Ascending projections from the solitary tract nucleus to the hypothalamus. A Phaseolus vulgaris lectin tracing study in the rat

Ascending projections from the solitary tract nucleus to the hypothalamus. A Phaseolus vulgaris lectin tracing study in the rat

Neuroscience Vol. 31, No. 3, pp. 18Sl91, Printed in Great Britain 1989 0306-4522/89 $3.00 + 0.00 Pergamon press plc 0 1989 IBRO ASCENDING PROJECTIO...

6MB Sizes 0 Downloads 76 Views

Neuroscience Vol. 31, No. 3, pp. 18Sl91, Printed in Great Britain

1989

0306-4522/89 $3.00 + 0.00 Pergamon press plc 0 1989 IBRO

ASCENDING PROJECTIONS FROM THE SOLITARY TRACT NUCLEUS TO THE HYPOTHALAMUS. A PHASEOLUS VULGARIS LECTIN TRACING STUDY IN THE RAT G.

J. TER HoR~T,*~

*Department IDepartment

P. DE BOER,$

P. G.

M.

LUITEN~

and J. D. VAN WILLIGEN*

of Neurobiology and Oral Physiology, Faculty of Medicine, University of Groningen, Bloemsingel 10, 9712 KZ Groningen, The Netherlands of Animal Physiology, University of Groningen, P.O. Box 14, 9750 AA Haren, The Netherlands

Abstract-The course visceral sensory part the hypothalamus in general, the posterior

of the ascending pathways originating from the anterior gustatory and posterior of the solitary tract nucleus and the topographic organization of the projections to the rat were studied with anterogradely transported Phuseolus vulgaris lectin. In visceral sensory part of the solitary tract nucleus has ascending projections as far

as the septum4iagonal band complex and gives rise to heavy input to the bed nucleus of the stria terminalis, and to the dorsomedial and paraventricular hypothalamic nuclei. A more moderate projection is aimed at a variety of other hypothalamic nuclei, to the medial and central amygdaloid nuclei and to the paraventricular nucleus of the thalamus. The ventromedial hypothalamic nucleus is strikingly missing an afferent input from the nucleus of the solitary tract. Furthermore, it was shown that whereas the caudal solitary tract nucleus has predominant long ascending connections, the projections from the anterior taste related region of the nucleus of the solitary tract have only limited forebrain _ nroiections which do not reach beyond the level of the anterior dorsal _ hypothalamic nucleus.

The hypothalamus is an important centre of control for many homeostatic processes. For example, this structure ensures the intake of food (feeding behaviour), it enables an organism to exist in its environment (defence behaviour), and guarantees the propagation of the species (sexual behaviour). For all these processes the hypothalamus acts as the outflow station to the autonomic nervous and pituitary.lO.29.30 The im_ system 10.11.18,26,29.30,32 portant role of the hypothalamus in the regulation of metabolic homeostasis is shown by the powerful effects of hypothalamic manipulation on food intake and pancreas hormone release: electrolytic lesions of the ventromedial (VMH) and lateral hypothalamic nuclei (LHA) give rise to hyper- and hypophagia lesions respectively. 22-24 In the case of VMH hyperphagia is accompanied by hyperinsulinemia. Moreover, the daily amount of food consumed is markedly enhanced after noradrenergic stimulation of the lateral, ventromedial and paraventricular (PVN) hypothalamic nuclei.~7~25 The connectivity of the hypothalamic nuclei related

tTo whom correspondence

should be addressed.

Abbreviations: DMH, dorsomedial hypothalamic nucleus; LHA, lateral hypothalamic nucleus; MFB, medial forebrain bundle; NTS, nucleus of the solitary tract; PHAL, Phase&s vulgaris leuco-agglutinin; PVM, magnocellular paraventricular hypothalamic nucleus; PVN, paraventricular hypothalamic nucleus; PVP, parvocellular paraventricular hypothalamic nucleus; TBS, Tris-buffered saline; VMH, ventromedial hypothalamic nucleus.

to the control of food intake and metabolic homeostasis, and the outflow pathways from these nuclei to the preganglionic cell groups of the autonomic 1n our nervous system, are known lO,ll.18,26,29-32.3' we emphasized the dominant previous papers position of the hypothalamus as the output gateway to endocrine organs (such as the pancreas and adrenal glands), either via neuroendocrine pathways or via neuronal circuits running from the hypothalamus to lower medullary or spinal autonomic preganglionic cell groups.‘0~“~29~32 Concerning the feeding behaviour, there is strong evidence that the noradrenergic systems innervating the hypothalamus are crucial in the control of feeding since stimulation of the VMH, LHA and PVN via noradrenergic pathways has strong effects on food intake.%7,23-25.37 The noradrenergic innervation of the hypothalamus is commonly considered to originate from noradrenergic neurons (especially the Al and A2 cell groups in the lower brainstem) which project to the hypothalamus via the ventral noradrenergic bundle.20,2’ Since the majority of the A2 noradrenergic cells 21,28is found in the nucleus of the solitary tract (NTS), we have traced the patterns of projections from the nucleus of the solitary tract to the hypothalamus. Moreover, since the NTS is one of the primary sensory sites for visceral and taste input in the brainstem,3~‘2~” we have also looked to see whether the subdivision of the NTS complex into a taste area and a visceral sensory area is reflected in a differentiation of ascending connectivity to the 785

786

G. J. TER HOR~T et al.

hypothalamus. The rationale of this view is the assumption that in hypothalamical control of food intake sensory information carried by the vagal nerve is more important than information about taste modalities, taste input being mainly integrated at various levels of the mesencephalon and the thalamus.‘3.14~‘9 Until now the descriptions of the ascending connections of the solitary tract nucleus to the hypothalamus have been based on retrograde36 and anterograde’r’3.‘8.‘9 tract tracing techniques. However, in these papers, due to the size of the tracer deposits, conclusions on the termination areas in the hypothalamus and the cellular origin in the solitary tract nucleus cannot be drawn. Therefore, in the present experiments we iontophoretically delivered minute doses of the lectin Phaseolus vulgaris leuco-agglutinin (PHA-L) in the NTS and traced the efferents from this nucleus. This procedure also enabled us to deposit small quantities of tracer into subdivisions of the NTS. The major advantage of this method, however, is the labelling of neurons from soma to presynaptic

ending.2.33.38 EXPERIMENTAL

PROCEDURES

The experiments were carried out on 34 male albino Wistar rats weighing 28&350 g. For the injection of PHA-L, the animals were anaesthetized with sodium pentobarbital (30mg/kg i.p.) and Hypnorm (Duphar) (0.4ml/kg) and placed in a Kopf stereotaxic frame. Deposits of PHA-L in the posterior aspects of the NTS were made according to the atlas of Pellegrino et al.” and in the anterior parts of the NTS according to Paxinos and Watson.i6 Bevelled glass micropipettes (tip diameter 17.5-25pm) filled with 2.5% PHA-L (Vector Labs) solution in Tris-buffered saline (TBS; pH = 7.4) were positioned in the brain and connected to the of a Midgard CS-3 constant current positive electrode

source. Iontophoretic delivery was achieved with a 5.5-PA current for 30 min in a 7-s half time on, half time off cycle. After iontophoresis the pipette was left in situ for 10 min to prevent leakage of tracer in the pipette track. After a postoperative survival time of 7 days the animals were perfused transcardially with a 2.5% glutaraldehyde, 0.5% paraformaldehyde solution after appropriate deep anaesthesia. Forty-micron sections cut on a freezing microtome were treated with subsequent incubations in goat-anti-PHA-L, rabbit-anti-goat IgG and goat peroxidase-antiperoxidase complex as described in detail in previous reports.“,2”3 After the incubations the peroxidase was visualized with a reaction on diaminobenzidine in the presence of H,O,; then the sections were counterstained with Cresyl Violet. For the control of the specificity of the immunoperoxidase reaction, in the first experiments some sections were incubated with goat anti-PHA-L alone, with rabbitanti-goat IgG alone, with goat peroxidase-antiperoxidase alone, and with solvent without any primary antisera (further procedures as above). All such control experiments appeared to be negative.

RESULTS

In 12 animals PHA-L injections were successfully placed within the boundaries of the NTS. The position and the extent of the deposits, and the resulting hypothalamic labelling, are summarized in Fig. 1 and Table 1. In Fig. 1, which is a bilateral horizontal projection of the NTS, the relationship between the injection sites in the NTS, and the projection areas of the visceral-sensory vagal nerve and the gustatory nerves are shown.3,4 Anterior solitary tract nucleus. Injections centred in the anterior NTS area that coincided with gustatory projection fields, caused dense labelling in the lower medullary region and parabrachial complex (Fig. 2C). From the gustatory area only a few

Abbreviations used in figures AC aha, AHy amb Zq bl bm cai, ic ce CnF DA DMC dmh, DM, DMH en f, F fi hi la lha, LH, LHA lhb lr me me5 Me5

anterior commissure anterior hypothalamic area nucleus ambiguus area postrema aqueduct basolateral amygdaloid nucleus basomedial amygdaloid nucleus capsula interna central amygdaloid nucleus cuneiform nucleus dorsal hypothalamic area dorsomedial hypothalamic nucleus, compacta dorsomedial hypothalamic nucleus endopyriform nucleus fornix fimbria hippocampus amygdalohippocampal area lateral amygdaloid nucleus lateral hypothalamic area lateral habenular nucleus lateral reticular nucleus medial amygdaloid nucleus. mesencephalic trigeminal tract mesencephalic trigeminal nucleus

nts ox

pars

P pmv Pv pvm, PVM pvp, PVP Pvt RCh re, Re rf SCP si sP5 St to vmh vsc v3 X XII zi, 21 4v

.

mamillothalamic tract solitary tract nucleus optic chiasm pyramidal tract ventral premammillary nucleus paraventricular nucleus magnocellar paraventricular nucleus parvocellular paraventricular nucleus thalamic paraventricular nucleus retrochiasmatic area nucleus reuniens rhinal fissure superior cerebellar peduncle substantia innominata spinal trigeminal nucleus stria terminalis optic tract ventromedial hypothalamic nucleus ventral spinocerebellar tract third ventricle dorsal motor nucleus of the vagus hypoglossal nucleus zona incerta fourth ventricle

Hypothalamic projections of solitary tract nucleus

Ainjection

sites

I r-2.5

_^

I P3.0

P3.5

I

W.0

24

I P4.5

P.5.0

vagus

n

II

P5.5

nerve

overlap

Fig. 1. Horizontal projection of the nucleus of the solitary tract adapted from Hamilton and Norgrenr Anterior is left, posterior is right. The horizontal scale refers to the anterior-posterior coordinates, posterior to the interaural line.” In panel A the position and size of the various PHA-L injections are indicated with reference to their experiment number as in the text and Table 1. The results of experiment 16 (shaded area in panel A) are illustrated in Fig. 5. In the bottom panel B the projection areas of the gustatory nerves and the visceral-sensory vagus nerve to the NTS according to Hamilton and Norgenr are indicated for comparison with the PHA-L injections.

efferents ran to the hypothalamic level, where they exclusively terminated in the dorsal hypothalamic area, lateral hypothalamic area and zona incerta. Posterior solitary tract nucleus. In contrast to projections arising from the anterior NTS, the posterior visceral part of the NTS was not only the source of considerable input to the parabrachial region (Fig. 3A and B), but also had widespread ascending projections to hypothalamic and forebrain regions. Bilaterally within the hypothalamus a clear pattern of labelling could be discerned. Projections to the paraventricular, dorsomedial and lateral hypothalamic nuclei were most prominent, whereas input to the perifornical area, supraoptic, medial preoptic arcuate and dorsal (Fig. 4B), retrochiasmatic, hypothalamic nuclei was less. In a few cases PHA-Lpositive labelling was observed in the suprachiasmatic nucleus and the anterior hypothalamic area. Although not in the scope of this study it is worth mentioning that a considerable NTS projection existed to extrahypothalamic areas: terminal bouton labelling was found in the medial and central amygdaloid nuclei, endopyriform nucleus, bed nucleus of the stria terminalis (Fig. 4A), lateral septum, diagonal band nuclei, nucleus accumbens and the thalamic paraventricular nucleus. Ascending

solitary

tract nucleus-hypothalamic

path-

way

Fibres labelled after injections of PHA-L into the anterior and posterior NTS initially ascended

787

through the NTS column, giving off fasicles of fibres innervating ventrolateral medullary targets such as the parvocellular reticular formation, facial nucleus, and the principal, spinal and motor trigeminal nuclei. The major contingent of the ascending fibres, however, passed the parvocellular reticular formation and then ran dorsolateral to the facial motor nucleus. In the parabrachial region, most fibres originating from the anterior NTS and some fibres arising from the posterior NTS terminated in the dorsal parabrachial nucleus and the locus coeruleus. However, the majority of the labelled fibres from the posterior NTS collected in the periaqueductal gray and adjacent tegmentum. In this region the fascicles split up into two pathways: one ran through the periventricular gray, the other took a tegmental trajectory in and around the medial lemniscus. Upon entering the diencephalon (Fig. 5F) the tegmental pathway continued in the dorsal and dorsolateral parts of the medial forebrain bundle (MFB). At the level of the hypothalamus (Fig. SF) the periventricular bundle bent ventrally and terminated in the periventricular nucleus. Some minor projections in the perinuclear fibres shell of the VMH were occasionally seen. The majority of the NTS projections to the hypothalamus reached their target nuclei via the MFB (Fig. 5). From the MFB, labelled fibres fanned out into the dorsomedial hypothalamic nucleus, perifornical area, dorsal hypothalamic area, paraventricular complex and arcuate, anterior, supraoptic and retrochiasmatic suprachiasmatic, hypothalamic nuclei. Throughout their course in the MFB, terminal labelling could be seen in the lateral hypothalamic area. Finally, the labelled fibres in the MFB gave off branches to the amygdaloid body and proceeded to the basal forebrain region, where they terminated in the forebrain structures mentioned above.

Projection

targets in the hypothalamus

In the hypothalamus the projections from the solitary tract nucleus were bilateral, with an ipsilateral dominance. The fibres that supplied the input to targets in the contralateral hypothalamus crossed at the medullary level, mainly in nuclei of the dorsal medulla oblongata, i.e. NTS, XII and prepositus hypoglossal nucleus-and for a minor part in the ventral brainstem. Ventromedial hypothalamic nucleus. A striking observation was that of the feeding related nuclei, the VMH was almost devoid of afferent input from the NTS. Only a few projections to the dendritic fibre shell that surrounds the VMH (Fig. 2A and A’) were sometimes seen arising from the posterior (visceral-sensory) NTS. The inner core of the VMH did not contain any PHA-L positive fibres, except in experiments 16,29 and 41. In experiments 16 and 29 (with PHA-L deposits in- the posterior NTS) a

788

G. J. TER HORST

Fig. 2. A. PHA-L positive fibres in the dendritic fibre shell of the VMH in the medial LHA. The labelled fibre indicated by the arrow is shown at higher magnification in A’. B. NTS projections onto cell bodies in the pars compacta of the dorsomedial hypothalamic nucleus. C. Dark-field photomicrograph of NTS projections in the caudal mesencephalic trigeminal nucleus, locus coeruleus and medial parabrachial nucleus. single fibre could be traced to the inner core of the ipsilateral intermediate and posterior VMH. Here, terminal bouton labelling was found in the ventrolateral parts of these subdivisions of the VMH.

One experiment (41) with a rostra1 NTS injection showed ipsilateral terminal bouton labelling in the dorsolateral part of the rostra1 VMH. Also in this experiment, labelled presynaptic endings were

Hypothalamic projections of solitary tract nucleus

7‘89

Fig. 3. A. Dark-field photomicrograph of terminal lahelling in the rostra1 mesencephalic trigeminal nucleus and the posterior periaqueductal gray. B. NTS projections in the dorsal parabrachial nucleus.

seen at a branching fibre that entered the VMH dorsolaterally. Dorsomedial hypothalamic nucleus. In contrast to the lack of NTS terminal bouton labelling in VMH, many projections were observed in the dorsomedial hypothalamic

nucleus (DMH), bilaterally. Lectin in-

jections into the posterior (visceral-sensory) NTS, in general, gave more dense terminal bouton labelling in the DMH than PHA-L deposits in the rostra1 solitary tract nucleus. In particular, the anterior and ventral parts of the posterior DMH were richly supplied with efferents from the posterior NTS, whereas these areas

G. J. TER HORST ef nl.

790

Fig. 4. A. NTS projections in the ventral part of the bed nucleus of the stria terminalis. B. Peculiar projection of NTS on bloodvessels in the medial preoptic area. C. Varicose PHA-L positive fibres in the dorsal

part of the rostra1 lateral

only received a modest input from the rostra1 solitary tract nucleus. Furthermore, both rostra1 and caudal NTS injections caused a few PHA-L positive fibres carrying terminal boutons in the dorsal part of the posterior DMH.

hypothalamic

area.

Projections to the cell dense pars compacta of the DMH (Figs 2B and 6) were traced from the posterior (visceral-sensory) NTS. No labelling was found in this area after lectin deposits in the rostra1 solitary tract nucleus.

Hypothalamic projections of solitary tract nucleus

791

Fig. 5. A series of transverse sections of the hypothalamus from anterior (A) to posterior (F) in which the anterogradely lahelled efferents are charted following a PHA-L injection in the visceral-sensory region of NTS. The PHA-L injection-site is indicated as the stippled area in the top right figure (experiment 16 from Fig. 1). The figures were adopted from the atlas of Paxinos and Watson.16 The bud-like thickening in the fibre labelling indicates labelled terminal boutons in the projection areas.

Lateral hypothalamic area. The LHA was found to be a target area for all subdivisions of the NTS. Tracer injections into the posterior visceral-sensory (vagal) subdivision, however, gave rise to more dense and more anteriorly reaching projections to the LHA than did the rostra1 (gustatory) NTS injections. All PHA-L deposits in the NTS gave a dense, bilateral, terminal bouton labelling in the dorsal and dorsolateral subdivisions of the posterior LHA

(Fig. SE and F). NTS projections to the intermediate area of the LHA were found in its medial and lateral aspects (Fig. SC and D), whereas the anterior LHA contained terminal bouton labelling in its lateral and ventral subdivisions (Fig. 5A and B). Tracer deposits in the rostra1 NTS always caused labelling of a small number of fibres in the intermediate and anterior subdivisions of the LHA (Fig. 4C); this phenomenon was independent of the diameter of the PHA-L

792

G. J. TER HO~WTer al.

Table 1. Semiquantitative listing of Phaseolus vulgaris leuco-agglutinin terminal labelling in the various hypothalamic nuclei after Phaseolus vulguris leuco-agglutinin injections in the nucleus of the solitary tract Experiment Anterior 52 39 Suprachiasmatic nucleus Retrochiasmatic area Supraoptic hypothalamic nucleus Anterior hypothalamic area Paraventricular hypothalamic nucleus Magnocellular Parvocellular Median eminence Periventricular hypothalamic nucleus Ventromedial hypothalamic nucleus Dendritic shell of the VMH Lateral hypothalamic area Anterior Intermediate Posterior Perifornical hypothalamic nucleus Anterior Intermediate Posterior Dorsomedial hypothalamic nucleus Anterior Intermediate Posterior Dorsal hypothalamic nucleus Anterior Intermediate Posterior Zona incerta Arcuate hypothalamic nucleus

46

+

54

38

+

+

+ +

+

28

29 + ++ ++ ++

17

+

+ ++

+ ++ ++ ++

++ + + +

+ ++

++ ++

++ +++ ++

+

+

++

+

+

+ +

++ + +

+ + +

+++ ++ ++ + ++ ++

+

+ + + +

Posterior 16

++ + +

+ +

++ + +

+ +

++

+++ +++ ++ ++ + + ++

++ +++ + + +

++

The experiment numbers are ordered with respect to their anterior-posterior position. The numbers correspond to the numbers in Fig. 1. + + +, dense terminal labelling; + + , modest terminal labelling; + , sparse lahelling.

injection. In contrast, heavy labelling of the intermediate and anterior subdivisions of the LHA was observed after injections of tracer into the posterior (visceral-sensory) subdivision of the NTS, where the number of labelled fibres in the LHA was related to the diameter of the injection. of the LHA, at all The central parts anterior-posterior levels, were always devoid of terminal bouton labeling. Paraoentricular hypothalamic nucleus. Inputs from the NTS to the PVN were found after traced deposits in the posterior (visceral-sensory) NTS (e.g. experiments 16, 29, 38, 54). In the parvo- (PVP) and magnocellular (PVM) PVN the projections appeared to be bilateral with an ipsilateral dominance. Here, efferent fibres from the solitary tract nucleus formed a network. Terminal bouton labelling in the paraventricular nucleus, however, was not evenly distributed over this fibre network. The majority of labelled presynaptic endings was concentrated in the transition zone of the PVP and PVM, the central part of the PVP (Figs 7 and 8A), and in the dorsal cap region of the parvocellular subdivision.

Other hypothalamic projections. Most other hypothalamic areas listed in Table 1 received inputs from the posterior (visceral-sensory) NTS; only the zona incerta was a target area for efferents from the rostra1 (gustatory) division of the NTS. Terminal bouton labelling in the zona incerta was found in the ventromedial part. The projection from the visceral-sensory NTS onto the arcuate nucleus was for the greater part found in the posterior subdivision of this nucleus. In the ventrolateral part of the anterior subdivision of the arcuate nucleus a modest terminal bouton labelling was demonstrated. These projections were the medial extensions of NTS efferents running in the ventral hypothalamus. The remaining hypothalamic areas listed in Table 1, e.g. the anterior hypothalamic area, perifomical, periventricular, supraoptic, retrochiasmatic and suprachiasmatic nuclei and median eminence, only received a modest input from the posterior NTS. These connections are beyond the scope of this investigation and are therefore not discussed in detail.

Hypothalamic projections of solitary tract nucleus

793

Fig. 6. Camera lucida drawing of PHA-L labelled projections to the pars compacta of the posterior DMH (DMC) after a lectin injection in the posterior NTS (a). Panel (b) shows by means of shading the regions in the posterior visceral-sensory NTS that give rise to projections to the pars compacta of the DMH. In panel (c) the position in the hypothalamus of drawing (a) is illustrated.

DISCUSSION

In the present study we describe the ascending pathways from the posterior visceral-sensory (vagal) and anterior gustatory divisions of the solitary tract nucleus projecting to the hypothalamus (Fig. 9). Projections from the posterior NTS could be demonstrated to all hypothalamic nuclei which are involved in regulation of feeding and body weight, with the exception of the VMH. In particular the PVN, LHA and DMH are recipients of a strong NTS input. Terminal bouton labelling after PHA-L injections in the anterior NTS for the greater part was found in the parabrachial area and zona incerta. In general, these findings are in agreement with previous descriptions of NTS efferents.1*9J3J*J9Our experiments, however, give more detailed information on the cellular source and the synaptic projection of NTS efferents since we applied the PHA-L method for labelling of NTS neurons. With this method, iontophoretic delivery of the lectin PHA-L permits deposits of tracer limited to the small areas of the NTS (Fig. 8B). Furthermore, with the PHA-L method, efferents are labelled up to the level of

presynaptic boutons. Therefore, a clear picture could be obtained of the course and composition of the ascending projections, which appeared to be composed of thin, fine-diameter and (probably) poorly myelinated fibres. The delicate nature of the ascending projections may also be a causal factor in the lack of labelling with tritiated amino acids in the hypothalamus in the tracing experiments of Norgren. I3 Support for our observation of finecalibre NTS-hypothalamic projections is given who demonstrated, with electroby Nosaka,” physiological techniques, monosynaptic projections with low conduction velocity to the PVN, DMH, and anterior hypothalamic areas. In the same study Nosaka gives a description of a projection to the VMH. This observation could not be confirmed by our experiments although a small NTS projection to the perinuclear fibre shell of the VMH was seen by us which may explain Nosaka’s findings. From a functional point of view, it is tempting to speculate on the meaning of a visceral (and probably noradrenergic) NTS input to the hypothalamic feeding-related nuclei. It is obvious that the main source of noradrenergic innervation to the hypo-

794

G. J. TER HORSTet al.

P2,3

33

4.3

53

Fig. 7. Camera lucida drawing of PHA-L iabelled projections to the paraventricular nucleus after a tracer injection in the posterior NTS (a). In panel (b) the regions in the NT’S that give rise to projections to the magnoeellular PVN (1) and to the parvocellular PVN (2) are indicated by the stippled regions in a horizontally projected NTS. Panel (c) indicates the hypothalamic position of drawing (a).

the preganghonic cell groups of the autonomic nervous system and the efferents of the NTS to the hypothalamus. In the model it can be seen that the major contingent of efferent fibres from the posterior The present tracing experiments, in combination with previous findings, give strong evidence that visceral sensory NTS to the hy~~alamus is aimed at targets in the DMH and PVN, the latter nucleus visceral NTS input is most important in the regubeing an important output-structure of the hypolation of food intake. This assumption is supported thalamus.‘1~29 It is known that the PVN has by the described powerful responses on food intake as a result of local administration of norepinephrine direct connections not only to preganglionic orthointo the PVN, VMH, DMH, and LHA.S-7*Q24*25~37 It and ~ras~pathetic cell groups (in the medulla remains, however, a matter of dispute as to how a oblongata and the spinal cord)” but also to the locally applied noradrenergic stimulus in the VMH pituitary.27 In our view the PVN controls the balance has such strong autonomic effects, since there is between ortho- and parasympathetic nervous activity almost no direct NTS (or noradrenergic) innervation in the autonomic nervous system and between autoof the VMH. On the other hand, a relatively dense nomic and neuroend~rine mechanisms for the regulation of metabolic homeostasis. We also speculate binding in the VMH of ligands with affinity to that the level of activity of the PVN is regulated by adrenergic receptors is shown.8 the DMH and NTS. In this way the DMH and On the basis of the present and previous NTS, indirectly, can influence the balance between experimentslo* 2+33we have designated an anatomical nervous activity in the model of feeding control (Fig. 10). In this model we ortho- and paras~pathetic autonomic nervous system via the PVN. Possibly, in have included the int~hy~thaiamic connections, descending pathways from the hypothalamus to this hierarchy of control, limbic and hypothalamic thalamus originates from the Al and A2 cell groups in the lower brainstem,21*28although involvement of other sources cannot entirely be excluded.

Hypothalamic projections of solitary tract nucleus

795

Fig. 8. A. Dark-field photomicrograph of PHA-L labelled fibres in the PVN after a tracer injection in the posterior NTS. B. Photomicrograph of a PHA-L injection in the posterior visceral sensory part of the NTS. Note the labelled somata in the NTS and a dense terminal labelling in the adjacent dorsal motor vagus nucleus (x), whereas the hypoglossal nucleus (XII) is devoid of NTS input. Scale bars: A, 200 pm; B, 2OOpm.

piluitafy

Fig. 9. Schematic diagram of target areas in the hypothalamus of the anterior-gustatory(left) and posteriorvisceral sensory- (right) solitary tract nucleus.

PP*CTePS

Fig. 10. Schematic diagram of nervous pathways for regulation of feeding and metabolic homeostasis by the hypothalamus. Line thickness indicates density of projections. Abbreviations: NTS, solitary tract nucleus; PVN, paraventricular nucleus; DMH, dorsomedial hypothalamic nucleus; LHA, lateral hypothalamic area; VMH, ventromedial hypothalamic nucleus; PAG, periaqueductal gray; RF, reticular formation; X, dorsal motor vagus nucleus; IML, intermediolateral cell group of the thoracic spinal cord. For more details see Refs 10, 29.

196

G. J. TER HORSTet al.

activities for the regulation of metabolic homeostasis are adjusted to the present status of the “milieu interieur” by the NTS. Physiological observations support this hypothesis. Lesions of the ventral noradrenergic bundle, which contains efferent noradrenergic fibres from the NTS en route to the hypothalamus, cause increased body weight.20,35 Noradrenergic stimulation of the DMH gives rise to increased levels of blood glucose and a simultaneous rise in plasma catecholamines (Gaffe, personal communication), whereas noradrenergic stimulation of the PVN (probably via the NTS-hypothalamic path-

way) evokes a feeding response the recurrent PVN-autonomic

which is mediated pathways.”

by

CONCLUSION

Visceral sensory afferents may evoke strong effects on the regulation of feeding and metabolic homeostasis by the hypothalamus via modulation of PVN nervous activity. Acknowledgements-We wish to thank Mrs Nieske Brouwer and Mrs Anja Kattenberg for processing histological material and Mrs Lidy Kingma for her secretarial assistance.

REFERENCES E. T. Jr and Sawchenko P. E. (1986) Local projections from the medial part of the nucleus of the solitary 1. Cunningham tract in the rat. Sec. Neurosci. Abstr. 12, 1055. neuroanatomical tracing method that shows the detailed 2. Gerfen C. R. and Sawchenko P. E. (1984) An anterograde morphology of neurons, their axons and terminals: immunocytochemical localization of an axonally transported plant lectin, Phaseolus vulgaris leuco-agglutinin (PHA-L). Bruin Res. 200, 219-238. 3. Hamilton R. B. and Norgren R. (1984) Central projections of gustatory nerves in the rat. J. camp. Neurol. 222,567-577. O., Lang R., Ganten D., Cue110 C. and Terenius L. (1984) Distribution 4. Kalia M., Fuxe K., Hiikfelt T., Johansson of neuropeptide immunoreactive nerve terminals within the subnuclei of the nucleus tractus solitarius of the rat. J. camp. Neurol. 222, 409444. nucleus: a primary site mediating adrenergic stimulation of feeding and drinking. 5. Leibowitz S. F. (1978) Paraventricular Pharmac. Biochem. Behav. 8, 163-175. injection 6. Leibowitz S. F., Hammer N. J. and Chang K. (1983) Feeding behaviour induced by central norepinephrine is attenuated by discrete lesions in the hypothalamic paraventricular nucleus. Pharmac. Biochem. Behau. 19, 945-950. I. Leibowitz S. F., Roossin P. and Rosenn M. (1984) Chronic norepinephrine injection into the hypothalamic paraventricular nucleus produces hyperphagia and increased body weight in the rat. Pharmac. Biochem. Behav. 21, 801-808. M., Dvorkin B. and Makman M. H. (1982) Distribution of cc-adrenergic, b-adrenergic 8. Leibowitz S. F., Jhanwar-Uniyal and dopaminergic receptors in discrete hypothalamic areas of rat. Brain Res. 233, 97-114. H. (1982) Organization of diencephalic and brainstem afferent projections 9. Luiten P. G. M., Kuipers F. and Schuitmaker to the lateral septum in the rat. Neurosci. Lett. 30, 211-216. intrinsic connections and outflow 10. Luiten P. G. M., Ter Horst G. J. and Steffens A. B. (1987) The hypothalamus, pathways to the endocrine system in relation to the control of feeding and metabolism. Prog. Neurobiol. 28, l-54. hypothalamic II. Luiten P. G. M., Ter Horst G. J., Karst H. and Steffens A. B. (1985) The course of paraventricular efferents to autonomic structures in medulla and spinal cord. Bruin Res. 289, 374378. 12. Nejad M. S. (1986) The neural activities of the greater superficial petrosal nerve of the rat in response to chemical stimulation of the palate. Chem. Senses 11, 283-293. 13. Norgren R. (1978) Projections from the nucleus of the solitary tract in the rat. Neuroscience 3, 2077218. 14. Norgren R. and Pfaffmann C. (1975) The pontine taste area in the rat. Brain Res. 91,99-117. vagal visceral input to the forebrain via a direct pathway in 15. Nosaka S. (1984) Solitary nucleus neurons transmitting rats. Expl Neurol. 85, 493-505. 16. Paxinos G. and Watson C. (1982) The Rat Brain in Stereotaxic Coordinates. Academic Press, New York. 17. Pellegrino L. J., Pellegrino A. S. and Cushman A. J. (1979) A Stereotaxtic Atlas of the Rut Brain. Plenum Press, New York. pathways involved in metabolic regulatory functions, as identified by tract-tracing 18. Ricardo J. A. (1983) Hypothalamic methods. In Advances in Metabolic Disorders, Vol. IO: CNS Regulation qf Carbohydrate Metabolism (ed. Szabo J.), pp. I-26.Academic Press, New York. evidence of direct projections from the nucleus of the solitary tract 19. Ricardo J. A. and Koh E. T. (1978) Anatomical to the hypothalamus, amygdala, and other forebrain structures in the rat. Brain Res. 153, l-26. 20. Sahakian B. J., Winn B. J., Robbins T. W., Deeley R. J., Everitt B. J., Dunn L. T., Wallace M. and James W. P. T. (1983) Changes in bodv weight and food related behavior induced bv destruction of the central or dorsal noradrenereic bundie in thi rat. Neurosci&ce 10,140>1420. of noradrenergic pathways from the brainstem to the 21. Sawchenko P. E. and Swanson L. W. (1982) The organization paraventricular and supraoptic nuclei in the rat. B&t Res. Rev. 4, 2755322.by the hypothalamus of glucagon and insulin secretion in rabbits: 22. Shimazu T. and Ishikawa K. (1981) Modulation studies with electrical and chemical stimulations. Endocrinology 108,605-611. role of the central nervous system on insulin and glucagon release during food 23. Steffens A. B. (1981) The regulatory intake in the rat. In Hormones and Cell Regulation, Vol. 5. (eds Dumont J. E. and Nunez J.), pp. 185-191. Elsevier/North-Holland, Amsterdam. of the central nervous system to the regulation of insulin 24. Steffens A. B. and Luiten P. G. M. (1986) The contribution release and stabilization of blood glucose. In Pathogenesis and Treatment of Diabetes Mellitus (eds Radder J. K., Lemkes H. M. P. J. and Krans M. M. J.), pp. 63-74. Martinus Nijhof, Boston. free fatty acids, insulin and 25. Steffens A. B., Damsma G., van der Gugten J. and Luiten P. G. M. (1984) Circulating glucose during chemical stimulation of the lateral and ventromedial areas of the hypothalamus in the rat. Am. J. Physiol. 247, E765-El71.

Hypothalamic projections of solitary tract nucleus

797

26. Swanson L. W. and Kuypers H. G. J. M. (1980) A direct projection from the ventromedial nucleus and retrochiasmatic area of the hypothalamus to the medulla and spinal cord of the rat. Neurosci. Let?. 17, 307-312. 27. Swanson L. W. and Sawchenko P. E. (1983) Hypothalamic integration: organization of the paraventricular and supraoptic nuclei. A. Rev. Neurosci. 6, 269-324. 28. Takagi H., Shiosaka S., Tohyama M., Senba E. and Sahanaka M. (1980) Ascending components of the medial forebrain bundle from the lower brainstem in the rat, with special reference to raphe and catecholamine cell groups, a study by the HRP method. Brain Res. 193, 315-337. 29. Ter Horst G. J. (1986) The hypothalamus, intrinsic connections and outflow pathways to the pancreas. Thesis, R. U. Groningen, The Netherlands. 30. Ter Horst G. J. and Luiten P. G. M. (1986) The projections of the dorsomedial hypothalamic nucleus in the rat. Brain Res. Bull. 16, 231-248. 31. Ter Horst G. J. and Luiten P. G. M. (1987) Phaseolus oulgaris leuco-agglutinin tracing of intrahypothalamic connections of lateral, ventromedial, dorsomedial and paraventricular nuclei of the hypothalamus in the rat. Bruin Res. BUN. 18, 191-203. 32. Ter Horst G. J., Luiten P. G. M. and Kuipers F. (1984) Descending pathways from hypothalamus to dorsal motor vagus and ambiguus nuclei in the rat. J. auton. nerv. Cyst. 11, 59-75. 33. Ter Horst G. J., Groenewegen H. J., Karst H. and Luiten P. G. M. (1984) Phuseolus vulguris leuco-agglutinin immunohistochemistry. A comparison between autoradiographic and lectin tracing of neuronal efferents. Bruin Res. 307, 379-383. 34. Travers S. P., Pfaffman C. and Norgren R. (1986) Convergence of lingual and palatal gustatory neural activity in the nucleus of the solitary tract. Brain Res. 356, 305-320. 35. Van der Gugten J. and Slangen J. L. (1977) Release of endogenous catecholamines from rat hypothalamus in uivo related to feeding and other behavior. Pharmac. Biochem. Behav. 7, 21 l-219. 36. Van der Kooy D., Koda L. Y., McGinty J. F., Gerfen C. R. and Bloom F. E. (1984) The organization of projections from the cortex, amygdala, and hypothalamus to the nucleus of the solitary tract in the rat. J. camp. Neural. 224, l-24.

37. Weiss G. F. and Leibowitz S. F. (1983) Dorsal paraventricular fiber system descending from paraventricular nucleus mediates feeding induced by norepinephrine and clonidine. Sot. Neurosci. Absrr. 9, 467. 38. Wouterlood F. G. and Groenewegen H. J. (1985) Neuroanatomical tracing by use of Phaseolus vulgaris leuco-agglutinin (PHA-L), electronmicroscopy of PHA-L filled neuronal somata, dendrites, axons and axon terminals. Brain Res. 326, 188-192. (Accepted 4 April 1989)