahama et al., 1987; Li et al., 1988; Panayotacopoulou et al., 1991; Zoli et al., 1993; Dudas and Merchenthaler, 2001; Panayotacopoulou et al., 2005; Dudas and Merchenthaler, 2006). However, whether these TH-immunoreactive (IR) neural elements represent dopaminergic, noradrenergic or adrenergic structures of the human hypothalamus has not been entirely elucidated yet. Since formaldehyde-induced fluorescence (Dahlstrom and Fuxe, 1964; Axelsson et al., 1973; Bjorklund and Nobin, 1973; Bjorklund et al., 1975) cannot differentiate between adrenergic and noradrenergic structures, phenylethanolamine-N-methyltransferase (PNMT), the rate limiting enzyme of the epinephrine synthesis, has been widely used to identify the synthesis of epinephrine in the brain of several species including human by using immunohistochemistry (Bohn et al., 1986; Ericson et al., 1989; Mezey, 1989; Palkovits et al., 1992) and enzyme activity assays (Nagatsu et al., 1977; Lew et al., 1977; Kopp et al., 1979; Moreno et al., 1992). Although epinephrine is widely distributed in the brain (Mefford, 1988), there is a general consensus that most of the adrenergic structures are fiber varicosities supplied by C1 and C2 cell groups located in the medulla oblongata (Dahlstrom and Fuxe, 1964; Kitahama et al., 1986, 1988; Astier et al., 1987; Sawchenko and Bohn, 1989; Cunningham et al., 1990). Additionally, PNMT-IR perikarya have been described in the limbic system (Mezey, 1989) and the posterior hypothalamus of rat (Ruggiero et al., 1985). Most of these studies provide data regarding the adrenergic elements in the rat hypothalamus (Lew et al., 1977; Ruggiero et al., 1985; Ericson et al., 1989; Mezey, 1989; Moreno et al., 1992; Palkovits et al., 1992). In human, the presence of epinephrine in the hypothalamic regions has been identified mainly by measuring PNMT enzyme activity (Nagatsu et al., 1977; Lew et al., 1977; Kopp et al., 1979). However, the precise distri-

a Neuroendocrine Organization Laboratory (NEO), Lake Erie College of Osteopathic Medicine (LECOM), Erie, PA 16509, USA b Children’s Memorial Research Center, Feinberg School of Medicine, Northwestern University, Chicago, IL 60614, USA c Department of Epidemiology and Preventive Medicine and Anatomy and Neurobiology, University of Maryland, Baltimore, MD 21201, USA

Abstract—Previous studies have demonstrated that catecholaminergic, tyrosine hydroxylase (TH)-immunoreactive (IR) perikarya and fibers are widely distributed in the human hypothalamus. Since TH is the key and rate-limiting enzyme for catecholaminergic synthesis, these IR neurons may represent dopaminergic, noradrenergic or adrenergic neural elements. However, the distribution and morphology of these neurotransmitter systems in the human hypothalamus is not entirely known. Since the different catecholaminergic systems can be detected by identifying the neurons containing the specific key enzymes of catecholaminergic synthesis, in the present study we mapped the catecholaminergic elements in the human hypothalamus using immunohistochemistry against the catecholaminergic enzymes, TH, dopamine beta-hydroxylase (DBH) and phenylethanolamine-N-methyltransferase (PNMT). Only a few, PNMT-IR, adrenergic neuronal elements were found mainly in the infundibulum and the periventricular zone. DBH-IR structures were more widely distributed in the human hypothalamus occupying chiefly the infundibulum/infundibular nucleus, periventricular area, supraoptic and paraventricular nuclei. Dopaminergic elements were detected by utilizing double label immunohistochemistry. First, the DBH-IR elements were visualized; then the TH-IR structures, that lack DBH, were detected with a different chromogen. In our study, we conclude that all of the catecholaminergic perikarya and the majority of the catecholaminergic fibers represent dopaminergic neurons in the human hypothalamus. Due to the extremely small number of PNMT-IR, adrenergic structures in the human hypothalamus, the DBH-IR fibers represent almost exclusively noradrenergic neuronal processes. These findings suggest that the juxtapositions between the TH-IR and numerous peptidergic systems revealed by previous reports indicate mostly dopaminergic synapses. © 2010 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: adrenaline, noradrenaline, epinephrine, norepinephrine, dopamine, diencephalon. *Corresponding author. Tel: ⫹1-814-866-8142; fax: ⫹1-814-866-8411. E-mail address: [email protected] (B. Dudas). Abbreviations: DAB, 3=-3= diaminobenzidine tetrahydrochloride; DAT, dopamine transporter; DBH, dopamine beta-hydroxylase; IR, immunoreactive; PNMT, phenylethanolamine-N-methyltransferase; PVN, paraventricular nuclei; SON, supraoptic nuclei; TH, tyrosine hydroxylase; TSA, tyramide signal amplification.

0306-4522/10 $ - see front matter © 2010 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2010.08.050

187

188

B. Dudas et al. / Neuroscience 171 (2010) 187–195

bution and morphology of PNMT-IR neural elements in the human hypothalamus have not been described previously. The noradrenergic system is composed of perikarya restricted to A1–A7 cell groups located in the rhombencephalon whose ascending and descending fibers innervate essentially all regions of the CNS (Dahlstrom and Fuxe, 1964; Nobin and Bjorklund, 1973; Olson et al., 1973a,b). However, noradrenergic neurons cannot be distinguished from adrenergic neurons by either formaldehyde-induced fluorescence or by immunohistochemistry to dopamine beta-hydroxylase (DBH) which is expressed in both the noradrenergic and adrenergic neurons (note that TH is the rate limiting enzyme in all three types of catecholaminergic neurons). The hypothalamus is believed to receive ascending fibers from both adrenergic and noradrenergic neurons from the brain stem (Swanson and Hartman, 1975; Ericson et al., 1989) but the contributions of each have not been clearly distinguished from DA fibers. Previous studies revealed DBH-IR structures in the rat hypothalamus (Swanson and Hartman, 1975; Ericson et al., 1989; Jansen et al., 2003), however, limited data exist regarding DBH-containing structures in human, or the contribution of DBH-expressing adrenergic structures. Although DBH enzyme activity has been described in humans (Nagatsu et al., 1977; Jansen et al., 2003), the pattern and morphology of DBH-IR and PNMT-IR neural elements in the human hypothalamus have not been entirely explored. In order to describe the adrenergic, noradrenergic, and dopaminergic system in the human hypothalamus, in the present study, we examined and mapped the TH-IR, DBH-IR and PNMT-IR neural elements using single label and double label immunohistochemistry. Additionally, the morphology and the putative connections of these catecholaminergic structures were also investigated.

chemical detection of PNMT and (iv) double label immunohistochemical detection of DBH (first label) and TH (second label).

Immunohistochemistry

Hypothalami (two adult women and two adult men, 55– 87 years of age) were obtained from autopsies at 24 – 48 h post mortem period in accordance with the regulation and permission of the Ethics Board of Lake Erie College of Osteopathic Medicine and the University of Szeged, Hungary. The clinical records of the individuals did not indicate any neurological and neuroendocrinological disorders or cognitive impairment.

Immunohistochemistry was performed using streptavidin-biotin (ABC) methods combined with silver intensification introduced by Gallyas and coworkers (Gallyas et al., 1982; Gallyas and Merchenthaler, 1988) and simultaneous detection of antigens described previously (Liposits et al., 1983). The samples were pretreated with 10% thioglycolic acid (Sigma-Aldrich, St. Louis, MO, USA) for 30 min to suppress the endogenous tissue argentophilia, then with 0.2% Triton X-100 for 20 min to increase permeability of immunoglobulins, and finally with 10% normal goat serum (NGS) in PBS for 1.0 h at room temperature to block nonspecific staining. Thereafter, the sections were incubated in a rabbit anti-rat PNMT (Bohn et al., 1987; dilution: 1:200), rabbit anti-rat DBH (Chemicon, Temecula, CA, USA; dilution: 1:1000) or rabbit anti-rat TH (Chemicon, Temecula, CA, USA; dilution 1:8000) sera, respectively, for 12 h. The antiserum diluent contained 10% NHS, PBS and 0.1% sodium azide. The sections were then incubated for 1.0 h with biotinylated goat anti-rabbit immunoglobulins (IgG) (Vectastain ABC Elite kit, Vector Laboratories, Burlingame, CA, USA) diluted 1:1000 in PBS. Following three washes in PBS for a total of 30 min, the tissue was incubated for 1.0 h in a solution of streptavidin-horseradish peroxidase (Vectastain ABC Elite kit) diluted 1:1000 in PBS. The cross-reactions of the secondary antibodies with the endogenous human IgGs were eliminated by the addition of 2% normal human serum to the secondary antiserum. The sections were washed 2⫻10 min in PBS and 10 min in Tris-buffered saline (TBS; pH 7.6) and then reacted with 3=-3= diaminobenzidine tetrahydrochloride (DAB), using nickel ammonium sulfate intensification (20 mg DAB, 100 mg nickel ammonium sulfate and 7.0 ␮l 30% H2O2 in 40 ml TBS). The resulting DAB polymer was silver-intensified in a solution containing 0.1% silver nitrate, 0.1% ammonium-nitrate, 1% silicotungstic acid and 0.2% formaldehyde for 3– 4 min (Gallyas et al., 1982; Gallyas and Merchenthaler, 1988). All of the steps of the immunohistochemistry were performed at 22 °C (room temperature). Simultaneous detection of DBH-IR and TH-IR structures was performed using double-label immunohistochemistry. First, DBH immunohistochemistry was carried out with the black DAB/nickel chromogen, and then the TH-containing neuronal structures were immunolabeled with the brown DAB chromogen. The anti-TH serum was used in a 1:4000 dilution in these studies. Since the first immunohistochemical signal, that was used to reveal the DBH-IR elements, has been silver-intensified, the second immunohistochemical signal reveals those TH-IR elements only that do not contain DBH, thus being dopaminergic. Olympus BX45 microscope with a 100⫻ oil immersion objective was used for the evaluation of the immunohistochemical staining.

Tissue preparation

Controls

Following the removal of the brains from the skull, the hypothalami were dissected and the tissue blocks were fixed by immersion in 0.1 M phosphate-buffered (pH 7.4; PB) 4% formaldehyde at 4 °C for 2– 8 weeks. Each block contained half of the hypothalamus divided in the midsagittal line. The samples were cryoprotected with 30% sucrose in phosphate buffer containing 0.9% sodium chloride (PBS) supplemented with 0.15% sodium-azide and then sectioned on a freezing microtome at 30 ␮m intervals in coronal planes. The sections were collected in four series of wells of plastic 24-compartment plates with PBS containing 0.2% sodiumazide, and stored at 4 °C until processing. The adjacent sections were processed as follows: (i) immunohistochemical detection of TH, (ii) immunohistochemical detection of DBH (iii) immunohisto-

In the control studies, the primary antibodies were omitted and replaced by non-IR vehicle used for diluting the antibodies. Immunoreaction was not observed. In order to test the PNMT antiserum on human brain tissue, sections from the upper medulla oblongata were stained by single label immunohistochemistry utilizing the same anti-PNMT antiserum that was used for the hypothalamic sections. The presence of PNMT-IR fibers in human brain stem were verified (Fig. 4D). Additionally, the lack of the DBH and PNMT perikarya detectable with the immunohistochemical methods we used was also verified using tyramide signal amplification (TSATM) system kit (PerkinElmer, Boston, MA 02118-2512, USA). No PNMT or DBH cell bodies were detected in the human hypothalamus.

EXPERIMENTAL PROCEDURES Brain samples

B. Dudas et al. / Neuroscience 171 (2010) 187–195

Computer-assisted mapping The hypothalamic sections were systematically scanned using a plain-scanner (Hewlett–Packard) and the outlines of the sections were traced with the CorelTrace software 4.0 (Corel Corporation, Ottawa, ON, Canada). The computer-generated superimposition of eight, consecutive sections was summarized in each of the slides of Figs. 2 and 3 denoted by letters from A to P. Each of these slides demonstrate a 960 ␮m thick section of the human diencephalon. The neurons and fibers were marked on these figures using an Olympus BX45 microscope with camera lucida, and Adobe Photoshop software (San Jose, CA, USA), version 5.0.

Terminology The terminology of the diencephalic structures was adapted from Braak and Braak (Braak and Braak, 1987) and Saper (Saper, 1990).

RESULTS The TH-IR elements of the diencephalon The distribution of catecholaminergic, TH-IR system of the human hypothalamus is in good agreement with our previous reports (Dudas and Merchenthaler, 2001, 2006). Briefly, the majority of the TH-IR perikarya are located in the ventral part of the preoptic and tuberal areas (Fig. 1A, C), while a few TH-IR cell bodies are present in the posterior hypothalamus. The medial-dorsal region of the septal area contains a small number of IR cells while no perikarya can be observed along the diagonal band of Broca and the lamina terminalis. TH-IR neurons are arranged periventricularly in the basal part of the preoptic

189

region and a few perikarya are located at the bottom of the optic recess on the dorsal surface of the optic chiasm. In the tuberal region, a substantial number of perikarya are present in the median eminence and in the periventricular area (Fig. 1A, C), while the infundibular region contains a relatively small number of TH-IR perikarya. The TH-IR cell bodies are densely packed in the paraventricular (PVN) and supraoptic nuclei (SON). In the posterior hypothalamus, the cells are arranged periventricularly, with few labeled neurons found around the mammillary body. Morphologically, the TH-IR neurons are mainly fusiform, with thin cell bodies and two processes emanating from their opposite poles. Multipolar neurons can be detected mainly in the paraventricular and supraotic nuclei. TH-IR fibers appear as a dense network primarily in the periventricular and infundibular areas (Fig. 1A, C). The fibers are mostly thick, heavily labeled axonal varicosities that often form well-defined terminal fields in the PVN and SON, previously described in our previous studies (Dudas et al., 2006). A small number of thinner TH-IR varicosities can be observed primarily in the periventricular and medial hypothalamic areas, although they often appeared to be obscured by the thicker fibers. Labeled axon varicosities can also be observed in the lateral hypothalamus and in the septal area. The DBH-IR elements of the diencephalon The DBH-IR structures are represented by thin fiber varicosities; no perikarya can be observed in the human diencephalon by immunohistochemistry combined with either

Fig. 1. Catecholaminergic, TH-IR (A, C) and noradrenergic/adrenergic, DBH-IR (B, D) neural elements in the PVN (A, B) and at the periventricular infundibular area (C, D) of the human hypothalamus. Inserts (A, B) and (C, D) illustrate catecholaminergic structures in consecutive sections surrounding the same blood vessel (asterisk). Scale bar: 40 ␮m.

190

B. Dudas et al. / Neuroscience 171 (2010) 187–195

Fig. 2. Distribution of the DBH-IR neuronal elements in the human hypothalamus. The noradrenergic or adrenergic structures are represented by axonal varicosities; no perikarya were detected in the hypothalamus. Each slide (A–P) is the superimposition of eight 30 ␮m thick sections. Arrows denote the major directions of the DBH-IR fibers. Ac, anterior commissure; Dbb, diagonal band of Broca; Fx, fornix; Mb, mammillary body; Och, optic chiasm; Ot, optic tract; Pvn, paraventricular nucleus; Son, supraoptic nucleus. Scale bar: 5000 ␮m.

silver intensification or TSATM methods (Figs. 2 and 5). The vast majority of the DBH-IR fibers are located in the

periventricular and medial hypothalamic zones of the preoptic and tuberal regions; only a few DBH-IR axon vari-

B. Dudas et al. / Neuroscience 171 (2010) 187–195

cosities can be detected in the lateral hypothalamic regions and in the posterior hypothalamus (Fig. 2). In front of the lamina terminalis, fibers can be detected along the diagonal band of Broca with their number gradually decreasing laterally and inferiorly (Fig. 2A–C). Fibers are also seen inferior to the medial part of the anterior commissure (Fig. 2A–D). No substantial number of the fibers can be detected in the lateral part of the septal area. Fibers are also seen in the lamina terminalis, generally oriented supero-inferiorly (Fig. 2D). In the periventricular zone of the preoptic area, a relatively dense fiber network can be detected with fibers commonly running parallel with the surface of the 3rd ventricle (Fig. 2E, F). In the tuberal region, the majority of the DBH-IR axonal varicosities can be found in the periventricular and medial hypothalamic zones, generally stretched between the infundibulum and the PVN (Fig. 2H–L). The PVN appeared to have relatively dense innervation with fibers running in every direction, while superior to the PVN, the periventricular fibers are gradually decreasing in number (Fig. 2G–I). In the infundibular region, the basal part of the infundibulum receives relatively rich innervation, fibers running parallel with the surface of the 3rd ventricle (Fig. 2H–L). A relatively dense fiber network is located in the infundibular nucleus, at the basal part of the medial hypothalamus below the fornix and at the lateral zone of the median eminence close to the optic tract (Fig. 2J–L). An additional, relatively dense fiber

191

network is present in the SON (Fig. 2F–I). Only few DBH-IR fibers can be seen in the medial part of the lateral hypothalamus, frequently at the lateral side of the fornix and in the basal hypothalamic region (Fig. 2E–L). In the posterior hypothalamus, the majority of the DBH-IR axonal varicosities are periventricularly arranged; scattered fibers are located medially and superiorly around the mammillary body (Fig. 2M–P). No substantial number of DBH-IR fibers can be detected in the mammillary nuclei, fornix and anterior commissure. The PNMT-IR elements of the diencephalon In the human hypothalamus, the adrenergic neural elements are represented by an exceptionally small number of thin, PNMT-IR fibers without any discernable perikarya detectable by immunohistochemistry combined with either silver intensification or TSATM methods (Figs. 3 and 4). Most of the PNMT-IR axonal varicosities are located in the basal part of the infundibular area running parallel with the surface of the 3rd ventricle (Figs. 3I–L and 4A, C). More superiorly, these fibers appeared to be arranged into two distinct subdivisions. (1) Periventricularly arranged PNMTIR fibers running parallel to, but not directly under the ependymal layer, are more numerous in the basal part of the periventricular area, while superiorly to the fornix, the detectable PNMT-IR axonal varicosities are located in the

Fig. 3. Distribution of the PNMT-IR neuronal elements in the human hypothalamus. The adrenergic structures are represented by axonal varicosities; no perikarya were detected in the hypothalamus. Each slide (F–M) is the superimposition of eight 30 ␮m thick sections. Arrows denote the major directions of the PNMT-IR fibers. Ac, anterior commissure; Fx, fornix; Mb, mammillary body; Och, optic chiasm; Ot, optic tract; Pvn, paraventricular nucleus; Son, supraoptic nucleus. Scale bar: 5000 ␮m.

192

B. Dudas et al. / Neuroscience 171 (2010) 187–195

Fig. 4. Catecholaminergic elements in the human hypothalamus. Adrenergic, PNMT-IR axonal varicosities are located mainly in the basal part of the periventricular area adjacent to but not directly under the ependymal layer (A), basal part of the medial hypothalamus (B), and in the infundibular region that mostly correspond to the infundibular nucleus (C). The third ventricle is denoted by the asterisk. For control, PNMT-IR fiber varicosities were detected in the human medulla oblongata using the same anti-PNMT antiserum as used for staining in the hypothalamus (D). Scale bar: 20 ␮m.

medial hypothalamic zone (Fig. 3I–L). The PVN contains negligible amount of PNMT-IR fibers. (2) A substantial number of PNMT-IR axons can be detected in the basal part of the medial hypothalamus. These fibers are seemingly the continuation of the ones in the basal infundibulum, and they are generally arranged either parallel with the surface of the 3rd ventricle reaching superiorly to the perifornical area or parallel with the basal surface of the hypothalamus, directly under the pia mater (Fig. 3I–L). Few PNMT-IR axons are located in the basal part of the lateral hypothalamus, where several of the detected fibers appeared to be rostrocaudally arranged (Fig. 3J–L). Associations between the catecholaminergic elements Double label immunohistochemistry visualizing first DBH-IR and then TH-IR neuronal elements in the hypothalamus revealed that DBH-IR fiber varicosities often surround TH-IR, but DBH-negative, dopaminergic perikarya in the PVN (Fig. 5A), periventricular zone, infundibulum/infundibular nucleus (Fig. 5B) and the supraoptic nucleus of the hypothalamus (Fig. 5C). Despite the frequent close appositions and the occasional juxtapositions between the TH-IR cell bodies and the DBH-IR abutting fibers, no substantial amount of juxtapositions can be revealed between the dopaminergic and the DBH-IR neurotransmitter systems. In addition, no close associations can be detected between the PNMT-IR fiber varicosities and DBH-IR or TH-IR neuronal elements (not shown).

present study is the first that provides a comprehensive and detailed map of the catecholaminergic structures in the human hypothalamus in which the three catecholaminergic phenotypes are distinguished. We observed only a small number of PNMT-IR, adrenergic fiber varicosities in the human hypothalamus (Fig. 3). These findings are in general consensus with previous studies reporting significantly smaller epinephrine content in the human hypothalamus comparing to that of other species (Mefford, 1988). Although previous studies described adrenergic cell bodies in the posterior hypothalamus of the rat (Ruggiero et al., 1985; Foster et al., 1985a), no apparent hypothalamic perikarya were detected in the human hypothalamus in the present study, suggesting discernible species differences regarding the distribution of the adrenergic structures. Previous observations reported adrenergic C1 and C2 cell groups in the rhombencephalon

DISCUSSION In the present study, we examined the distribution of the catecholaminergic elements in the human hypothalamus using immunohistochemical detection of the three key enzymes required for synthesis of dopamine, norepinephrine or epinephrine. Although the morphology of the catecholaminergic system has been described by earlier studies in the rat, little is known about the pattern of the catecholaminergic neural elements in the human. The

Fig. 5. Double label immunohistochemistry revealed DBH-IR fiber varicosities (black) surrounding dopaminergic perikarya (brown) in the PVN (A), infundibulum (B) and the SON (C). Since the first immunohistochemical signal, that was used to reveal the DBH-IR elements, was silver-intensified, the second immunohistochemical signal reveals those TH-IR elements only that do not contain DBH, thus being dopaminergic. Scale bar: 20 ␮m.

B. Dudas et al. / Neuroscience 171 (2010) 187–195

of several species (Hokfelt et al., 1973; Howe et al., 1980; Ruggiero et al., 1985; Kitahama et al., 1985, 1986; Foster et al., 1985b; Sawchenko and Bohn, 1989), indicating that the source of the PNMT-IR fiber varicosities in the human hypothalamus is the brain stem. Similarly to the noradrenergic system, these ascending adrenergic fibers may also reach the hypothalamus via the medial forebrain bundle that is a rather elusive, poorly defined network of fibers running antero-posteriorly through the lateral hypothalamus. In the present study, the PNMT-IR fibers appeared to be accumulated into two major pathways that converge towards the medial forebrain bundle. The periventricularlyarranged fibers seemingly targeting the infundibulum/median eminence appeared to turn around the fornix superiorly and to run mediolaterally; some fibers were observed to run at the basal hypothalamus at the basal surface of the median eminence (Fig. 3). Both fiber networks appeared to converge at the basal part of the lateral hypothalamus indicating that they continued their path as part of the medial forebrain bundle. Indeed, we have observed PNMT-IR fiber varicosities that often appeared to be in oblique or cross section in the basal part of the lateral hypothalamic area. The phenotype of neurons innervated by adrenergic fibers are yet to be identified. The majority of the hypothalamic PNMT-IR fibers converge towards the infundibulum and the pituitary stalk (Fig. 3) where numerous superoinferiorly arranged fiber varicosities were observed (Fig. 4C), indicating that these fibers (1) may modulate neurotransmitter systems in the infundibulum/infundibular nucleus, (2) may terminate around portal vessels and release epinephrine into the portal blood and/or (3) may continue in the pituitary stalk and terminate in the posterior pituitary. Additionally, an extremely small number of PNMT-IR fibers were present in the PVN and the superior part of the periventricular area (Fig. 3). These fibers are apparently not forming well-defined terminal fields similar to the catecholaminergic, TH-IR terminal fields reported previously (Dudas and Merchenthaler, 2001); instead they run straight, often parallel with the surface of the 3rd ventricle forming numerous varicosities along their course. The extremely small number of PNMT-IR axonal varicosities in the human hypothalamus suggests that, (1) the far majority of DBH-IR fibers represent noradrenergic structures, and (2) the volume of the hypothalamic structures innervated by the adrenergic system is rather limited. Thus, these observations suggest that the adrenergic system does not play a pivotal role in the regulation of vital hypothalamic functions. The greater majority of DBH-IR, primarily noradrenergic neuronal elements is represented by thin fiber varicosities forming a loose network that is concentrated mostly in the medial and basal hypothalamic areas and the medial zone of the lateral hypothalamic area (Fig. 2). Only few thick fibers were detected, arranged mostly periventricularly, and in the median eminence. Since the volume of the PNMT-IR fibers is rather limited in the human hypothalamus, it is conceivable that most of the DBH-IR fiber varicosities represent noradrenergic structures. Previous stud-

193

ies have revealed that noradrenergic perikarya are restricted to the pontine and medullary tegmental regions in several species, including humans. Similarly to the adrenergic system, there is a general consensus that the ascending DBH-IR fibers reach the hypothalamus via the medial forebrain bundle located in the lateral hypothalamus. Our study is in agreement with these observations as numerous DBH-IR fiber varicosities were detected in the medial part of the lateral hypothalamus. The DBH-IR axonal varicosities formed a relatively dense network in the PVN, SON and in the basal part of the infundibulum that includes the infundibular nucleus (Figs. 1B, D, 2 and 5). These results are in agreement with previous findings (Jansen et al., 2003). The abundance of the DBH-IR neuronal elements in these regions, although not comparable to the density of the TH-IR structures in the same areas, suggests that DBH-IR fibers, representing primarily the noradrenergic system, play a role in the regulation of several neurotransmitter systems located in the PVN, SON and infundibulum. Additionally, similarly to the adrenergic fibers, the noradrenergic axonal varicosities may terminate around portal vessels or may continue in the pituitary stalk and terminate in the pituitary in human. The synaptic targets of these axonal varicosities are yet to be revealed. Previous studies reported that the posterior pituitary contains TH-IR fibers only without any detectable DBH-IR in pig (Leshin et al., 1996), suggesting that DBH-IR fibers may target the adenohypophysis. Since the PVN contained a negligible number of PNMT-IR axons and these axons were not detected at all in the SON, the DBH-IR fibers in these regions are likely to exclusively represent noradrenergic structures. Similarly, the diagonal band of Broca and the lamina terminalis contained a significant number of DBH-IR axonal varicosities, but no PNMT-IR structures, indicating that these fibers are indeed noradrenergic. In our previous studies, we have described TH-IR fiber varicosities in the PVN and SON, forming fiber baskets around the contours of fusiform neurons that were apparently not TH positive (Dudas and Merchenthaler, 2001). In the present study, we did not detect PNMT-IR structures participating in the formation of such terminal fields (Fig. 4) suggesting that the fiber baskets described earlier were either noradrenergic or dopaminergic. Since we have shown dense TH-IR innervation often forming such terminal fields around oxytocinergic (Semeniken et al., 2009) and vasopressinergic neurons (Dudas et al., 2006), it is likely that these fiber baskets represent the morphological substrate of norepinephrine- or dopamine-modulated oxytocin and vasopressin release. Our findings regarding the distribution and morphology of the TH-IR axonal varicosities are in good consensus with previous studies describing TH-IR and dopamine transporter (DAT)-IR fibers in the human hypothalamus (Ciliax et al., 1999). Previous studies revealed that hypothalamic neurons showed no detectable DAT-immunoreactivity indicating that (1) at least some of these dopaminergic, DAT-IR fibers may originate from mesencephalic areas and/or some hypothalamic neurons may have differ-

194

B. Dudas et al. / Neuroscience 171 (2010) 187–195

ent mechanisms to regulate their extracellular dopamine levels (Ciliax et al., 1999). In our studies, TH-IR fibers appeared primarily as relatively thick fiber varicosities (Fig. 1A, C). Additionally, thin, TH-IR fibers were also detected primarily in the basal part of the periventricular and medial hypothalamic areas (Fig. 1A and C ), in the SON and in the medial zone of the lateral hypothalamus, although these were often obscured by the thicker fibers and the TH-IR perikarya. Since DBH-IR and PNMT-IR neuronal elements were almost exclusively represented by thin fiber varicosities (Figs. 1B, D, 4 and 5), it is conceivable that the thick TH-IR fibers represent mostly dopaminergic axonal varicosities while the thin ones are either noradrenergic or adrenergic. Moreover, previous studies utilizing various techniques identified three catecholaminergic cell groups A11, A12, A14 —in the hypothalamus (Ungerstedt, 1971; Bjorklund and Nobin, 1973; Felten et al., 1974; Felten, 1976; Felten and Sladek, 1983). Since no DBH and PNMT-IR perikarya were observed in the human hypothalamus, it is plausible that these TH-IR cell bodies described also in the present study are all dopaminergic.

CONCLUSION In summary, the present study is the first that provides a comprehensive map of the key enzymes of catecholaminergic synthesis in the human hypothalamus. Since these enzymes each are required for a specific step in the catecholaminergic synthesis, it was possible to definitely distinguish structures representing dopaminergic, noradrenergic or adrenergic neuronal elements. Understanding the distribution and morphology of these elements is pivotal for revealing the processes regulated by each catecholamine on key hypothalamic functions, including reproduction, growth, feeding and stress.

REFERENCES Astier B, Kitahama K, Denoroy L, Jouvet M, Renaud B (1987) Immunohistochemical evidence for the adrenergic medullary longitudinal bundle as a major ascending pathway to the hypothalamus. Neurosci Lett 78:241–246. Axelsson S, Bjorklund A, Falck B, Lindvall O, Svensson LA (1973) Glyoxylic acid condensation: a new fluorescence method for the histochemical demonstration of biogenic monoamines. Acta Physiol Scand 87:57– 62. Bjorklund A, Lindvall O, Nobin A (1975) Evidence of an incerto-hypothalamic dopamine neurone system in the rat. Brain Res 89: 29 – 42. Bjorklund A, Nobin A (1973) Fluorescence histochemical and microspectrofluorometric mapping of dopamine and noradrenaline cell groups in the rat diencephalon. Brain Res 51:193–205. Bohn MC, Dreyfus CF, Friedman WJ, Markey KA (1987) Glucocorticoid effects on phenylethanolamine N-methyltransferase (PNMT) in explants of embryonic rat medulla oblongata. Brain Res 465:257–266. Bohn MC, Goldstein M, Black IB (1986) Expression and development of phenylethanolamine N-methyltransferase (PNMT) in rat brain stem: studies with glucocorticoids. Dev Biol 114:180 –193. Braak H, Braak E (1987) The hypothalamus of the human adult: chiasmatic region. Anat Embryol (Berl) 175:315–330. Ciliax BJ, Drash GW, Staley JK, Haber S, Mobley CJ, Miller GW, Mufson EJ, Mash DC, Levey AI (1999) Immunocytochemical local-

ization of the dopamine transporter in human brain. J Comp Neurol 409:38 –56. Cunningham ET Jr, Bohn MC, Sawchenko PE (1990) Organization of adrenergic inputs to the paraventricular and supraoptic nuclei of the hypothalamus in the rat. J Comp Neurol 292:651– 667. Dahlstrom A, Fuxe K (1964) Localization of monoamines in the lower brain stem. Experientia 20:398 –399. Dudas B, Merchenthaler I (2001) Catecholaminergic axons innervate LH-releasing hormone immunoreactive neurons of the human diencephalon. J Clin Endocrinol Metab 86:5620 –5626. Dudas B, Merchenthaler I (2006) Three-dimensional representation of the neurotransmitter systems of the human hypothalamus: inputs of the gonadotrophin hormone-releasing hormone neuronal system. J Neuroendocrinol 18:79 –95. Dudas B, Semeniken KR, Merchenthaler I (2006) Morphological substrate of the catecholaminergic input of the vasopressin neuronal system in humans. J Neuroendocrinol 18:895–901. Ericson H, Blomqvist A, Kohler C (1989) Brainstem afferents to the tuberomammillary nucleus in the rat brain with special reference to monoaminergic innervation. J Comp Neurol 281:169 –192. Felten DL (1976) Catecholamine neurons in the squirrel monkey hypothalamus. J Neural Transm 39:269 –280. Felten DL, Laties AM, Carpenter MB (1974) Monoamine-containing cell bodies in the squirrel monkey brain. Am J Anat 139:153–165. Felten DL, Sladek JR Jr (1983) Monoamine distribution in primate brain V. Monoaminergic nuclei: anatomy, pathways and local organization. Brain Res Bull 10:171–284. Foster GA, Hokfelt T, Coyle JT, Goldstein M (1985a) Immunohistochemical evidence for phenylethanolamine-N-methyltransferasepositive/tyrosine hydroxylase-negative neurones in the retina and the posterior hypothalamus of the rat. Brain Res 330:183–188. Foster GA, Schultzberg M, Goldstein M, Hokfelt T (1985b) Ontogeny of phenylethanolamine N-methyltransferase- and tyrosine hydroxylase-like immunoreactivity in presumptive adrenaline neurones of the foetal rat central nervous system. J Comp Neurol 236:348 – 381. Gallyas F, Gorcs T, Merchenthaler I (1982) High-grade intensification of the end-product of the diaminobenzidine reaction for peroxidase histochemistry. J Histochem Cytochem 30:183–184. Gallyas F, Merchenthaler I (1988) Copper-H2O2 oxidation strikingly improves silver intensification of the nickel-diaminobenzidine (NiDAB) end-product of the peroxidase reaction. J Histochem Cytochem 36:807– 810. Hokfelt T, Fuxe K, Goldstein M, Johansson O (1973) Evidence for adrenaline neurons in the rat brain. Acta Physiol Scand 89:286 – 288. Hokfelt T, Johansson O, Fuxe K, Goldstein M, Park D (1976) Immunohistochemical studies on the localization and distribution of monoamine neuron systems in the rat brain. I. Tyrosine hydroxylase in the mes- and diencephalon. Med Biol 54:427– 453. Howe PR, Costa M, Furness JB, Chalmers JP (1980) Simultaneous demonstration of phenylethanolamine N-methyltransferase immunofluorescent and catecholamine fluorescent nerve cell bodies in the rat medulla oblongata. Neuroscience 5:2229 –2238. Jansen AS, Schmidt ED, Voorn P, Tilders FJ (2003) Substance induced plasticity in noradrenergic innervation of the paraventricular hypothalamic nucleus. Eur J Neurosci 17:298 –306. Kitahama K, Denoroy L, Berod A, Jouvet M (1986) Distribution of PNMT-immunoreactive neurons in the cat medulla oblongata. Brain Res Bull 17:197–208. Kitahama K, Denoroy L, Goldstein M, Jouvet M, Pearson J (1988) Immunohistochemistry of tyrosine hydroxylase and phenylethanolamine N-methyltransferase in the human brain stem: description of adrenergic perikarya and characterization of longitudinal catecholaminergic pathways. Neuroscience 25:97–111. Kitahama K, Luppi PH, Berod A, Goldstein M, Jouvet M (1987) Localization of tyrosine hydroxylase-immunoreactive neurons in the cat

B. Dudas et al. / Neuroscience 171 (2010) 187–195 hypothalamus, with special reference to fluorescence histochemistry. J Comp Neurol 262:578 –593. Kitahama K, Pearson J, Denoroy L, Kopp N, Ulrich J, Maeda T, Jouvet M (1985) Adrenergic neurons in human brain demonstrated by immunohistochemistry with antibodies to phenylethanolamine-Nmethyltransferase (PNMT): discovery of a new group in the nucleus tractus solitarius. Neurosci Lett 53:303–308. Kopp N, Denoroy L, Renaud B, Pujol JF, Tabib A, Tommasi M (1979) Distribution of adrenaline-synthesizing enzyme activity in the human brain. J Neurol Sci 41:397– 409. Leshin LS, Kraeling RR, Kineman RD, Barb CR, Rampacek GB (1996) Immunocytochemical distribution of catecholamine-synthesizing neurons in the hypothalamus and pituitary gland of pigs: tyrosine hydroxylase and dopamine-beta-hydroxylase. J Comp Neurol 364:151–168. Lew JY, Matsumoto Y, Pearson J, Goldstein M, Hokfelt T, Fuxe K (1977) Localization and characterization of phenylethanolamine N-methyl transferase in the brain of various mammalian species. Brain Res 119:199 –210. Li YW, Halliday GM, Joh TH, Geffen LB, Blessing WW (1988) Tyrosine hydroxylase-containing neurons in the supraoptic and paraventricular nuclei of the adult human. Brain Res 461:75– 86. Liposits Z, Gorcs T, Torok A, Domany S, Setalo G (1983) Simultaneous localization of two different tissue antigens based on the silver intensified PAP-DAB and on the traditional PAP-DAB method. Acta Morphol Hung 31:365–369. Mefford IN (1988) Epinephrine in mammalian brain. Prog Neuropsychopharmacol Biol Psychiatry 12:365–388. Mezey E (1989) Phenylethanolamine N-methyltransferase-containing neurons in the limbic system of the young rat. Proc Natl Acad Sci U S A 86:347–351. Moreno ML, Villanua MA, Esquifino AI (1992) Serum prolactin and luteinizing hormone levels and the activities of hypothalamic monoamine oxidase A and B and phenylethanolamine-N-methyl transferase are changed during sexual maturation in male rats treated neonatally with melatonin. J Pineal Res 13:167–173. Nagatsu T, Kato T, Numata Y, Ikuta K, Sano M (1977) Phenylethanolamine N-methyltransferase and other enzymes of catecholamine metabolism in human brain. Clin Chim Acta 75:221–232. Nobin A, Bjorklund A (1973) Topography of the monoamine neuron systems in the human brain as revealed in fetuses. Acta Physiol Scand Suppl 388:1– 40. Olson L, Boreus LO, Seiger A (1973a) Histochemical demonstration and mapping of 5-hydroxytryptamine- and catecholamine-contain-

195

ing neuron systems in the human fetal brain. Z Anat Entwicklungsgesch 139:259 –282. Olson L, Nystrom B, Seiger A (1973b) Monoamine fluorescence histochemistry of human post mortem brain. Brain Res 63:231–247. Palkovits M, Mezey E, Skirboll LR, Hokfelt T (1992) Adrenergic projections from the lower brainstem to the hypothalamic paraventricular nucleus, the lateral hypothalamic area and the central nucleus of the amygdala in rats. J Chem Neuroanat 5:407– 415. Panayotacopoulou MT, Guntern R, Bouras C, Issidorides MR, Constantinidis J (1991) Tyrosine hydroxylase-immunoreactive neurons in paraventricular and supraoptic nuclei of the human brain demonstrated by a method adapted to prolonged formalin fixation. J Neurosci Methods 39:39 – 44. Panayotacopoulou MT, Malidelis Y, van Heerikhuize J, Unmehopa U, Swaab D (2005) Individual differences in the expression of tyrosine hydroxylase mRNA in neurosecretory neurons of the human paraventricular and supraoptic nuclei: positive correlation with vasopressin mRNA. Neuroendocrinology 81:329 –338. Ruggiero DA, Ross CA, Anwar M, Park DH, Joh TH, Reis DJ (1985) Distribution of neurons containing phenylethanolamine N-methyltransferase in medulla and hypothalamus of rat. J Comp Neurol 239:127–154. Saper CB (1990) Hypothalamus. In: The human nervous system, (Paxinos G, ed), pp 389 – 413 San Diego, CA: Academic Press. Sawchenko PE, Bohn MC (1989) Glucocorticoid receptor-immunoreactivity in C1, C2, and C3 adrenergic neurons that project to the hypothalamus or to the spinal cord in the rat. J Comp Neurol 285:107–116. Semeniken K, Merchenthaler I, Hu W, Dudas B (2009) Catecholaminergic input to the oxytocin neurosecretory system in the human hypothalamus. J Chem Neuroanat 37:229 –233. Swanson LW, Hartman BK (1975) The central adrenergic system. An immunofluorescence study of the location of cell bodies and their efferent connections in the rat utilizing dopamine-beta-hydroxylase as a marker. J Comp Neurol 163:467–505. Ungerstedt U (1971) Stereotaxic mapping of the monoamine pathways in the rat brain. Acta Physiol Scand Suppl 367:1– 48. Zoli M, Agnati LF, Tinner B, Steinbusch HW, Fuxe K (1993) Distribution of dopamine-immunoreactive neurons and their relationships to transmitter and hypothalamic hormone-immunoreactive neuronal systems in the rat mediobasal hypothalamus. A morphometric and microdensitometric analysis. J Chem Neuroanat 6:293–310.

(Accepted 21 August 2010) (Available online 27 August 2010)