Brain Research, 241 (1982) 11-29 Elsevier Biomedical Press
11
Efferent Projections of the A1 Catecholamine Cell Group in the Rat: An Autoradiographic Study S. McKELLAR and A. D. LOEWY* Department of Anatomy and Neurobiology, Washington University School of Medicine, 660 South Euclid Avenue, St. Louis, MO 63110 (U.S.A.) (Accepted November 3rd, 1981) Key words: A1 cell group - - catecholamines - - norepinephrine - - spinal cord - - intermediolateral cell column - - meduUa - hypothalamus - - median eminence - - supraoptic nucleus - - paraventricular nucleus
Efferent connections of the region of the A1 catecholamine cell group were investigated by the anterograde autoradiographic method in rats, some of which had been pretreated with intraventricular injections of 6-hydroxydopamine (6-OHDA). Spinal projections were further studied by combining histofluorescence and horseradish peroxidase staining in the same sections. Projections from the A1 region ascend through the lateral hypothalamus to the bed nucleus of the stria terminalis, the medial preoptic area, and several hypothalamic nuclei: the dorsomedial nucleus, the dorsal hypothalamic area, the paraventricular and supraoptic nuclei, and the median eminence. These projections are predominantly ipsilateral. Since they are sensitive to 6-OHDA, they presumably arise from catecholamine cells of the A1 group. Other pathways are not noticeably affected by 6-OHDA. These include projections through the reticular formation to the contralateral nucleus ambiguus, the ipsilateral facial nucleus, and the KSlliker-Fuse nuclei, the parabrachial nuclei and the periaqueductal gray on both sides, as well as the intralaminar nuclei of the thalamus. These pathways probably do not arise from the A1 group, although a minor noradrenergic component cannot be ruled out. Spinal projections extend to the intermediolateral cell column and the ventral horn, and especially to the phrenic motor nucleus. However, these projections arise almost exclusively from non-catecholaminecells. Only a small minority of the fluorescent A1 cells are retrogradely labeled after injections of HRP into the upper thoracic spinal cord. They lie at the level of the pyramidal decussation. Likewise only a few fluorescent cells of the A2 group are labeled. Although the A1 and A2 groups were long thought to be principal sources of spinal norepinephrine, a review of the literature shows that this belief was based on equivocal evidence. INTRODUCTION The A1 catecholamine cell group was first described by D a h l s t r 6 m a n d Fuxe 20, who regarded it as a principal source of spinal catecholamines 21. Later studies added only f r a g m e n t a r y i n f o r m a t i o n a b o u t its connections. I n m a n y cases the A1 group was n o t clearly separated from other noradrenergic cell groups, except for the locus coerttleus. As a result, the n o n - c o e r u l e a n cell groups are sometimes treated as a single 'lateral tegmental' system 62. Catecholamine pathways are often studied by means of lesions followed by m i c r o p u n c h biochemical assays or histofluorescence. However, lesions m a y involve fibers or cell bodies of more t h a n one
* To whom correspondence should be addressed. 0006-8993/82/0000-0000/$02.75 © Elsevier Biomedical Press
catecholamine system. I n addition, lesions m a y change regional catecholamine levels in unpredictable ways n o t directly related to the pathways severed, due to sprouting or other secondary effects 66. In order to avoid these problems, we have used the anterograde autoradiographic t e c h n i q u O s to study the efferent projections of the A1 cell group. Tritiated a m i n o acids were injected in a n d a r o u n d the region of the A1 group in the ventrolateral m e d u l l a of the rat. These experiments were repeated in rats which h a d received intraventricular injections of 6 - h y d r o x y d o p a m i n e (6-OHDA), a n e u r o t o x i n relatively selective for catecholamine cells 31,42. Spinal projections were further studied using horse-
12 radish peroxidase (HRP) as a retrograde tracer, in combination with histofluorescence8. MATERIALS AND METHODS For the autoradiographic experiments a mixture of tritiated leucine, lysine, and proline (New England Nuclear; Boston, MA; spec. act. 26-57 Ci/mmol) was dried and redissolved in sterile saline to a total concentration of 45 #Ci/#l (15/~Ci/#l for each amino acid). This mixture was injected via a 1 #1 syringe into the medullas of male Holtzman rats, anesthetized with chloral hydrate (350 mg/kg). Injection volumes ranged from 25 to 100 nl. After survival times of 5-8 days, the rats were perfused through the heart with saline followed by fixative. Usually the fixative was 10 ~ formol saline; in some cases it was 'Faglu' (see below). The brains and spinal cords were cut in coronal sections and processed for autoradiography, using 6 week exposures. Most of this material was embedded in paraffin and processed as described previouslylS,SL In the cases perfused with Faglu, serial Vibratome sections through the injection sites were mounted on subbed slides and examined with a fluorescence microscope to verify the position of fluorescent cells. This tissue was then dehydrated and processed for autoradiography like the paraffin material. Out of a larger series, 28 cases were selected for analysis. In 5 of the autoradiographic cases, the rats had been pretreated at least two weeks earlier with intraventricular injections of 6-hydroxydopamine (Sigma, St. Louis, MO; two injections of 250 #g each, dissolved in 10/zl saline with 0.1 ~ ascorbic acid). The initial retrograde transport studies used a technique devised by Blessing et al. s for combining horseradish peroxidase (HRP) with histofluorescence. Four 1 #1 injections of HRP (Sigma, Type VI; 33 ~/o in saline) were made into the spinal cords of male Holtzman rats at about the Tl level. Two or three days later the rats were perfused with Faglu z0 ( 4 ~ formaldehyde and 0.5 ~ glutaraldehyde in 0.1 M phosphate buffer, pH = 7.0). Vibratome sections of the brainstem were processed as described elsewhere s, except that the H R P reaction (using tetramethylbenzidine) was performed according to the protocol of deOlmos et al. 25. With this technique
one can photograph fluorescent catecholamine cells and later compare the photomicrographs with retrograde labeling in the same sections. Later a variation of this procedure made it possible to view the H R P reaction simultaneously with the histofluorescencO 0. The following procedure is a modification of one described by Blessing et al. 9. The H R P injections and the perfusions were performed as before. Vibratome sections were collected in ice-cold Faglu and given five 1 min rinses in cold phosphate buffer (0.1 M; pH = 7.3), followed by a brief rinse in cold saline. The sections were then stained for H R P according to deOlmos et al. 25 except that the H202 concentration was increased fourfold. After no more than 30 min of incubation, the sections were rinsed for 5 min in an ice-cold solution of 4 % formaldehyde and 0.5 % glutaraldehyde, with the pH adjusted to 3.3 with HC1. Then they were transferred to cold Faglu and mounted onto subbed slides. The slides were dried briefly with a hair dryer (cool setting) and then in vacuo over P205 for 1 h or more. They could then be cover-slipped with mineral oil and viewed with a fluorescence microscope. Since prolonged soaking weakens the fluorescence, it is important to finish the procedure quickly. Before the first section is cut, the rinses and the solutions needed for the H R P reaction should be prepared and put on ice (except for the TMB solution, which jells if chilled too long). All rinses and incubations should be ice-cold. In particular, the aldehyde rinses may cause non-specific fluorescence if they are not cold. Frozen sections (rather than Vibratome sections) also give satisfactory results (J. Neil, unpublished results). Three control rats received intravenous injections of 1.3 mg H R P dissolved in 0.1 ml heparinized saline. After two days they were perfused ( 3 ~ glvtaraldehyde, 0 . 5 ~ paraformaldehyde and 1 sucrose in 0.1 M phosphate buffer, pH = 7.3). Frozen sections of the lower medulla were reacted for H R P by the method of deOlmos et al. 25. RESULTS Nomenclature As used here, the term 'A1 group' refers only to noradrenergic cells. Tlae nearby adrenergic or epi-
13 nephrine-containing cells will be termed the 'C1 group TM. They cannot be distinguished in the present study, since they are resistant to 6-OHDA ~9,72 and do not fluoresce under the conditions used here 3s. Although the A 1 group is sometimes depicted as part of the lateral reticular nucleus (e.g. ref. 67), in our experience only an occasional fluorescent cell is actually within this nucleus. Most of the A1 cells are just outside the dorsal or lateral boundary of the lateral reticular nucleus, with a few ventral or medial to it. Similar findings have been reported in the cat 7. This discrepancy probably reflects an ambiguity in nomenclature. As used here, the term 'lateral reticular nucleus' refers to the cerebellar relay nucleus described, for example, by Walberg vg. It extends roughly from the caudal pole of the inferior olive to the caudal pole of the retrofacial nucleus. Just rostral to the lateral reticular nucleus, and ventral to the retrofacial nucleus, lies a more amorphous region which contains the most rostral cells of the A1 group. It has been called the 'nucleus reticularis lateralis'11, 5s. Since this term is the exact Latin equivalent of 'lateral reticular nucleus', it has led to considerable confusion. For the sake of clarity, we shall use the term 'nucleus paragigantocellularis lateralis '1 & The nucleus ambiguus, as defined here in the rat, lies dorsal to most of the lateral reticular nucleus. At its rostral pole it forms a prominent spindle-shaped mass between the medial and lateral portions of the lateral reticular nucleus. At still more rostral levels it is replaced by the nucleus paragigantocellularis lateralis and by the retrofacial nucleus. We use the term 'nucleus ambiguus' because that is the name ordinarily applied to this cell group (though the term is often extended to include the retrofacial nucleus). However, this usage is not meant to imply anything about the projections of these cells. In particular, they may not correspond to the efferent neurons of the vagus and glossopharyngeal nerves.
Injections of the A1 group Figure 1 illustrates a moderately large (50 nl) injection of tritiated amino acids into the rostral half of the A1 group. The injection also involved the nucleus ambiguus, the lateral reticular nucleus (es-
pecially the subtrigeminal division), and the most caudal portion of the nucleus paragigantocellularis lateralis. This case probably provides a fairly complete picture of the A1 system, although other pathways were labeled as well. Spinal projections are considered in a separate section (see below). In the lower medulla, labeled fibers passed ventral to the inferior olive, to or through the nucleus raphe paUidus, and apparently terminated in the lateral part of the B 1 serotonergic cell group °-°. Other fibers, apparently crossing just ventral to the hypoglossal nucleus, heavily innervated the contralateral nucleus ambiguus. Fibers of uncertain trajectory terminated in the caudal medial accessory olive, just medial to its lateral tip. Short projections are difficult to distinguish from diffusion of the isotope, but there appeared to be diffuse labeling throughout the caudal ipsilateral solitary complex, with heavier labeling at the level of the obex in the ventrolateral and intermediate subnuclei. The labeling of the contralateral solitary complex was somewhat lighter. The ipsilateral hypoglossal nucleus received a moderate projection. There may also be a projection to the ipsilateral inferior olive, roughly mirroring that seen on the opposite side. Many labeled fibers passed around the trigeminal complex into the inferior cerebellar peduncle, eventually entering the cerebellum. Many other cases showed a similar projection. Since these fibers are presumably derived from the lateral reticular nucleus they will not be described further. Likewise, a few fibers crossed to the contralateral cerebellum, and may be attributed to slight involvement of the inferior olive. A few labeled fibers ascended in the spinal trigeminal nucleus, presumably reflecting slight involvement of pars caudalis of the spinal trigeminal nucleus 43. Most fibers, however, ascended bilaterally in the reticular formation, especially in the nucleus paragigantocellularis lateralis. Others ascended contralaterally between the inferior olive and the medial longitudinal fasciculus. The nucleus raphe paltidus and the B1 serotonergic cell groups were still labeled at this level. The rostral pole of each hypoglossal nucleus was moderately labeled in the most medial corner; this labeling extended into the nucleus prepositus as well.
14
..~.-'.: .].
H c IC .GP
:"
:.." "r ~;$Cn
"." "
AD
TLO
P81
',.GP •
E
'
.'"
•
VH
".
• ""
"
J
•
•
•
:'~
T6
Fig. 1. Normal rat with injection of tritiated amino acids in the region of rostral A1 cell group. In this and the succeeding drawings, the dots represent the distribution of silver grains, and the injection site appears black. See list of abbreviations after text.
Light, diffuse labeling was present in both facial nuclei. In addition, the lateral third of the ipsilateral facial nucleus received a dense projection. Light labeling in the genu and root of the facial nerve indicates that the injection may have involved salivatory cells 63. Labeled fibers continued to ascend in the reticular formation, passing mostly dorsal to the facial nucleus and superior olive. In the pons they passed medial to the motor trigeminal nucleus to innervate the nucleus tegmenti laterodorsalis bilaterally and
the locus coeruleus ipsilaterally. The K611iker-Fuse nucleus was heavily labeled on both sides, with lighter labeling in the parabrachial nuclei. Some of the fibers ascending in the midbrain tegmentum evidently supplied a prominent innervation of the periaqueductal gray; this labeling continued to the level of the posterior commissure. Other fibers collected into the medial lemniscus, especially contralaterally. A few others reached the external nucleus of the inferior colliculus or the midlayers of the superior colliculus (stratum griseum intermediale
15 and stratum album intermediale). In the contralateral thalamus, some labeling was found in the parafascicular, centrolateral, and paracentral nuclei, as well as the paraventricular thalamic nucleus. Another patch of labeling lay medial to the medial geniculate nucleus, in what may be part of the posterior complex. Finally, a light band of labeling was present contralaterally in the lateral portion of the ventrobasal thalamus. Several hypothalamic areas were labeled. Fibers ascended bilaterally in the zona incerta and fields of Forel; after apparently innervating the edges of the entopeduncular nucleus they ascended further in the dorsal part of the lateral hypothalamus. From that position, labeled fibers arched over the fornix to innervate the dorsomedial nucleus and the dorsal hypothalamic area on both sides, and the paraventricular nucleus mainly on the ipsilateral side. The projection to the paraventricular nucleus has been described in detail elsewhereSL Although it extended to most of the subdivisions of this nucleus, it was largely excluded from the periventricular division, and was heaviest in the medial parvocellular and dorsomedial divisions. The ventral aspect of the ipsilateral supraoptic nucleus was also labeled. Light labeling was present in the retrochiasmatic area and in the internal lamina of the median eminence. Labeled fibers continued diffusely through the lateral hypothalamus and preoptic area, with distinct patches of labeling in the median preoptic nucleus and in the ventral or subcommissural half of the bed nucleus of the stria terminalis. A few faintly labeled fibers could be seen passing around the genu of the corpus callosum into the ipsilateral cingulate bundle. When the injections covered more caudal parts of the A1 region, many of the same pathways were labeled. These included projections to the facial nucleus, the K611iker-Fuse nucleus, the parabrachial nuclei, and the periaqueductal gray. The intralaminar nuclei of the thalamus were generally labeled in small patches which were difficult to compare from one brain to the next. The hypothalamus and the bed nucleus of the stria terminalis were consistently labeled whenever the injection involved the A1 group. Many other pathways were not seen with more
caudal injections. These included projections to the locus coeruleus, the inferior and superior colliculi, the most rostral parts of the periaqueductal gray, the entopeduncular nucleus, and the cingulate bundle. These projections may tentatively be attributed to the ventral medulla roughly at the level of the retrofacial nucleus 52. It is also possible that they reflect a topographic organization of the A1 group. A few areas were variably labeled. The nucleus prepositus, the nucleus tegmenti laterodorsalis, the paraventricular nucleus of the thalamus, and the median eminence (Fig. 2C) were never labeled more than lightly, and often not at all. The inferior olive was sometimes unlabeled, or labeled in different places. The ventrobasal thalamus was not usually labeled (see Discussion). The median preoptic nucleus was not labeled in any other case. In addition, with injections of the caudal A1 region there was a dense band of labeling at the border of the area postrema, especially rostrally. This projection, though present, was not so striking in the case described above.
A 1 injections of 6-OHDA-treated animals One group of rats was pretreated with intraventricular injections of 6-OHDA, a neurotoxin relatively selective for catecholamine pathways zl,42. In 5 of these rats the A1 region was successfully injected with tritiated amino acids. One such case is illustrated in Fig. 3. In the thalamus, brainstem, and spinal cord, the labeling was very similar to that seen in normal animals with comparable injections. The hypothalamus, however, was unlabeled except for light labeling in the lateral hypothalamus. The bed nucleus of the stria terminalis was also unlabeled. The other cases gave similar results. The 6-OHDA had no apparent effect except in the hypothalamus and in the bed nucleus of the stria terminalis. In these locations labeling was either absent or greatly attenuated, compared to normal controls. However, the projection to the supraoptic nucleus was sometimes less affected by the 6-OHDA, perhaps because it is relatively remote from the ventricular system.
Injections involving the reticular formation Fig. 4 shows an injection dorsal to the lateral reticular nucleus, involving the caudal portion of the nucleus ambiguus. Labeled fibers crossed to termi-
16
T-
£
/
:QI
• ~.'.
IV|I[Z,L
,
3rd V E N T R I C L E
H~,
;.,"-
:."
"..
,"
'~
. -
i'
~ %.:"
" .~
.'.
~
:
~
"?
v,~-le
.: .-
5
: 4 •`
ME e x t "
:
• o:
i
i!
.
Fig. 2. A: fluorescent cells of the caudal A1 group after injections of H R P into the upper thoracic spinal cord. Granules of H R P reaction product can be seen against the fluorescence ( × 150). B: same cells under brightfield illumination to show the retrograde labeling more clearly ( x 150). C: darkfield photomicrograph of the median eminence in a normal rat after an injection of PHlamino acids into the rostral A1 region. Note the light concentration of silver grains in the internal lamina (ME-int) compared to the external lamina (ME-ext) ( × 200).
nate in the opposite nucleus ambiguus and inferior olive. Light, diffuse labeling was present in the nucleus of the solitary tract, the dorsal motor nucleus of the vagus, and the ipsilateral hypoglossal
nucleus. Labeled fibers heavily innervated the medial and lateral thirds of the ipsilateral facial nucleus, with sparser labeling in the middle third and in the contralateral facial nucleus. Some fibers conti-
17 nued rostrally in the reticular formation to the K611iker-Fuse nucleus, the parabrachial nuclei, and the periaqueductal gray. A few labeled fibers ascended within the trigeminal complex, apparently reflecting slight involvement of the medial aspect of the spinal trigeminal nucleus, pars caudalis. Others ascended in the medial lemniscus and terminated in two regions of the thalamus. One region was just medial to the contralateral medial geniculate nucleus; the other was the
A
lateral portion of the ventrobasal thalamus on both sides. The hypothalamus was not labeled. Except for the labeling in the ventrobasal thalamus, a similar pattern was seen whenever the injection extended dorsal to the lateral reticular nucleus. Often there was light labeling at the medial edge of the nucleus of the solitary tract at rostral levels; this area may actually be the dorsal motor nucleus of the vagus 17,24, which is indistinct at these levels. The pattern of labeling varied in the facial nucleus (see
J
Fig. 3.6-OHDA-treated rat with injection in A1 region. Note near absence of labeling in hypotha!amus. See list of abbreviations after text.
18
A
E
I
10N
LGV~
~
~P
MN
~tML
L1
Fig. 4. Normal rat with injection in lateral reticular formation and caudal nucleus ambiguus. Note that the hypothalamus is unlabeled. See list of abbreviations after text. Discussion). The hypothalamus and the bed nucleus of the stria terminalis were labeled only when the injections involved the A1 group. The most dorsomedially placed injection in the reticular formation shows a distinctive set of projections. There was heavy labeling in the ipsilateral hypoglossal nucleus. This labeling did not represent non-specific spread from the injection site, since it extended for some distance rostrally and caudally, and fell off sharply at the midline. This case was the
only one in the series to show any substantial labeling in the motor trigeminal nucleus or in the middle third of the facial nucleus. No labeling was detectable in the parabrachial region nor at more rostral levels.
Injections of the spinal trigeminal nucleus Three injections heavily involved the magnocellufar layers of the spinal trigeminal nucleus, pars caudalis, without invading the marginal layer. La-
19 beled fibers ascended within and medial to the trigeminal complex to the lateral third of the facial nucleus and to the principal sensory trigeminal nucleus. A few fibers invaded the parabrachial region. Others crossed to innervate rostral parts of the contralateral inferior olive or ascended in the medial lemniscus to the medial portion of the ventrobasal thalamus. The other two injections were placed more caudally; the projections were broadly similar to those illustrated. The thalamus was not labeled, though some labeling was detectable in the medial lemniscus. Both cases also showed a projection to the edge of the area postrema, similar to that seen with more medial injections. In some other cases where the injection invaded the marginal layer of the trigeminal complex, there was labeling in the dorsal half of the submedial thalamic nucleus. This projection was best seen in the 6OHDA-treated rat described above (Fig. 3). Though this case is perhaps not representative due to the drug treatment, the same projection was also seen in normal rats.
Rostral injections
Several injections were made in the ventral medulla just lateral to the rostral third of the inferior olive. These findings have been reported ~2 and will not be described in detail here. In the present context, the most relevant finding was a set of projections to the hypothalamus and the bed nucleus of the stria terminalis, very similar to those arising from the A1 region. The projections from this rostral area, however, were resistant to 6OHDA. The same region also projects to the intermediolateral cell column aT.
Projections to spinal cord - - autoradiography
Most of the cases reported here showed projections to the spinal cord. The details of the labeling patterns could not be closely correlated with injection placement. Most cases, however, fell into two groups with broadly different spinal projections. In one group injections were placed roughly at the level of the retrofacial nucleus, just lateral to the inferior olive. More caudal injections involved the reticular formation dorsal to the lateral reticular nucleus.
Both groups showed diffuse labeling in the ventral horn and intermediate gray at cervical levels. With rostral injections this labeling tended to be ipsilateral; with caudal injections it was bilateral. Both groups also showed labeled fibers descending in the ipsilateral dorsolateral funiculus (sometimes with light labeling in the opposite dorsolateral funiculus) to the intermediolateral cell column. After rostral injections this labeling was bilateral, and extended all the way to upper lumbar levels; at midthoracic levels it was relatively sparse. This projection is thought to be largely serotonergic 47. After caudal injections the projection to the intermediolateral column was restricted to upper thoracic levels, and was almost completely ipsilateral. Sometimes there was a light projection to the most ventral part of the thoracic ventral horn apparently descending in the lateral funiculus. The lumbar ventral horn was also labeled. After rostral injections, this labeling was found at the L3 level and below, mainly in the lateral part of the ipsilateral ventral horn. Sometimes the labeling extended into sacral segments. After caudal injections the upper lumbar ventral horn was labeled, bilaterally or contralaterally. The most distinctive labeling in the spinal cord occurred when injections were made at intermediate levels, involving the nucleus ambiguus. These injections produced intense labeling in the midcervical ventral horn concentrated in the phrenic motor nucleus 44. This nucleus was also labeled, though much more lightly, whenever the injection encroached slightly on the nucleus ambiguus. The fibers appeared to descend in the lateral and ventral funicull. The projection was ordinarily bilateral. In a few cases it was predominantly contralateral; the ipsilateral fibers may have been severed by the injection needle. Injections heavily involving the nucleus ambiguus also showed bilateral projections to all levels of the intermediolateral column. Labeling in the spinal cord was not detectably changed by 6-OHDA. Furthermore, no projection could be clearly attributed to the A1 region. In one case a large injection blanketed the caudal A1 region containing the spinally projecting A1 cells (see below). Labeling was present bilaterally in the ventral funiculi and ipsilaterally in the dorsolateral funiculus to midthoracic levels, but except for some
20
Fig. 5. A: low-power photomicrograph of the medulla oblongata (~: 18). The area enclosed by the rectangle is shown at higher magnification in B. B: this photomicrograph shows the nucleus ambiguus heavily labeled after spinal injections of HRP (× 180). Note that fluorescent A1 cells are not labeled at this level. C: fluorescent cells of the A2 group after spinal injections of HRP ( × 270). Some cells are lightly labeled (black arrows) and other cells are not labeled (white arrows). In no case are cells of the A1 or A2 group heavily labeled, although labeling elsewhere is both heavy and extensive, labeling in the upper cervical ventral horn, no labeling was detectable in the gray matter.
Retrograde labeling I n an attempt to retrogradely label the A1 group f r o m the spinal cord, we combined H R P with histofluorescence. The new technique 10 allowed the histofluorescence and the retrograde labeling to be viewed simultaneously, instead o f successively as in earlier studies48,49, 56. After large injections of H R P into the upper thoracic spinal cord, a small n u m b e r of fluorescent cells o f the A1 group were lightly labeled with H R P (Fig. 2A, B). Almost all of these cells were at the level o f the pyramidal decussation, but even in this region the retrogradely labeled cells
were in the minority. The great majority of fluorescent cells were not labeled. Non-fluorescent cells in m a n y areas were more heavily labeled, as in the nucleus ambiguus (Fig. 5B). Likewise, a few fluorescent cells were lightly labeled in the A2 group (Fig. 5C), in the commissural portion o f the nucleus o f the solitary tract. Most fluorescent cells were not labeled. A few non-fluorescent cells were also labeled in the same region; typically they were slightly larger than the fluorescent cells. Three control rats received i.v. injections o f the same a m o u n t o f H R P ; n o labeled cells were f o u n d in the A1 or A 2 groups. Hence the slight labeling seen after spinal injections was not the result of
21 endogenous peroxidase, nor of blood-borne HRP transported from the circumventricular organs. It remains possible that anterograde labeling, appearing in axosomatic terminals, was mistaken for light retrograde labeling. It might be argued that the A1 group is peculiarly resistant to labeling by HRP. However, similar experiments have labeled the A1 group from the medial preoptic area za. The A1 group might project mainly to levels below the injection site. However, the large injections used (4 × 1 /A) apparently labeled fibers of passage as well as terminals, since the nucleus tegmenti laterodorsalis was routinely labeled; this nucleus projects to the sacral intermediolateral cell column 51. It remains possible that the A1 group projects mainly to cervical levels. Previously we had used a technique whereby photomicrographs of fluorescent cells were compared with retrograde labeling in the same sections 56. Despite an intensive search throughout the A1 and A2 groups, no convincing examples of retrogradely labeled fluorescent ceils were found. The same technique readily and reliably demonstrated many spinally projecting cells in the A5, A6, and A7 groups 48,49,56, and in dopaminergic cells of the diencephalon (unpublished results; see also refs. 5, 6 and 36). In retrospect it is clear why these earlier experiments were unsuccessful. The cells in question are few in number, lightly labeled, and surrounded by anterograde labeling in fiber tracts. These conditions make it nearly impossible to conclusively identify a small clump of labeling with a fluorescent cell unless the fluorescence can be seen simultaneously. Usually, however, the earlier (and more time-consuming) technique is more sensitive, because it permits maximum labeling with the HRP. When the fluorescence is viewed simultaneously, the HRP reaction product in a heavily labeled cell can obscure the fluorescence. Even if the fluorescence is visible, it can be difficult to document photographically. DISCUSSION Projections to the forebrain The septal region receives noradrenergic fibers from at least two sources46, 61. Fine caliber fluorescent fibers originate in the locus coeruleus. Coarser
fibers originate outside the locus coeruleus, and have a different distribution. They are most densely concentrated in the ventral half of the bed nucleus of the stria terminalis, with lesser concentrations in several other regions 46,~1. The projection to the bed nucleus of the stria terminalis almost certainly arises at least in part from the A1 group. The projection shown by autoradiography matches the location of noradrenergic fibers4~,61,76 and is sensitive to 6-OHDA. Fluorescent cells in the A1 group (and A2 group as well) can be retrogradely labeled by injections centered in the bed nucleus of the stria terminalis (unpublished results). Other areas receiving the coarse, non-coeruleus fibers include the lateral septal nucleus and the nucleus of the diagonal band46,6L No projections to these areas were seen in our autoradiographic material. The autoradiographic method may not be sensitive enough to reveal these sparser projections. Retrograde transport studies indicate that the A 1 group projects to the medial preoptic area 23. In our material this projection appears rather sparse at best, and is not always detectable. According to immunohistochemical evidence as well, noradrenergic fibers are relatively sparse in this region7% Only in one case (Fig. 1) was there a substantial projection to the median preoptic nucleus (including the nucleus triangularis septi as termed by Lindvall and Stenevi46). This labeling does not correspond well to fluorescent fibers; it may arise from cells near the rostral end of the A1 group. This region was not fully explored in the present study. The A1 region projects to several hypothalamic sites: the dorsomedial nucleus, the dorsal hypothalamic area, the paraventricular nucleus, the supraoptic nucleus, and the median eminence. These projections correspond to noradrenergic terminal fields7~, and are sensitive to 6-OHDA. Studies combining retrograde tracers with catecholamine markers have shown that the A1 group projects to hypothalamic levels 2z,77 (also unpublished results). Furthermore, lesions of the A1 group lead to substantial losses of hypothalamic norepinephrine 69. For all these reasons, it is likely that the hypothalamic projections from the A1 region arise principally, if not exclusively, from catecholamine cells. Hypothalamic projections also arise from the
22 ventral medullaSL Since this system is resistant to 6O H D A (but not to 5,7-dihydroxytryptamine - - a serotonin neurotoxin) 52, it is probably distinct from the A1 group. In addition, it may include the epinephrine-containing cells of the C1 group 34, which reportedly project to the hypothalamus 68 and are resistant to 6-OHDA 39,72. Previous reports suggest that the A1 group projects to the arcuate nucleus and the ventromedial nucleus 69. The present results do not support these conclusions, though light projections cannot be excluded. The ventromedial nucleus receives relatively little noradrenergic innervation 76. The arcuate nucleus may receive noradrenergic innervation from the A2 group of the dorsomedial medulla 73. An H R P study of the median eminence found no retrogradely labeled cells in the medulla sl. This negative result may be related to the very small injection volumes used. Although a projection to the internal lamina of the median eminence was evident in our autoradiographic material (Fig. 2C), it was detectable only after the largest injections, and was never very heavy. The neurons responsible are probably either few in number or widely scattered. In the thalamus, labeling was often found in a wedge-shaped region between the medial geniculate nucleus and the pretectum, contralateral to the injection (Fig. 3C). This region may be part of the posterior complex of the thalamus. The pathway apparently arises from the lateral part or" the reticular formation, and possibly from the marginal layer of the spinal trigeminal nucleus. The magnocellular layers show little projection to the thalamus2L In one case with labeling in the medial portion of the ventrobasat thalamus, the injection extended into pars interpolaris, which does project to the thalamus 27. Surprisingly, 3 cases have labeling in the lateral portion of the ventrobasal thalamus (e.g. Figs. 1, 4). Since this lateral region represents the body rather than the face z2, this pathway can hardly be attributed to the trigeminal system. An HRP study suggests that it arises in 'pars dorsalis' of the lateral reticular nucleus zg. The projection to the submedial nucleus apparently arises principally, if not exclusively, from the marginal layer of the spinal trigeminal nucleus 19. It is bilateral but predominantly contralateral.
In many cases there was labeling in the paracentral and centrolateral nuclei, and in parts of the parafascicular nucleus. However, this labeling was patchy and difficult to compare from one brain to the next. The paraventricular nucleus of the thalamus was labeled only rarely, and very lightly; the origin of this pathway is unclear. The lateral dorsal thalamic nucleus receives a projection from the ventral medulla 52. Brain stem projections Although the present study found no projections of the A1 group within the brainstem, this result should not be taken at face value. Some projections may have been obscured by non-catecholaminergic pathways. Furthermore, the 6-OHDA experiments must be interpreted cautiously. Not only does this drug have non-specific effects32, but even its specific effects are more complete in some regions than in others4L For example, the projection to the supraoptic nucleus was relatively resistant to 6-OHDA, possibly because this pathway is remote from the ventricular system. H R P studies have suggested that the A1 group projects to the locus coeruleus 15,16. Our material shows such a projection only with fairly rostral injections. There is no evidence that this projection is noradrenergic, but it may include adrenergic fibers from the C1 group a4. One system of brainstem projections was common to most of the cases reported here. Labeled fibers ascended bilaterally through the reticular formation, especially ventral to the retrofacial nucleus, to innervate the ipsilateral facial nucleus. Sometimes the opposite facial nucleus was lightly labeled as well. Passing dorsally over the facial nucleus and superior olive, labeled fibers continued rostrally to the parabrachial region. Most of these fibers passed medial to the motor trigeminal nucleus. A smaller contingent took a more ventrolateral course through the region of the A5 catecholamine cell group; these fibers looped rostrally around the principal sensory nucleus of the trigeminal nerve to the parabrachial region. Here the heaviest labeling lay in the K611iker-Fuse nucleus, with light to moderate labeling in the medial and lateral parabrachial nuclei. Some fibers may have entered the cerebellum along with the ventral spinocerebellar tract.
23 Some labeled fibers continued past the parabrachial region in the central tegmental tract, to innervate the lateral part of the periaqueductal gray. Although the other brain stem projections were predominantly ipsilateral, the projection to the periaqueductal gray was heavier - - and slightly more dorsal - - on the contralateral side. Broadly similar findings have been reported in the cat 37,41. On the basis of HRP studies in both the cata3, 41 and the rat 2, this system can largely be attributed to the reticular formation dorsM to the lateral reticular nucleus. There may also be a contribution from the trigeminal complex to the parabrachial region 13. Projections to the facial nucleus were topographically organized. As reported by Erzurumlu and Killackey z6, magnocellular layers of the spinal trigeminal nucleus, pars caudalis project to the lateral division (using the nomenclature of Martin and Lodge55), and possibly to the dorsal division. The marginal layer appeared to project to the dorsal division in particular (Fig. 3; see also ref. 13). Injections just medial to the trigeminal complex labeled not only the lateral and dorsal divisions but also the medial and ventromedial divisions, with sparse labeling in the opposite facial nucleus. Similar projections were reported in a degeneration study, but were attributed to a spinofacial pathway z6. With more medial injections which involved the reticular formation, the medial division of the facial nucleus was labeled more heavily; the most medial injections showed projections almost exclusively to the medial and ventromedial divisions. Only the most dorsomedial injection showed a substantial projection to the intermediate division of the facial nucleus. This case also showed prominent labeling in the ipsilateral hypoglossal nucleus and in both motor trigeminal nuclei, but not at more rostral levels. Many cases showed a projection to the contralateral inferior olive, arising either from the spinal trigeminal nucleus or from cells just medial to it. No projections to the inferior olive were seen from injection sites medial to the nucleus ambiguus. The details of the labeling varied, but two areas were most commonly labeled: the medial edge of the dorsal accessory inferior olive and a small region of the caudal medial accessory inferior olive, just me-
dial to the lateral tip. Sometimes the medial accessory inferior olive was labeled bilaterally; in other cases the ipsilateral projection either was absent oi' could not be distinguished from the nearby injection site. In the cat, roughly comparable regions of the inferior olive receive a projection from the spinal trigeminal nucleus, pars caudalis 3. In the rat, however, this projection reportedly arises not from pars caudalis but from pars interpolaris1% Unfortunately, the present study cannot resolve this question, since the labeling may also have arisen from the lateral reticular nucleus 12. The lateral reticular nucleus projects so heavily to the cerebellum that it would have obscured any contribution from the A1 group. Such a projection is highly unlikely, however. In one rat, multiple 1/~1 injections of H R P into the cerebellum failed to label any cells in the A1 region outside of the lateral reticular nucleus and the trigeminal complex (unpublished results). Similar findings have been reported using antibodies to dopamine beta-hydroxylase as a retrograde tracer 70. Senba et al. 75 have reported that the A1 group projects to the spinal trigeminal nucleus. The present study offers little support for this conclusion. When there was labeling in more rostral parts of the spinal trigeminal nucleus, it could be attributed to the well known intranuclear projection arising from the spinal trigeminal nucleus, pars caudalis 4a, which was invaded by some injections. The H R P injections of Senba et al. 75 evidently spilled either into the cerebellum or into the inferior cerebellar peduncle, since the contralateral inferior olive was retrogradely labeled (see their Fig. 3). Likewise, the labeled cells assumed to belong to the A1 group may have belonged instead to the lateral reticular nucleus, since no evidence was offered that they were catecholaminergic 75. The nucleus of the solitary tract and the dorsal motor nucleus of the vagus consistently received a light projection after injections of [3H]amino acids in the A1 region. Typically this labeling was slightly heavier on the ipsilateral side, and heaviest in the ventrolateral and intermediate subnuclei. In many cases there was a dense concentration of labeling at the boundary of the area postrema, especially near its rostral pole. This finding was somewhat variable,
24 however, and the source of the projection could not be identified. Many cases also showed light to moderate labeling in the ipsilateral hypoglossal nucleus. Although there is a dense catecholaminergic innervation of the solitary complex which arises from the A2 and A5 cell groups48,v s, our results regarding whether the A1 cell group projects to this area are inconclusive. [~H]amino acid injections in the region of the A1 cell group also involve neurons of the nucleus ambiguus. This complicates the interpretation of the present results because the axons of the ambiguus neurons course dorsomedially toward the nucleus of the solitary tract before curving laterally to exit the medulla in the vagus nerve 14. The projection from the A1 region to the dorsomedial medulla was not noticeably affected by 6OHDA. Hence, the present study cannot determine whether the A1 group projects to the solitary complex, as reported by Blessing et al. 9, who found that A1 cells could be retrogradely labeled after H R P injections in the dorsomedial medulla.
Spinal cord Dahlstr6m and Fuxe were the first to report a spinal projection from the A1 group zl. After spinal lesions, some cells in the rostral part of the AI group (and some in the A2 group as well) fluoresced more brightly than usual. This response was assumed to be a retrograde reaction to axotomy. Since other cell groups did not show such a response, Dahlstr6m and Fuxe proposed that the A1 and A2 groups were the principal or exclusive sources of spinal catecholamines 21. In retrospect, however, this conclusion was not completely consistent with their data. The supposed reaction to axotomy was associated mainly with lesions of the ventral white matter, which supplies catecholamines to the ventral horn. Lesions of the dorsolateral funiculus, which supplies the intermediolateral column and the dorsal horn, did not elicit this reaction. Hence other cell groups presumably contributed to this more dorsal pathway. Later studies confirmed that the A5, A6, and A7 groups all project to the spinal cord10,48,49,56,65,74, as do some of the dopaminergic cells of the diencephalon 5,6,36. In particular, the catecholamine fibers in the ventral white matter arise principally, if not
exclusively, from the locus coeruleus and the subcoeruleus group 65. Hence there is little reason to regard lesion-induced increases of fluorescence in the A1 group as a result of axotomy. While such an interpretation was reasonable at first, the increased fluorescence probably reflected some less specific reaction. For example, a similar increase in fluorescence can result from chemical sympathectomy 53. Satoh et al. briefly noted a spinal projection from the A1 group, using a combination of H R P and monoamine oxidase staining TM. As noted elsewhere 50, this report is difficult to interpret because the monoamine oxidase stain is not specific for catecholamine cells. In our hands the Al-spinal projection proved difficult to demonstrate by retrograde transport, even though the same techniques readily demonstrated spinal projections of other catecholamine cell groups4S,49, 56. After these results were first published49, 56, others reported similar findingslO, 80. In the rabbit, the spinal projection arises from the rostral rather than the caudal portion of the A1 group, and apparently not at all from the A2 group TM. In the rat, experiments using antibodies to dopamine beta-hydroxylase as a retrograde label failed to detect any spinal projection from either cell group 80, although some of the epinephrine-containing cells might be expected to transport the antibodies 35. In any case, it is now clear that the A1 and A2 groups provide at most a minor contribution to spinal norepinephrine. Opposite results have been reported in the cat, on the basis of lesion studies 2s. We find the evidence unconvincing, since lesions of the A1 region probably destroy descending fibers from other noradrenergic cell groups. Only one cat received a lesion of the A5 group, and this lesion was placed according to the atlas of Poitras and Parent vl. However, in several respects this atlas is at variance with the observations of others 7,45 (also unpublished results). In the cat, the A5 group extends further caudally than it does in the rat, and lies mostly lateral and ventral to the facial nucleus 45. Hence the lesion reported 2s may have been too rostral. While it has often been assumed that the A1 group projects to the intermediolateral cell column, there is no evidence for this assumption. The present study demonstrates a projection to the intermedio-
25 lateral column, but not from catecholamine cells. It is resistant to 6-OHDA, and is present in cases in which the injection avoids the A1 group (Fig. 4). It probably arises in or near the nucleus ambiguus. This pathway may be equivalent to one reported in the opossum 54. The spinal termination of the A1 fibers will be extremely difficult to identify. Lesions of the A1 group will almost certainly destroy fibers of passage from more rostral catecholamine cells, such as the A5 cell group 48. The autoradiographic method is not selective for catecholamines, and may not be sensitive enough for such a sparse pathway. The projection to the phrenic motor nucleus apparently arises from the nucleus ambiguus, as defined here. This nucleus is probably part of the ventral respiratory group, which has been studied in the cat by electrophysiological means4,~9, 60. If so, it would correspond to the 'nucleus para ambiguus' of Kalia 40. Its connections in the rat appear similar to those of inspiratory cells in the cat 6°, except that the spinal projection is bilateral rather than predominantly crossed. If this correspondence is confirmed physiologically, it may be necessary to abandon the nomenclature used here. The 'nucleus ambiguus' is ordinarily assumed to be a source of efferent fibers in the vagus and glossopharyngeal nerves. However, the efferent neurons in the rat do not fit neatly into
well-defined nucleP TM, and their relationship to the spinally-projecting cells of the nucleus ambiguus is unknown. They are unlikely to be the same cells 4, but this question has evidently not been addressed in the rat. The functional role of the A1 group is unknown. However, the projection to magnocellular neurosecretory cells of the hypothalamus ~7 suggests that the A1 group has a rather direct influence on the secretion of oxytocin or vasopressin. Indeed, it may be no accident that tb_e A1 cells are mixed together with cardiovagal and respiratory neurons. If they receive some of the same afferent inputs, the A1 cells would be in an ideal position to relay information from baroreceptors and chemoreceptors 82, and possibly from other visceral receptors, to the neuroendocrine system.
ABBREVIATIONS
DTN EN F FI FN G GP IC Ic ICA ICol III IML ION IP IV IVN KF LC LD LGD LGV LH
AC AD AM Arab AP ARH AV A1 A5 BST CG CL C1 CM Cm CPu D DBB DLL DMH DR
anterior commissure anterodorsal nucleus anteromedial nucleus nucleus ambiguus area postrema arcuate nucleus of the hypothalamus anteroventral nucleus A1 catecholamine cell group A5 catecholamine cell group bed nucleus of the stria terminalis central gray centrolateral nucleus lateral cuneate nucleus centre median nucleus medial cuneate nucleus caudate-putamen nucleus of Darkschewitsch diagonal band of Broca dorsal nucleus of the lateral lemniscus dorsomedial nucleus of the hypothalamus dorsal raphe nucleus
ACKNOWLEDGEMENTS This work was supported by USPHS grant HL25449 and a grant-in-aid from the American Heart Association with funds contributed in part by the Missouri Heart Association. Scott McKellar was supported by N I H grant HL-07275. We thank Tinh Hon Ha and Sang Van Pham for technical help, and W. W. Blessing for providing an account of his HRP/histofluorescence procedure. A.D.L. is an Established Investigator of the American Heart Association.
dorsaltegmental nucleus entopeduncularnucleus fornix fimbria facial nucleus gracile nucleus globus pallidus internal capsule nucleus intercalatus interstitialnucleus of Cajal inferiorcolliculus oculomotor nucleus intermediolateral cell column inferiorolivary nucleus interpeduncularnucleus trochlear nucleus inferiorvestibular nucleus KOlliker-Fusenucleus locus coeruleus laterodorsalnucleus dorsallateral geniculate nucleus ventrallateral geniculate nucleus lateral habenula
26 LM LP LRN LS LVN MD ME MeV MGN MgPO MH ML MM1 MMp MnPO MoV MPN MS MV NPC NTS PBG PB1 PBm PC PF PG PH PMN POM PON POT PSV PT PTA PVH PVT RB
lateral mammillary nucleus lateral posterior nucleus lateral reticular nucleus lateral septum lateral vestibular nucleus medial dorsal nucleus median eminence mesencephalic trigeminal nucleus medial geniculate nucleus magnocellular preoptic nucleus medial habenular nucleus medial lemniscus medial mammillary nucleus, pars lateralis medial mammillary nucleus, pars posterior median preoptic nucleus motor trigeminal nucleus medial preoptic nucleus medial septum medial vestibular nucleus nucleus of the posterior commissure nucleus of the solitary tract parabigeminal nucleus lateral parabrachial nucleus medial parabrachial nucleus paracentral nucleus parafascicular nucleus pontine gray nucleus prepositus hypoglossi phrenic motor nucleus posterior complex, medial division preolivary nucleus posterior complex of the thalamus principal sensory nucleus of the trigeminal paratenial nucleus pretectal area paraventricular nucleus of the hypothalamus paraventricular nucleus of the thalamus restiform body
Re
RF Rh RM RN RT SCn SCol SCR SF SM SNc SN1 SNr SNV SO SON STn SUM SuV SV TBN TLD TRP VA VB VI VIIg VL VLL VM VMb VMH VTN X XII ZI
REFERENCES 1 Andrezik, J. A., Chan-Palay, V. and Palay, S. L., The nucleus paragigantocellularis lateralis in the rat: conformation and cytology, Anat. Embryol., 161 (1981) 355-371. 2 Andrezik, J. A., Chan-Palay, V. and Palay, S. L., The nucleus paragigantocellularis lateralis in the rat: demonstration of afferents by the retrograde transport of horseradish peroxidase, Anat. Embryol., 161 (1981) 373-390. 3 Berkley, K. J. and Hand, P. J., Projections to the inferior olive of the cat: comparisons of input from the gracile, cuneate and the spinal trigeminal nuclei, J. eomp. NeuroL, 180 (1978) 253-264. 4 Bianchi, A. L., Localisation et 6rude des neurones respiratoires bulbaires. Mise en antidromique stimulation ou Vagale, J. Physiol. (Paris), 63 (1971) 5-40. 5 Bj6rklund, A. and Skagerberg, G., Evidence for a major spinal cord projection from the diencephalic A l l dopamine cell group in the rat using transmitter-specific fluorescent retrograde tracing, Brain Research, 177 (1979) 170-175. 6 Blessing, W. W. and Chalmers, J. P., Direct projection of catecholamine (presumably dopamine)-containing neu-
7 8
9
10
11 12
nucleus reuniens retrofacial nucleus rhomboid nucleus nucleus raphe magnus red nucleus reticular nucleus suprachiasmatic nucleus superior colliculus superior central nucleus of the raphe septofimbrial nucleus submedial nucleus substantia nigra, pars cornpacta substantia nigra, pars lateralis substantia nigra, pars reticularis spinal trigeminal nucleus superior olive supraoptic nucleus subthalamic nucleus supramammillary nucleus subtrigeminal nucleus superior vestibular nucleus nucleus of the trapezoid body nucleus tegmenti laterodorsalis pontine reticulotegmental nucleus ventral anterior nucleus ventrobasal complex abducens nucleus internal genu of the facial nerve ventrolateral nucleus ventral nucleus of the lateral lemniscus ventromedial nucleus ventrornedial complex, basal nucleus ventromedial nucleus of the hypothalamus ventral tegmental nucleus dorsal motor nucleus of the vagus hypoglossal nucleus zona incerta
rons from hypothalamus to spinal cord, Neurosci. Lett., 11 (1979) 35-40. Blessing, W. W., Frost, P. and Furness, J. B., Catecholamine cell groups of the cat medulla oblongata, Brain Research, 192 (1980) 69-75. Blessing, W. W., Furness, J. B., Costa, M. and Chalmers, J. P., Localization of catecholamine fluorescence and retrogradely transported horseradish peroxidase within the same nerve cell, Neurosei. Lett., 9 (1978) 311-315. Blessing, W. W., Furness, J. B., Costa, M., West, M. J. and Chalmers, J. P., Projection of ventrolateral medullary (AI) catecholamine neurons toward nucleus tractus solitarius, Cell Tiss. Res., 220 (1981) 27-40. Blessing, W. W., Goodchild, A. K., Dampney, R. A. L. and Chalrners, J. P., Cell groups in the lower brain stem of the rabbit projecting to the spinal cord, with special reference to catecholarnine-containing neurons, Brain Research, 221 (1981) 35-55. Brodal, A., The Reticular Formation of the Brain Stem: Anatomical Aspects and Functional Correlations, Oliver and Boyd, London, 1957. Brown, J. T., Chan-Palay, V. and Palay, S. L., A study of afferent input to the inferior olivary complex in the rat by
27 retrograde axonal transport of horseradish peroxidase, J. comp. NeuroL, 176 (1977) 1-22. 13 Burton, H., Craig, A. D., Jr., Poulos, D. A. and Molt, J. T., Efferent projections from temperature sensitive recording loci within the marginal zone of the nucleus caudalis of the spinal trigeminal complex in the cat, J. comp. Neurol., 183 (1979) 753-778. 14 Cajal, S., R.am6n, Y., Histologie du Syst~me Nerveux de l'Homme et des Vert6br6s, VoL I, Maloine, Paris, 1909. 15 Cedarbaum, J. M. and Aghajanian, G. K., Afferent projections to the rat locus coeruleus as determined by a retrograde tracing technique, J. comp. NeuroL, 178 (1978) 1-16. 16 Clavier, R. M., Afferent projections to the self-stimulation regions of the dorsal pons, including the locus coeruleus, in the rat as demonstrated by the horseradish peroxidase technique, Brain Res. Bull., 4 (1979) 497-504. 17 Contreras, R. J., Gomez, M. M. and Norgren, R., Central origins of cranial nerve parasympathetic neurons in the rat, J. comp. Neurol., 190 (1980) 373-394. 18 Cowan, W. M., Gottlieb, D. I., Hendrickson, A. E., Price, J. L. and Woolsey, T. A., The autoradiographic demonstration of axonal connections in the central nervous system, Brain Research, 37 (1972) 21-51. 19 Craig, A. D., Jr. and Burton, H., Spinal and medullary lamina I projection to nucleus submedius in medial thalamus: a possible pain center, J. NeurophysioL, 45 (1981) 443-466. 20 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine-containing neurons in the central nervous system: I. Demonstration of monoamines in the cell bodies of brain stem neurons, Acta physioL scand., 62, Suppl. 232 (1964) 1-55. 21 Dahlstr~Sm, A. and Fuxe, K., Evidence for the existence of monoamine neurons in the central nervous system: II. Experimentally induced changes in the intraneuronal amine levels of bulbospinal neuron systems, Acta physiol. scand., 64, Suppl. 247 (1965) 1-32. 22 Davidson, N., The projection of afferent pathways on the thalamus of the rat, J. comp. NeuroL, 124 (1965) 377-390. 23 Day, T. A., Blessing, W. and Willoughby, J. O., Noradrenergic and dopaminergic projections to the medial preoptic area of the rat. A combined horseradish peroxidase/catecholamine fluorescence study, Brain Research, 193 (1980) 543-548. 24 Dennison, S. J., Merritt, V. E., Aprison, M. H. and Felten, D. L., Redefinition of the location of the dorsal (motor) nucleus of the vagus in the rat, Brain Res. Bull., 6 (1981) 77-81. 25 deOlmos, J., Hardy, H. and Helmet, L., The afferent connections of the main and the accessory olfactory bulb formations in the rat: an experimental HRP study, J. comp. Neurol., 181 (1978) 213-244. 26 Erzurumlu, R. S. and Killackey, H. P., Efferent connections of the brainstem trigeminal complex with the facial nucleus of the rat, J. comp. Neurol., 188 (1979) 75-86. 27 Erzurumlu, R. S. and Killackey, H. P., Diencephalic projections of the subnucleus interpolaris of the brainstem trigeminal complex in the rat, Neuroscience, 5 (1980) 1891-1901. 28 Fleetwood-Walker, S. M. and Coote, J. H., The contribution of brain stem catecholamine cell groups to the innervation of the sympathetic lateral cell column, Brain Research, 205 (1981) 141-155.
29 Fukushima, T. and Kerr, F. W. L., Organization of trigeminothalamic tracts and other thalamic afferent systems of the brainstem in the rat: presence of gelatinosa neurons with thalamic connections, J. comp. Neurol., 183 (1979) 169-184. 30 Furness, J. B., Costa, M. and Blessing, W. W., Simultaneous fixation and production of catecholamine fluorescence in central nervous tissue by perfusion with aldehydes, Histoehem. J., 9 (1977) 745-750. 31 Garattini, S. and Samanin, R., Selective neurotoxicity of 6-hydroxydopamine. In L. Roizin, H. Shiraki and N. Gr~evi6 (Eds.), Neurotoxicology, Raven Press, New York, 1977, pp. 15-23. 32 Hedreen, J., Neuronal degeneration caused by lateral ventricular injection of 6-hydroxydopamine. I. The ventral premammillary nucleus and other cell groups, Brain Research, 157 (1978) 129-135. 33 Henkel, C. K. and Edwards, S. B., The superior colliculus control of pinna movements in the cat: possible anatomical connections, J. comp. NeuroL, 182 (1978) 763-776. 34 H6kfelt, T., Fuxe, K., Goldstein, M. and Johansson, O., Immunohistochemical evidence for the existence of adrenaline neurons in the rat brain, Brain Research, 66 (1974) 235-251. 35 H6kfelt, T., Goldstein, M., Fuxe, K., Johansson, O., Verhofstad, A., Steinbusch, H., Penke, B. and Vargas, J., Histochemical identification of adrenaline containing ceils with special reference to neurons. In K. Fuxe, M. Goldstein, B. Hokfelt and T. Hokfelt (Eds.), Central Adrenaline Neurons: Basic Aspects and Their Role in Cardiovascular Functions, Pergamon Press, Oxford, 1979, pp. 19-47. 36 H6kfelt, T., Phillipson, O. and Goldstein, M., Evidence for a dopaminergic pathway in the rat descending from the A l l cell group to the spinal cord, Acta physiol. scand., 107 (1979) 393-395. 37 Holstege, G., Kuypers, H. G. J. M. and Pekker, .L J., The organization of the bulbar fibre connections to the trigeminal, facial, and hypoglossal motor nuclei. II. An autoradiographic tracing study in cat, Brain, 100 (1977) 265-286. 38 Howe, P. R. C., Costa, M., Furness, J. B. and Chalmers, J. P., Simultaneous demonstration of phenylethanolamine N-methyltransferase immunofluorescent and catecholamine fluorescent nerve cell bodies in the rat medulla oblongata, Neuroscience, 5 (1980) 2229-2238. 39 Jonsson, G., Fuxe, K., I-I6kfelt, T. and Goldstein, M., Resistance of central phenylethanolamine-N-methyl transferase containing neurons to 6-hydroxydopamine, Med. Biol., 54 (1976) 421-426. 40 Kalia, M. P., Anatomical organization of central respiratory neurons, Ann. Rev. Physiol., 43 (1981) 105-120. 41 King, G. W., Topology of ascending brainstem projections to nucleus parabrachialis in the cat, J. comp. NeuroL, 191 (1980) 615-638. 42 Kostrzewa, R. M. and Jacobowitz, D. M., Pharmacological actions of 6-hydroxydopamine, PharmacoL Rev., 28 (1974) 199-288. 43 Kruger, L., Saporta, S. and Feldman, S. G., Axonal transport studies of the sensory trigeminal complex. In D. J. Anderson and B. M. Matthews (Eds.), Pain in the Trigeminal Region, Elsevier/North-Holland Biomedical Press, Amsterdam, 1977, pp. 191-202. 44 Kuzuhara, S. and Chou, S. M., Localization of the
28 phrenic nucleus in the rat: a HRP study, Neurosei. Lett., 16 (1980) 119-124. 45 Lackner, K. J., Mapping of monoamine neurones and fibres in the cat lower brainstem and spinal cord, Anat. EmbryoL, 161 (1980) 169-195. 46 Lindvall, O. and Stenevi, U., Dopamine and noradrenaline neurons projecting to the septal area in the rat, Cell Tiss. Res., 190 (1978) 383-407. 47 Loewy, A. D. and McKellar, S., Serotonergic projections from the ventral medulla to the intermediolateral cell column in the rat, Brain Research, 211 (1981) 146-152. 48 Loewy, A. D., McKellar, S. and Saper, C. B., Direct projections from the A5 catecholamine cell group to the intermediolateral cell column, Brain Research, 174 (1979) 309-314. 49 Loewy, A. D., McKellar, S., Swensson, E. E. and Panneton, W. M., Onset of hypertension in spontaneously hypertensive rats despite the depletion of spinal cord catecholamines, Brain Research, 185 (1980) 449-454. 50 Loewy, A. D. and Neil, J. J., The role of descending monoaminergic systems in the central control of blood pressure, Fed. Proc., 40 (1981) 2778-2785. 51 Loewy, A. D., Saper, C. B. and Baker, R. P., Descending projections from the pontine micturition center, Brain Research, 172 (1979) 533-538. 52 Loewy, A. D., Wallach, J. H. and McKellar, S., Efferent connections of the ventral medulla oblongata in the rat, Brain Res. Rev., 3 (1981) 63-80. 53 Lorez, H. P., Catecholarnine fluorescence in the brains of chemically sympathectomized adult rats : increased fluorescence in bulbospinal neurons, Brain Research, 128 (1977) 361-368. 54 Martin, G. F., Humbertson, A. O., Laxson, C. and Panneton, W. M., Evidence for direct bulbospinal projections to laminae IX, X and the intermediolateral cell column. Studies using axonal transport techniques in the North American opossum, Brain Research, 170 (1979) 165-171. 55 Martin, M. R. and Lodge, D., Morphology of the facial nucleus of the rat, Brain Research, 123 (1977) 1-12. 56 McKellar, S. and Loewy, A. D., Spinal projections of norepinephrine-containing neurons in the rat, Neurosci. Abstr., 5 (1979) 344. 57 McKellar, S. and Loewy, A. D., Organization of some brain stem afferents to the paraventricular nucleus of the hypothalamus in the rat, Brain Research, 217 (1981) 351-357. 58 Meessen, H. and Olszewski, J., A Cytoarchitectonic Atlas of the Rhombencephalon of the Rabbit, Karger, Basel, 1949. 59 Merrill, E. G., The lateral respiratory neurones of the medulla: their associations with nucleus ambiguus, nucleus retroambigualis, the spinal accessory nucleus and the spinal cord, Brain Research, 24 (1970) 11-28. 60 Merrill, E. G., Finding a respiratory function for the medullary respiratory neurons. In R. Bellairs and E. G. Gray (Eds.), Essays on the Nervous System: A Festschri~t for Professor at. Z. Young, Clarendon Press, Oxford, 1974, pp. 451-486. 61 Moore, R. Y., Catecholamine innervation of the basal forebrain: I. The septal area, J. comp. NeuroL, 177 (1978) 665-683. 62 Moore, R. Y. and Bloom, F. E., Central catecholamine neuron systems: Anatomy and physiology of the norepinephrine and epinephrine systems, Ann. Rev. Neurosci., 2 (1979) 113-168.
63 Nicholson, J. E. and Severin, C. M., The superior and inferior salivatory nuclei in the rat, Neurosci. Lett., 21 (1981) 149-154. 64 Nosaka, S., Yamamoto, T. and Yasunaga, K., Localization of vagal cardioinhibitory preganglionic neurons within rat brain stem, J. comp. Neurol., 186 (1979) 79-92. 65 Nygren, L. and Olson, L., A new major projection from locus coeruleus: the main source of noradrenergic nerve terminals in the ventral and dorsal columns of the spinal cord, Brain Research, 132 (1977) 85-93. 66 O'Donohue, T. L. Crowley, W. R. and Jacobowitz, D. M., Biochemical mapping of the noradrenergic ventral bundle projection sites : evidence for a noradrenergic-dopaminergic interaction, Brain Research, 172 (1979) 87-100. 67 Palkovits, M. and Jacobowitz, D. M., Topographic atlas of catecholamine and acetylcholinesterase-containing neurons in the rat brain, J. eomp. Neurol., 157 (1974) 29-42. 68 Palkovits, M., Mezey, E., Zaborszky, L., Feminger, A., Versteeg, D. H. G., Wijnen, H. J. L. M., De Jong, W., Fekete, M. I. K., Herman, J. P. and Kanyicska, B., Adrenergic innervation of the rat hypothalamus, Neurosei. Lett., 18 (1980) 237-243. 69 Palkovits, M., Zaborszky, L., Feminger, A., Mezey, E., Fekete, M. I. K., Herman, J. P., Kanyicska, B. and Szabo, D., Noradrenergic innervation of the rat hypothalamus: Experimental biochemical and electron microscopic studies, Brain Research, 191 (1980) 161-171. 70 Pasquier, D. A., Gold, M. A. and Jacobowitz, D. M., Noradrenergic perikarya (A5-A7, subcoeruleus) projections to the rat cerebellum, Brain Research, 196 (1980) 270-275. 71 Poitras, D. and Parent, A., Atlas of the distribution of monoamine-containing nerve cell bodies in the brain stem of the cat, J. camp. Neurol., 179 (1978) 699-718. 72 Reid, J. L., Zivin, J. A. and Kopin, I. J., The effects of spinal cord transection and intracisternal 6-hydroxydopamine on phenylethanolamine-N-methyltransferase (PNMT) activity in rat brain stem and spinal cord, J. Neurochem., 26 (1976) 629-631. 73 Ricardo, J. A. and Koh, E. T., Anatomical evidence of direct projections from the nucleus of the solitary tract to the hypothalamus, amygdala, and other forebrain structures in the rat, Brain Research, 153 (1978) 1-26. 74 Satoh, K., Tohyama, M., Yamamoto, K., Sakumoto, T. and Shimizu, N., Noradrenaline innervation of the spinal cord studied by the horseradish peroxidase method combined with monoamine oxidase staining, Exp. Brain Res., 30 (1977) 175-186. 75 Senba, E., Tohyama, M., Shiosaka, S., Tagaki, H., Sakanaka, M., Matsuzaki, T., Takahashi, Y. and Shimizu, N., Experimental and morphological studies of the noradrenaline innervations in the nucleus tractus spinalis nervi trigemini of the rat with special reference to their fine structures, Brain Research, 206 (1981) 39-50. 76 Swanson, L. W. and I-/artman, B. K., The central adrenergic system. An immunofluorescence study of the location of cell bodies and their efferent connections in the rat utilizing dopamine-/3-hydroxylase as a marker, J. comp. Neurol., 163 (1975) 467-506. 77 Tagaki, H., Shiosaka, S., Tohyama, M., Senba, E. and Sakanaka, M., Ascending components of the medial forebrain bundle from the lower brain stem in the rat, with special reference to raphe and catecholamine cell
29 groups. A study by the I-IRP method, Brain Research, 193 (1980) 315-337. 78 Takahashi, Y., Satoh, K., Sakumoto, T., Tohyama, M. and Shimizu, N., A major source of catecholamine terminals in the nucleus tractus solitarii, Brain Research, 172 (1979) 372-377. 79 Walberg, F., The lateral reticular nucleus of the medulla oblongata in mammals. A comparative anatomical study, J. comp. NeuroL, 96 (1952) 283-343. 80 Westlund, K. N., Bowker, R. M., Ziegler, M. G. and
Coulter, J. D., Origins of spinal noradrenergic pathways demonstrated by retrograde transport of antibody to dopamine-/%hydroxylase, Neurosci. Lett., 25 (1981) 243-249. 81 Wiegand, S. J. and Price, J. L., Cells of origin of the afferent fibers to the median eminence in the rat, J. comp. Neurol., 192 (1980) 1-19. 82 Yamashita, H., Effect of baro- and chemoreceptor activation on supraoptic nuclei neurons in the hypothalamus, Brain Research, 126 (1977) 551-556.