Brain Research Bulletin,Vol. 29, pp. 479-491, Printed in the USA. All rights reserved.
1992 Copyright
0361-9230/92 $5.00 + .OO 0 1992 Pergamon Press Ltd.
Ultrastructure of Cholinergic Neurons in the Laterodorsal Tegmental Nucleus of the Rat: Interaction With Catecholamine Fibers YOSHIYUKI
KUBOTA,’
ESTHER
LEUNG AND STEVEN R. VINCENT2
Kinsmen Laboratory qfNeurologica1 Research, Department of Psychiatry, University of British Columbia, Vancouver, B.C., Canada V6T 123
Received 27 January 1992; Accepted 9 April 1992 KUBOTA, Y., E. LEUNG AND S. R. VINCENT. Uiltrastruclure of cholinergic neurons in the laterodorsal tegmental nucleus oj BRAIN RES BULL 29(3/4) 479-49 I, 1992.-The ultrastructure of choline acetyltransferase (ChAT)-immunoreactive neurons in the laterodorsal tegmental nucleus (TLD) of the rat was investigated by immunohistochemical techniques. The immunoreactive neurons were medium to large in size, with a few elongated dendrites, contained well-developed cytoplasm, and a nucleus with deep infoldings. They received many nonimmunoreactive, mostly asymmetric synaptic inputs on their soma and dendrites. ChAT-immunoreactive, usually myelinated, axons were occasionally seen in TLD. Only one immunoreactive axon terminal was observed within TLD, and it made synaptic contact with a nonimmunoreactive neuronal perikaryon. The synaptic interactions between ChAT-immunoreactive neurons and tyrosine hydroxylase (TH)-immunoreactive fibers in the TLD were investigated with a double immunohistochemical staining method. ChAT-immunoreactivity detected with a P-galactosidase method was light blue-green in the light microscope and formed dot-like electron dense particles at the electron microscopic level. TH-immunoreactivity, visualized with a nickel-enhanced immunoperoxidase method, was dark blue-black in the light microscope and diffusely opaque in the electron microscope. Therefore, the difference between these two kinds of immunoreactivity could be quite easily distinguished at both light and electron microscopic levels. In the light microscope, TH-positive fibers were often closely apposed to ChAT-immunoreactive cell bodies and dendrites in TLD. In the electron microscope, the cell soma and proximal dendrites of ChAT-immunoreactive neurons received synaptic contacts from TH-immunoreactive axon terminals. These results provide a morphological basis for catecholaminergic regulation of the cholinergic reticular system. the rat Interaction wifh calecholaminejibers.
Cholinergic neurons nucleus
Catecholamines
Synaptic interactions
Immunocytochemistry
Laterodorsal tegmental
The cholinergic neurons of the mesopontine tegmentum are thought to interact with brain stem aminergic systems to regulate behavioural state (16,25,35). Thus, the cholinergic neurons in TLD have been the focus of numerous pharmacological, physic logical, biochemical, and anatomical studies (8, I 1,16- 19,22,293 1,34,40,41,43,44). However, their ultrastructural morphology has not yet been described. In this report we present the ultrastructure of the cholinelgic neurons in the TLD of the rat using immunoh&chemical methods. In addition, evidence is provided for a direct innervation of these cholinergic neurons by catecholamine fibers.
A large group of cholinergic neurons is found in the mesopontine tegmentum in the laterodorsal tegmental nucleus (TLD) and pedunculopontine tegmental nucleus (PPT) (2,19,29). These neurons project to many regions, including the medial frontal cortex, basal forebrain, thalamus, tectum, and pontine reticular formation (7, IO,3 I ,44). Afferents to the TLD arise from the medial prefrontal cortex, diagonal band, thalamus, hypothalamus and reticular formation (3,7,14,30). Although the neurochemical nature of these inputs has not yet been analyzed, at the light microscopic level, fibers containing substance P, neurotensin, somatostatin, enkephalin, neuropeptide Y, corticotropin-releasing factor, galanin, cholecystokinin, dynorphin, vasoactive intestinal polypeptide, calcitonin gene-related peptide, atrial natriuretic peptide, GABA, acetylcholine, and catecholamines have been identified in the TLD (37). The choline& neurons of the mesopontine tegmentum are thought to interact with brain stem aminergic systems to regulate
METHOD Tissue Preparation
Six male Sprague-Dawley rats (150-200 g) rats were anesthetized with pentobarbital through the heart with saline, followed by 300 formaldehyde, 15% saturated picric acid and
’ Present address: Neural Systems Laboratory, Frontier Research Program, RIKEN, Wako, Saitama 35 l-01, Japan. ’ Requests for reprints should be addressed to Dr. Steven R. Vincent.
479
were used. The and perfused ml of 4% para0.05% glutar-
480
KlJBOTA.
LEUNG
AND
V1NCEN-I
FIG. I. Light micrograph ofthe ChAT-immunoreactive neurons in the laterodorsal tegmental nucleus (TLD) detected with the immunoperoxldase method. This coronal. vibratome section has ken treated with osmium and embedded in plastic. 4~: 4th ventricle.. Scale har: 100 urn.
aldehyde in 0.1 J1 sodium phosphate bufher. pH 7.4. The brains were removed and postfixed in the same fixative for 3 hours at 4°C. Coronal sections containing the TLD were cut with a vibratome at 50 pm thickness. The tissue sections were put in glass tubes in 15%’sucrose in 0. I M phosphate buffer for I hour and then frozen with liquid nitrogen and thawed at room temperature.
To examine the ultrastructure of the cholinergic neurons in the TLD, an indirect immunoperoxidase method was used. The sections were washed with 0. I M phosphate-buffered saline (PBS) several times and incubated with 10% normal goat serum for I hour at room temperature. They were then incubated for 2 days at 4°C in monoclonal rat antiChAT IgG (Boehringer Mannheim), diluted to I pg/ml in PBS. After washing 3 X 20 min with PBS, biotinylated goat anti-rat IgG (1:200: Vector Laboratories) was used as secondary antibody for 6 hours at room temperature. The sections were then washed 3 X 20 min in PBS and incubated in the ABC complex (avidin-DH and biotinylatedhorseradish peroxidase complex: 1:200: Vector) overnight at 4°C. After several washings with PBS, the sections were rinsed with 0.5 M Tris HCI buffer (pH 7.6) for 5 min and processed with the glucose oxidase-diaminobenzidine-nickel method (33). in 100 ml of 0.1 hf acetate buffer (pH 6.0). containing 2.5 g of nickel ammonium sulphate, 200 mg of /3-D-glucose, 40 mg of ammonium chloride, I mg of glucose oxidase, and 50 mg of 3,3’-diaminobenzidine tetrachloride. The reaction was stopped by rinsing in acetate buffer. A double immunohistochemical method (6,20,27) was used to detect both cholinergic neurons and catecholamine libers at the light and electron microsopic levels. Sections were first incubated in the rat anti-ChAT IgG. as described above. Then. after 3 X 20 min washing with PBS, they were incubated with biotinylated goat anti-rat IgG (Vector, I :200) for 3 hours at room temperature, then washed 3 X 20 min and incubated in avidinDH (Vector. 1:200) for I hour. After washing 3 X 20 min. the
sections wcrc incubated with biotinylated &galactosidase (Vcctor. I: 100) for 3 hours at room temperature. After washing 3 X 20 min. the sections were incubated for the &galactosidase reaction overnight at 37°C. The incubation solution consisted of 25 ~1 of solution A and 2.276 ml of solution B. Solution A contained IO mg of 5-bromo-4-chloro-3-indolyl-@-D-galactoside (Sigma. St. Louis, MO) in 0.5 ml of dimethyl formamide. and solution B contained I. I m.l/ MgC12 in 7 ml of0.02 ,U phosphate buffered saline and I65 mg of potassium ferricyanide and 2 I 1 mg of potassium ferrocyanide in 20 ml distilled water. The sections were finally washed with PBS several times to stop the reaction. To detect TH immunoreactivity. a peroxidase-antipcroxidase (PAP) method was used. Sections were incubated in rabbit anti-TH serum (Eugene Tech Intl. 1:50) overnight at 4°C and then washed 3 X 20 min. Goat anti-rabbit IgG (Zymed Lab. I :5.000) was used as secondary serum for 6 hours at room temperature. After washing with PBS 3 X 20 min. the sections were incubated with rabbit PAP complex (DAKO. 1:5,000) overnight. After washing with PBS, the sections were rinsed with 0.05 :\I Tris HCI buffer (pH 7.6) for 5 min and processed with the glucose oxidase-diaminobenzidine-nickel method described above.
In the preparations for electron microscopy. the sections were treated with I’i;’ osmic acid in 0.1 111phosphate buffer for I hour at room temperature. Subsequently. the sections were dehydrated with graded ethanol and stained with I% uranyl acetate in 70”% ethanol. then embedded with Epon between glass slides and coverslips coated with silicone. After observation and photography under the light microscope. blocks of TLD containing the ChAT immunoreactive neuronal profiles were dissected and cut into ultrathin sections. The ultrathin sections were collected on one-hole slot grids coated with formvar and stained with lead citrate. A Philips 201 electron microscope was used.
CHOLINERGIC-CATECHOLAMINE
481
INTERACTIONS
FIG. 2. lmmunoperoxidase staining of cholinergic neurons in TLD. The ChAT-immunoreactive neuron between the asterisks in part A was sectioned and observed in the electron microscope (part B). The asterisks indicate corresponding capillaries in parts A and B. In part B, arrowheads indicate the infolding of the nucleus in the cholinergic neuron. Part C shows a higher magnification of the square in part B, illustrating an asymmetric synaptic contact onto the cholinergic cell body from a nonimmunoreactive axon terminal containing many small vesicles (open arrow). Scale bars: A = 10 Frn, B = 5 pm. C = 0.5 Wm.
Immltnohisto~hemicul
Controls
Control incubations were conducted in which normal rat or rabbit sera were substituted for the ChAT or TH primary antisera, respectively. With other sections, the primary antiserum was left out. In addition, the crossreactivity of the secondary antibodies was examined. When the biotinylated goat anti-rat IgG was used secondary to the rabbit anti-TH serum, no staining was detected with the immunoperoxidase method. Likewise, when the biotinylated goat anti-rabbit serum was used with the rat anti-ChAT IgG, no reaction was observed with the P-galactosidase technique. Therefore, the P-galactosidase and glucose-oxidase DAB-nickel methods could be used together to detect ChAT- and TH-immunoreactivities, respectively. RESULTS The ChAT-positive neurons of the TLD are located between the locus ceruleus (LC) and the dorsal tegmental nucleus (Gudden) within the central gray. Rostrally they con-
tinue into the PPT in the lateral tegmentum. At the most rostra1 level of LC, the ChAT-positive neurons of the TLD extend ventrolaterally across the medial longitudinal fasciculus through the central tegmental tract (Fig. 1). ChAT-immunoreactivity was present in most of the neurons in the TLD. Their cell bodies were medium to large in size, showed oval or fusiform shapes and had a few major processes. Extensive networks of ChAT-immunoreactive dendrites were also present throughout the TLD. Individual immunoreactive dendrites were well stained and could be followed for over 150 Km. Neurons closely apposed to capillaries were frequently observed. In the sections stained with the immunoperoxidase method, the electron-dense reaction product was diffusely distributed throughout the ChAT-immunoreactive neuronal cytoplasm (Figs. 2, 3). The ChAT-immunoreactive neurons had large nuclei which showed infoldings. Indeed, in some ultrathin sections these invaginations were very deep, almost dividing the nucleus into two parts. These neurons had large amounts of cytoplasm which contained the usual complement
482
KUBOTA.
LEUNG
AND VINCENT
FIG. 3. Immunoperoxidase staining of a ChAT-immunoreactive neuron in TLD. (A) The nucleus contains a large nucleolus and lies at one pole of the cell. Arrowheads indicate infoldings of the nucleus. The abundant cytoplasm contains the common subcellular organelles. (B and C) High magnification electron micrographs of the lower (part B) and upper squares (part C) in part A. illustrating two asymmetric. nonimmunoreactive synaptic inputs onto the cell soma (open arrows). Both axon terminals contain many small vesicles. and in part B a few large cored vesicles are also present. Scale bars: A = I pm. B. C y 0.S pm.
CHOLINERGIC-CATECHOLAMINE
483
INTERACTIONS
FIG. 4. lmmunoperoxidase staining of ChAT-immunoreactive dendrites in TLD in longitudinal (part A) and crosssection (part B). The immunoreactivity is associated with microtubules. In part A, three asymmetric synaptic contacts from nonimmunoreactive axons are seen (open arrows). These nonimmunoreactive terminals have many small, round vesicles and some large, cored vesicles and wide, well-developed membrane specializations. A small ChAT-immunoreactive profile, which probably corresponds to a spine, receives synaptic contact from a nonimmunoreactive axon (arrow). In part B, a dense plaque resembling a subsynaptic bar beneath the membrane specialization is present (black arrow). This dendrite also receives a synapse from a large terminal containing pleomorphic vescicles. Scale bars: 0.5 pm.
of subcellular organelles, i.e., mitochondria, endoplasmic reticulum, Golgi apparatus, microtubules. They also contained cored vesicles. The ChAT-immunoreactive cell bodies received synaptic contacts from nonimmunoreactive axon terminals (Figs. 2, 3). In some instances, one to four synaptic contacts were observed on a perikaryon in one ultrathin section, suggesting that many axo-somatic synaptic contacts with nonimmunoreactive axon
terminals occur. These terminals contained many round or oval small vesicles and sometimes large cored vesicles. The synaptic membrane specialization was well developed, wide, and usually asymmetric (Figs. 2C, 3B, C). ChAT-immunoreactive dendrites ranging in size from 0.1 to 1.5 pm in diameter (Figs. 4, 7A) were present throughout the TLD. The electron-dense immunoreactivity was associated primarily with microtubules. These dendrites received
484
KUBOTA.
LEUNG
AND
VINCENT
FIG. 5. Immunoperoxidase staining of a ChAT-immunoreactive dendrite showing a close apposition with a pericyte (P) adjacent to a capillary (C). Arrowhead in part B shows a synaptic contact on the ChAT-immunoreactive dendrite. Scale bars: A = I pm, B = 0.5 pm.
many minals served profile
axon tersynaptic inputs from nonimmunoreactive ob). Two to four synaptic contacts were sometimes dendritic close tog ,ether on a ChAT-immunoreactive many round (Fig. 4). These axons usually contained
or oval small vesicles. and the membrane specialization was usually asymmetric and often occupied a wide area. In some instances, a dense plaque resembling a subsynaptic bar was present beneath the membrane specialization (Fig. 4A. B). A
CHOLINERGIC-CATECHOLAMINE
INTERACTIONS
FIG. 6. lmmunoperoxidase staining of myelinated ChAT-immunoreactive axons in TLD (black asterisks). Nonimmunoreactive myelinated axons are also observed (open asterisks). A large, ChAT-immunoreactive dendrite receiving synaptic input from a nonimmunoreactive terminal is also shown in part A (open arrow). Scale bars: A = 0.5 Km, B = 0.5 Wm.
485
486
KIJBO’IA.
LEUNG
AND
VINCENT
FIG. 7. Immunoperoxidase staining of a ChAT-immunoreactive axon making synaptic contact with a nonimmunoreactive cell in TLD. Parts B and Care successive sections from the square in part A. The ChAT-immunoreactive axon shows a synaptic membrane specialization in part B (arrow) and an aggregation of small vesicles at the corresponding site in part C (arrow). Scale bars: A = I pm, B and C 7 0.5 urn.
few symmetric contacts onto cholinergic neurons from terminals containing pleomorphic vesicles were also observed (Fig. 4B). Small, ChAT-immunoreactive spines were also observed in contact with nonimmunoreactive axon terminals (Fig. 4A). ChAT-immunoreactive processes were also observed closely associated with pericytes surrounding capillaries (Fig. 5A). Indeed, direct contacts between ChAT-immunoreactive dendrites and pericytes could occasionally be seen (Fig. 5B).
ChAT-immunoreactive myelinated axons were frequently seen containing immunoreactive. electron-dense material associated with neurofilaments and microtubules (Fig. 6). The diameters of the ChAT-immunoreactive axons ranged from 0.5 to 1.O pm. Only one case of a ChAT-immunoreactive axon terminal was seen. and it made synaptic contact with a nonimmunoreactive perikaryon (Fig. 7). This axon terminal had many small round vesicles, a few large, cored vesicles and an asymmetric synaptic membrane specialization. The active zone was
CHOLINERGIC-CATECHOLAMINE
487
INTERACTIONS
FIG. 8. Double immunohistochemistry in TLD showing a ChAT-immunoreactive neuron (open arrow), which was stained light blue-green with the P-galactosidase method in the light microscope. Many TH-immunoreactive terminals stained dark blue-black with the immunoperoxidase method are seen (arrows), some of which show close appositions with the ChATimmunoreactive neuron (arrow heads). Scale bar: IO Frn.
short, about 170 nm, and the innervated neuron was small (about I5 pm diameter). The /3-galactosidase immunohistochemical technique was less sensitive than the peroxidase method. The immunoreactivity was only observed within 5 to 10 pm from the surface of the vibratome sections. ChAT-immunoreactive, light blue-green staining was seen in the neuronal perikarya and dendrites in TLD. In contrast, TH-immunoreactive structures detected with the glucose-oxidase DAB-nickel reaction showed almost black immunoreactive material. Thus, these two different kinds of immunoreactivity could be easily distinguished in the light microscope. Great numbers of TH-immunoreactive neuronal perikarya were seen in the LC and the lateral part of TLD, and TH-immunoreactive fibers extended from the LC into the TLD. Moderate numbers ofTH-immunoreactive fibers were seen throughout the TLD (Fig. 8). The thin, TH-immunoreactive fibers sometimes showed varicose swellings. About one-third of the ChAT-immunoreactive neurons had close appositions with THimmunoreactive fibers on their perikarya and/or proximal dendrites. When the fl-galactosidase method was used to detect the choline& neurons, the immunoreaction product formed small, dot-like, electron-dense material and was distributed throughout the labelled neuronal profiles (Fig. 9). At high magnification, the reaction product showed moderate electron density. In contrast, the TH-immunoreactive material detected with the immunoperoxidase method was diffusely distributed throughout the labelled neuronal profiles and the electron density was much greater than that of the /3-galactosidase reaction (Fig. 9). Therefore, differentiation of these two kinds of immunoreaction on the same ultrathin section in the electron microscope was quite easy and reliable.
Many TH-immunoreactive fibers were present in the TLD, and occasionally, TH-immunoreactive axon terminals made synaptic contacts with nonimmunoreactive neuronal profiles. The membrane specialization was symmetrical and the synapse was very narrow and small (SO-100 nm). Seven ChAT-immunoreactive neurons in TLD showing close apposition with THimmunoreactive profiles in the light microscope were subsequently examined in the electron microscope. All of them showed close appositions with TH-immunoreactive axons in the electron microscope, and some of them received specific synaptic contacts from TH-immunoreactive axon terminals (Fig. 10). Usually one or two TH-immunoreactive terminals were seen in contact with a ChAT-immunoreactive neuron in a single thin section. In one case, eight TH-immunoreactive terminals were observed in apposition with one ChAT-immunoreactive cell body and its proximal dendrite (Figs. 9. IO). DISCUSSION
The ultrastructure of ChAT-immunoreactive neurons in the TLD was observed in the present study. The distribution of cholinergic neurons in the mesopontine tegmentum corresponded with previous reports (2,19,40). Morphologically, the ChAT-immunoreactive neurons in TLD appeared to form a homogeneous group. Most of the neurons in TLD showed strong ChAT-immunoreactivity, and a dense network of ChAT-immunoreactive dendrites was also present. The ChAT-positive cell bodies were medium to large in size and had two to four elongated major dendrites with extensive arborizations. The ChAT-immunoreactive neurons had indented nuclei and well-developed cytoplasm containing the ordinary subcellular organs. Most of the cholinergic neurons received synaptic contacts from nonimmunoreactive axon
488
KUBOTA.
FIG. deep close (large bars:
9. Ultrastructure
LEUNG
AND VINCENT
of the ChAT-immunoreactive neuron in Fig. 8. showing at low magnihcation the nucleus with a very infolding (arrow heads) and a single nucleolus. Seven TH-immunoreactive axon terminals (large arrows) are seen in apposition to the cell soma. Part B presents a higher magnitication of five of the TH-immunoreactive axon terminals arrows). The small arrows indicate the &galactosidasc reaction product in the ChAT-immunoreactivc neuron. Scale A = 2 pm. B = 0.5 pm.
CHOLINERGIC-CATECHOLAMINE
INTERACTIONS
FIG. IO. Serial thin sections of the TH-immunoreactive terminal labelled (arrow a) in Fig. 9B. Parts A and B are three sections apart and part C is adjacent to part B. The terminal forms a small synapse with the ChAT-immunoreactive cell soma. The arrows indicate /3-galactosidase reaction product in the ChATimmunoreactive neuron. Scale bar: 0.5 pm.
489
KIJBOTA.
490
terminals on their cell bodies and proximal dendrites. The overall morphology of the ChAT-immunoreactive neurons in the TLD was. thus. very similar to that of the cholinergic neurons in PPT described in previous reports (23.36). ChAT-immunoreactive axon fibers were occasionally seen in TLD, and these were primarily myelinated. This is consistent with the rather fast conduction velocities seen previously in rat and cat TLD t&17,34), although another report found a somewhat slower conduction speed (I I). Only one ChAT-immunoreactive axon terminal was observed, and it made synaptic contact with a nonimmunoreactive neuronal cell body. The axon terminal appeared similar to the ChAT-immunoreactive axon terminals observed previously in various regions innervated by the mesopontine cholinergic neurons (4,5,13,23,24). The paucity of ChAT-immunoreactive axon terminals in TLD observed in this study is consistent with the ChAT-immunoreactive neurons being projection neurons and not local circuit neurons. Previous ultrastructural studies on the basal forebrain cholinergic neurons have also noted an absence of labelled terminals ( I ,26). ChAT-immunoreactive cell bodies and dendrites sometimes showed close appositions with pericytes surrounding blood vessels. ChAT-immunoreactive neuronal structures in the basal forebrain also appear to engage in such contacts (I). These observations are consistent with biochemical, pharmacological. and physiological work demonstrating cholinergic regulation of central blood vessels (9). Together, these observations suggest that acetylcholine has an important role in the modulation of blood flow in the central nervous system. The mesopontine cholinergic neurons have recently been shown to contain high levels of nitric oxide synthase (I 5.39.42). Thus, nitric oxide produced in these cells may also act on these closely apposed blood vessels. The /Sgalactosidase immunohistochemical method was introduced by Bondi et al. (6). and an electron microscope method was also described for double immunohistochemical staining in combination with immunoperoxidase methods (20,27). The immunoreaction product of the &galactosidase stain is easy to distinguish from that of peroxidase reactions by the color difference in the light microscope and the different deposits seen in the electron microscope. Therefore, this double immunohistochemical staining method is especially useful for the study of synaptic interactions on neuronal perikarya and proximal dendrites. It is quite a reliable and simple technique. although the sensitivity of the /Sgalactosidase immunohistochemical method is lower than that of the peroxidase reaction, and synaptic interactions at distal dendrites are difficult to discern with this procedure. The distribution and ultrastructural features of the TH-positive structures in TLD coincided well with previous studies of catecholamine fibers. At the light microscopic level, TH-immunoreactive fibers have been noted within the TLD (37).
LEUNG
AND
VINCENT
sometimes showing swellings close to cholinergic neurons (30). Ultrastructurally, the TH-immunoreactive fibers in TLD rarely showed synaptic membrane specializations. Generally the synaptic active zones of TH-immunoreactive axons are quite narrow and small throughout the central nervous system (2 I). Although TH-immunoreactive axons were observed in serial. thin sections. only a few TH-immunoreactive axons showing synaptic active zones could be identified. The major finding in the present study was that ChAT-immunoreactive neurons in the TLD receive synaptic input from TH-immunoreactive axons. What might be the origin of these TH-positive fibers innervating the TLD? The TLD is known to receive input from the substantia nigra (3.7,30). However. this input arises principally from nondopaminergic neurons in the pars reticulata (30). The TH-positive neurons of the locus ceruleus have been shown to innervate the TLD (30). Many axons and dendrites from locus ceruleus extend medially into the adjacent TLD (32). Positive fibers have also been described in TLD in immunohistochemical studies with antisera against dopamine-@-hydroxylase ( 16). These fibers appear to be derived from the noradrenergic neurons in locus ceruleus (I 6). Jones (16) also suggested the possible interaction between cholinergic neurons and noradrenergic neuron in TLD by the observation that varicose DBHpositive fibers are found close to ChAT-positive TLD cells. Autoradiographic binding studies have localized (Yeadrenergic receptors in TLD (38). and noradrenaline h,yperpolarizes identified cholinergic TLD neurons in vitro (Wtlhams and Reiner. personal communication). Together these observations are consistent with a direct input of TLD cholinergic cells by the noradrenergic locus ceruleus. In addition, phenylethanolamine-Nmethyl-transferase-immunoreactive fibers originating in the Cl adrenaline cell group ofthe nucleus paragigantocellularis lateralis have been reported in TLD (12.28). It is not yet clear whether the TH-immunoreactive fibers observed in the present study represent dopaminergic-. noradrenergic-. or adrenaline-containing fibers. However, there is physiological evidence consistent with reciprocal interactions between brain stem cholinergic and noradrenergic neurons. Together, these two brain stem systems are thought to play a major role in the regulation of the sleep-wake cycle (35). The TH-immunoreactive terminals observed in the present study may. therefore, represent the morphological basis for the direct regulation of the cholinergic neurons of the TLD by noradrenergic fibers. ,ACKNOWLEDGEMENTS
The authors thank Drs. P. B. Reiner. K. Shinado. and J. Kohno fat fruitful discussions, and Ms. Sandra Sturgeon for secretarial assistance. This work was supported by the Medical Research Council of Canada. S.R.V. is an MRC Scientist.
REFERENCES Armstrong. D. M. Ultrastructural characterization of choline acetyltransferase-containing neurons in the basal forebrain of rat: Evidence for a cholinergic
innervation of intracerebral 1986.
blood vessels.
J. Comp. Neural. 250:81-92:
Armstrong, D. M.; Saper. C. B.; Levey, A. L.: Wainer, B. H.: Terry, R. D. Distribution of cholinergic neurons in the rat brain: Demonstrated by the immunocytochemical localization of choline acetyltransferase. J. Comp. Neurol. 216:53-68: 1983. Beckstead. R. M.: Domesick. V. B.; Nauta, W. J. H. Efferent connections of the substantia nigra and ventral tegmental area in the rat. Brain Res. 175:191-217: 1979.
Beninato. M.; Spencer. R. F. The cholinergic innervation of the rat substantia nigra: A light and electron microscopic immunohistochemical study. Exp. Brain Res. 72: 178-184: 1988. Bolam. J. P.; Francis. C. M.: Henderson, Z. Cholinergic input to dopaminergic neurons in the substatia nigra: A double immunocytochemical study. Neuroscience 4 1:483-494: 199 I. Bondi. A.: Chieregatti, G.; Eusebi, V.; Fulcheri, E.: Bussolati, G. The use of@-galactosidase as a tracer in immunocytochemistry. Histochemistry 76: 15%158: 1982. 7. Cornwall, J.: Cooper. J. D.: Phillipson. 0. T. Afferent and efferent connections of the laterodorsal tegmental nucleus in the rat. Brain Res. Bull. 25:271-384: 1990.
CHOLINERCIC-CATECHOLAMINE
491
INTERACTIONS
8. El Mansari, 9.
10.
Il.
12.
13.
14.
15.
16. 17.
18.
19. 20.
21.
22.
23.
24.
25. 26.
27.
M.; Sakai, K.; Jouvet, M. Unitary characteristics of presumptive cholinergic tegmental neurons during the sleep waking cycle in freely moving cats. Exp. Brain Res. 76:5 19-529; 1989. Estrada, C.; Hamel, E.; Krause, D. N. Biochemical evidence for cholinergic innervation of intracerebral blood vessels. Brain Res. 266:261-270; 1983. Gould, E.; Woolf, N. J.; Butcher, L. L. Cholinergic projections to the substantia nigra from the pedunculopontine and laterodorsal tegmental nuclei. Neuroscience 28:61 l-623: 1989. Grant, S. J.; Highfield, D. A. Extracellular characteristics of putative cholinergic neurons in the rat laterodorsal tegmental nucleus. Brain Res. 559:64-74; 199 I. Guyenet, P. G.; Young, B. S. Projections of nucleus paragigantocellularis lateralis to locus coeruleus and other structures in rat. Brain Res. 406:171-184; 1987. Hallanger, A. E.; Wainer, B. H. Ultrastruture of ChAT-immunoreactive synaptic terminals in the thalamic reticular nucleus of the rat. J. Comp. Neurol. 278:486-497: 1988. Higo, S.; Ito, K.; Fuchs, D.; McCarley. R. W. Anatomical interconnections of the pedunculopontine tegmental nucleus and the nucleus prepositus hypoglossi in the cat. Brain Res. 536:79-85: 1990. Hope, B. T.: Michael, G. J.; Knigge, K. M.; Vincent, S. R. Neuronal NADPH diaphorase is a nitric oxide synthase. Proc. Natl. Acad. Sci. USA 88:281 I-2814: 1991. Jones, B. E. Paradoxical sleep and its chemical/structural substances in the brain. Neuroscience 40:637-656: I99 I. Kayama, Y.; Ogawa, T. Electrophysiology of ascending. possibly cholinergic neurons in the rat laterodorsal tegmental nucleus: Comparison with monoamine neurons. Neurosci. Lett. 77:277-282: 1987. Kayama, Y.; Takagi, M.; Ogawa. T. Cholinergic influence of the laterodorsal tegmental nucleus on neuronal activity in the rat lateral geniculate nucleus. J. Neurophysiol. 56: I297- 130 I; 1986. Kimura, H.: Maeda, T. Aminergic and cholinergic systems in the dorsolateral pontine tegmentum..Brain Res. Bull.~9:493-499; 1982. Kohno. J.: Shinoda. K.: Kawai. Y.: Ohuchi. T.: Ono. K.: Shiotani. Y. Interaction between adrenergic fibers and intermediate choline@ neurons in the rat spinal cord: A new double-immunostaining method for correlated light and electron microscopic observations. Neuroscience 25:113-121: 1988. Liu. X.-B.; Jones, E. G. The fine structure of serotonin and tyrosine hydroxylase immunoreactive terminals in the ventral posterior thalamic nucleus of cat and monkey. Exp. Brain Res. (in press). Luebke. J. I.; Greene, R. W.: Semba, K.; Kamondi, A.; McCarley, R. W.; Reiner, P. B. Serotonin hyperpolarizes cholinergic low threshold burst neurons in the rat laterodorsal temental nucleus in vitro. Proc. Natl. Acad. Sci. USA 89:743-747; 1992. Martinez-Murillo, R.; Villalba, R.; Montero-Caballero, M. 1.; Rodrigo, J. Cholinergic somata and terminals in the rat substantia nigra: An immunocytochemical study with optical and electron microscopic techniques. J. Comp. Neurol. 28 1:397-415; 1989 Martinez-Murillo, R.; Villalba, R. M.; Rodrigo, J. Electron microscopic localization of cholinergic terminals in the rat substantia nigra: An immunocytochemical study. Neurosci. Lett. 96: I2 I - 126; 1989. McCormick, D. A. Cholinergic and noradrenergic modulation of thalamocortical processing. Trends Neurosci. I2:2 15-22 I ; 1989. Palacios, G.; Garcia-Ladona, J.; Codina, M. Ultrastrutural study of cholinergic neurons in the medial septal nucleus and vertical limb of the diagonal band of Broca in the basal forebrain of the rat. J. Chem. Neuroanat. 4:205-22 I ; 199 1. Peng, Y.; Kohno, J.; Shinoda, K.; Kawai, Y.; Ono, K.; Jiang, W.; Shiotani. Y. Interactions between cholinergic neurons and substance
28.
29.
30.
31.
32.
33.
34.
35. 36.
31.
38.
39. 40.
41.
42.
43.
44.
P or calcitonin gene-related peptide terminals in the rat sacral intermediolateral nucleus: Double immunostaining at the light and electron microscopic levels. J. Comp. Neurol. 285: 1-8; 1989. Pieribone, V. A.; Aston-Jones, G. Adrenergic innervation of the rat nucleus locus coeruleus arises predominantly from the Cl adrenergic cell group in the rostra1 medulla. Neuroscience 4 I :525-542; I99 I. Reiner, P. B.: Vincent, S. R. Topographic relations of cholinergic and noradrenergic neurons in the feline pontomescencephalic tegmentum: An immunohistochemical study. Brain Res. Bull. 19:705714: 1987. Semba, K.; Fibiger, H. C. Afferent connections of the laterodorsal and the pedunculopontine tegmental nuclei in the rat: A retro- and anterograde transport and immunohistochemical study. J. Comp. Neural. (in press). Semba, K.: Reiner, P. B.; Fibiger. H. C. Single cholinergic mesopontine tegmental neurons project to both the pontine reticular formation and the thalamus in the rat. Neuroscience 38:643-654; 1990. Shimizu. N.: Ohnishi, S.: Satoh, K.; Tohyama, M. Cellular organization of locus coeruleus in the rat as studied by Golgi method. Arch. Histol. Jpn. 41:103-l 12; 1978. Shu, S.: Ju. G.: Fan. L. The glucose oxidase-DAB-nickel method in peroxidase histochemistry ofthe nervous system. Neurosci. Lett. 85: 169-171: 1988. Steriade. M.: Datta. S.; Oakson, G.: Dossi. R. C. Neuronal activities in brain-stem cholinergic nuclei related to tonic activation processes in the thalamocortical systems. J. Neurosci. 10:2541-2559; 1990. Steriade, M.; McCarley. R. W. Brainstem control of wakefulness and sleep. New York: Plenum: 1990. Sugimoto. T.: Mizukawa, K.; Hattori. T.; Konishi. A.; Kaneko, T.; Mizuno. N. Cholinergic neurons in the nucleus tegmenti pedunculopontinus pars compacta and the caudoputamen of the rat: A light and electron microscopic immunocytochemical study using a monoclonal antibody to choline acetyltransferase. Neurosci. Lett. 51:l 13-I 17: 1984. Sutin. E. L.: Jacobowitz, D. M. lmmunocytochemical localization of peptides and other neurochemicals in the rat laterodorsal tegmental nucleus and adjacent area. J. Comp. Neurol. 270:243-270; 1988. Unnerstall. J. R.: Kopajitic, T. A.; Kuhar, M. J. Distribution of a1r agonist binding sites in the rat and human central nervous system: Analysis of some functional, anatomic correlates of the pharmacologtc effects of clonidine and related adrenergic agents. Brain Res. Rev. 7:69-101; 1984. Vincent, S. R.; Hope, B. T. Neurons that say NO. Trends Neurosci. 15:108-l 13: 1992. Vincent. S. R.: Satoh, K.; Armstrong, D. M.: Panula, P.; Vale, W.; Fibiger. H. C. Neuropeptides and NADPH-diaphorase activity in the ascending cholinergic reticular system of the rat. Neuroscience 17:167-182: 1986. Vincent, S. R.: Satoh. K.; Armstrong, D. M.; Fibiger. H. C. Substance P in the ascending cholinergic reticular system. Nature 306:688691: 1983. Vincent. S. R.; Satoh, K.; Armstrong, D. M.: Fibiger, H. C. NADPHdiaphorase: A selective histochemical marker for the cholinergic neurons of the pontine reticular formation. Neurosci. Lett. 43:3236: 1983. Wilcox, K. S.; Grant, S. J.; Burkhart. B. A.: Christoph. G. R. In vitro electrophysiology of neurons in the lateral dorsal tegmental nucleus. Brain Res. Bull. 22:557-560: 1989. Woolf, N. J.: Butcher, L. L. Cholinergic systems in the rat brain. III. Projections from the pontomesencephalic tegmentum to the thalamus, tectum, basal ganglia, and basal forebrain. Brain Res. Bull. 16:603-637: 1986.
1992 Copyright
0361-9230/92 $5.00 + .OO 0 1992 Pergamon Press Ltd.
Ultrastructure of Cholinergic Neurons in the Laterodorsal Tegmental Nucleus of the Rat: Interaction With Catecholamine Fibers YOSHIYUKI
KUBOTA,’
ESTHER
LEUNG AND STEVEN R. VINCENT2
Kinsmen Laboratory qfNeurologica1 Research, Department of Psychiatry, University of British Columbia, Vancouver, B.C., Canada V6T 123
Received 27 January 1992; Accepted 9 April 1992 KUBOTA, Y., E. LEUNG AND S. R. VINCENT. Uiltrastruclure of cholinergic neurons in the laterodorsal tegmental nucleus oj BRAIN RES BULL 29(3/4) 479-49 I, 1992.-The ultrastructure of choline acetyltransferase (ChAT)-immunoreactive neurons in the laterodorsal tegmental nucleus (TLD) of the rat was investigated by immunohistochemical techniques. The immunoreactive neurons were medium to large in size, with a few elongated dendrites, contained well-developed cytoplasm, and a nucleus with deep infoldings. They received many nonimmunoreactive, mostly asymmetric synaptic inputs on their soma and dendrites. ChAT-immunoreactive, usually myelinated, axons were occasionally seen in TLD. Only one immunoreactive axon terminal was observed within TLD, and it made synaptic contact with a nonimmunoreactive neuronal perikaryon. The synaptic interactions between ChAT-immunoreactive neurons and tyrosine hydroxylase (TH)-immunoreactive fibers in the TLD were investigated with a double immunohistochemical staining method. ChAT-immunoreactivity detected with a P-galactosidase method was light blue-green in the light microscope and formed dot-like electron dense particles at the electron microscopic level. TH-immunoreactivity, visualized with a nickel-enhanced immunoperoxidase method, was dark blue-black in the light microscope and diffusely opaque in the electron microscope. Therefore, the difference between these two kinds of immunoreactivity could be quite easily distinguished at both light and electron microscopic levels. In the light microscope, TH-positive fibers were often closely apposed to ChAT-immunoreactive cell bodies and dendrites in TLD. In the electron microscope, the cell soma and proximal dendrites of ChAT-immunoreactive neurons received synaptic contacts from TH-immunoreactive axon terminals. These results provide a morphological basis for catecholaminergic regulation of the cholinergic reticular system. the rat Interaction wifh calecholaminejibers.
Cholinergic neurons nucleus
Catecholamines
Synaptic interactions
Immunocytochemistry
Laterodorsal tegmental
The cholinergic neurons of the mesopontine tegmentum are thought to interact with brain stem aminergic systems to regulate behavioural state (16,25,35). Thus, the cholinergic neurons in TLD have been the focus of numerous pharmacological, physic logical, biochemical, and anatomical studies (8, I 1,16- 19,22,293 1,34,40,41,43,44). However, their ultrastructural morphology has not yet been described. In this report we present the ultrastructure of the cholinelgic neurons in the TLD of the rat using immunoh&chemical methods. In addition, evidence is provided for a direct innervation of these cholinergic neurons by catecholamine fibers.
A large group of cholinergic neurons is found in the mesopontine tegmentum in the laterodorsal tegmental nucleus (TLD) and pedunculopontine tegmental nucleus (PPT) (2,19,29). These neurons project to many regions, including the medial frontal cortex, basal forebrain, thalamus, tectum, and pontine reticular formation (7, IO,3 I ,44). Afferents to the TLD arise from the medial prefrontal cortex, diagonal band, thalamus, hypothalamus and reticular formation (3,7,14,30). Although the neurochemical nature of these inputs has not yet been analyzed, at the light microscopic level, fibers containing substance P, neurotensin, somatostatin, enkephalin, neuropeptide Y, corticotropin-releasing factor, galanin, cholecystokinin, dynorphin, vasoactive intestinal polypeptide, calcitonin gene-related peptide, atrial natriuretic peptide, GABA, acetylcholine, and catecholamines have been identified in the TLD (37). The choline& neurons of the mesopontine tegmentum are thought to interact with brain stem aminergic systems to regulate
METHOD Tissue Preparation
Six male Sprague-Dawley rats (150-200 g) rats were anesthetized with pentobarbital through the heart with saline, followed by 300 formaldehyde, 15% saturated picric acid and
’ Present address: Neural Systems Laboratory, Frontier Research Program, RIKEN, Wako, Saitama 35 l-01, Japan. ’ Requests for reprints should be addressed to Dr. Steven R. Vincent.
479
were used. The and perfused ml of 4% para0.05% glutar-
480
KlJBOTA.
LEUNG
AND
V1NCEN-I
FIG. I. Light micrograph ofthe ChAT-immunoreactive neurons in the laterodorsal tegmental nucleus (TLD) detected with the immunoperoxldase method. This coronal. vibratome section has ken treated with osmium and embedded in plastic. 4~: 4th ventricle.. Scale har: 100 urn.
aldehyde in 0.1 J1 sodium phosphate bufher. pH 7.4. The brains were removed and postfixed in the same fixative for 3 hours at 4°C. Coronal sections containing the TLD were cut with a vibratome at 50 pm thickness. The tissue sections were put in glass tubes in 15%’sucrose in 0. I M phosphate buffer for I hour and then frozen with liquid nitrogen and thawed at room temperature.
To examine the ultrastructure of the cholinergic neurons in the TLD, an indirect immunoperoxidase method was used. The sections were washed with 0. I M phosphate-buffered saline (PBS) several times and incubated with 10% normal goat serum for I hour at room temperature. They were then incubated for 2 days at 4°C in monoclonal rat antiChAT IgG (Boehringer Mannheim), diluted to I pg/ml in PBS. After washing 3 X 20 min with PBS, biotinylated goat anti-rat IgG (1:200: Vector Laboratories) was used as secondary antibody for 6 hours at room temperature. The sections were then washed 3 X 20 min in PBS and incubated in the ABC complex (avidin-DH and biotinylatedhorseradish peroxidase complex: 1:200: Vector) overnight at 4°C. After several washings with PBS, the sections were rinsed with 0.5 M Tris HCI buffer (pH 7.6) for 5 min and processed with the glucose oxidase-diaminobenzidine-nickel method (33). in 100 ml of 0.1 hf acetate buffer (pH 6.0). containing 2.5 g of nickel ammonium sulphate, 200 mg of /3-D-glucose, 40 mg of ammonium chloride, I mg of glucose oxidase, and 50 mg of 3,3’-diaminobenzidine tetrachloride. The reaction was stopped by rinsing in acetate buffer. A double immunohistochemical method (6,20,27) was used to detect both cholinergic neurons and catecholamine libers at the light and electron microsopic levels. Sections were first incubated in the rat anti-ChAT IgG. as described above. Then. after 3 X 20 min washing with PBS, they were incubated with biotinylated goat anti-rat IgG (Vector, I :200) for 3 hours at room temperature, then washed 3 X 20 min and incubated in avidinDH (Vector. 1:200) for I hour. After washing 3 X 20 min. the
sections wcrc incubated with biotinylated &galactosidase (Vcctor. I: 100) for 3 hours at room temperature. After washing 3 X 20 min. the sections were incubated for the &galactosidase reaction overnight at 37°C. The incubation solution consisted of 25 ~1 of solution A and 2.276 ml of solution B. Solution A contained IO mg of 5-bromo-4-chloro-3-indolyl-@-D-galactoside (Sigma. St. Louis, MO) in 0.5 ml of dimethyl formamide. and solution B contained I. I m.l/ MgC12 in 7 ml of0.02 ,U phosphate buffered saline and I65 mg of potassium ferricyanide and 2 I 1 mg of potassium ferrocyanide in 20 ml distilled water. The sections were finally washed with PBS several times to stop the reaction. To detect TH immunoreactivity. a peroxidase-antipcroxidase (PAP) method was used. Sections were incubated in rabbit anti-TH serum (Eugene Tech Intl. 1:50) overnight at 4°C and then washed 3 X 20 min. Goat anti-rabbit IgG (Zymed Lab. I :5.000) was used as secondary serum for 6 hours at room temperature. After washing with PBS 3 X 20 min. the sections were incubated with rabbit PAP complex (DAKO. 1:5,000) overnight. After washing with PBS, the sections were rinsed with 0.05 :\I Tris HCI buffer (pH 7.6) for 5 min and processed with the glucose oxidase-diaminobenzidine-nickel method described above.
In the preparations for electron microscopy. the sections were treated with I’i;’ osmic acid in 0.1 111phosphate buffer for I hour at room temperature. Subsequently. the sections were dehydrated with graded ethanol and stained with I% uranyl acetate in 70”% ethanol. then embedded with Epon between glass slides and coverslips coated with silicone. After observation and photography under the light microscope. blocks of TLD containing the ChAT immunoreactive neuronal profiles were dissected and cut into ultrathin sections. The ultrathin sections were collected on one-hole slot grids coated with formvar and stained with lead citrate. A Philips 201 electron microscope was used.
CHOLINERGIC-CATECHOLAMINE
481
INTERACTIONS
FIG. 2. lmmunoperoxidase staining of cholinergic neurons in TLD. The ChAT-immunoreactive neuron between the asterisks in part A was sectioned and observed in the electron microscope (part B). The asterisks indicate corresponding capillaries in parts A and B. In part B, arrowheads indicate the infolding of the nucleus in the cholinergic neuron. Part C shows a higher magnification of the square in part B, illustrating an asymmetric synaptic contact onto the cholinergic cell body from a nonimmunoreactive axon terminal containing many small vesicles (open arrow). Scale bars: A = 10 Frn, B = 5 pm. C = 0.5 Wm.
Immltnohisto~hemicul
Controls
Control incubations were conducted in which normal rat or rabbit sera were substituted for the ChAT or TH primary antisera, respectively. With other sections, the primary antiserum was left out. In addition, the crossreactivity of the secondary antibodies was examined. When the biotinylated goat anti-rat IgG was used secondary to the rabbit anti-TH serum, no staining was detected with the immunoperoxidase method. Likewise, when the biotinylated goat anti-rabbit serum was used with the rat anti-ChAT IgG, no reaction was observed with the P-galactosidase technique. Therefore, the P-galactosidase and glucose-oxidase DAB-nickel methods could be used together to detect ChAT- and TH-immunoreactivities, respectively. RESULTS The ChAT-positive neurons of the TLD are located between the locus ceruleus (LC) and the dorsal tegmental nucleus (Gudden) within the central gray. Rostrally they con-
tinue into the PPT in the lateral tegmentum. At the most rostra1 level of LC, the ChAT-positive neurons of the TLD extend ventrolaterally across the medial longitudinal fasciculus through the central tegmental tract (Fig. 1). ChAT-immunoreactivity was present in most of the neurons in the TLD. Their cell bodies were medium to large in size, showed oval or fusiform shapes and had a few major processes. Extensive networks of ChAT-immunoreactive dendrites were also present throughout the TLD. Individual immunoreactive dendrites were well stained and could be followed for over 150 Km. Neurons closely apposed to capillaries were frequently observed. In the sections stained with the immunoperoxidase method, the electron-dense reaction product was diffusely distributed throughout the ChAT-immunoreactive neuronal cytoplasm (Figs. 2, 3). The ChAT-immunoreactive neurons had large nuclei which showed infoldings. Indeed, in some ultrathin sections these invaginations were very deep, almost dividing the nucleus into two parts. These neurons had large amounts of cytoplasm which contained the usual complement
482
KUBOTA.
LEUNG
AND VINCENT
FIG. 3. Immunoperoxidase staining of a ChAT-immunoreactive neuron in TLD. (A) The nucleus contains a large nucleolus and lies at one pole of the cell. Arrowheads indicate infoldings of the nucleus. The abundant cytoplasm contains the common subcellular organelles. (B and C) High magnification electron micrographs of the lower (part B) and upper squares (part C) in part A. illustrating two asymmetric. nonimmunoreactive synaptic inputs onto the cell soma (open arrows). Both axon terminals contain many small vesicles. and in part B a few large cored vesicles are also present. Scale bars: A = I pm. B. C y 0.S pm.
CHOLINERGIC-CATECHOLAMINE
483
INTERACTIONS
FIG. 4. lmmunoperoxidase staining of ChAT-immunoreactive dendrites in TLD in longitudinal (part A) and crosssection (part B). The immunoreactivity is associated with microtubules. In part A, three asymmetric synaptic contacts from nonimmunoreactive axons are seen (open arrows). These nonimmunoreactive terminals have many small, round vesicles and some large, cored vesicles and wide, well-developed membrane specializations. A small ChAT-immunoreactive profile, which probably corresponds to a spine, receives synaptic contact from a nonimmunoreactive axon (arrow). In part B, a dense plaque resembling a subsynaptic bar beneath the membrane specialization is present (black arrow). This dendrite also receives a synapse from a large terminal containing pleomorphic vescicles. Scale bars: 0.5 pm.
of subcellular organelles, i.e., mitochondria, endoplasmic reticulum, Golgi apparatus, microtubules. They also contained cored vesicles. The ChAT-immunoreactive cell bodies received synaptic contacts from nonimmunoreactive axon terminals (Figs. 2, 3). In some instances, one to four synaptic contacts were observed on a perikaryon in one ultrathin section, suggesting that many axo-somatic synaptic contacts with nonimmunoreactive axon
terminals occur. These terminals contained many round or oval small vesicles and sometimes large cored vesicles. The synaptic membrane specialization was well developed, wide, and usually asymmetric (Figs. 2C, 3B, C). ChAT-immunoreactive dendrites ranging in size from 0.1 to 1.5 pm in diameter (Figs. 4, 7A) were present throughout the TLD. The electron-dense immunoreactivity was associated primarily with microtubules. These dendrites received
484
KUBOTA.
LEUNG
AND
VINCENT
FIG. 5. Immunoperoxidase staining of a ChAT-immunoreactive dendrite showing a close apposition with a pericyte (P) adjacent to a capillary (C). Arrowhead in part B shows a synaptic contact on the ChAT-immunoreactive dendrite. Scale bars: A = I pm, B = 0.5 pm.
many minals served profile
axon tersynaptic inputs from nonimmunoreactive ob). Two to four synaptic contacts were sometimes dendritic close tog ,ether on a ChAT-immunoreactive many round (Fig. 4). These axons usually contained
or oval small vesicles. and the membrane specialization was usually asymmetric and often occupied a wide area. In some instances, a dense plaque resembling a subsynaptic bar was present beneath the membrane specialization (Fig. 4A. B). A
CHOLINERGIC-CATECHOLAMINE
INTERACTIONS
FIG. 6. lmmunoperoxidase staining of myelinated ChAT-immunoreactive axons in TLD (black asterisks). Nonimmunoreactive myelinated axons are also observed (open asterisks). A large, ChAT-immunoreactive dendrite receiving synaptic input from a nonimmunoreactive terminal is also shown in part A (open arrow). Scale bars: A = 0.5 Km, B = 0.5 Wm.
485
486
KIJBO’IA.
LEUNG
AND
VINCENT
FIG. 7. Immunoperoxidase staining of a ChAT-immunoreactive axon making synaptic contact with a nonimmunoreactive cell in TLD. Parts B and Care successive sections from the square in part A. The ChAT-immunoreactive axon shows a synaptic membrane specialization in part B (arrow) and an aggregation of small vesicles at the corresponding site in part C (arrow). Scale bars: A = I pm, B and C 7 0.5 urn.
few symmetric contacts onto cholinergic neurons from terminals containing pleomorphic vesicles were also observed (Fig. 4B). Small, ChAT-immunoreactive spines were also observed in contact with nonimmunoreactive axon terminals (Fig. 4A). ChAT-immunoreactive processes were also observed closely associated with pericytes surrounding capillaries (Fig. 5A). Indeed, direct contacts between ChAT-immunoreactive dendrites and pericytes could occasionally be seen (Fig. 5B).
ChAT-immunoreactive myelinated axons were frequently seen containing immunoreactive. electron-dense material associated with neurofilaments and microtubules (Fig. 6). The diameters of the ChAT-immunoreactive axons ranged from 0.5 to 1.O pm. Only one case of a ChAT-immunoreactive axon terminal was seen. and it made synaptic contact with a nonimmunoreactive perikaryon (Fig. 7). This axon terminal had many small round vesicles, a few large, cored vesicles and an asymmetric synaptic membrane specialization. The active zone was
CHOLINERGIC-CATECHOLAMINE
487
INTERACTIONS
FIG. 8. Double immunohistochemistry in TLD showing a ChAT-immunoreactive neuron (open arrow), which was stained light blue-green with the P-galactosidase method in the light microscope. Many TH-immunoreactive terminals stained dark blue-black with the immunoperoxidase method are seen (arrows), some of which show close appositions with the ChATimmunoreactive neuron (arrow heads). Scale bar: IO Frn.
short, about 170 nm, and the innervated neuron was small (about I5 pm diameter). The /3-galactosidase immunohistochemical technique was less sensitive than the peroxidase method. The immunoreactivity was only observed within 5 to 10 pm from the surface of the vibratome sections. ChAT-immunoreactive, light blue-green staining was seen in the neuronal perikarya and dendrites in TLD. In contrast, TH-immunoreactive structures detected with the glucose-oxidase DAB-nickel reaction showed almost black immunoreactive material. Thus, these two different kinds of immunoreactivity could be easily distinguished in the light microscope. Great numbers of TH-immunoreactive neuronal perikarya were seen in the LC and the lateral part of TLD, and TH-immunoreactive fibers extended from the LC into the TLD. Moderate numbers ofTH-immunoreactive fibers were seen throughout the TLD (Fig. 8). The thin, TH-immunoreactive fibers sometimes showed varicose swellings. About one-third of the ChAT-immunoreactive neurons had close appositions with THimmunoreactive fibers on their perikarya and/or proximal dendrites. When the fl-galactosidase method was used to detect the choline& neurons, the immunoreaction product formed small, dot-like, electron-dense material and was distributed throughout the labelled neuronal profiles (Fig. 9). At high magnification, the reaction product showed moderate electron density. In contrast, the TH-immunoreactive material detected with the immunoperoxidase method was diffusely distributed throughout the labelled neuronal profiles and the electron density was much greater than that of the /3-galactosidase reaction (Fig. 9). Therefore, differentiation of these two kinds of immunoreaction on the same ultrathin section in the electron microscope was quite easy and reliable.
Many TH-immunoreactive fibers were present in the TLD, and occasionally, TH-immunoreactive axon terminals made synaptic contacts with nonimmunoreactive neuronal profiles. The membrane specialization was symmetrical and the synapse was very narrow and small (SO-100 nm). Seven ChAT-immunoreactive neurons in TLD showing close apposition with THimmunoreactive profiles in the light microscope were subsequently examined in the electron microscope. All of them showed close appositions with TH-immunoreactive axons in the electron microscope, and some of them received specific synaptic contacts from TH-immunoreactive axon terminals (Fig. 10). Usually one or two TH-immunoreactive terminals were seen in contact with a ChAT-immunoreactive neuron in a single thin section. In one case, eight TH-immunoreactive terminals were observed in apposition with one ChAT-immunoreactive cell body and its proximal dendrite (Figs. 9. IO). DISCUSSION
The ultrastructure of ChAT-immunoreactive neurons in the TLD was observed in the present study. The distribution of cholinergic neurons in the mesopontine tegmentum corresponded with previous reports (2,19,40). Morphologically, the ChAT-immunoreactive neurons in TLD appeared to form a homogeneous group. Most of the neurons in TLD showed strong ChAT-immunoreactivity, and a dense network of ChAT-immunoreactive dendrites was also present. The ChAT-positive cell bodies were medium to large in size and had two to four elongated major dendrites with extensive arborizations. The ChAT-immunoreactive neurons had indented nuclei and well-developed cytoplasm containing the ordinary subcellular organs. Most of the cholinergic neurons received synaptic contacts from nonimmunoreactive axon
488
KUBOTA.
FIG. deep close (large bars:
9. Ultrastructure
LEUNG
AND VINCENT
of the ChAT-immunoreactive neuron in Fig. 8. showing at low magnihcation the nucleus with a very infolding (arrow heads) and a single nucleolus. Seven TH-immunoreactive axon terminals (large arrows) are seen in apposition to the cell soma. Part B presents a higher magnitication of five of the TH-immunoreactive axon terminals arrows). The small arrows indicate the &galactosidasc reaction product in the ChAT-immunoreactivc neuron. Scale A = 2 pm. B = 0.5 pm.
CHOLINERGIC-CATECHOLAMINE
INTERACTIONS
FIG. IO. Serial thin sections of the TH-immunoreactive terminal labelled (arrow a) in Fig. 9B. Parts A and B are three sections apart and part C is adjacent to part B. The terminal forms a small synapse with the ChAT-immunoreactive cell soma. The arrows indicate /3-galactosidase reaction product in the ChATimmunoreactive neuron. Scale bar: 0.5 pm.
489
KIJBOTA.
490
terminals on their cell bodies and proximal dendrites. The overall morphology of the ChAT-immunoreactive neurons in the TLD was. thus. very similar to that of the cholinergic neurons in PPT described in previous reports (23.36). ChAT-immunoreactive axon fibers were occasionally seen in TLD, and these were primarily myelinated. This is consistent with the rather fast conduction velocities seen previously in rat and cat TLD t&17,34), although another report found a somewhat slower conduction speed (I I). Only one ChAT-immunoreactive axon terminal was observed, and it made synaptic contact with a nonimmunoreactive neuronal cell body. The axon terminal appeared similar to the ChAT-immunoreactive axon terminals observed previously in various regions innervated by the mesopontine cholinergic neurons (4,5,13,23,24). The paucity of ChAT-immunoreactive axon terminals in TLD observed in this study is consistent with the ChAT-immunoreactive neurons being projection neurons and not local circuit neurons. Previous ultrastructural studies on the basal forebrain cholinergic neurons have also noted an absence of labelled terminals ( I ,26). ChAT-immunoreactive cell bodies and dendrites sometimes showed close appositions with pericytes surrounding blood vessels. ChAT-immunoreactive neuronal structures in the basal forebrain also appear to engage in such contacts (I). These observations are consistent with biochemical, pharmacological. and physiological work demonstrating cholinergic regulation of central blood vessels (9). Together, these observations suggest that acetylcholine has an important role in the modulation of blood flow in the central nervous system. The mesopontine cholinergic neurons have recently been shown to contain high levels of nitric oxide synthase (I 5.39.42). Thus, nitric oxide produced in these cells may also act on these closely apposed blood vessels. The /Sgalactosidase immunohistochemical method was introduced by Bondi et al. (6). and an electron microscope method was also described for double immunohistochemical staining in combination with immunoperoxidase methods (20,27). The immunoreaction product of the &galactosidase stain is easy to distinguish from that of peroxidase reactions by the color difference in the light microscope and the different deposits seen in the electron microscope. Therefore, this double immunohistochemical staining method is especially useful for the study of synaptic interactions on neuronal perikarya and proximal dendrites. It is quite a reliable and simple technique. although the sensitivity of the /Sgalactosidase immunohistochemical method is lower than that of the peroxidase reaction, and synaptic interactions at distal dendrites are difficult to discern with this procedure. The distribution and ultrastructural features of the TH-positive structures in TLD coincided well with previous studies of catecholamine fibers. At the light microscopic level, TH-immunoreactive fibers have been noted within the TLD (37).
LEUNG
AND
VINCENT
sometimes showing swellings close to cholinergic neurons (30). Ultrastructurally, the TH-immunoreactive fibers in TLD rarely showed synaptic membrane specializations. Generally the synaptic active zones of TH-immunoreactive axons are quite narrow and small throughout the central nervous system (2 I). Although TH-immunoreactive axons were observed in serial. thin sections. only a few TH-immunoreactive axons showing synaptic active zones could be identified. The major finding in the present study was that ChAT-immunoreactive neurons in the TLD receive synaptic input from TH-immunoreactive axons. What might be the origin of these TH-positive fibers innervating the TLD? The TLD is known to receive input from the substantia nigra (3.7,30). However. this input arises principally from nondopaminergic neurons in the pars reticulata (30). The TH-positive neurons of the locus ceruleus have been shown to innervate the TLD (30). Many axons and dendrites from locus ceruleus extend medially into the adjacent TLD (32). Positive fibers have also been described in TLD in immunohistochemical studies with antisera against dopamine-@-hydroxylase ( 16). These fibers appear to be derived from the noradrenergic neurons in locus ceruleus (I 6). Jones (16) also suggested the possible interaction between cholinergic neurons and noradrenergic neuron in TLD by the observation that varicose DBHpositive fibers are found close to ChAT-positive TLD cells. Autoradiographic binding studies have localized (Yeadrenergic receptors in TLD (38). and noradrenaline h,yperpolarizes identified cholinergic TLD neurons in vitro (Wtlhams and Reiner. personal communication). Together these observations are consistent with a direct input of TLD cholinergic cells by the noradrenergic locus ceruleus. In addition, phenylethanolamine-Nmethyl-transferase-immunoreactive fibers originating in the Cl adrenaline cell group ofthe nucleus paragigantocellularis lateralis have been reported in TLD (12.28). It is not yet clear whether the TH-immunoreactive fibers observed in the present study represent dopaminergic-. noradrenergic-. or adrenaline-containing fibers. However, there is physiological evidence consistent with reciprocal interactions between brain stem cholinergic and noradrenergic neurons. Together, these two brain stem systems are thought to play a major role in the regulation of the sleep-wake cycle (35). The TH-immunoreactive terminals observed in the present study may. therefore, represent the morphological basis for the direct regulation of the cholinergic neurons of the TLD by noradrenergic fibers. ,ACKNOWLEDGEMENTS
The authors thank Drs. P. B. Reiner. K. Shinado. and J. Kohno fat fruitful discussions, and Ms. Sandra Sturgeon for secretarial assistance. This work was supported by the Medical Research Council of Canada. S.R.V. is an MRC Scientist.
REFERENCES Armstrong. D. M. Ultrastructural characterization of choline acetyltransferase-containing neurons in the basal forebrain of rat: Evidence for a cholinergic
innervation of intracerebral 1986.
blood vessels.
J. Comp. Neural. 250:81-92:
Armstrong, D. M.; Saper. C. B.; Levey, A. L.: Wainer, B. H.: Terry, R. D. Distribution of cholinergic neurons in the rat brain: Demonstrated by the immunocytochemical localization of choline acetyltransferase. J. Comp. Neurol. 216:53-68: 1983. Beckstead. R. M.: Domesick. V. B.; Nauta, W. J. H. Efferent connections of the substantia nigra and ventral tegmental area in the rat. Brain Res. 175:191-217: 1979.
Beninato. M.; Spencer. R. F. The cholinergic innervation of the rat substantia nigra: A light and electron microscopic immunohistochemical study. Exp. Brain Res. 72: 178-184: 1988. Bolam. J. P.; Francis. C. M.: Henderson, Z. Cholinergic input to dopaminergic neurons in the substatia nigra: A double immunocytochemical study. Neuroscience 4 1:483-494: 199 I. Bondi. A.: Chieregatti, G.; Eusebi, V.; Fulcheri, E.: Bussolati, G. The use of@-galactosidase as a tracer in immunocytochemistry. Histochemistry 76: 15%158: 1982. 7. Cornwall, J.: Cooper. J. D.: Phillipson. 0. T. Afferent and efferent connections of the laterodorsal tegmental nucleus in the rat. Brain Res. Bull. 25:271-384: 1990.
CHOLINERCIC-CATECHOLAMINE
491
INTERACTIONS
8. El Mansari, 9.
10.
Il.
12.
13.
14.
15.
16. 17.
18.
19. 20.
21.
22.
23.
24.
25. 26.
27.
M.; Sakai, K.; Jouvet, M. Unitary characteristics of presumptive cholinergic tegmental neurons during the sleep waking cycle in freely moving cats. Exp. Brain Res. 76:5 19-529; 1989. Estrada, C.; Hamel, E.; Krause, D. N. Biochemical evidence for cholinergic innervation of intracerebral blood vessels. Brain Res. 266:261-270; 1983. Gould, E.; Woolf, N. J.; Butcher, L. L. Cholinergic projections to the substantia nigra from the pedunculopontine and laterodorsal tegmental nuclei. Neuroscience 28:61 l-623: 1989. Grant, S. J.; Highfield, D. A. Extracellular characteristics of putative cholinergic neurons in the rat laterodorsal tegmental nucleus. Brain Res. 559:64-74; 199 I. Guyenet, P. G.; Young, B. S. Projections of nucleus paragigantocellularis lateralis to locus coeruleus and other structures in rat. Brain Res. 406:171-184; 1987. Hallanger, A. E.; Wainer, B. H. Ultrastruture of ChAT-immunoreactive synaptic terminals in the thalamic reticular nucleus of the rat. J. Comp. Neurol. 278:486-497: 1988. Higo, S.; Ito, K.; Fuchs, D.; McCarley. R. W. Anatomical interconnections of the pedunculopontine tegmental nucleus and the nucleus prepositus hypoglossi in the cat. Brain Res. 536:79-85: 1990. Hope, B. T.: Michael, G. J.; Knigge, K. M.; Vincent, S. R. Neuronal NADPH diaphorase is a nitric oxide synthase. Proc. Natl. Acad. Sci. USA 88:281 I-2814: 1991. Jones, B. E. Paradoxical sleep and its chemical/structural substances in the brain. Neuroscience 40:637-656: I99 I. Kayama, Y.; Ogawa, T. Electrophysiology of ascending. possibly cholinergic neurons in the rat laterodorsal tegmental nucleus: Comparison with monoamine neurons. Neurosci. Lett. 77:277-282: 1987. Kayama, Y.; Takagi, M.; Ogawa. T. Cholinergic influence of the laterodorsal tegmental nucleus on neuronal activity in the rat lateral geniculate nucleus. J. Neurophysiol. 56: I297- 130 I; 1986. Kimura, H.: Maeda, T. Aminergic and cholinergic systems in the dorsolateral pontine tegmentum..Brain Res. Bull.~9:493-499; 1982. Kohno. J.: Shinoda. K.: Kawai. Y.: Ohuchi. T.: Ono. K.: Shiotani. Y. Interaction between adrenergic fibers and intermediate choline@ neurons in the rat spinal cord: A new double-immunostaining method for correlated light and electron microscopic observations. Neuroscience 25:113-121: 1988. Liu. X.-B.; Jones, E. G. The fine structure of serotonin and tyrosine hydroxylase immunoreactive terminals in the ventral posterior thalamic nucleus of cat and monkey. Exp. Brain Res. (in press). Luebke. J. I.; Greene, R. W.: Semba, K.; Kamondi, A.; McCarley, R. W.; Reiner, P. B. Serotonin hyperpolarizes cholinergic low threshold burst neurons in the rat laterodorsal temental nucleus in vitro. Proc. Natl. Acad. Sci. USA 89:743-747; 1992. Martinez-Murillo, R.; Villalba, R.; Montero-Caballero, M. 1.; Rodrigo, J. Cholinergic somata and terminals in the rat substantia nigra: An immunocytochemical study with optical and electron microscopic techniques. J. Comp. Neurol. 28 1:397-415; 1989 Martinez-Murillo, R.; Villalba, R. M.; Rodrigo, J. Electron microscopic localization of cholinergic terminals in the rat substantia nigra: An immunocytochemical study. Neurosci. Lett. 96: I2 I - 126; 1989. McCormick, D. A. Cholinergic and noradrenergic modulation of thalamocortical processing. Trends Neurosci. I2:2 15-22 I ; 1989. Palacios, G.; Garcia-Ladona, J.; Codina, M. Ultrastrutural study of cholinergic neurons in the medial septal nucleus and vertical limb of the diagonal band of Broca in the basal forebrain of the rat. J. Chem. Neuroanat. 4:205-22 I ; 199 1. Peng, Y.; Kohno, J.; Shinoda, K.; Kawai, Y.; Ono, K.; Jiang, W.; Shiotani. Y. Interactions between cholinergic neurons and substance
28.
29.
30.
31.
32.
33.
34.
35. 36.
31.
38.
39. 40.
41.
42.
43.
44.
P or calcitonin gene-related peptide terminals in the rat sacral intermediolateral nucleus: Double immunostaining at the light and electron microscopic levels. J. Comp. Neurol. 285: 1-8; 1989. Pieribone, V. A.; Aston-Jones, G. Adrenergic innervation of the rat nucleus locus coeruleus arises predominantly from the Cl adrenergic cell group in the rostra1 medulla. Neuroscience 4 I :525-542; I99 I. Reiner, P. B.: Vincent, S. R. Topographic relations of cholinergic and noradrenergic neurons in the feline pontomescencephalic tegmentum: An immunohistochemical study. Brain Res. Bull. 19:705714: 1987. Semba, K.; Fibiger, H. C. Afferent connections of the laterodorsal and the pedunculopontine tegmental nuclei in the rat: A retro- and anterograde transport and immunohistochemical study. J. Comp. Neural. (in press). Semba, K.: Reiner, P. B.; Fibiger. H. C. Single cholinergic mesopontine tegmental neurons project to both the pontine reticular formation and the thalamus in the rat. Neuroscience 38:643-654; 1990. Shimizu. N.: Ohnishi, S.: Satoh, K.; Tohyama, M. Cellular organization of locus coeruleus in the rat as studied by Golgi method. Arch. Histol. Jpn. 41:103-l 12; 1978. Shu, S.: Ju. G.: Fan. L. The glucose oxidase-DAB-nickel method in peroxidase histochemistry ofthe nervous system. Neurosci. Lett. 85: 169-171: 1988. Steriade. M.: Datta. S.; Oakson, G.: Dossi. R. C. Neuronal activities in brain-stem cholinergic nuclei related to tonic activation processes in the thalamocortical systems. J. Neurosci. 10:2541-2559; 1990. Steriade, M.; McCarley. R. W. Brainstem control of wakefulness and sleep. New York: Plenum: 1990. Sugimoto. T.: Mizukawa, K.; Hattori. T.; Konishi. A.; Kaneko, T.; Mizuno. N. Cholinergic neurons in the nucleus tegmenti pedunculopontinus pars compacta and the caudoputamen of the rat: A light and electron microscopic immunocytochemical study using a monoclonal antibody to choline acetyltransferase. Neurosci. Lett. 51:l 13-I 17: 1984. Sutin. E. L.: Jacobowitz, D. M. lmmunocytochemical localization of peptides and other neurochemicals in the rat laterodorsal tegmental nucleus and adjacent area. J. Comp. Neurol. 270:243-270; 1988. Unnerstall. J. R.: Kopajitic, T. A.; Kuhar, M. J. Distribution of a1r agonist binding sites in the rat and human central nervous system: Analysis of some functional, anatomic correlates of the pharmacologtc effects of clonidine and related adrenergic agents. Brain Res. Rev. 7:69-101; 1984. Vincent, S. R.; Hope, B. T. Neurons that say NO. Trends Neurosci. 15:108-l 13: 1992. Vincent. S. R.: Satoh, K.; Armstrong, D. M.: Panula, P.; Vale, W.; Fibiger. H. C. Neuropeptides and NADPH-diaphorase activity in the ascending cholinergic reticular system of the rat. Neuroscience 17:167-182: 1986. Vincent, S. R.: Satoh. K.; Armstrong, D. M.; Fibiger. H. C. Substance P in the ascending cholinergic reticular system. Nature 306:688691: 1983. Vincent. S. R.; Satoh, K.; Armstrong, D. M.: Fibiger, H. C. NADPHdiaphorase: A selective histochemical marker for the cholinergic neurons of the pontine reticular formation. Neurosci. Lett. 43:3236: 1983. Wilcox, K. S.; Grant, S. J.; Burkhart. B. A.: Christoph. G. R. In vitro electrophysiology of neurons in the lateral dorsal tegmental nucleus. Brain Res. Bull. 22:557-560: 1989. Woolf, N. J.: Butcher, L. L. Cholinergic systems in the rat brain. III. Projections from the pontomesencephalic tegmentum to the thalamus, tectum, basal ganglia, and basal forebrain. Brain Res. Bull. 16:603-637: 1986.
