0306-4522/91 $3.00 + 0.00 Pergamon Press plc
Neuroscience Vol. 45, No. 1, pp. 153-160, 1991 Printed in Great Britain
0 1991IBRO
LOCUS
COERULEUS PROJECTIONS TO THE DORSAL MOTOR VAGUS NUCLEUS IN THE RAT G. J. TERHORST,G. J. TOESand J. D. VAN WILLIGEN
Department of Neurobiology and Oral Physiology, University of Groningen, Bloemsingel 10, 9712 KZ Groningen, The Netherlands Abatrac&-The origin of the noradrenergic innervation of the preganglionic autonomic nuclei in the medulla oblongata and spinal cord is still controversial. In this investigation descending connections of the locus coeruleus to the dorsal motor vagus nucleus in the rat are studied with Phase&s oulgaris leucoagglutinin and horseradish peroxidase as neuroanatomical tracers. Locus coeruleus projections in the motor vagus nucleus are found in the medial part at rostra1 levels and in the lateral part at intermediate levels of this nucleus. The terminal labeling in the lateral intermediate part of the vagus nucleus appears in an area where possibly preganglionic parasympathetic cardiac neurons are located, suggesting that the locus coeruleus might be involved in regulation of cardiovascular functions. After small iontophoretic injections of horseradish peroxidase in the motor vagus nucleus, retrogradely labeled cells aie found in the ventral part of the locus coeruleus and occasionally in the dorsal part of the nucleus. The results show that the locus coeruleus-dorsal motor vagus nucleus pathway may participate in the inhibition of the cardiac preganglionic neurons in the dorsal motor vagus nucleus by the hypothalamic paraventricular nucleus.
The hypothalamic paraventricular nucleus (PVN) is an important structure for limbic control of autonomic functions.‘6*32 The PVN may exert its autonomic responses via direct projections to the preganglionic autonomic cells in the dorsal motor vagus nucleus (DMnX), the nucleus ambiguus and the intermediolateral cell groups in the thoracic spinal cord.17*24,31 In this paradigm, indirect pathways from the PVN to the preganglionic autonomic nuclei could modulate thereafter the initial response caused by the direct innervation, because of longer signal transduction times. Nuclei with autonomic functions receiving a PVN innervation are the ventromedial, lateral and dorsomedial hypothalamic nuclei, the periaqueductal gray and the caudal parvocellular reticular formation. I73 However, the PVN also innervates the locus coeruleus (LC),” a noradrenergic nucleus with widespread projections to numerous forebrain,21 brainstem and spinal areas6,’ and probably very global functions. LC involvement in autonomic regulatory functions is controversial. The coeruleo-DMnX projection has been the focus of many studies in a variety of species, yet considerable disagreement still exists about this pathway. In monkeys, LC-DMnX connections were found with autoradiographic tract-tracing techniques and dopaimmunohistochemistry.39 In mine-/3-hydroxylase Abbreuiarions: DMnX, dorsal motor vagus nucleus; HRP,
horseradish peroxidase; LC, locus coeruleus; PAP, peroxidase-antiperoxidase; PHA-L, Phaseolus vulgaris leuPVN, hypothalamic paraventricular coagglutinin; nucleus; TBS, Tris-buffered saline; TBS-T, Tris-buffered saline with Triton X-100; TH, tyrosine hydroxylase; TMB, tetramethylbenzidine; WGA-HRP, wheat germ agglutinin-conjugated horseradish peroxidase.
the rat, the LC-DMnX pathway was identified with catecholamine autofluorescence in LC-lesioned animals,13 wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP)” and anterograde transport of tritiated proteins.’ However, tracing of DMnX afferents with HRP in the same species, revealed labeled cells near and not in the LC.37 Moreover, the connection was also not revealed with more sensitive immunocytochemical tracing techniques.6x7 Tract-tracing techniques have limitations which may be relevant when they are used for identifying connections of small-sized nuclei like the DMnX and LC. In autoradiographic tracing one gets an indirect image of the labeling in the form of silver grains in the overlying photographic emulsion. Precise identification of target areas in autoradiographic preparations is difficult because labeled fibers and terminal presynaptic endings cannot readily be distinguished from each other and, in sparsely labeled areas, from the background labeling. In retrograde tracing of connections with HRP, the size of the injection, which is particularly relevant for a smallsized nucleus like the DMnX, and undesired uptake of HRP by damaged passing fibers, impede proper interpretation of labeled connections. The immunocytochemical Phaseolus oulgaris leucoagglutinin (PHA-L) tracing method proved to be very effective for the identification of short and long distance projectionss~36 because it is not hampered by haphazard spreading of injected tracer or aspecific debris and precipitates. With this technique it is possible to make small iontophoretic injections. Moreover, the PHA-L method, in contrast to autoradiographic and HRP tracing, yields completely
153
G. J. TEH HORS-~ et ul. NEGATIVE
POSITIVE
F
Fig. t. Series of transverse sections from the rostra1 (A) to the caudal (F) level of the LC showing the location and dimensions of the PHA-L injections into the LC region. The LC and nucleus sulxoeruleus are shaded and the different cases are indicated with the numbers. Injections causing terminal bouton labeling in the dorsal DMnX (positive) are illustrated on the right side. Control (negative) injections are shown on the left side. 2. Series of transverse sections showing the location of retrogradely labeled cells (black dots) in the LC after a small HRP injection into the ipsilaterai DMnX.
Abbreviation
Bar DTg 37 lc, LC LVe meT mlf moT, MOT
Barrington’s nucleus dorsal tegmental nucleus germ of the facial nerve locus coeruleus lateral vestibular nucleus mesencephalic trigeminal nucleus medial Ion~t~in~ fasci4us motor trigeminal nucleus
used in the figures
n7, 7 Pb PHA-L :I ts X
XII
SOL
facial nerve parabrachial nucleus PhaseoIus vulgaris ie~a~lutinin superior cerebellar peduncle nucleus of the solitary tract solitary tract dorsal motor vagus nucieus hypoglossal nucleus
155
Locus coeruleus projections
filled axons that can be traced from the labeled cell body to the target area where the characteristic varicosities and terminal boutons appear that indicate the site of presynaptic terminations.8*Nv40 The immunocytochemical PI-IA-L technique gave optimal results with survival times of seven to 10 days in our hand~.~ Longer survival times caused weaker signals in the target areas, indicating that for identification of small projections, like the LC-DMnX pathway, the post-operative survival time is an important factor. Therefore, in addition to Fritschy and Grazanna,’ not showing LC-DMnX projections with survival times of 14-21 days, we have studied efferent connections of the LC to the dorsal motor vagus nucleus, seven days after the injection of PHA-L in the LC. Because the injections of PHA-L always labeled cells near the LC, small iontophoretic injections of HRP into the DMnX were used as controls. The noradrenergic nature of the HRP-labeled cells in the LC region was demonstrated with tyrosine hydroxylase (TH) staining of the adjacent sections.
EXPR-AL
PROCEDURES
The PHA-L experiments were conducted on 12 male Wistar rats (locally bred at the CDL, Univ. Groningen) of 250-3001. For injecting PHA-L, the animals were anesthetized with halothane and placed in a Kopf stereotaxic frame. Deposits of PHA-L in the LC were made according to the coordinate system of Pax&s and Watson.25The glass micropipette was inserted in the brain at a 20” angle to avoid the transverse sinus. Bevelled glass micropipettes (IO-25pm) were l&d with 2.5% PHA-L (Vector Labs, Burlingame, CA, U.S.A.) solution in 0.05 M Tris-buffered saline PBS; pH 7.4). After positioning the pipette in the LC it was connected to the positive electrode of a Midprd CS-3 constant current source. Iontophoretic delivery was achieved with a 5 PA current for 30 min in a 7 8 half time on-half time off cycle. After iontophoresis, the pipette was left in situ for 10 min to avoid leakage of tracer in the pipette track. Following a post-operative survival time of seven days the rats were deeply anesthetized with 4 ml/kg 6% sodium pentobarbital and perfused tramly, after a short saline pre-rinse, with 2.5% glutaraldehyde, 0.5% paraformaldehyde, and 4% sucrose solution in 0.05 M phosphate buffer (PH 7.4). Brains were removed and stored overnight at 4°C in a 30% sucrose solution in 0.05 M phosphate buffer (PH 7.4). Serial 4Oqm sections were cut on a freezing microtome and collected in TBS. Every second section was incubated for 48 h with rabbit anti-PHA-L (Dakopatts, Denmark, 1: 2000) solution in 0.05M Tris bufFer to which 0.5M sodium chloride, 20% T&on X-100 (TBS-T; pH 7.4) and 250~1 normal goat serum were added. Next, the sections were thoroughly rinsed in TBST and ~hccd for 12-l 5 h in a noat anti-rabbii IgG (Sigma, 1: 150)-&tion in TBS-T. A&r a thorough rinsing in TBS-T the sections were incubated for 4 h with a rabGt peroxidase-antiperoxidase (PAP; Dakopatts, Denmark, 1: 800) solution in TBS-T. After rinsing in 0.05 M Tris-HCl (pH 7.4), the presence of ueroxidase was revealed with 100 &l of a i).04%6diaminobe&dine solution in Tris-HCl to which 0.8ml of a 1.5% H,O. solution was added. Then the sections were rinsed in T&&l, mounted onto gelatin-coated slides, air-dried, counterstained with Cresyl Violet and coverslipped. Iontophoretic injections of HRP were made in seven Wistar rats. Glass micropipettes (tip diameter 20-25pm)
were filled with 10% HRP (Sigma, type VI) solution in 0.01 M sodium chloride and placed in the DMnX stereotactically, using the coordinate system of the atlas of hllegrino et a1.mThe iontophoretic procedure and subsequent fixation and histochemical methods have been described in detail in previous articles.3’37 In brief, the following procedure was &d. Twenty-four hours a!‘ter the iontophoretic injection of HRP (0.5 uA. 30 min) the rats were deeDlv anesthetized with sod& p&dbarbii and perfused t&&ardially with the Iixing solution described in the PHA-L procedure. After overnight dehydration in a 30% sucrose solution, serial 40-ym sections were cut on a freezing microtome and collected in a cold 0.05 M phosphate buffer solution @H 7.4). Every second section was incubated for 4Omin in a 0.01 M acetate buffer solution (pH 3.3) containing 0.005% tetramethylbenzidine (TMB), 0.04% sodium nitroferricyanide and 0.009% H,O1. The sections were rinsed in 0.05 M phosphate buffer (PH 7.4) at room temperature and placed for 30 min in a 0.05 M Tris-HCl (PH 7.4) solution containing 0.04% diaminobenzindine, 0.02% cobalt chloride and 0.0015% H,O,. The peroxidase reaction was stopped by placing the sections in Tris-HCl. Then, the sections were mounted onto gelatin-coated slides, air-dried, counterstained with Safranin-O/Neutral Red and coverslipped. The adjacent sections were stained to reveal the presence of norepinephrine, using TH as a marker. The &mumcytoche&a~procedure for revealing the presence of TH was described in detail elsewhere.” For the control of the specificity of the immunoperoxidase reaction, in parallel exp&ments some sections Were incubated solely with rabbit anti-PHA-L. rabbit anti-TH. goat anti-rabbit_IgG, rabbit PAP, and w&h TBS-T or PI&&T without any primary antisera (further procedures as above). All such control experiments appeared to be negative.
RESULTS Limiting the PHA-L injections to the LC was extremely difficult; of the 12 injections placed, only four were in the center of the LC. In all experiments, lectin-positive cells were found in areas adjacent to the LC; the pontine periaqueductal gray, Barrington’s nucleus, the medial parabrachial nucleus, the mesencephalic trigeminal nucleus or the nucleus subcoeruleus. Projections to the DMnX were found when lectin-positive cells were present in the LC; injections into the nucleus subcoeruleus and the other areas adjacent to the LC were negative (Fig. 1.1).
Pathways from locus coeruleus to the dorsal motor vagus nucleus In general, injections of lectin into the LC (Fig. 2A) resulted in labeling of fibers which descended (i) medially, (ii) laterally and (iii) dorsomedially in the reticular formation. Descending fibers from the LC innervating the dorsal motor vagus nucleus took the medial and dorsomedial trajectory. (i) Processes taking the medial route emerged ventrally from the LC, traversed the pontine reticular formation and coursed medially of the motor trigeminal nucleus to the ventromedial medulla oblongata. In the area located laterally of the raphe magnus and medially of the facial nucleus, the medially descending fibers from the LC curved in posterior direction and could be followed in this ventromedial position to the caudal
156
G J. TFK HORSTPI d
Fig. 2. Photomicrographs of (A) a small PHA-L injection into the LC (scale bar =250pm); (B) noradrenalin-positive cells in the LC and nucleus suhcoeruleus (courtesy of Dr Dries Kalsbeek) (scale bar = 200 pm); (C) PHA-L-labeled fibers in the lateral part at the intermediate levels of the DMnX (scale bar = 120 pm); (D) a small iontophoretic HRP injection into the vague nucleus as seen in diaminobenzidine-stained sections (scale bar = 300 pm); (E) high power magnification of the boxed area in C showing PHA-L-labeled fibers in the lateral part of the DMnX after an injection into the LC (scale bar = 80 pm); (F) high power magnification of a HRP-labeled cell (arrow) in the ventral part of the LC after dn iontophoretic peroxidase injection into the DMnX (scale bar = 40 pm).
157
Locus coeruleus projections
.-
F -
-_ -
_-
> _-
--
-_
-_
-
I
< i
Fig. 3. DorsaI view of a stereometric reconstruction of the right (R) DMnX. The scale indicates the anterior-posterior coordinates as posterior to the interaural line as given by FWegrino ei aLm Lane 2 shows the location in the DMnX of the p~ganglioni~ vagal neurons ~nne~ating the pancreas fl-cells (according to Luiten et a~.“}. Lane 3 is a camera lucida drawing of the DMnX giving, at the different anterior-posterior levels, the pattern of innervation after a PHA-L injection into the LC. Lane 4 gives the location in the DMnX of preganglionic parasympathetic cells with axons running through the cardiac branch of the vagus nerve (according to Nosaka et ~1.~~).
oblcmgata. In the posterior medulla obiongata, beginning at the rostra1 level of the DMnX and extending up to the most posterior level of this preganglionic parasympathetic autonomic nucleus, medially descending fibers from the LC curved in a dorsal direction and ran to the DMnX in the medial or lateral reticular formation. These fibers innervated the medial part of the rostra1 DMnX and the lateral part of the middle and posterior DMnX, respectively. (ii) Fibers emerging ventrolaterally from the injection in the LC and taking a lateral course in the reticular formation innervated the dorsolateral parts of the parvocellular reticular formatian and the spinal trigeminal nuclei. {iii) A small number of labeled fibers from the LC took a descending course in the dorsomedial medulla. These fibers left tbe PHA-L injection site medially and coursed, ventral of the medulla
fourth psntine
ventricle, to the dorsomedial part of the periaqueductal gray where they bent in a
posterior direction. In the rostra1 medulla oblongata, dorsomedially descending fibers from the LC were seen in the ventromedial part of the prepositus hypoglossal nucleus. At more caudal levels in the brainstem, some dorsom~i~ly descending fibers from the LC curved in a lateral direction and innervated the medial part of the rostra1 third of the DMnX. Fibers not innervating the DMnX continued with the descending course and entered the cervical spinal cord, running caudally adjacent to the spinal canal. Efferents from the LC innervated the DMnX bilaterally. Crossing fibers were found in the ventral medulla oblongata and in the pontine periaqueductal gray. Target areas in the dorsal
m&r
vugus nucleus
Bilaterally, the PHA-L-positive varicose fibers and terminal boutons from the LC were found medially in the rostra1 third and laterally in the posterior
Ci.
158
J. TER HORST ef
two-thirds of the DMnX (Fig. 3). In the most caudal parts of the DMnX a scattered pattern of innervation was found. The DMnX innervation from the LC was more intense ipsilaterally but the difference with the contralateral side was small. At the level of the area postrema (Figs 2C, E, 3) in the DMnX, the innervation from the LC was most intense. Ventrolaterally in the DMnX, at this anterior-posterior level, a network of labeled fibers was found enclosing some of the preganglionic cells in this area. The PHA-L-labeled fibers have boutons en passage and run from lateral to medial in the DMnX. In the LC experiments, a few lectin-positive varicose fibers were found in the ventral part of the hypoglossal nucleus and, dorsal of the DMnX, in the medial and lateral part of the nucleus of the solitary tract. Innervations of the hypoglossal and solitary tract nuclei were also found in the control experiments but in these cases the DMnX was not labeled. Control experiments
After small iontophoretic HRP injections into the DMnX (Fig. 2D) bilaterally in the LC the retrogradely labeled cells were found. The number of labeled neurons in the LC was small. In one experiment, with a very small HRP injection into the lateral part of the DMnX, ipsilaterally eight and contralaterally three labeled cells were found. These labeled cells were located ventrally (Figs 1.2, 2F) in the middle LC and dorsally at the most rostra1 and caudal levels of the LC. In the dorsal LC the neurons were faintly labeled. DlSCUSSION
In the present investigation efferent connections from the LC to the dorsal motor nucleus of the vagus were shown with PHA-L as a neuroanatomical tracer. The target areas in the DMnX were the lateral part at the level of the area postrema and the medial part at the rostra1 levels. The connection was conf%med with control HRP injections into the DMnX. The PHA-L tracing of LC efferents to the DMnX corroborated previous studies where this connection was identified with catecholamine autofluorescence and electrolytic lesioning of the LC,r3 WGA-HRP retrograde tracing of afferents of the nucleus of the solitary tract3’ and autoradiographic tracing of LC efferents.39 In the rat, Jones and Yang’ also found a few silver grains in the caudal part of the DMnX after tritiated leucine injections into the LC region but we could not corroborate these observations in a previous HRP investigation of DMnX afferents.” Moreover, the LC-DMnX pathway is not described in a recent PHA-L investigation giving the distribution of LC efferents in the brainstem of the rat.6,7 The methods used for the identification of LC efferents may account for the differences seen between the investigations. In the autoradiographic tracing
ul.
experiments not all the efferents of the LC’ may be revealed. The location of the LC near the lateral aspect of the fourth ventricle and the small dimensions of this noradrenergic nucleus impede the proper targeting of tritiated amino acids into the LC. Injection sites, after autoradiographic processing of the sections, appear as accumulations of silver grains in the overlying photographic emulsion without a distinct morphological basis for identifying the cells transporting the tracer. In general, neuroanatomists assume that all the cells in the area with a high density of silver grains are labeled, but this may not be the case. Cells transporting the marker, in the PHA-L procedure for tracing of neuronal efferents, are stained from the soma up to the terminal presynaptic endings8.36.40and enable an accurate identification of the cellular origin of the projections. In the HRP experiments, the chromagen and the incubation parameters used for revealing the presence of the peroxidase are important for the effectiveness of the reaction. In the present investigation we used the TMB procedure of Mesulam2’ which in comparative studies was shown to be superior to all the other methods for revealing HRP at the level of the light microscopic examination. 20.22In our previous HRP investigation of DMnX afferents,” labeled cells in the LC were probably not detected because we used the benzidine dihydrochloride method of De Olmos and Heimer,s which revealed, in the comparative study,20,22only 3&50% of the retrogradely labeled cells seen in the TMB-stained sections. Although similar pathways from the LC are revealed in the brainstem with PHA-L as a neuroanatomical tracer, Fritschy and Grazanna’ did not find the projections from the LC to the DMnX. At least two factors may be responsible for the observed difference; the unequal survival times (14-21 vs seven days) and/or the cells in the LC projecting to the DMnX were not labeled with PHA-L. Survival times of seven to 10 days gave the best results in our experiments. Longer periods caused much weaker signals in the target areas36 suggesting that, with survival times of more than 14 days, in Fritschy and Grazanna’s study the signal in the DMnX might have been below the detection level. However, it is also possible that the DMnX projecting cells in the LC were not labeled with PHA-L because Fritschy and Grazanna’ used very small iontophoretic injections for tracing LC efferents. In their 18 experiments, the number of labeled neurons in the LC ranged from 20 to 100. For a survey of all efferent connections of approximately 1500 cells33 in the LC, the 18 small injections were probably not sufficient. After small HRP injections into the DMnX, retrogradely labeled cells were found in the ventral part of the middle LC and in the dorsal part. Similar distributions of efferent projections from subpopulations of the LC were reported by Loughlin et al.‘4J5 who, by using large HRP injections, describe projections from the ventromedial parts of the LC to the cerebel-
Locus coeruleus projections lum and spinal cord. However, Loughlin’s
investigation was not complete, only areas known for receiving noradrenergic inputs were injected with HRP. Injections into the dorsomedial brainstem or DMnX were not made. In addition, in the present HRP inv~ti~a~on of DMnX alIerents the retrogradely labeled cells were found in the rostra1 part of the LC, in the subpopulation projecting to the hypothalamus.‘4*15 Collateralization of LC efferents projecting to the brainstem and hypothalamus needs to be studied. The main target areas in the DMnX for the noradrenergic LC efferents were the rostromedial parts and, at the level of the area postrema, the lateral parts of the DMnX. Immunohistochemical studies, using TH and dopamine-b-hydroxylase for labeling norad~ner~c nerve fibers in the DMnX,i”J9 showed abundance of noradrenalin-positive fibers in the lateral part of the DMnX at the area postrema and obex levels and a modest noradrenergic fiber labeling in the medial DMnX. These immunohistochemical observations have a remarkable resemblance to the pattern of fiber labeling in the DMnX seen after PHA-L injections into the LC and show that the LC is indeed an important source of noradrenergic fibers in the lateral and medial DMnX. The preganglionic parasympathetic cells in the medial parts of the motor vagus nucleus innervate the stomach’” and the pancreas’J*2s (Fig. 3), whereas the preganglionic cells in the lateral part of the DMnX project to the heartiQ3 (Fig. 3). The projections from the LC to the DMnX were most dense in the lateral, cardiac part of the nucleus, suggesting that the LC is involved in the
regulation of cardiovascular functions. Physiological studies corroborate this suggestion by showing widespread changes in cardiovascular dynamics after elec-
159
trical stim~ation of the region of the LC.38 Occasionally, a bradycardia was seen after the cessation of the stimulus in the LC region.‘” Involvement of the LC in autonomic regulation of insulin release is not very likely. The medial parts of the DMnX, containing the small-sized perikarya of the preganglionic parasympathetic cells for autonomic regulation of insulin-producing pancreatic /Icells’* (Fig. 3), were not innervated by the efferents from the LC. The dendrites of these preganglionic neurons also probably extend in regions lacking a substantial LC inneffation because they are short and mainly oriented horizontally along the mediolatera1 axis of the DM~IX.‘**~~ The paraventricular hypothalamic nucleus is important for limbic control of autonomic functions; it is probably the main h~~alamic output center to the preganglionic autonomic nuclei.‘6,‘7 The parvocellular part of the PVN has direct projections to the para- and orthosympathetic preganglionic neurons in the brainstem and spinal cord17*t1*3’ and also innervates the locus coeruleus.” Participation of the PVN in food intake behavior,rz pancreas hormone release” and cardiovascular control3 is demonstrated. With respect to the cardiovascular control of the PVN, it has been shown that electrical stimulation of the paraventricular nucleus elicits cardioacceleration (tachycardia~~” possibly mediated by vagal inhibition3,28,zg and sympathetic excitation.3 The LCDMnX pathway described in the present investigation may participate in the inhibition of the cardiac preganglionic neurons in the DMnX by the PVN. Acknowfedgemetrts-The authors wish to thank Mrs Nieske Brouwer, Anja Dreijer-Kattenberg and Liza Mast for the processing of the histological sections.
REFERENCES 1. Berthoud H.-R., Fox E. A. and Powley T. L. (1990) Localization of vagal preganglionics that stimulate insulin and glucagon secretion. Am. J. Physiol. 258, R160-R168. 2. Berthoud H.-R. and Powley T. L. (1990) Identification of vagal pregangiionics that mediate cephalic phase insulin response. Am. J. Fhysiol. 258, R523-R530. 3. Ciriello J. and Calaresu F. R. (1980) Role of paraventricular and supraoptic nuclei in central cardiovascular regulation in the cat. Am. J. Physiof. 239, Rl37-R142. 4. Copray J. C. V. M., Liem R. S. B., Ter Horst G. J. and Van Will&n J. D. (1990) Dopaminergic afferents to the mesencephalic trigeminal nucleus of the rat: a light and electron microscope immunocytcchemistry study. Brain Res. 514, 343-348. 5. De Ohnos J. and Heimer L. (1977) Mapping of collateral projections with the HRP-method. Neurosci. Lat. 6,107-i 14. 6. Fritschy J. M. and Graxanna R. (1990) Demonstration of two separate descending noradrenergic pathways to the rat spinal cord: evidence for an intragriseal trajectory of locus coeruleus axons in the superficial layers of the dorsal horn. J. camp. Neural. 291, 533-582. 7. Fritschy J. M. and Grazanna R. (1990) Distribution of locus coeruleus axons within the rat brainstem demonstrated by Phuseolus udgaris leucoagglutinin anterograde tracing in combination with dopamine-fl-hydroxylase immunofluorescence. J. camp. Newol. 293, 616-631. 8. Gerfen C. R. and Sswchenko P. E. (1984) An anterograde ne~~nato~~ tracing method that shows the detailed morphology of neurons, their axons and terminals: immunohist~he~~l localization of an axonally transported plant lectin, Phaseofus vulgaris, leucoagglutinin (PHA-L). Bruin Res. 290, 219-238. 9. Jones B. E. and Yang T. Z. (1985) The efferent proiections from the reticular formation and the locus coeruleus studied by anterograde and-retrograde axonal transport in the rat. J. camp. Neurol. 242, X-92. 10. Kalia M., Fuxe K. and Goldstein M. (1985) Bat medulla oblongata. II. Dopaminergic, noradrenergic (Al and A2) and adrenergic neurons, nerve fibers and presumptive terminal processes, J. camp. Neural. 233, 308-332.
160
‘2. .I fm
HORSTrl (I/
11. Kaha M. and Me&am M. M. (1980) Brainstem projections of sensory and motor components of the vagus complex in the cat: II. Laryngeal, tracheobronchial, pulmonary, cardiac and gastrointestinal branches. J. camp. Ncuml. 193, 467-508. 12. Leibowitz S. F., Hammer N. J. and Chang K. (1981) Hypothalamic paraventricular nucleus lesions produce overeatmg and obesity in the rat. Physiol. Behav. 27, 1031-1040. 13. Loizou L. A. (1969) Projections of the nucleus locus coeruleus in the albino rat. Bruin Res. 15, 5633565. 14. Loughlin S. E., Foote S. L. and Bloom F. E. (1986) Efferent projections of nucleus locus coeruleus: morphologrc subpopulations have different efferent targets. Neuroscience 18, 307--319. 15. Loughlin S. E., Foote S. L. and Bloom F. E. (1986) Efferent projections of nucleus locus coeruleus: topographic organization of cells of origin demonstrated by three-dimensional reconstruction. Neuroscience 18, 291~ 3 19. 16. Luiten P. G. M., Ter Horst G. J. and Steffens A. B. (1987) The hypothalamus, intrinsic connections and outflow pathways to the endocrine system in relation to the control of feeding and metabolism. Prog. Neurobiol. 28, 1-54. 11. Luiten P. G. M., Ter Horst G. J., Karst H. and Steffens A. B. (1985) The course of paraventricular hypothalamic efferents to autonomic structures in medulla and spinal cord. Bruin Res. 329, 374378. 18. Luiten P. G. M., Ter Horst G. J., Koopmans S. J., Rietberg M. and Steffens A. B. (1984) Preganglionic Innervation of the pancreas islet cells in the rat. J. aufon. nerz:. Sysr. 10, 2742. 19. Manier M., Feuerstein C., Passagia J. G., Mouchet P., Mons N., Geffard M. and Thibault J. (1990) Evidence for the existence of L-DOPA- and dopamine-immunoreactive nerve cell bodies in the caudal part of the dorsal motor nucleus of the vagus nerve. J. Chem. Neurounat. 3, 193-205. 20. Mesulam M. M. (1982) Principles of horseradish peroxidase neurohistochemistry and their applications for tracing of neural pathways-axonal transport, enzyme histochemistry and light microscopic analysis. In Trueing Neural Connections with Horseradish Peroxidase (ed. Mesulam M. M.), pp. I-153. John Wiley, New York. 21. Moore R. Y. and Card J. P. (1984) Noradrenalin containing neuron systems. In Handbook of Chemical Neurounutom.y. Vol. 2: Classical Transmitters in the CNS (eds Bjdrklund A. and Hijkfelt T.), pp. 123-156. Elsevier, Amsterdam. 22. Morrell J. I., Greenberger L. M. and Pfaff D. W. (1981) Comparison of horseradish peroxidase visualization methods: quantitative results and further technical specifics. J. Histochem. Cytochem. 29, 903-916. 23. Nosaka S., Yamamoto T. and Yasunaga K. (1979) Localization of vagal cardio-inhibitory preganglionic neurons within the rat brainstem. J. camp. Neurol. 186, 79992. 24. Ono T., Nishino H., Sasaki K., Muramoto K., Yano 1. and Simson A. (1978) Paraventricular nucleus connections to spinal cord and pituitary. Neurosci. Lett. 10, 141~146. 25. Paxinos G. and Watson C. (1986) The Rut Bruin in Stereotuxic Coordinates. Academic Press, New York. 26. Pellegrino L. J., Pellegrino A. S. and Cushman A. J. (1979) A Stereotuxic Atlas of the Rat Bruin. Plenum Press, New York. 27. Pittman Q. J. and Franklin L. G. (1985) Vasopressin antagonist in nucleus tractus solitarius/vagal area reduces pressor and tachycardia responses to paraventricular nucleus stimulation in rats. Neurosci. Lett. 56, 1555160. 28. Rogers R. C. and Hermann G. E. (1985) Vagal afferent stimulation-evoked gastric secretion suppressed by paraventricular nucleus lesion. J. auton. new. Sysi 13, 191-199. 29. Rogers R. C. and Nelson D. 0. (1984) Neurons of the vagal division of the solitary nucleus activated bv the par~ventricular nucleus of the hypothalamus. J. uuton. nerv. cyst. 10, 193-197. _ 30. Rye D. B., Lee H. J., Saper C. B. and Wainer B. H. (1988) Medullary and spinal efferents of the pedunculopontine tegmental nucleus and adjacent mesopontine tegmentum in the rat. J. camp. Neural. 269, 315-341. 31. Swanson L. W. and Kuypers H. G. J. M. (1980) The paraventricular nucleus of the hypothalamus: cytoarchitectonic subdivisions and organization of projections to the pituitary, dorsal vagal complex and spinal cord as demonstrated by retrograde fluorescent double labeling methods. J. camp. Neurol. 194, 5555570. 32. Swanson L. W. and Sawchenko P. E. (1983) Hypothalamic integration: organization of the paraventricular and supraoptic nuclei. A. Rev. Neurosci. 6, 269.-324. 33. Swanson L. W. (1976) The locus coeruleus: a cytoarchitectonic, Golgi and immunohistochemical study in the albino rat. Bruin Res. 110, 39-56. 34. Ter Horst G. J. and Luiten P. G. M. (1987) Phaseolus vulgaris leucoagglutinin tracing of intrahypothalamic connections of the lateral, ventromedial, dorsomedial and paraventricular nuclei of the hypothalamus in the rat. Bruin Res. Bull. 18, 191-203.
35. Ter Horst G. J., Copray J. C. V. M., Van Willigen J. D. and Liem R. S. B. (1990) Contralateral projections of cells in the motor trigeminal nucleus of the rat. Neurosci. Lett. 113, 26&266. 36. Ter Horst G. J., Groenewegen H. J., Karst H. and Luiten P. G. M. (1984) Phuseolus uulguris leucoagghrtinin immunohistochemistry. A comparison between autoradiographic and lectin tracing of neuronal efferents. Rruin Res. 307, 3799383. 37. Ter Horst G. J., Luiten P. G. M. and Kuipers F. (1984) Descending pathways from the hypothalamus to dorsal motor vagus and ambiguus nuclei in the rat. J. auton. new. Syst. 11, 59-75. 38. Ward D. G. and Gunn C. G. (1976) Locus coeruleus complex: elicitation of a pressor response and a brainstem region necessary for its occurrence. Bruin Res. 107, 401-406. 39. Westlund K. N. and Coulter J. D. (1980) Descending projections of the locus coeruleus and subcoeruleus/medial parabranchial nuclei in monkey: axonal transport studies and dopamine-b-hydroxylase immunocytochemistry. Bruin Res. Rev. 2, 235-264.
40. Wouterlood F. G. and Groenewegen H. J. (1985) Neuroanatomical tracing by use of Phaseolus vulguris leucoagglutinin (PHA-L), electron microscopy of PHA-L filled neutonal somata, dendrites, axons and axon terminals. Bruin Res. 326, 188-192.
41. Yoshimatsu H., Niijima A., Oomura Y., Yamabe K. and Katafuchi T. (1984) Effects of hypothalamic lesion on pancreatic autonomic nerve activity in the rat. Bruin Res. 303, 1477152. (Accepted
9 April 1991)
Neuroscience Vol. 45, No. 1, pp. 153-160, 1991 Printed in Great Britain
0 1991IBRO
LOCUS
COERULEUS PROJECTIONS TO THE DORSAL MOTOR VAGUS NUCLEUS IN THE RAT G. J. TERHORST,G. J. TOESand J. D. VAN WILLIGEN
Department of Neurobiology and Oral Physiology, University of Groningen, Bloemsingel 10, 9712 KZ Groningen, The Netherlands Abatrac&-The origin of the noradrenergic innervation of the preganglionic autonomic nuclei in the medulla oblongata and spinal cord is still controversial. In this investigation descending connections of the locus coeruleus to the dorsal motor vagus nucleus in the rat are studied with Phase&s oulgaris leucoagglutinin and horseradish peroxidase as neuroanatomical tracers. Locus coeruleus projections in the motor vagus nucleus are found in the medial part at rostra1 levels and in the lateral part at intermediate levels of this nucleus. The terminal labeling in the lateral intermediate part of the vagus nucleus appears in an area where possibly preganglionic parasympathetic cardiac neurons are located, suggesting that the locus coeruleus might be involved in regulation of cardiovascular functions. After small iontophoretic injections of horseradish peroxidase in the motor vagus nucleus, retrogradely labeled cells aie found in the ventral part of the locus coeruleus and occasionally in the dorsal part of the nucleus. The results show that the locus coeruleus-dorsal motor vagus nucleus pathway may participate in the inhibition of the cardiac preganglionic neurons in the dorsal motor vagus nucleus by the hypothalamic paraventricular nucleus.
The hypothalamic paraventricular nucleus (PVN) is an important structure for limbic control of autonomic functions.‘6*32 The PVN may exert its autonomic responses via direct projections to the preganglionic autonomic cells in the dorsal motor vagus nucleus (DMnX), the nucleus ambiguus and the intermediolateral cell groups in the thoracic spinal cord.17*24,31 In this paradigm, indirect pathways from the PVN to the preganglionic autonomic nuclei could modulate thereafter the initial response caused by the direct innervation, because of longer signal transduction times. Nuclei with autonomic functions receiving a PVN innervation are the ventromedial, lateral and dorsomedial hypothalamic nuclei, the periaqueductal gray and the caudal parvocellular reticular formation. I73 However, the PVN also innervates the locus coeruleus (LC),” a noradrenergic nucleus with widespread projections to numerous forebrain,21 brainstem and spinal areas6,’ and probably very global functions. LC involvement in autonomic regulatory functions is controversial. The coeruleo-DMnX projection has been the focus of many studies in a variety of species, yet considerable disagreement still exists about this pathway. In monkeys, LC-DMnX connections were found with autoradiographic tract-tracing techniques and dopaimmunohistochemistry.39 In mine-/3-hydroxylase Abbreuiarions: DMnX, dorsal motor vagus nucleus; HRP,
horseradish peroxidase; LC, locus coeruleus; PAP, peroxidase-antiperoxidase; PHA-L, Phaseolus vulgaris leuPVN, hypothalamic paraventricular coagglutinin; nucleus; TBS, Tris-buffered saline; TBS-T, Tris-buffered saline with Triton X-100; TH, tyrosine hydroxylase; TMB, tetramethylbenzidine; WGA-HRP, wheat germ agglutinin-conjugated horseradish peroxidase.
the rat, the LC-DMnX pathway was identified with catecholamine autofluorescence in LC-lesioned animals,13 wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP)” and anterograde transport of tritiated proteins.’ However, tracing of DMnX afferents with HRP in the same species, revealed labeled cells near and not in the LC.37 Moreover, the connection was also not revealed with more sensitive immunocytochemical tracing techniques.6x7 Tract-tracing techniques have limitations which may be relevant when they are used for identifying connections of small-sized nuclei like the DMnX and LC. In autoradiographic tracing one gets an indirect image of the labeling in the form of silver grains in the overlying photographic emulsion. Precise identification of target areas in autoradiographic preparations is difficult because labeled fibers and terminal presynaptic endings cannot readily be distinguished from each other and, in sparsely labeled areas, from the background labeling. In retrograde tracing of connections with HRP, the size of the injection, which is particularly relevant for a smallsized nucleus like the DMnX, and undesired uptake of HRP by damaged passing fibers, impede proper interpretation of labeled connections. The immunocytochemical Phaseolus oulgaris leucoagglutinin (PHA-L) tracing method proved to be very effective for the identification of short and long distance projectionss~36 because it is not hampered by haphazard spreading of injected tracer or aspecific debris and precipitates. With this technique it is possible to make small iontophoretic injections. Moreover, the PHA-L method, in contrast to autoradiographic and HRP tracing, yields completely
153
G. J. TEH HORS-~ et ul. NEGATIVE
POSITIVE
F
Fig. t. Series of transverse sections from the rostra1 (A) to the caudal (F) level of the LC showing the location and dimensions of the PHA-L injections into the LC region. The LC and nucleus sulxoeruleus are shaded and the different cases are indicated with the numbers. Injections causing terminal bouton labeling in the dorsal DMnX (positive) are illustrated on the right side. Control (negative) injections are shown on the left side. 2. Series of transverse sections showing the location of retrogradely labeled cells (black dots) in the LC after a small HRP injection into the ipsilaterai DMnX.
Abbreviation
Bar DTg 37 lc, LC LVe meT mlf moT, MOT
Barrington’s nucleus dorsal tegmental nucleus germ of the facial nerve locus coeruleus lateral vestibular nucleus mesencephalic trigeminal nucleus medial Ion~t~in~ fasci4us motor trigeminal nucleus
used in the figures
n7, 7 Pb PHA-L :I ts X
XII
SOL
facial nerve parabrachial nucleus PhaseoIus vulgaris ie~a~lutinin superior cerebellar peduncle nucleus of the solitary tract solitary tract dorsal motor vagus nucieus hypoglossal nucleus
155
Locus coeruleus projections
filled axons that can be traced from the labeled cell body to the target area where the characteristic varicosities and terminal boutons appear that indicate the site of presynaptic terminations.8*Nv40 The immunocytochemical PI-IA-L technique gave optimal results with survival times of seven to 10 days in our hand~.~ Longer survival times caused weaker signals in the target areas, indicating that for identification of small projections, like the LC-DMnX pathway, the post-operative survival time is an important factor. Therefore, in addition to Fritschy and Grazanna,’ not showing LC-DMnX projections with survival times of 14-21 days, we have studied efferent connections of the LC to the dorsal motor vagus nucleus, seven days after the injection of PHA-L in the LC. Because the injections of PHA-L always labeled cells near the LC, small iontophoretic injections of HRP into the DMnX were used as controls. The noradrenergic nature of the HRP-labeled cells in the LC region was demonstrated with tyrosine hydroxylase (TH) staining of the adjacent sections.
EXPR-AL
PROCEDURES
The PHA-L experiments were conducted on 12 male Wistar rats (locally bred at the CDL, Univ. Groningen) of 250-3001. For injecting PHA-L, the animals were anesthetized with halothane and placed in a Kopf stereotaxic frame. Deposits of PHA-L in the LC were made according to the coordinate system of Pax&s and Watson.25The glass micropipette was inserted in the brain at a 20” angle to avoid the transverse sinus. Bevelled glass micropipettes (IO-25pm) were l&d with 2.5% PHA-L (Vector Labs, Burlingame, CA, U.S.A.) solution in 0.05 M Tris-buffered saline PBS; pH 7.4). After positioning the pipette in the LC it was connected to the positive electrode of a Midprd CS-3 constant current source. Iontophoretic delivery was achieved with a 5 PA current for 30 min in a 7 8 half time on-half time off cycle. After iontophoresis, the pipette was left in situ for 10 min to avoid leakage of tracer in the pipette track. Following a post-operative survival time of seven days the rats were deeply anesthetized with 4 ml/kg 6% sodium pentobarbital and perfused tramly, after a short saline pre-rinse, with 2.5% glutaraldehyde, 0.5% paraformaldehyde, and 4% sucrose solution in 0.05 M phosphate buffer (PH 7.4). Brains were removed and stored overnight at 4°C in a 30% sucrose solution in 0.05 M phosphate buffer (PH 7.4). Serial 4Oqm sections were cut on a freezing microtome and collected in TBS. Every second section was incubated for 48 h with rabbit anti-PHA-L (Dakopatts, Denmark, 1: 2000) solution in 0.05M Tris bufFer to which 0.5M sodium chloride, 20% T&on X-100 (TBS-T; pH 7.4) and 250~1 normal goat serum were added. Next, the sections were thoroughly rinsed in TBST and ~hccd for 12-l 5 h in a noat anti-rabbii IgG (Sigma, 1: 150)-&tion in TBS-T. A&r a thorough rinsing in TBS-T the sections were incubated for 4 h with a rabGt peroxidase-antiperoxidase (PAP; Dakopatts, Denmark, 1: 800) solution in TBS-T. After rinsing in 0.05 M Tris-HCl (pH 7.4), the presence of ueroxidase was revealed with 100 &l of a i).04%6diaminobe&dine solution in Tris-HCl to which 0.8ml of a 1.5% H,O. solution was added. Then the sections were rinsed in T&&l, mounted onto gelatin-coated slides, air-dried, counterstained with Cresyl Violet and coverslipped. Iontophoretic injections of HRP were made in seven Wistar rats. Glass micropipettes (tip diameter 20-25pm)
were filled with 10% HRP (Sigma, type VI) solution in 0.01 M sodium chloride and placed in the DMnX stereotactically, using the coordinate system of the atlas of hllegrino et a1.mThe iontophoretic procedure and subsequent fixation and histochemical methods have been described in detail in previous articles.3’37 In brief, the following procedure was &d. Twenty-four hours a!‘ter the iontophoretic injection of HRP (0.5 uA. 30 min) the rats were deeDlv anesthetized with sod& p&dbarbii and perfused t&&ardially with the Iixing solution described in the PHA-L procedure. After overnight dehydration in a 30% sucrose solution, serial 40-ym sections were cut on a freezing microtome and collected in a cold 0.05 M phosphate buffer solution @H 7.4). Every second section was incubated for 4Omin in a 0.01 M acetate buffer solution (pH 3.3) containing 0.005% tetramethylbenzidine (TMB), 0.04% sodium nitroferricyanide and 0.009% H,O1. The sections were rinsed in 0.05 M phosphate buffer (PH 7.4) at room temperature and placed for 30 min in a 0.05 M Tris-HCl (PH 7.4) solution containing 0.04% diaminobenzindine, 0.02% cobalt chloride and 0.0015% H,O,. The peroxidase reaction was stopped by placing the sections in Tris-HCl. Then, the sections were mounted onto gelatin-coated slides, air-dried, counterstained with Safranin-O/Neutral Red and coverslipped. The adjacent sections were stained to reveal the presence of norepinephrine, using TH as a marker. The &mumcytoche&a~procedure for revealing the presence of TH was described in detail elsewhere.” For the control of the specificity of the immunoperoxidase reaction, in parallel exp&ments some sections Were incubated solely with rabbit anti-PHA-L. rabbit anti-TH. goat anti-rabbit_IgG, rabbit PAP, and w&h TBS-T or PI&&T without any primary antisera (further procedures as above). All such control experiments appeared to be negative.
RESULTS Limiting the PHA-L injections to the LC was extremely difficult; of the 12 injections placed, only four were in the center of the LC. In all experiments, lectin-positive cells were found in areas adjacent to the LC; the pontine periaqueductal gray, Barrington’s nucleus, the medial parabrachial nucleus, the mesencephalic trigeminal nucleus or the nucleus subcoeruleus. Projections to the DMnX were found when lectin-positive cells were present in the LC; injections into the nucleus subcoeruleus and the other areas adjacent to the LC were negative (Fig. 1.1).
Pathways from locus coeruleus to the dorsal motor vagus nucleus In general, injections of lectin into the LC (Fig. 2A) resulted in labeling of fibers which descended (i) medially, (ii) laterally and (iii) dorsomedially in the reticular formation. Descending fibers from the LC innervating the dorsal motor vagus nucleus took the medial and dorsomedial trajectory. (i) Processes taking the medial route emerged ventrally from the LC, traversed the pontine reticular formation and coursed medially of the motor trigeminal nucleus to the ventromedial medulla oblongata. In the area located laterally of the raphe magnus and medially of the facial nucleus, the medially descending fibers from the LC curved in posterior direction and could be followed in this ventromedial position to the caudal
156
G J. TFK HORSTPI d
Fig. 2. Photomicrographs of (A) a small PHA-L injection into the LC (scale bar =250pm); (B) noradrenalin-positive cells in the LC and nucleus suhcoeruleus (courtesy of Dr Dries Kalsbeek) (scale bar = 200 pm); (C) PHA-L-labeled fibers in the lateral part at the intermediate levels of the DMnX (scale bar = 120 pm); (D) a small iontophoretic HRP injection into the vague nucleus as seen in diaminobenzidine-stained sections (scale bar = 300 pm); (E) high power magnification of the boxed area in C showing PHA-L-labeled fibers in the lateral part of the DMnX after an injection into the LC (scale bar = 80 pm); (F) high power magnification of a HRP-labeled cell (arrow) in the ventral part of the LC after dn iontophoretic peroxidase injection into the DMnX (scale bar = 40 pm).
157
Locus coeruleus projections
.-
F -
-_ -
_-
> _-
--
-_
-_
-
I
< i
Fig. 3. DorsaI view of a stereometric reconstruction of the right (R) DMnX. The scale indicates the anterior-posterior coordinates as posterior to the interaural line as given by FWegrino ei aLm Lane 2 shows the location in the DMnX of the p~ganglioni~ vagal neurons ~nne~ating the pancreas fl-cells (according to Luiten et a~.“}. Lane 3 is a camera lucida drawing of the DMnX giving, at the different anterior-posterior levels, the pattern of innervation after a PHA-L injection into the LC. Lane 4 gives the location in the DMnX of preganglionic parasympathetic cells with axons running through the cardiac branch of the vagus nerve (according to Nosaka et ~1.~~).
oblcmgata. In the posterior medulla obiongata, beginning at the rostra1 level of the DMnX and extending up to the most posterior level of this preganglionic parasympathetic autonomic nucleus, medially descending fibers from the LC curved in a dorsal direction and ran to the DMnX in the medial or lateral reticular formation. These fibers innervated the medial part of the rostra1 DMnX and the lateral part of the middle and posterior DMnX, respectively. (ii) Fibers emerging ventrolaterally from the injection in the LC and taking a lateral course in the reticular formation innervated the dorsolateral parts of the parvocellular reticular formatian and the spinal trigeminal nuclei. {iii) A small number of labeled fibers from the LC took a descending course in the dorsomedial medulla. These fibers left tbe PHA-L injection site medially and coursed, ventral of the medulla
fourth psntine
ventricle, to the dorsomedial part of the periaqueductal gray where they bent in a
posterior direction. In the rostra1 medulla oblongata, dorsomedially descending fibers from the LC were seen in the ventromedial part of the prepositus hypoglossal nucleus. At more caudal levels in the brainstem, some dorsom~i~ly descending fibers from the LC curved in a lateral direction and innervated the medial part of the rostra1 third of the DMnX. Fibers not innervating the DMnX continued with the descending course and entered the cervical spinal cord, running caudally adjacent to the spinal canal. Efferents from the LC innervated the DMnX bilaterally. Crossing fibers were found in the ventral medulla oblongata and in the pontine periaqueductal gray. Target areas in the dorsal
m&r
vugus nucleus
Bilaterally, the PHA-L-positive varicose fibers and terminal boutons from the LC were found medially in the rostra1 third and laterally in the posterior
Ci.
158
J. TER HORST ef
two-thirds of the DMnX (Fig. 3). In the most caudal parts of the DMnX a scattered pattern of innervation was found. The DMnX innervation from the LC was more intense ipsilaterally but the difference with the contralateral side was small. At the level of the area postrema (Figs 2C, E, 3) in the DMnX, the innervation from the LC was most intense. Ventrolaterally in the DMnX, at this anterior-posterior level, a network of labeled fibers was found enclosing some of the preganglionic cells in this area. The PHA-L-labeled fibers have boutons en passage and run from lateral to medial in the DMnX. In the LC experiments, a few lectin-positive varicose fibers were found in the ventral part of the hypoglossal nucleus and, dorsal of the DMnX, in the medial and lateral part of the nucleus of the solitary tract. Innervations of the hypoglossal and solitary tract nuclei were also found in the control experiments but in these cases the DMnX was not labeled. Control experiments
After small iontophoretic HRP injections into the DMnX (Fig. 2D) bilaterally in the LC the retrogradely labeled cells were found. The number of labeled neurons in the LC was small. In one experiment, with a very small HRP injection into the lateral part of the DMnX, ipsilaterally eight and contralaterally three labeled cells were found. These labeled cells were located ventrally (Figs 1.2, 2F) in the middle LC and dorsally at the most rostra1 and caudal levels of the LC. In the dorsal LC the neurons were faintly labeled. DlSCUSSION
In the present investigation efferent connections from the LC to the dorsal motor nucleus of the vagus were shown with PHA-L as a neuroanatomical tracer. The target areas in the DMnX were the lateral part at the level of the area postrema and the medial part at the rostra1 levels. The connection was conf%med with control HRP injections into the DMnX. The PHA-L tracing of LC efferents to the DMnX corroborated previous studies where this connection was identified with catecholamine autofluorescence and electrolytic lesioning of the LC,r3 WGA-HRP retrograde tracing of afferents of the nucleus of the solitary tract3’ and autoradiographic tracing of LC efferents.39 In the rat, Jones and Yang’ also found a few silver grains in the caudal part of the DMnX after tritiated leucine injections into the LC region but we could not corroborate these observations in a previous HRP investigation of DMnX afferents.” Moreover, the LC-DMnX pathway is not described in a recent PHA-L investigation giving the distribution of LC efferents in the brainstem of the rat.6,7 The methods used for the identification of LC efferents may account for the differences seen between the investigations. In the autoradiographic tracing
ul.
experiments not all the efferents of the LC’ may be revealed. The location of the LC near the lateral aspect of the fourth ventricle and the small dimensions of this noradrenergic nucleus impede the proper targeting of tritiated amino acids into the LC. Injection sites, after autoradiographic processing of the sections, appear as accumulations of silver grains in the overlying photographic emulsion without a distinct morphological basis for identifying the cells transporting the tracer. In general, neuroanatomists assume that all the cells in the area with a high density of silver grains are labeled, but this may not be the case. Cells transporting the marker, in the PHA-L procedure for tracing of neuronal efferents, are stained from the soma up to the terminal presynaptic endings8.36.40and enable an accurate identification of the cellular origin of the projections. In the HRP experiments, the chromagen and the incubation parameters used for revealing the presence of the peroxidase are important for the effectiveness of the reaction. In the present investigation we used the TMB procedure of Mesulam2’ which in comparative studies was shown to be superior to all the other methods for revealing HRP at the level of the light microscopic examination. 20.22In our previous HRP investigation of DMnX afferents,” labeled cells in the LC were probably not detected because we used the benzidine dihydrochloride method of De Olmos and Heimer,s which revealed, in the comparative study,20,22only 3&50% of the retrogradely labeled cells seen in the TMB-stained sections. Although similar pathways from the LC are revealed in the brainstem with PHA-L as a neuroanatomical tracer, Fritschy and Grazanna’ did not find the projections from the LC to the DMnX. At least two factors may be responsible for the observed difference; the unequal survival times (14-21 vs seven days) and/or the cells in the LC projecting to the DMnX were not labeled with PHA-L. Survival times of seven to 10 days gave the best results in our experiments. Longer periods caused much weaker signals in the target areas36 suggesting that, with survival times of more than 14 days, in Fritschy and Grazanna’s study the signal in the DMnX might have been below the detection level. However, it is also possible that the DMnX projecting cells in the LC were not labeled with PHA-L because Fritschy and Grazanna’ used very small iontophoretic injections for tracing LC efferents. In their 18 experiments, the number of labeled neurons in the LC ranged from 20 to 100. For a survey of all efferent connections of approximately 1500 cells33 in the LC, the 18 small injections were probably not sufficient. After small HRP injections into the DMnX, retrogradely labeled cells were found in the ventral part of the middle LC and in the dorsal part. Similar distributions of efferent projections from subpopulations of the LC were reported by Loughlin et al.‘4J5 who, by using large HRP injections, describe projections from the ventromedial parts of the LC to the cerebel-
Locus coeruleus projections lum and spinal cord. However, Loughlin’s
investigation was not complete, only areas known for receiving noradrenergic inputs were injected with HRP. Injections into the dorsomedial brainstem or DMnX were not made. In addition, in the present HRP inv~ti~a~on of DMnX alIerents the retrogradely labeled cells were found in the rostra1 part of the LC, in the subpopulation projecting to the hypothalamus.‘4*15 Collateralization of LC efferents projecting to the brainstem and hypothalamus needs to be studied. The main target areas in the DMnX for the noradrenergic LC efferents were the rostromedial parts and, at the level of the area postrema, the lateral parts of the DMnX. Immunohistochemical studies, using TH and dopamine-b-hydroxylase for labeling norad~ner~c nerve fibers in the DMnX,i”J9 showed abundance of noradrenalin-positive fibers in the lateral part of the DMnX at the area postrema and obex levels and a modest noradrenergic fiber labeling in the medial DMnX. These immunohistochemical observations have a remarkable resemblance to the pattern of fiber labeling in the DMnX seen after PHA-L injections into the LC and show that the LC is indeed an important source of noradrenergic fibers in the lateral and medial DMnX. The preganglionic parasympathetic cells in the medial parts of the motor vagus nucleus innervate the stomach’” and the pancreas’J*2s (Fig. 3), whereas the preganglionic cells in the lateral part of the DMnX project to the heartiQ3 (Fig. 3). The projections from the LC to the DMnX were most dense in the lateral, cardiac part of the nucleus, suggesting that the LC is involved in the
regulation of cardiovascular functions. Physiological studies corroborate this suggestion by showing widespread changes in cardiovascular dynamics after elec-
159
trical stim~ation of the region of the LC.38 Occasionally, a bradycardia was seen after the cessation of the stimulus in the LC region.‘” Involvement of the LC in autonomic regulation of insulin release is not very likely. The medial parts of the DMnX, containing the small-sized perikarya of the preganglionic parasympathetic cells for autonomic regulation of insulin-producing pancreatic /Icells’* (Fig. 3), were not innervated by the efferents from the LC. The dendrites of these preganglionic neurons also probably extend in regions lacking a substantial LC inneffation because they are short and mainly oriented horizontally along the mediolatera1 axis of the DM~IX.‘**~~ The paraventricular hypothalamic nucleus is important for limbic control of autonomic functions; it is probably the main h~~alamic output center to the preganglionic autonomic nuclei.‘6,‘7 The parvocellular part of the PVN has direct projections to the para- and orthosympathetic preganglionic neurons in the brainstem and spinal cord17*t1*3’ and also innervates the locus coeruleus.” Participation of the PVN in food intake behavior,rz pancreas hormone release” and cardiovascular control3 is demonstrated. With respect to the cardiovascular control of the PVN, it has been shown that electrical stimulation of the paraventricular nucleus elicits cardioacceleration (tachycardia~~” possibly mediated by vagal inhibition3,28,zg and sympathetic excitation.3 The LCDMnX pathway described in the present investigation may participate in the inhibition of the cardiac preganglionic neurons in the DMnX by the PVN. Acknowfedgemetrts-The authors wish to thank Mrs Nieske Brouwer, Anja Dreijer-Kattenberg and Liza Mast for the processing of the histological sections.
REFERENCES 1. Berthoud H.-R., Fox E. A. and Powley T. L. (1990) Localization of vagal preganglionics that stimulate insulin and glucagon secretion. Am. J. Physiol. 258, R160-R168. 2. Berthoud H.-R. and Powley T. L. (1990) Identification of vagal pregangiionics that mediate cephalic phase insulin response. Am. J. Fhysiol. 258, R523-R530. 3. Ciriello J. and Calaresu F. R. (1980) Role of paraventricular and supraoptic nuclei in central cardiovascular regulation in the cat. Am. J. Physiof. 239, Rl37-R142. 4. Copray J. C. V. M., Liem R. S. B., Ter Horst G. J. and Van Will&n J. D. (1990) Dopaminergic afferents to the mesencephalic trigeminal nucleus of the rat: a light and electron microscope immunocytcchemistry study. Brain Res. 514, 343-348. 5. De Ohnos J. and Heimer L. (1977) Mapping of collateral projections with the HRP-method. Neurosci. Lat. 6,107-i 14. 6. Fritschy J. M. and Graxanna R. (1990) Demonstration of two separate descending noradrenergic pathways to the rat spinal cord: evidence for an intragriseal trajectory of locus coeruleus axons in the superficial layers of the dorsal horn. J. camp. Neural. 291, 533-582. 7. Fritschy J. M. and Grazanna R. (1990) Distribution of locus coeruleus axons within the rat brainstem demonstrated by Phuseolus udgaris leucoagglutinin anterograde tracing in combination with dopamine-fl-hydroxylase immunofluorescence. J. camp. Newol. 293, 616-631. 8. Gerfen C. R. and Sswchenko P. E. (1984) An anterograde ne~~nato~~ tracing method that shows the detailed morphology of neurons, their axons and terminals: immunohist~he~~l localization of an axonally transported plant lectin, Phaseofus vulgaris, leucoagglutinin (PHA-L). Bruin Res. 290, 219-238. 9. Jones B. E. and Yang T. Z. (1985) The efferent proiections from the reticular formation and the locus coeruleus studied by anterograde and-retrograde axonal transport in the rat. J. camp. Neurol. 242, X-92. 10. Kalia M., Fuxe K. and Goldstein M. (1985) Bat medulla oblongata. II. Dopaminergic, noradrenergic (Al and A2) and adrenergic neurons, nerve fibers and presumptive terminal processes, J. camp. Neural. 233, 308-332.
160
‘2. .I fm
HORSTrl (I/
11. Kaha M. and Me&am M. M. (1980) Brainstem projections of sensory and motor components of the vagus complex in the cat: II. Laryngeal, tracheobronchial, pulmonary, cardiac and gastrointestinal branches. J. camp. Ncuml. 193, 467-508. 12. Leibowitz S. F., Hammer N. J. and Chang K. (1981) Hypothalamic paraventricular nucleus lesions produce overeatmg and obesity in the rat. Physiol. Behav. 27, 1031-1040. 13. Loizou L. A. (1969) Projections of the nucleus locus coeruleus in the albino rat. Bruin Res. 15, 5633565. 14. Loughlin S. E., Foote S. L. and Bloom F. E. (1986) Efferent projections of nucleus locus coeruleus: morphologrc subpopulations have different efferent targets. Neuroscience 18, 307--319. 15. Loughlin S. E., Foote S. L. and Bloom F. E. (1986) Efferent projections of nucleus locus coeruleus: topographic organization of cells of origin demonstrated by three-dimensional reconstruction. Neuroscience 18, 291~ 3 19. 16. Luiten P. G. M., Ter Horst G. J. and Steffens A. B. (1987) The hypothalamus, intrinsic connections and outflow pathways to the endocrine system in relation to the control of feeding and metabolism. Prog. Neurobiol. 28, 1-54. 11. Luiten P. G. M., Ter Horst G. J., Karst H. and Steffens A. B. (1985) The course of paraventricular hypothalamic efferents to autonomic structures in medulla and spinal cord. Bruin Res. 329, 374378. 18. Luiten P. G. M., Ter Horst G. J., Koopmans S. J., Rietberg M. and Steffens A. B. (1984) Preganglionic Innervation of the pancreas islet cells in the rat. J. aufon. nerz:. Sysr. 10, 2742. 19. Manier M., Feuerstein C., Passagia J. G., Mouchet P., Mons N., Geffard M. and Thibault J. (1990) Evidence for the existence of L-DOPA- and dopamine-immunoreactive nerve cell bodies in the caudal part of the dorsal motor nucleus of the vagus nerve. J. Chem. Neurounat. 3, 193-205. 20. Mesulam M. M. (1982) Principles of horseradish peroxidase neurohistochemistry and their applications for tracing of neural pathways-axonal transport, enzyme histochemistry and light microscopic analysis. In Trueing Neural Connections with Horseradish Peroxidase (ed. Mesulam M. M.), pp. I-153. John Wiley, New York. 21. Moore R. Y. and Card J. P. (1984) Noradrenalin containing neuron systems. In Handbook of Chemical Neurounutom.y. Vol. 2: Classical Transmitters in the CNS (eds Bjdrklund A. and Hijkfelt T.), pp. 123-156. Elsevier, Amsterdam. 22. Morrell J. I., Greenberger L. M. and Pfaff D. W. (1981) Comparison of horseradish peroxidase visualization methods: quantitative results and further technical specifics. J. Histochem. Cytochem. 29, 903-916. 23. Nosaka S., Yamamoto T. and Yasunaga K. (1979) Localization of vagal cardio-inhibitory preganglionic neurons within the rat brainstem. J. camp. Neurol. 186, 79992. 24. Ono T., Nishino H., Sasaki K., Muramoto K., Yano 1. and Simson A. (1978) Paraventricular nucleus connections to spinal cord and pituitary. Neurosci. Lett. 10, 141~146. 25. Paxinos G. and Watson C. (1986) The Rut Bruin in Stereotuxic Coordinates. Academic Press, New York. 26. Pellegrino L. J., Pellegrino A. S. and Cushman A. J. (1979) A Stereotuxic Atlas of the Rat Bruin. Plenum Press, New York. 27. Pittman Q. J. and Franklin L. G. (1985) Vasopressin antagonist in nucleus tractus solitarius/vagal area reduces pressor and tachycardia responses to paraventricular nucleus stimulation in rats. Neurosci. Lett. 56, 1555160. 28. Rogers R. C. and Hermann G. E. (1985) Vagal afferent stimulation-evoked gastric secretion suppressed by paraventricular nucleus lesion. J. auton. new. Sysi 13, 191-199. 29. Rogers R. C. and Nelson D. 0. (1984) Neurons of the vagal division of the solitary nucleus activated bv the par~ventricular nucleus of the hypothalamus. J. uuton. nerv. cyst. 10, 193-197. _ 30. Rye D. B., Lee H. J., Saper C. B. and Wainer B. H. (1988) Medullary and spinal efferents of the pedunculopontine tegmental nucleus and adjacent mesopontine tegmentum in the rat. J. camp. Neural. 269, 315-341. 31. Swanson L. W. and Kuypers H. G. J. M. (1980) The paraventricular nucleus of the hypothalamus: cytoarchitectonic subdivisions and organization of projections to the pituitary, dorsal vagal complex and spinal cord as demonstrated by retrograde fluorescent double labeling methods. J. camp. Neurol. 194, 5555570. 32. Swanson L. W. and Sawchenko P. E. (1983) Hypothalamic integration: organization of the paraventricular and supraoptic nuclei. A. Rev. Neurosci. 6, 269.-324. 33. Swanson L. W. (1976) The locus coeruleus: a cytoarchitectonic, Golgi and immunohistochemical study in the albino rat. Bruin Res. 110, 39-56. 34. Ter Horst G. J. and Luiten P. G. M. (1987) Phaseolus vulgaris leucoagglutinin tracing of intrahypothalamic connections of the lateral, ventromedial, dorsomedial and paraventricular nuclei of the hypothalamus in the rat. Bruin Res. Bull. 18, 191-203.
35. Ter Horst G. J., Copray J. C. V. M., Van Willigen J. D. and Liem R. S. B. (1990) Contralateral projections of cells in the motor trigeminal nucleus of the rat. Neurosci. Lett. 113, 26&266. 36. Ter Horst G. J., Groenewegen H. J., Karst H. and Luiten P. G. M. (1984) Phuseolus uulguris leucoagghrtinin immunohistochemistry. A comparison between autoradiographic and lectin tracing of neuronal efferents. Rruin Res. 307, 3799383. 37. Ter Horst G. J., Luiten P. G. M. and Kuipers F. (1984) Descending pathways from the hypothalamus to dorsal motor vagus and ambiguus nuclei in the rat. J. auton. new. Syst. 11, 59-75. 38. Ward D. G. and Gunn C. G. (1976) Locus coeruleus complex: elicitation of a pressor response and a brainstem region necessary for its occurrence. Bruin Res. 107, 401-406. 39. Westlund K. N. and Coulter J. D. (1980) Descending projections of the locus coeruleus and subcoeruleus/medial parabranchial nuclei in monkey: axonal transport studies and dopamine-b-hydroxylase immunocytochemistry. Bruin Res. Rev. 2, 235-264.
40. Wouterlood F. G. and Groenewegen H. J. (1985) Neuroanatomical tracing by use of Phaseolus vulguris leucoagglutinin (PHA-L), electron microscopy of PHA-L filled neutonal somata, dendrites, axons and axon terminals. Bruin Res. 326, 188-192.
41. Yoshimatsu H., Niijima A., Oomura Y., Yamabe K. and Katafuchi T. (1984) Effects of hypothalamic lesion on pancreatic autonomic nerve activity in the rat. Bruin Res. 303, 1477152. (Accepted
9 April 1991)
