Brain Research, 546 (1991) 190-194 © 1991 Elsevier Science Publishers B.V. 0006-8993/91/$03.50 ADONIS 000689939116492Q
190
BRES 16492
Microinjection of oxytocin into the dorsal vagal complex decreases pancreatic insulin secretion P. Siaud 1, R. P u e c h e, I. A s s e n m a c h e r I a n d G. A l o n s o 1 i Laboratoire de Neurobiologie Endocrinologique, URA 1197 CNRS, Universit~ de Montpellier H and 2Laboratoire de Pharmacologie et de Pharmacodynamie, URA 85 CNRS, Universit~ de Montpellier L Montpellier (France)
(Accepted 30 October 1990) Key words: Oxytocin; Insulin; Parasympathetic nervous system; Medulla oblongata; Rat
Microinjections of oxytocin and of an oxytocin antagonist into the dorsal vagal complex of the medulla oblongata were performed in order to study the possible role of the oxytocin containing axons that innervate this region in the regulation of pancreatic insulin secretion. No significant effect was produced by the intrameduUary injection of the oxytocin vehicle alone or of 0.04 pM oxytoein. Injections of 4 and 20 pM oxytocin produced a reversible decrease of plasmatic insulin levels which fall to 59% of basal levels 15 min after the injection. Such an effect was abolished when 4 pM oxytocin was injected to animals which have been previously bilaterally vagotomized. In contrast to oxytocin, intramedullary injection of a specific antagonist of oxytocin to intact animals induced a marked increase of plasmatic insulin levels which raised 131% of basal levels 15 min after the injection. In animals receiving such an injection of oxytocin antagonist, a secondary injection of 4 pM oxytocin produced a slight but not significant decrease of plasmatic insulin levels. These data strongly suggest that the hypothalamic neurons producing oxytocin that densely project to the dorsal vagal complex may be involved in an inhibitory control of the vagal preganglionic neurons that innervate the pancreas. INTRODUCTION It has now been clearly demonstrated that the vagus nerve conveys inputs that stimulate the pancreatic secretion of insulin 8 and that the preganglionic parasympathetic neurons responsible for this control are mainly located in the dorsal motor nucleus of the vagus nerve (dmnX) in the dorsal vagal complex of the medulla oblongata ~8'23. It is also well known that the hypothalamus plays a major role in the central regulation of the pancreatic secretion of insulin w'18'2°. For instance, several earlier experiments clearly demonstrated that basal insulin secretion was dramatically modified both by the electrical stimulation or electrolytic lesion of hypothalamic nuclei 1'4'5"1°'z°, strongly suggesting the occurrence of direct neuronal connections between the hypothalamus and the autonomic centers innervating the pancreas. Recently, this idea of a hypothalamic-islet axis has been reinforced by a series of neuroanatomical data indicating that a n u m b e r of hypothalamic perikarya directly project to the dorsal vagal complex of the medulla 13'21"24'25. A m o n g the several types of hypothalamic neurons that have been shown to project to the dorsal vagal complex, oxytocin (OT)-producing neurons located in the parvo-
cellular part of the paraventricular nucleus (PVN) certainly represent the system which provides t h e densest innervation of this medullary nucleus 15'21"25. Additional data indicating that OT-containing axons establish frequent synaptic contacts with the perikarya of this medullary region 22'29 and that the dorsal vagal complex and more specifically the d m n X contain high densities of OT-binding sites 28, strongly suggest that hypothalamic O T neurons could be involved in the regulation of visceral functions under the control of the vagus. Accordingly, previous data indicated that the application of O T to the dorsal medullary region modifies both the electrical activity of neurons of the d m n X 6 and the secretion of gastric acidity 19. The present study was therefore undertaken to study the possible role of O T on the medullary control of the secretion of pancreatic insulin. MATERIALS AND METHODS Fourty-two male adult Sprague-Dawley rats (200-250 g) were used in this study. They were housed in fight- and temperaturecontrolled rooms (12 h fight starting at 7.00 h and 12 h darkness, at 21 + 1 °C) and had free access to standard dry food and tap water.
Correspondence: P. Siaud, Laboratoire de Neurobiologie Endocrinologique, URA 1197 CNRS, Universit~ de Montpellier II, Place E. Bataillon, 34095 Montpeilier Cedex 05, France.
191 Under deep Equithesine anesthesia, a polyethylene cannula filled with heparinized (250 U/ml) saline buffer was inserted into the left carotid artery, which was carefully exposed to avoid damage of the nearby vagus nerve. The animals were placed in a stereotaxic device with the head in the nose down position. The dorsal cervical musculature was resected and the obex region of the dorsal medulla was exposed after removing the occipital skull plate. In an other group of rats, after cannulation of the carotid artery, bilateral vagotomy was performed by cutting the vagal trunks running along the carotids. The intramedullary microinjections were performed by means of the 1 /zl Hamilton microsyringe fitted with a glass micropipette with an outside tip diameter of 40-50/~m. Both the stereotaxic coordinates of the injection site and the volume of the injection were determined in order to obtain the impregnation of the dorsal vagal complex throughout its rostro-caudal extent. According to data obtained in a parallel neuroanatomical study based on the injection of tritiated amino acids into the dorsal vagal
complex, this condition appeared fulfilled by positioning the tip of the micropipette at about 0.1 mm anterior, 0.9 mm lateral to the obex and 0.2-0.3 mm ventral to the brain surface and by injecting a volume of 40 nl (see Fig. 1). The substances injected in the present study were: oxytocin (Bachem) (0.04, 4 and 20 pM OT/injection) and an OT antagonist (AOT = d(CH2) 5 (Tyr(Me)2, Thr4, TyrNHE9)OVT)(kindly provided by Dr. C. Barberis, Montpellier, France) (2 pM AOT/injection). They were infused in 40 nl phosphate-buffered saline (PBS) over 1 rain on the side opposite to the cannulated carotid. The micropipette was left in place an additional 2 rain before withdrawal to minimize back diffusion along the cannulae tract. For single intramedullary injections of PBS, OT, or AOT, 100/~l blood samples were collected in heparinized tubes 10 min before the injection and 5, 15, 25 and 35 min after the injection. Double intramedullary injections consisted of an injection of 2 pM AOT followed 15 rain after by an injection of 4 pM OT. For these double injection experiments, blood samples were collected 10 rain after the injection of AOT and then 15 min after the OT administration. After centrifugation, aliquots of plasma were stored at -30 °C. Insulin was assayed by the method of Herbert et al. 11 using an antibody supplied by Miles laboratories (Paris, France). [125I]insufin was obtained from international CIS (Gif/Yvette, France). The standard used was pure rat insulin (Novo, Copenhagen, Denmark) whose biological activity is 22.3/~U/ng. The intra- and inter-assay coefficients of variations were 9.7% and 14.4% respectively, and the sensitivity was 0.2 ng/ml. Because of the relatively large individual variations of the basal levels, the effects of the different types of intramedullary injections were expressed for each animal as percentage of the value measured in a blood sample obtained in the same animal 10 min before the intrameduUary injection. One-way ANOVA statistical programme with analysis of variance (F-Test) was used for statistical analysis. RESULTS
..............
I
"-- ...........
f ............
l~
x,,
....
"_-t
- . - , v ,,---- "1 ,Io :i /
Fig. 1. Radioautographic determination of the intramedullary injection site. Fourty nl of PBS containing 2/~Ci of [3H]Leu were injected as described in Materials and Methods. Fourty minutes after the injection, animals were intracardially perfused with 10% formalin. After rinsing in PBS-sucrose, the medulla was cut frontally in a cryostat into 10 ~tm serial thick sections which were subsequently treated for radioautography. In Fig. 1A, the location of the radioautographic reaction clearly indicates that the injection site (arrow) is centered in the dorsal NTS and that the injected tracer spread over both the whole NTS and the dmnX. The different compartments of the dorsal vagal complex included in the micrograph Fig. 1A are delineated in the schematic representation Fig. lB. G = x70. Abbreviations: AP, area postrema; CC, central canal; dmnX, dorsal motor nucleus of the vagus; Gr, gracile nucleus; NTS, nucleus of the tractus solitarius; XII, nucleus of the hypoglossal nerve.
The m e a n basal plasmatic insulin levels o b t a i n e d in the 42 p e n t o b a r b i t a l - a n e s t h e s i z e d intact rats prior to intram e d u l l a r y injection was 4.16 + 0.94 ng/ml. T h e effects of the different microinjection e x p e r i m e n t s are shown in Figs. 2 and 3. N o effects on baseline secretion were o b s e r v e d after injections of the O T vehicule (PBS) alone (Fig. 2A). Microinjections of 4 and 20 p M O T into the dorsal vagal complex of the rat m e d u l l a o b l o n g a t a resulted in a clear decrease in the basal plasmatic insulin levels (Fig. 2B), the response being m a x i m u m 15 min after the injection (59% of basal levels). In contrast, no significant effect was o b s e r v e d after the injection o f 0.04 p M O T (Fig. 2B). In accordance with previous studies, bilateral vagoto m y induced a slight, but non-significant decrease of basal levels of plasmatic insulin (3.43 + 0.81 ng/ml vs 4.28 + 1.03 ng/ml in intact animals). In such v a g o t o m i z e d rats, the basal levels of insulin were n o t m o d i f i e d at 15 min and up to 45 min after an i n t r a m e d u l l a r y injection of PBS or of 4 p M O T (Fig. 3). In contrast with the effects o b s e r v e d after intramedullary injections of OT, a clear increase of the p l a s m a insulin level (131 + 11% o f basal levels) was o b s e r v e d 5 min after an i n t r a m e d u l l a r y injection o f A O T (2 p M ) , which persisted up to 35 min after the injection (Fig. 2C).
192 In such animals pretreated with the OT antagonist, the intramedullary injection of O T produced a slight decrease of plasmatic insulin levels, but which was not significantly different from the corresponding levels measured after the injection of A O T alone (Fig. 2C).
150 -A
PBS
50
150
I
I
i
I
DISCUSSION
I
"B
100
! T ' i % " 50
I
t
f xx I
--~ I
I
x
x
150
J
lOO
f
i
~ ' ~ / ~
OT ~OT
50
o
r
I
~
1~
A
~
Time (mln) Fig. 2. Effects of intrameduUary injections of PBS (A), OT (B) and AOT (C) in intact animals. Arrows indicate time of injection. (A) injection of PBS ( H ) (control, n = 6); (B) injection of OT;
(0---------0)0.04 oM (n = 6); (r-i ....... Fq) 4 pM (n = 8); (O O) 20 pM (n = 8); + and ++ = significantly different from control at P < 0.05 and P < 0.01. (C) injection of AOT; (11--------41)2 pM (n = 6) (+ and ++ = significantlydifferent from control at P < 0.05 and P < 0.01) and double injection of AOT 2 pM and OT 4 pM (~7 ~7) (n = 6) (10 and 20 rain after the injection of OT, plasma insulin levels are not significantly different from levels measured in animals injected with AOT alone). 100-
=
n:4
n=4
I xx
.j
n=8
T
3"--
o so-
E.o _oo_ I:~_E
~E o O-
PE, S OT Fig. 3. Comparison of the effect of intramedullary injection of PBS and OT in intact and vagotomized animals. In intact animals (white columns), results represent the insulin levels measured 15 rain after the injections of PBS or of 4 pM OT and are compared to basal values measured 10 min before the injections. In vagotomized animals (dashed columns) results represent the insulin levels measured 15 min after the injections of PBS or 4 pM OT and are compared to values measured in the same animals 15 min after vagotomy and 10 min before the injections. + + = significantly different from basal values at P < 0.01.
During the past few years, a large number of studies have been devoted to the characterization of the neuronal systems involved in one of the major functions of the hypothalamus, i.e. the control of visceral autonomic functions. Recently, the improvement of neurobiological research tools has led to the identification of various hypothalamic neuron types that project to the medullary dorsal vagal complex 17 and may mediate functions of the vagus, the largest nerve of the parasympathetic nervous system which innervates most of the cephalic, thoracic and upper abdominal viscera under autonomic control 12. According to series of neuroanatomical and physiological studies, it is generally assumed that, among other hypothalamic neuron types, neurons of the parvocellular PVN producing O T represent good candidates for such a role. Indeed, OT-immunoreactive axons originating from the parvocellular PVN have been shown to terminate both in the dmnX which contain the vagal preganglionic neurons and, more densely, in the dorsal part of the nucleus of the tractus solitarius (NTS) 2'22 which is a major visceral afferent nucleus that projects massively to the dmnX 14'16'26. These OT-containing axon terminals were found to establish typical synaptic contacts with neuronal dendrites of the dorsal vagal complex 22'29 and in vitro experiments demonstrated that these synapses were able to release their OT content 3 The present results indicate that the O T axons terminating in the dmnX may participate in the control of the pancreatic secretion of insulin. Under the present experimental conditions, insulin plasma levels were found to be rapidly and reversibly decreased when pmol quantities of OT were injected into the dorsal vagal complex. The present study also provides clear indications that in intact animals, intramedullary injections of a specific OT antagonist produce in contrast a marked increase of plasma insulin levels which persisted up to 35 min after the injection. The significance of such an effect of A O T could be that, under basal conditions, the secretion of insulin is tonically inhibited by the O T innervation of the dorsal vagal complex. The long lasting effect of A O T which clearly contrasts with the reversible effect of OT could be explained by (i) the higher affinity of this OT antagonist than O T for O T receptors 7 and (ii) the slower inactivation of this molecule by O T specific peptidases at the site of injection. Finally, since the AOT-induced
193 increase of plasma insulin was not significantly modified by an intramedullary injection of OT, it can be assumed that in the dorsal medulla, OT and AOT binds to the same receptor types, i.e. OT or OT-like receptors. Since such an inhibition of insulin secretion mimics the effects of the stimulation of the sympathetic motor neurons of the spinal cord 1°'2°, it could be assumed that the effects observed here after intrameduUary injections of OT actually result from a sympathetic activation. Indeed, although the control experiments clearly indicate that our injection spreads were restricted to the NTS and dmnX medullary regions (Fig. 1), it is possible that sympathetic neurons may be influenced by a direct or indirect projection of NTS neurons. However, although this possibility cannot be completely discarded, the present observation that the effects of intramedullary OT injections were abolished by previous sectioning of the two vagus nerves rather suggest that they mainly result from an inhibition of the activity of motor neurons of the vagus. However, the present statement that OT inhibits preganglionic vagal neurons controlling the secretion of insulin appears in contradiction with two previous series of studies indicating that (i) microinjection of OT into the dmnX stimulates the gastric acid secretion 19 and (ii) application of OT on brain slices in vitro stimulates the electrical activity of dmnX neurons 6. The reasons for this discrepancy might be due to marked differences in the sites of action of administered OT. Indeed, although our injection spreads included the dmnX, they were mainly centered on the dorsal NTS which has been shown to contain the largest density in OT synapses throughout the dorsal vagal complex 22'29. Since NTS neurons are known to project massively to the dmnX 14'16'26, it could thus be reasonably assumed that at least part of the physiological response determined by our OT injections resulted from indirect effects of OT on preganglionic vagal neurons via a synaptic neuronal relay. Such a hypothesis is strongly supported by series of recent data indicating that (i) in the rat OT axons are synaptically connected to adrenergic neurons of the dorsal NTS 9'22, the axons of which form very numerous synapses with preganglionic vagal neurons innervating both the stomach and the pancreas 23 and (ii) injections of nM quantities of adrenaline in the dorsal vagal complex produce a decrease of insulin secretion which is very comparable to that obtained after an injection of OT 23. On the contrary, it is probable that the effects reported by Rogers and Dreifuss in fact resulted essentially from direct action of OT on preganglionic vagal neurons. The injections performed by Rogers 19 were indeed centered to the dmnX and it is very likely that the very small volumes injected (1 nl vs. 40 nl for our injections) do not allow the OT to diffuse outside of this nucleus. The data reported by Dreifuss 6 were obtained in
vitro from parasagittal slices of the brainstem, 300-400 /~m thick, including the dmnX. Thus it is probable that under these conditions a number of neurons of the dorso-lateral NTS region which receive OT synaptic contacts (and namely the adrenergic neurons of the C2 region) were not included within the slices or that at least the descending axons coursing from these neurons to the dmnX have been damaged during the dissection. That under these conditions the stimulatory effects of OT on preganglionic vagal neurons do not involve synaptic coupling was indeed attested by their persistence when synaptic transmission was suppressed by incubation of the slices with a solution with a reduced concentration of calcium and an excess of magnesium or by adding tetrodotoxin to the incubation bath. It could thus be postulated that O T axons of the dorsal vagal complex are composed of two functionally different types: (i) some axons which terminate directly on the preganglionic vagal neurons of the dmnX and stimulate the activity of these neurons and (ii) a majority of axons which terminate on neurons of the NTS and which produce an inhibition of the activity of preganglionic neurons by means of a neuronal synaptic relay. The recent literature provides numerous examples of mismatches between the distributions of various types of nerve terminals identified by the immunocytochemical detection of their neurotransmitter and of the corresponding receptors mapped by radioautography. This is the case in the dorsal medulla oblongata where high densities of vasopressin or oxytocin binding were found in the NTS and the dmnX respectively, whereas corresponding synapses were scarce in these regions. Although the significance of such mismatches remains unclear, the radioautographic location of receptors in the medulla oblongata interestingly provides another possible support to the proposition that different sites of injection of OT within the medullary region may produce different effects. Indeed, although less efficiently than vasopressin, vasopressin receptors also bind OT. Thus, the inhibition of motor neurons of the vagus could possibly result, indirectly, from a binding of O T to the vasopressin receptors of the NTS, whereas binding of OT on OT receptors of the dmnX would stimulate these motor neurones. Although OT neurons are at the moment the best characterized of the hypothalamic neuronal system that establish direct connections with the dorsal vagal complex, the importance of this OT system in the control exerted by the hypothalamus on the secretion of insulin remains to be estimated. Namely it has been shown that neurons directly projecting into the dorsal vagal complex are also included in the ventromedial-arcuate nucleus region 24 and the lateral hypothalamus region (Siaud, P.,
194 u n p u b l i s h e d observations) which are hypothalamic regions, well k n o w n to strongly influence the insulin secretion. However, n e u r o n s located in both of these hypothalamic regions are also known to project to the parvocellular PVN 27. Thus, one may assume that besides
the ventromedial and lateral hypothalamus may influence the secretion of insulin indirectly through the control of the O T n e u r o n s of the PVN that innervate the dorsal vagal complex. C o m p l e m e n t a r y studies are now under way in our laboratory to test this hypothesis.
a direct control of the dorsal vagal complex, neurons of REFERENCES 1 Berthoud, H.R. and Jeanrenaud, B., Acute hyperinsulinemia and its reversal by vagotomy after lesions of ventromedial hypothalamus in anesthetized rats, Endocrinology, 105 (1979) 146-151. 2 Buijs, R.M., Intra- and extrahypothalamic vasopressin and oxytocin pathways in the rat. Pathways to limbic system, medulla oblongata and spinal cord, Cell Tiss. Res., 192 (1978) 423-436. 3 Buijs, R.M. and Van Heerikhuize, J.J., Vasopressin and oxytocin release in the brain. A synaptic event, Brain Research, 252 (1982) 71-76. 4 Bray, G.A. and York, D.A., Hypothalamic and genetic obesity in experimental animals: an autonomic and endocrine hypothesis, Physiol. Rev., 59 (1979) 719-809. 5 Bray, G.A., Inoue, S. and Nishizawa, Y., Hypothalamic obesity. The autonomic hypothesis and the lateral hypothalamus, Diabetologia, 20 (1981) 366-377. 6 Dreifuss, J.J., Raggenbass, M., Charpak, S., Dubois-Dauphin, M. and Tribollet, E., A role of central oxytocin in autonomic functions: its action in the motor nucleus of the vagus nerve, Brain Res. Bull., 20 (1988) 765-770. 7 Elands, J., Barberis, C., Jard, S., Tribollet, E., Dreifuss, J.J., Bankowski, K., Manning, M. and Sawyer, H.W., 125I-labelled d(CH2)5 (Tyr(Me)2, Thr4, Tyr-NH29)OVT: a selective oxytocin receptor ligand, Eur. J. Pharmacol., 147 (1987) 197-207. 8 Frohman, L.A., Ezdinli, E.Z. and Javid, R., Effects of vagotomy and vagal stimulation on insulin secretion, Diabetes, 16 (1967) 443-448. 9 Hancock, M.B. and Nicholas, A.P., Oxytocin immunoreactive projections onto medullary adrenaline neurons, Brain Res. Bull., 18 (1987) 213-219. 10 Helman, A., Marre, M., Bobbioni, E., Poussier, P., Reach, G. and Assan, R., The brain islet axis: the nervous control of the endocrine pancreas, Diab~te et M~tabolisme (Paris), 8 (1982) 53-64. 11 Herbert, V., Lou, K.S., Gottlieb, C.W. and Blickert, S.J., Coated carcoal immunoassay of insulin, J. Clin. Endocrinol. Metal., 25 (1965) 1375-1384. 12 Kalia, M. and Mesulam, M.M., Brainstem projections of sensory and motor components of the vagus complex in the cat: II. laryngeal, tracheo-bronchial, pulmonary, cardiac and gastrointestinal branches, J. Comp. Neurol., 193 (1980) 467-508. 13 Luiten, P.G.M., Ter Horst, G.J., Karst, H. and Steffens, A.B., The course of paraventricular hypothalamic efferent to autonomic structures in the medulla and the spinal cord, Brain Research, 329 (1985) 374-378. 14 Morest, D.K., Experimental study of the projections of the nucleus of the tractus solitarius and the area postrema in the rat, J. Comp. Neurol., 130 (1987) 277-290. 15 Nilaver, G., Zimmerman, E.A., Wilkins, J., Hoffman, D. and Silverman, A.J., Magnocellular hypothalamic projections to the lower brainstem and spinal cord of the rat. Immunocytochemical evidence for predominance of the oxytocin-neurophysin system compared to the vasopressin-neurophysin system, Neuroendocrinology, 30 (1980) 150-158. 16 Norgren, R., Projections from the nucleus of the solitary tract,
Neuroscience, 3 (t978) 207-218. 17 Palkovits, M., Distribution of neuroactive substances in the dorsal vagal complex of the medulla oblongata, Neurochem. Int., 7 (1985) 213-219. 18 Powley, T.L. and Laughton, W., Neural pathways involved in the hypothalamic integration of autonomic responses, Diabetologia, 20 (1981) 378-387. 19 Rogers, R.C. and Hermann, G.E., Dorsal medullary oxytocin, vasopressin, oxytocin antagonist and TRH effects on gastric acid secretion and heart rate, Peptides, 6 (1985) 1143-1148. 20 Rohner-Jeanrenaud, E and Jeanrenaud, B., Interactions entre le syst~me nerveux central, le pancrras endocrine et le mrtabolisme, Ann. Endoc., 48 (1987) 400-406. 21 Sawchenko, P.E. and Swanson, L.W., Immunocytochemical identification of neurons in the paraventricular nucleus of the hypothalamus that project to the medulla or to the spinal cord in the rat, J. Comp. Neurol., 205 (1982) 260-272. 22 Siaud, P., Denoroy, L., Assenmacher, I. and Alonso, G., Comparative immunocytochemical study of the catecholaminergic and peptidergic afferent innervation to the dorsal vagal complex in rat and guinea pig, J. Comp. Neurol., 290 (1989) 323-335. 23 Siaud, P., Puech, R., Assenmacher, I. and Alonso, G., The adrenergic innervation of the dorsal motor nucleus of the vagus: possible involvement in the inhibitory control of gastric acid and pancreatic insulin secretions, Cell Tiss. Res., 259 (1990) 535-542. 24 Sofroniew, M.V., Vascular and neural projections of hypothalamic neurons producing neurohypophysial of ACTH related peptides. In Baertschi, A.J. and Dreifuss, J.J. (Eds.), Neuroendocrinology of vasopressin, corticoliberin and opiomelanocortins, New York Academic Press, 1982, pp. 73-86. 25 Sofroniew, M.V. and Schrell, U., Evidence for a direct projection from oxytocin and vasopressin neurons in the hypothalamic paraventricular nucleus of the medulla oblongata: immunohistochemical visualization of both the horseradish peroxidase transported and the peptide produced by the same neurons, Neurosci. Lett., 22 (1981) 211-217. 26 Ter Horst, G.J., De Boer, P., Luiten, P.G.M. and Van Willigen, J.D., Ascending projections from the solitary tract nucleus to the hypothalamus. A phaseolus vulgaris lectin tracing study in the rat, Neuroscience, 31 (1989) 785-797. 27 Ter Horst, G.J. and Luiten, P.G.M., Phaseolus vulgaris leucoagglutinin tracing of intra-hypothalamic connections of the lateral, ventromedial and paraventricular hypothalamic nuclei in the rat, Brain Res. Bull., 18 (1987) 191-203. 28 Tribollet, E., Barberis, C., Jard, S., Elands, J., DuboisDauphin, M., Marguerat, A. and Dreifuss, J.J., Mapping and analysis of receptors for neurohypophyseal peptides present in the brain. In B.T. Picketing, J.B. Wakerley and A.J.S. Summerlee (Eds.), Neurosecretion: cellular aspects of the production and release of neuropeptides, Plenum, New York, 1988, pp. 81-89. 29 Voorn, E. and Buijs, R.N., An immuno-electronmicroscopal study comparing vasopressin, oxytocin, substance P and enkephalin containing nerve terminals in the nucleus of the solitary tract of the rat, Brain Research, 270 (1983) 169-173.
190
BRES 16492
Microinjection of oxytocin into the dorsal vagal complex decreases pancreatic insulin secretion P. Siaud 1, R. P u e c h e, I. A s s e n m a c h e r I a n d G. A l o n s o 1 i Laboratoire de Neurobiologie Endocrinologique, URA 1197 CNRS, Universit~ de Montpellier H and 2Laboratoire de Pharmacologie et de Pharmacodynamie, URA 85 CNRS, Universit~ de Montpellier L Montpellier (France)
(Accepted 30 October 1990) Key words: Oxytocin; Insulin; Parasympathetic nervous system; Medulla oblongata; Rat
Microinjections of oxytocin and of an oxytocin antagonist into the dorsal vagal complex of the medulla oblongata were performed in order to study the possible role of the oxytocin containing axons that innervate this region in the regulation of pancreatic insulin secretion. No significant effect was produced by the intrameduUary injection of the oxytocin vehicle alone or of 0.04 pM oxytoein. Injections of 4 and 20 pM oxytocin produced a reversible decrease of plasmatic insulin levels which fall to 59% of basal levels 15 min after the injection. Such an effect was abolished when 4 pM oxytocin was injected to animals which have been previously bilaterally vagotomized. In contrast to oxytocin, intramedullary injection of a specific antagonist of oxytocin to intact animals induced a marked increase of plasmatic insulin levels which raised 131% of basal levels 15 min after the injection. In animals receiving such an injection of oxytocin antagonist, a secondary injection of 4 pM oxytocin produced a slight but not significant decrease of plasmatic insulin levels. These data strongly suggest that the hypothalamic neurons producing oxytocin that densely project to the dorsal vagal complex may be involved in an inhibitory control of the vagal preganglionic neurons that innervate the pancreas. INTRODUCTION It has now been clearly demonstrated that the vagus nerve conveys inputs that stimulate the pancreatic secretion of insulin 8 and that the preganglionic parasympathetic neurons responsible for this control are mainly located in the dorsal motor nucleus of the vagus nerve (dmnX) in the dorsal vagal complex of the medulla oblongata ~8'23. It is also well known that the hypothalamus plays a major role in the central regulation of the pancreatic secretion of insulin w'18'2°. For instance, several earlier experiments clearly demonstrated that basal insulin secretion was dramatically modified both by the electrical stimulation or electrolytic lesion of hypothalamic nuclei 1'4'5"1°'z°, strongly suggesting the occurrence of direct neuronal connections between the hypothalamus and the autonomic centers innervating the pancreas. Recently, this idea of a hypothalamic-islet axis has been reinforced by a series of neuroanatomical data indicating that a n u m b e r of hypothalamic perikarya directly project to the dorsal vagal complex of the medulla 13'21"24'25. A m o n g the several types of hypothalamic neurons that have been shown to project to the dorsal vagal complex, oxytocin (OT)-producing neurons located in the parvo-
cellular part of the paraventricular nucleus (PVN) certainly represent the system which provides t h e densest innervation of this medullary nucleus 15'21"25. Additional data indicating that OT-containing axons establish frequent synaptic contacts with the perikarya of this medullary region 22'29 and that the dorsal vagal complex and more specifically the d m n X contain high densities of OT-binding sites 28, strongly suggest that hypothalamic O T neurons could be involved in the regulation of visceral functions under the control of the vagus. Accordingly, previous data indicated that the application of O T to the dorsal medullary region modifies both the electrical activity of neurons of the d m n X 6 and the secretion of gastric acidity 19. The present study was therefore undertaken to study the possible role of O T on the medullary control of the secretion of pancreatic insulin. MATERIALS AND METHODS Fourty-two male adult Sprague-Dawley rats (200-250 g) were used in this study. They were housed in fight- and temperaturecontrolled rooms (12 h fight starting at 7.00 h and 12 h darkness, at 21 + 1 °C) and had free access to standard dry food and tap water.
Correspondence: P. Siaud, Laboratoire de Neurobiologie Endocrinologique, URA 1197 CNRS, Universit~ de Montpellier II, Place E. Bataillon, 34095 Montpeilier Cedex 05, France.
191 Under deep Equithesine anesthesia, a polyethylene cannula filled with heparinized (250 U/ml) saline buffer was inserted into the left carotid artery, which was carefully exposed to avoid damage of the nearby vagus nerve. The animals were placed in a stereotaxic device with the head in the nose down position. The dorsal cervical musculature was resected and the obex region of the dorsal medulla was exposed after removing the occipital skull plate. In an other group of rats, after cannulation of the carotid artery, bilateral vagotomy was performed by cutting the vagal trunks running along the carotids. The intramedullary microinjections were performed by means of the 1 /zl Hamilton microsyringe fitted with a glass micropipette with an outside tip diameter of 40-50/~m. Both the stereotaxic coordinates of the injection site and the volume of the injection were determined in order to obtain the impregnation of the dorsal vagal complex throughout its rostro-caudal extent. According to data obtained in a parallel neuroanatomical study based on the injection of tritiated amino acids into the dorsal vagal
complex, this condition appeared fulfilled by positioning the tip of the micropipette at about 0.1 mm anterior, 0.9 mm lateral to the obex and 0.2-0.3 mm ventral to the brain surface and by injecting a volume of 40 nl (see Fig. 1). The substances injected in the present study were: oxytocin (Bachem) (0.04, 4 and 20 pM OT/injection) and an OT antagonist (AOT = d(CH2) 5 (Tyr(Me)2, Thr4, TyrNHE9)OVT)(kindly provided by Dr. C. Barberis, Montpellier, France) (2 pM AOT/injection). They were infused in 40 nl phosphate-buffered saline (PBS) over 1 rain on the side opposite to the cannulated carotid. The micropipette was left in place an additional 2 rain before withdrawal to minimize back diffusion along the cannulae tract. For single intramedullary injections of PBS, OT, or AOT, 100/~l blood samples were collected in heparinized tubes 10 min before the injection and 5, 15, 25 and 35 min after the injection. Double intramedullary injections consisted of an injection of 2 pM AOT followed 15 rain after by an injection of 4 pM OT. For these double injection experiments, blood samples were collected 10 rain after the injection of AOT and then 15 min after the OT administration. After centrifugation, aliquots of plasma were stored at -30 °C. Insulin was assayed by the method of Herbert et al. 11 using an antibody supplied by Miles laboratories (Paris, France). [125I]insufin was obtained from international CIS (Gif/Yvette, France). The standard used was pure rat insulin (Novo, Copenhagen, Denmark) whose biological activity is 22.3/~U/ng. The intra- and inter-assay coefficients of variations were 9.7% and 14.4% respectively, and the sensitivity was 0.2 ng/ml. Because of the relatively large individual variations of the basal levels, the effects of the different types of intramedullary injections were expressed for each animal as percentage of the value measured in a blood sample obtained in the same animal 10 min before the intrameduUary injection. One-way ANOVA statistical programme with analysis of variance (F-Test) was used for statistical analysis. RESULTS
..............
I
"-- ...........
f ............
l~
x,,
....
"_-t
- . - , v ,,---- "1 ,Io :i /
Fig. 1. Radioautographic determination of the intramedullary injection site. Fourty nl of PBS containing 2/~Ci of [3H]Leu were injected as described in Materials and Methods. Fourty minutes after the injection, animals were intracardially perfused with 10% formalin. After rinsing in PBS-sucrose, the medulla was cut frontally in a cryostat into 10 ~tm serial thick sections which were subsequently treated for radioautography. In Fig. 1A, the location of the radioautographic reaction clearly indicates that the injection site (arrow) is centered in the dorsal NTS and that the injected tracer spread over both the whole NTS and the dmnX. The different compartments of the dorsal vagal complex included in the micrograph Fig. 1A are delineated in the schematic representation Fig. lB. G = x70. Abbreviations: AP, area postrema; CC, central canal; dmnX, dorsal motor nucleus of the vagus; Gr, gracile nucleus; NTS, nucleus of the tractus solitarius; XII, nucleus of the hypoglossal nerve.
The m e a n basal plasmatic insulin levels o b t a i n e d in the 42 p e n t o b a r b i t a l - a n e s t h e s i z e d intact rats prior to intram e d u l l a r y injection was 4.16 + 0.94 ng/ml. T h e effects of the different microinjection e x p e r i m e n t s are shown in Figs. 2 and 3. N o effects on baseline secretion were o b s e r v e d after injections of the O T vehicule (PBS) alone (Fig. 2A). Microinjections of 4 and 20 p M O T into the dorsal vagal complex of the rat m e d u l l a o b l o n g a t a resulted in a clear decrease in the basal plasmatic insulin levels (Fig. 2B), the response being m a x i m u m 15 min after the injection (59% of basal levels). In contrast, no significant effect was o b s e r v e d after the injection o f 0.04 p M O T (Fig. 2B). In accordance with previous studies, bilateral vagoto m y induced a slight, but non-significant decrease of basal levels of plasmatic insulin (3.43 + 0.81 ng/ml vs 4.28 + 1.03 ng/ml in intact animals). In such v a g o t o m i z e d rats, the basal levels of insulin were n o t m o d i f i e d at 15 min and up to 45 min after an i n t r a m e d u l l a r y injection of PBS or of 4 p M O T (Fig. 3). In contrast with the effects o b s e r v e d after intramedullary injections of OT, a clear increase of the p l a s m a insulin level (131 + 11% o f basal levels) was o b s e r v e d 5 min after an i n t r a m e d u l l a r y injection o f A O T (2 p M ) , which persisted up to 35 min after the injection (Fig. 2C).
192 In such animals pretreated with the OT antagonist, the intramedullary injection of O T produced a slight decrease of plasmatic insulin levels, but which was not significantly different from the corresponding levels measured after the injection of A O T alone (Fig. 2C).
150 -A
PBS
50
150
I
I
i
I
DISCUSSION
I
"B
100
! T ' i % " 50
I
t
f xx I
--~ I
I
x
x
150
J
lOO
f
i
~ ' ~ / ~
OT ~OT
50
o
r
I
~
1~
A
~
Time (mln) Fig. 2. Effects of intrameduUary injections of PBS (A), OT (B) and AOT (C) in intact animals. Arrows indicate time of injection. (A) injection of PBS ( H ) (control, n = 6); (B) injection of OT;
(0---------0)0.04 oM (n = 6); (r-i ....... Fq) 4 pM (n = 8); (O O) 20 pM (n = 8); + and ++ = significantly different from control at P < 0.05 and P < 0.01. (C) injection of AOT; (11--------41)2 pM (n = 6) (+ and ++ = significantlydifferent from control at P < 0.05 and P < 0.01) and double injection of AOT 2 pM and OT 4 pM (~7 ~7) (n = 6) (10 and 20 rain after the injection of OT, plasma insulin levels are not significantly different from levels measured in animals injected with AOT alone). 100-
=
n:4
n=4
I xx
.j
n=8
T
3"--
o so-
E.o _oo_ I:~_E
~E o O-
PE, S OT Fig. 3. Comparison of the effect of intramedullary injection of PBS and OT in intact and vagotomized animals. In intact animals (white columns), results represent the insulin levels measured 15 rain after the injections of PBS or of 4 pM OT and are compared to basal values measured 10 min before the injections. In vagotomized animals (dashed columns) results represent the insulin levels measured 15 min after the injections of PBS or 4 pM OT and are compared to values measured in the same animals 15 min after vagotomy and 10 min before the injections. + + = significantly different from basal values at P < 0.01.
During the past few years, a large number of studies have been devoted to the characterization of the neuronal systems involved in one of the major functions of the hypothalamus, i.e. the control of visceral autonomic functions. Recently, the improvement of neurobiological research tools has led to the identification of various hypothalamic neuron types that project to the medullary dorsal vagal complex 17 and may mediate functions of the vagus, the largest nerve of the parasympathetic nervous system which innervates most of the cephalic, thoracic and upper abdominal viscera under autonomic control 12. According to series of neuroanatomical and physiological studies, it is generally assumed that, among other hypothalamic neuron types, neurons of the parvocellular PVN producing O T represent good candidates for such a role. Indeed, OT-immunoreactive axons originating from the parvocellular PVN have been shown to terminate both in the dmnX which contain the vagal preganglionic neurons and, more densely, in the dorsal part of the nucleus of the tractus solitarius (NTS) 2'22 which is a major visceral afferent nucleus that projects massively to the dmnX 14'16'26. These OT-containing axon terminals were found to establish typical synaptic contacts with neuronal dendrites of the dorsal vagal complex 22'29 and in vitro experiments demonstrated that these synapses were able to release their OT content 3 The present results indicate that the O T axons terminating in the dmnX may participate in the control of the pancreatic secretion of insulin. Under the present experimental conditions, insulin plasma levels were found to be rapidly and reversibly decreased when pmol quantities of OT were injected into the dorsal vagal complex. The present study also provides clear indications that in intact animals, intramedullary injections of a specific OT antagonist produce in contrast a marked increase of plasma insulin levels which persisted up to 35 min after the injection. The significance of such an effect of A O T could be that, under basal conditions, the secretion of insulin is tonically inhibited by the O T innervation of the dorsal vagal complex. The long lasting effect of A O T which clearly contrasts with the reversible effect of OT could be explained by (i) the higher affinity of this OT antagonist than O T for O T receptors 7 and (ii) the slower inactivation of this molecule by O T specific peptidases at the site of injection. Finally, since the AOT-induced
193 increase of plasma insulin was not significantly modified by an intramedullary injection of OT, it can be assumed that in the dorsal medulla, OT and AOT binds to the same receptor types, i.e. OT or OT-like receptors. Since such an inhibition of insulin secretion mimics the effects of the stimulation of the sympathetic motor neurons of the spinal cord 1°'2°, it could be assumed that the effects observed here after intrameduUary injections of OT actually result from a sympathetic activation. Indeed, although the control experiments clearly indicate that our injection spreads were restricted to the NTS and dmnX medullary regions (Fig. 1), it is possible that sympathetic neurons may be influenced by a direct or indirect projection of NTS neurons. However, although this possibility cannot be completely discarded, the present observation that the effects of intramedullary OT injections were abolished by previous sectioning of the two vagus nerves rather suggest that they mainly result from an inhibition of the activity of motor neurons of the vagus. However, the present statement that OT inhibits preganglionic vagal neurons controlling the secretion of insulin appears in contradiction with two previous series of studies indicating that (i) microinjection of OT into the dmnX stimulates the gastric acid secretion 19 and (ii) application of OT on brain slices in vitro stimulates the electrical activity of dmnX neurons 6. The reasons for this discrepancy might be due to marked differences in the sites of action of administered OT. Indeed, although our injection spreads included the dmnX, they were mainly centered on the dorsal NTS which has been shown to contain the largest density in OT synapses throughout the dorsal vagal complex 22'29. Since NTS neurons are known to project massively to the dmnX 14'16'26, it could thus be reasonably assumed that at least part of the physiological response determined by our OT injections resulted from indirect effects of OT on preganglionic vagal neurons via a synaptic neuronal relay. Such a hypothesis is strongly supported by series of recent data indicating that (i) in the rat OT axons are synaptically connected to adrenergic neurons of the dorsal NTS 9'22, the axons of which form very numerous synapses with preganglionic vagal neurons innervating both the stomach and the pancreas 23 and (ii) injections of nM quantities of adrenaline in the dorsal vagal complex produce a decrease of insulin secretion which is very comparable to that obtained after an injection of OT 23. On the contrary, it is probable that the effects reported by Rogers and Dreifuss in fact resulted essentially from direct action of OT on preganglionic vagal neurons. The injections performed by Rogers 19 were indeed centered to the dmnX and it is very likely that the very small volumes injected (1 nl vs. 40 nl for our injections) do not allow the OT to diffuse outside of this nucleus. The data reported by Dreifuss 6 were obtained in
vitro from parasagittal slices of the brainstem, 300-400 /~m thick, including the dmnX. Thus it is probable that under these conditions a number of neurons of the dorso-lateral NTS region which receive OT synaptic contacts (and namely the adrenergic neurons of the C2 region) were not included within the slices or that at least the descending axons coursing from these neurons to the dmnX have been damaged during the dissection. That under these conditions the stimulatory effects of OT on preganglionic vagal neurons do not involve synaptic coupling was indeed attested by their persistence when synaptic transmission was suppressed by incubation of the slices with a solution with a reduced concentration of calcium and an excess of magnesium or by adding tetrodotoxin to the incubation bath. It could thus be postulated that O T axons of the dorsal vagal complex are composed of two functionally different types: (i) some axons which terminate directly on the preganglionic vagal neurons of the dmnX and stimulate the activity of these neurons and (ii) a majority of axons which terminate on neurons of the NTS and which produce an inhibition of the activity of preganglionic neurons by means of a neuronal synaptic relay. The recent literature provides numerous examples of mismatches between the distributions of various types of nerve terminals identified by the immunocytochemical detection of their neurotransmitter and of the corresponding receptors mapped by radioautography. This is the case in the dorsal medulla oblongata where high densities of vasopressin or oxytocin binding were found in the NTS and the dmnX respectively, whereas corresponding synapses were scarce in these regions. Although the significance of such mismatches remains unclear, the radioautographic location of receptors in the medulla oblongata interestingly provides another possible support to the proposition that different sites of injection of OT within the medullary region may produce different effects. Indeed, although less efficiently than vasopressin, vasopressin receptors also bind OT. Thus, the inhibition of motor neurons of the vagus could possibly result, indirectly, from a binding of O T to the vasopressin receptors of the NTS, whereas binding of OT on OT receptors of the dmnX would stimulate these motor neurones. Although OT neurons are at the moment the best characterized of the hypothalamic neuronal system that establish direct connections with the dorsal vagal complex, the importance of this OT system in the control exerted by the hypothalamus on the secretion of insulin remains to be estimated. Namely it has been shown that neurons directly projecting into the dorsal vagal complex are also included in the ventromedial-arcuate nucleus region 24 and the lateral hypothalamus region (Siaud, P.,
194 u n p u b l i s h e d observations) which are hypothalamic regions, well k n o w n to strongly influence the insulin secretion. However, n e u r o n s located in both of these hypothalamic regions are also known to project to the parvocellular PVN 27. Thus, one may assume that besides
the ventromedial and lateral hypothalamus may influence the secretion of insulin indirectly through the control of the O T n e u r o n s of the PVN that innervate the dorsal vagal complex. C o m p l e m e n t a r y studies are now under way in our laboratory to test this hypothesis.
a direct control of the dorsal vagal complex, neurons of REFERENCES 1 Berthoud, H.R. and Jeanrenaud, B., Acute hyperinsulinemia and its reversal by vagotomy after lesions of ventromedial hypothalamus in anesthetized rats, Endocrinology, 105 (1979) 146-151. 2 Buijs, R.M., Intra- and extrahypothalamic vasopressin and oxytocin pathways in the rat. Pathways to limbic system, medulla oblongata and spinal cord, Cell Tiss. Res., 192 (1978) 423-436. 3 Buijs, R.M. and Van Heerikhuize, J.J., Vasopressin and oxytocin release in the brain. A synaptic event, Brain Research, 252 (1982) 71-76. 4 Bray, G.A. and York, D.A., Hypothalamic and genetic obesity in experimental animals: an autonomic and endocrine hypothesis, Physiol. Rev., 59 (1979) 719-809. 5 Bray, G.A., Inoue, S. and Nishizawa, Y., Hypothalamic obesity. The autonomic hypothesis and the lateral hypothalamus, Diabetologia, 20 (1981) 366-377. 6 Dreifuss, J.J., Raggenbass, M., Charpak, S., Dubois-Dauphin, M. and Tribollet, E., A role of central oxytocin in autonomic functions: its action in the motor nucleus of the vagus nerve, Brain Res. Bull., 20 (1988) 765-770. 7 Elands, J., Barberis, C., Jard, S., Tribollet, E., Dreifuss, J.J., Bankowski, K., Manning, M. and Sawyer, H.W., 125I-labelled d(CH2)5 (Tyr(Me)2, Thr4, Tyr-NH29)OVT: a selective oxytocin receptor ligand, Eur. J. Pharmacol., 147 (1987) 197-207. 8 Frohman, L.A., Ezdinli, E.Z. and Javid, R., Effects of vagotomy and vagal stimulation on insulin secretion, Diabetes, 16 (1967) 443-448. 9 Hancock, M.B. and Nicholas, A.P., Oxytocin immunoreactive projections onto medullary adrenaline neurons, Brain Res. Bull., 18 (1987) 213-219. 10 Helman, A., Marre, M., Bobbioni, E., Poussier, P., Reach, G. and Assan, R., The brain islet axis: the nervous control of the endocrine pancreas, Diab~te et M~tabolisme (Paris), 8 (1982) 53-64. 11 Herbert, V., Lou, K.S., Gottlieb, C.W. and Blickert, S.J., Coated carcoal immunoassay of insulin, J. Clin. Endocrinol. Metal., 25 (1965) 1375-1384. 12 Kalia, M. and Mesulam, M.M., Brainstem projections of sensory and motor components of the vagus complex in the cat: II. laryngeal, tracheo-bronchial, pulmonary, cardiac and gastrointestinal branches, J. Comp. Neurol., 193 (1980) 467-508. 13 Luiten, P.G.M., Ter Horst, G.J., Karst, H. and Steffens, A.B., The course of paraventricular hypothalamic efferent to autonomic structures in the medulla and the spinal cord, Brain Research, 329 (1985) 374-378. 14 Morest, D.K., Experimental study of the projections of the nucleus of the tractus solitarius and the area postrema in the rat, J. Comp. Neurol., 130 (1987) 277-290. 15 Nilaver, G., Zimmerman, E.A., Wilkins, J., Hoffman, D. and Silverman, A.J., Magnocellular hypothalamic projections to the lower brainstem and spinal cord of the rat. Immunocytochemical evidence for predominance of the oxytocin-neurophysin system compared to the vasopressin-neurophysin system, Neuroendocrinology, 30 (1980) 150-158. 16 Norgren, R., Projections from the nucleus of the solitary tract,
Neuroscience, 3 (t978) 207-218. 17 Palkovits, M., Distribution of neuroactive substances in the dorsal vagal complex of the medulla oblongata, Neurochem. Int., 7 (1985) 213-219. 18 Powley, T.L. and Laughton, W., Neural pathways involved in the hypothalamic integration of autonomic responses, Diabetologia, 20 (1981) 378-387. 19 Rogers, R.C. and Hermann, G.E., Dorsal medullary oxytocin, vasopressin, oxytocin antagonist and TRH effects on gastric acid secretion and heart rate, Peptides, 6 (1985) 1143-1148. 20 Rohner-Jeanrenaud, E and Jeanrenaud, B., Interactions entre le syst~me nerveux central, le pancrras endocrine et le mrtabolisme, Ann. Endoc., 48 (1987) 400-406. 21 Sawchenko, P.E. and Swanson, L.W., Immunocytochemical identification of neurons in the paraventricular nucleus of the hypothalamus that project to the medulla or to the spinal cord in the rat, J. Comp. Neurol., 205 (1982) 260-272. 22 Siaud, P., Denoroy, L., Assenmacher, I. and Alonso, G., Comparative immunocytochemical study of the catecholaminergic and peptidergic afferent innervation to the dorsal vagal complex in rat and guinea pig, J. Comp. Neurol., 290 (1989) 323-335. 23 Siaud, P., Puech, R., Assenmacher, I. and Alonso, G., The adrenergic innervation of the dorsal motor nucleus of the vagus: possible involvement in the inhibitory control of gastric acid and pancreatic insulin secretions, Cell Tiss. Res., 259 (1990) 535-542. 24 Sofroniew, M.V., Vascular and neural projections of hypothalamic neurons producing neurohypophysial of ACTH related peptides. In Baertschi, A.J. and Dreifuss, J.J. (Eds.), Neuroendocrinology of vasopressin, corticoliberin and opiomelanocortins, New York Academic Press, 1982, pp. 73-86. 25 Sofroniew, M.V. and Schrell, U., Evidence for a direct projection from oxytocin and vasopressin neurons in the hypothalamic paraventricular nucleus of the medulla oblongata: immunohistochemical visualization of both the horseradish peroxidase transported and the peptide produced by the same neurons, Neurosci. Lett., 22 (1981) 211-217. 26 Ter Horst, G.J., De Boer, P., Luiten, P.G.M. and Van Willigen, J.D., Ascending projections from the solitary tract nucleus to the hypothalamus. A phaseolus vulgaris lectin tracing study in the rat, Neuroscience, 31 (1989) 785-797. 27 Ter Horst, G.J. and Luiten, P.G.M., Phaseolus vulgaris leucoagglutinin tracing of intra-hypothalamic connections of the lateral, ventromedial and paraventricular hypothalamic nuclei in the rat, Brain Res. Bull., 18 (1987) 191-203. 28 Tribollet, E., Barberis, C., Jard, S., Elands, J., DuboisDauphin, M., Marguerat, A. and Dreifuss, J.J., Mapping and analysis of receptors for neurohypophyseal peptides present in the brain. In B.T. Picketing, J.B. Wakerley and A.J.S. Summerlee (Eds.), Neurosecretion: cellular aspects of the production and release of neuropeptides, Plenum, New York, 1988, pp. 81-89. 29 Voorn, E. and Buijs, R.N., An immuno-electronmicroscopal study comparing vasopressin, oxytocin, substance P and enkephalin containing nerve terminals in the nucleus of the solitary tract of the rat, Brain Research, 270 (1983) 169-173.
