- Email: [email protected]
Bulbospinal projections to the intermediolateral cell column; a neuroanatomical study
Journal of the Autonomic Nervous System, 1 (1979) 103--117
I03
© Elsevier/North-Holland Biomedical Press
BULBOSPINAL PROJECTIONS TO THE INTERMEDIOLATERAL CELL COLUMN; A NEUROANATOMICAL STUDY
K L A U S AMENDT, JI)RGEN CZACHURSKI, K L A U S D E M B O W S K Y and H O R S T SELLER I. Physiologisches lnstitut o f the University of Heidelberg, lm Neuenheimer F~ld 326. D-6900 Heidelberg 1 (G.F.R.)
(Received April 21st, 1979) (Accepted June 5th, 1979)
K e y w o r d s : Horseradish peroxidase -- intermediolateral column -- buibo-
spinal neurones
ABSTRACT
The location of those neurones in the brain stem that project to the i~termediolateral column (ILC) from which preganglionic sympathetic neurones have their origin was studied by the m e t h o d of retrograde Lransport of horseradish peroxidase (HRP). In cats 30--50 nl of a 30% :~RP solution was injected into the region of the ILC at T3 or L~ on one side. After a survival periox of 72 h the lower brain stem from CI to the inferior colliculi was sectioned and prepared for histological study under bright:'ield illumination. Ncurones stained with exogenous HRP were found in three regions: (a) in the ipsilateral, dorsomedial part of the nucleus of the solitary tract (NTS) (43% of all labelled neurones), in the cranial part of the NT:~, and also on the contralateral side (7%); (b) in the ventrolateral reticular for raation begiiming at the level of the obex up to 8 mm cranial to the obex (2,i% ipsilateral, 3% contralateral); and (c) in the ventral part of the raphe nuch,i (postpyramidal and inferior central nucleus) from 2 to 9 mm cranial to the c,bex (22%).
INTRODUCTION
The existence of a vasomotor centre in the lower brain stem has been investigated for more than a hundred years. In 1871 Ow;jannikow [23] showed that arterial blood pressure was unaffected by complete transection of the brain stem at the level of the inferior colliculi. Blood pressure, however, decreased following further transections at levels lower in the brain
104 stem. It fell to its lowest level, about 40 m m Hg~ after a transection 4 mm cranial to the obex. These results were later confirmed by Dittmar [13] in 1873. By dorsal hemisections Dittmar also found that the area which was essential for the maintenance of normal blood pressure was restricted to the ventral half of the medulla. From transection studies Alexander [ 1] showed that the lower brain stem was a source of the tonic activation of the sympathetic outflow. Since there is a partial recovery of sympathetic activity after a further transection at C2, Alexander postulated the existence of a depressor as well as a pressor area in the caudal brmn stem. Recent experiments with restricted bilateral lesions of the ventral surface of the medulla caudal to the trapezoid bodies have demonstrated more precisely that there is a circumscribed region important for the maintenance and regulation of normal blood pressure [14]. Electrical stimulation within the brain stem has not given consistent results relative to the location of pressor and depressor areas [4]. Electrical stimulation is, however, an inappropriate m e t h o d for the functional identification of brain stem areas [10] since there is always an uncertainty as to whether one stimulates a mixed population of antagonistic neurones or ascending and/or descending fibres within a particular region of the nervous system. Another method which has been used in attempts to identify bulbospinal neurones that terminate on spinal sympathetic neurones has been to antidromicaUy activate them by stimuli applied to their terminals in the intermediolateral column {ILC) [16]. Recordings of field potentials and single neuronal activity in the brain stem, however, have revealed a widespread dis. tribution of the cell bodies of bulbospinal neurones thus activated [16]. Again the precision effected by the methodology is questionable. Another method of greater promise, however, is now available. Neurones with axons descending to the spinal cord have been identified in the brain stem by means of retrograde cell degeneration [8] and by the retrograde axonal transport of horseradish peroxidase {HRP) [18]. The bulbospinal neurones thus identified have been located in the gigantocellular, magnocellular nuclei and lateral tegmental field, in the lateral reticular nucleus, in the nucleus of the solitm-y tract and in the nuclei of the raphe (nomenclature of medullary ~uclei according to Berman [ 6] ). These observations suggest that bulbospinal neurones whose axons project to the intermediolateral columns of the cord (ILC) are probably located somewhere within the limits of these medullary nuclei. It was the aim of the present investigation to identify these neurones by injection of small voltlmes of HRP into the region of the ILC. METHODS Twenty-three cats of either sex (body weight !.8--3.5 kg) were anaesthetized by intrapei~toneal injection of 35--40 mg/kg of pentobarbitone. A catheter was introduced in~o a cutaneous vein of the right foreleg. Cats were
105
intubated to permit artificial respiration. The left 3rd thoracic white ramus (WR-T3) was dissected and cut close to its junction with the sympathetic chain. A laminectomy was performed at the third thoracic segment. The tip of the cannula for the injection of HRP was positioned in the intermediolateral column by two different procedures: (a) the WR-T3 was stimulated and the antidromic field potential in the cord was recorded by means of a glass micropipette filled with KC1 solution (3 mol/litre). Then the recording electrode was removed and a glass pipette (tip diameter 30--40 pro) filled with HRP solution was inserted into the same position from which the identifying field potential had been recorded. (b) A glass pipette or a hypodermic needle (outer diameter 150 ~m) filled with HRP was positioned into the intermediolateral column (ILC) according to the stereotaxic coordinates elaborated by Lebedev and Mikhallova [19]. In each experiment three injections of 30--50 nl HRP solution were performed at a distance of i mm from each other in the cranio-caudal direction. The latter m e t h o d was als¢; used for injection of HRP into the ILC at the first lumbar segment. The localization and the e x t e n t of injected HRP was assessed histologically in each experimem;. Thirty t~ fifty nl of a solution of 30% HRP (Boehringer I) in Tris buffer (pH 7.6) was injected at each injection site by pressure according to the technique of Price et al. [24]. It has been shown that an injection of this volume of HRP solution is confined to an area of 200--500 ~m in diameter from which HRP is taken up for retrograde transport [25]. The evidence q m t e d was obtained from cerebral cortex injections but injections into cord ,;issue were similarly confined. After the injection the glass pipette or the hypodermic needle was l i f t in place for more than 5 min to allow the enzyme to diffuse from the tip before removing the pipette or needle and closing the wound. The animals were allowed to recover from anaesthesia and kept alive Zor a period of 72 h. They were then anaesthetized by pentobarbitone, a catheter was introduced into the abdominal aorta for perfusion and the inferior vena cava opened just caudal to the diaphragm. The cats were perfused first with 2 litres of Ringer solution then with 2 litres of a fixative solution which consisted of 1% paraformaldehyde, 2.5% glutaraldehyde and 1% sucrose in 0.1 mol/litre phosphate buffer at pH = 7.4, according to the method described by Kim and Strick [17]. The brain stem between C~ and the inferior colliculi and the spinal cord s,egments at T3 or L~ were removed immediately af'~er the perfusion. The tissue was placed in fresh fixative solution -- which was identical to that used in the perfusion -- for 12 h at 4°C and then transferred to a 20% sucrose solution for 24 h. The tissue was then cut into 40 tam thick sections with a freezing microtome. The histochemical identification of HRP was accomplished according to the m e t h o d of Mesulam [21] which consists principally of an incubation of the sections in a solution of benzidine dihydrochloride and sodium nitroferricyanide to which after 20 rain H202 is added.
106
T h e outer shapes of the slices and the exact position of each neurone stained by HRP were drawn b y means of a Leitz microscope drawing device. These drawings were transferred to the plates of a brain stem atlas [6]. If n o t stated differently the dots plotted in the figures represent all labelled cells seen on serial sections within half of the quoted distance between the next caudal and rostral plate. By comparing the distance between two points on the surface of the spinal cord before and after fixation, the longitudinal tissue shrinkage was estimated to be about 10%. RESULTS
H R P was injected into the intermediolateral column (ILC) at Ts in 18 experiments and at L~ in 5 experiments. In 15 experiments the injection pipette was inserted vertically into the ][LC at T3, while in a further 3 experiments ir, order to avoid retrograde transport by injured fibres in the dorsolateral funiculus or in the dorsal horn, the ILC at Ts wa~ approached at an
~
o~
-t8mm 2turn Fig. I. TwG, single charts of origh, al serial sections frc~m one e~periment at ~.0 mm di~° tance cranial and 1.8 ~ r caudal from the obe:r. Each d,)t represen~ oP.e labs~led neu~'one i~ these single sections. CI, central inferior m~c~eus; CUC, cuneate nucleus, caudal divisi~n; FTL, lateral tegmentai field; FTM, magaocetlular te~menta~ ~ie]d; GRR, ~vacile nucleus, rostral di~dsion; P, pyra~nidat tract; PPR, postpyramidat nucleus of the raphe; SM, medial nucleus of the solitary tract; 5ST, spinal trigen~ina] tract; 7L, facial nucleus, lateral division; 7M, facial nucleus, mediaJ division; 12~ hypoglossat ~l~cleu~.
~
, 4mm
,
~
)
,.~..,. ::. ~ ~
Injection at L1 (n-3}
.42~m
......
~4.gmm
......
.~ig, 2. Sequences of superimposed serial sections of the brain stem at different distances from the obex (+ is crapJa| and ~o-is caudal to Q ;he obex) from several different preparations. Each dot represents one neurone labelled w~th ~ x u ~ o u ~ ;~RP. -~
njeotion at T3 (n=5}
mm
,L
1.4
4.$
• "... ::
"- .!: ".:!-..
108 angle passing through the dorsal medical sulcus. The data presented are based on the results of 11 experiments (8 at T3, 3 at L~). In 3 of the remaining 12 experiments no HRP-stained neurone could be detected within the brain stem and 9 experiments were excluded from analysis because the neurones were not stained with sufficient intensity to permit an unequivocal identification. The main criterion used for identification was the presence of intracellular granulated reaction products of ?l~e enzyme. Histological examination of the location and the extent of spread of the injected HRP in the thoracic or lumbar spinal cord always produced evidence that the centre of the injection spot was within the ILC and largely confined there (Fig. 2). After injection of HRP into the ILC at T3 specifically stained neurones could be detected in three separate regions of the brain stem: a dorsomedial region, a ventrolateral region, and a ventromedial region (Fig. 1). These results were uniformly obtained and were independent from the various metb, ods of injection.
Dorsomedial region The dorsomedial regi,~m containing stained neurones e x t e n d e d from 4 m m caudal to 3.5 mm crani~fl to the obex (Figs. 3 and 4). In its caudal extension this region was close to and dorsolateral to the central canal; it deviated to the lateral side in its cranial prolongation. The neurones in the caudal part of this region, up to 1.5 m m cranial to the obex, were found ipsilateral to the site of the injection. Neurones in the cranial third of this region were located mainly contralateral to the site of the injection. The impregnated neurones in the dorsomedial region were mostly round with a diameter of 2(}--40 ~m (see Fig. 6C). According to tt~e atlas of Berman [6] most of the neurones were located in the medial nucleus of the solitary tract (Fig. 5). Less than 10% of all neurones were located in the adjacent gracile nucleus.
VentroIateral region Eighty percent of all labelled neurones in this region were located in an area extending from the obtx up to 8 m m in the cranial direction on the side ipsilateral to the injecti¢,,~ (Figs. 3 and 4). Only a few neurones were found caudal to the obex. Approximately 10% of the labelled neurones in this ventrolateral region were found contralateral to the site of the injection. The medio to lateral ,extent of this region was from 1.7 to 4.5 m m lateral to the nddline. Most of the neurones were located from 450 to 1150 pm beneath the ventral surface of the brain stem (Figs. 3, 5 mid 6). The shape of the neurones was mostly polygonal, the largest diameter being 40--70 pro. This region containing these HRP-identified neurones lies in the cranial prolongation of the lateral reticul~u¢ nucleus and is ventrotateral to ~ghemagnocellular tegmental field ( Fig. 5).
Veu tromedial re~,ion The neurones in this region were located in a narrow bead 500 um on
o
w
""
° "/ •
\
+4.9mm~
+5.6 mm
"
•
" ~
-r...:. !:.. o
~
+6.5m
+8.1mm
+8.9
,
10
~
mm
!
Fig. 3. Summary of the loea|ization of HRP-labelled neurones from 5 experiments after' injection at T3.
- ~.2 mm
= 1.9mm\~';-'"'~""'"/ ~~"
- 0.6mm • . -~--o - ' " ~ ' ) "~
+0.6mm
+I.4m
+29mm :
~
c
)
0
4
3ram
VENTROMED|AL REGION
3ram
t o
i
ee
.'.:-
VE~JTROLATERAL REGION
..
Fig. 4 L o c a l i z a t i o n o f H R P - | a b e | l e d ~ e u r o r l e s f r o m 5 e x p e r i m e n t s a f t e r i n j e c t i o n o f H R P a t T 3 in p r o j e c t i o n t o t h e d o r s a l s u r f a c e o f ~he l o w e r b r a i n stern.
DOR,~OM~D|AL ~I~/G|ON
°:°{
/
ti -i
III
A
~
/
4:,
~f i
q_
S
112
lO00pm
200~m
Fig. 6. T y p e s o f l a b e l l e d n e u r o n e s in t h e t h r e e d i f f e r e n t areas, A: v e n W o m e d i a l area; R : v e n t r o l a t e r a ] ; C: d o r s o m e d i a ~ r e g i o n .
three regions mentioned; a small number, however, (mean 20%) were difi~asely scattered throughout the brain stem. The concentration or dispersal of neurones thus identified within these regions, however, varied from experi-
;c
~mm 6
~mm
- - 1 9 mrrl
+ 4 , 9 mm
Fig. 7. D i s t r i b u t i o n o f l a b e l l e d n e u r o n e s a f t e r i n j e c t i o n o f t-IRP a t t h e I L C at T3 o n t h e left side. A - - D r e p r e s e n t 4 d i f f e r e n t e x p e r i m e n t s with the same amount of injected HRP and the same su~ival period.
\
: 7.0 mm
114 m e n t t o experiment (Fig. 7). For all experiments the distribution expressed as a percentage of the total number of neurones within file three regions was 50% in the dorsomedial, 28% in the ventrolateral, and 22% in the ventromedial region. Seventy-two hours after injection of HRP at L~ labelled neurones could be found in the three regions described above, but the total n u m b e r of labelled neurones per experiment was smaller than after injection at T3 (mean value 60 neurones) (Fig. 2). The distribution of neurones among the three regions was: dorsomedial region, 47%; ventrolateral region, 26%; ventromedial region, 27%. DISCUSSION The method of retrograde transport of HRP permitted identification of three brain stem areas which contain neurones with spinal descending projection to the ILC. This analysis contributes to our knowledge of autonomic: system control by at least identifying the pathways available for exerting excitatory and inhibitory tone and reflex control over preganglionic neurones. The method of retrograde labelling of neurones with HRP presents some general problems and limitations, therefore the interpretation of the results requires discussion. It has been demonstrated that retrograde axonal transport of HRP occurs if the HRP is taken up by nerve terminals or by injured axons [ 15,22]. HRP can ~,A~oenter axons en passant by diffusion but this HRP is not involved in the retrograde transport [15,22]. The neurones which have been identified in the brain stem in the present experiments by means of exogenous HRP therefore may have taken up the HRP through either injured ,axons or their nerve terminals. The injury of fibres could have been produced by the insertion of the injection cannula. This would have involved fibres which run through the dorsolateral funiculus and lateral parts of the dorsal horn. Therefore it must be considered possible that the population of bulbospina] neurones which has been demonstrated in these experiments might contain not only neurones which terminate within the ILC but also
some
,v,,,,.,,
~ , , , . , ~
in
~c
dorsal
H ~ U - -r- l l [~~ J'
. l? ~ .~AL'I,[I~L'[IIU[¢~ . . . . ~ . . . . . . . . L,I[1}.~5 ~ : . . . .~. o. D I ~ A D ' -. -
tion might include neurones which send axons through the dorsolateral funiculus and terminate in the ventral horn at the same or more caudal segmental level. However, experiments in which the insertion of the injection cannula was performed at an angle through the dor~aJ funiculus revealed the same findings as after vertical insertion making it likely that the presented observations are not due to uptake of HRP by damaged ~_xons in the injection canal but due to uptake by nerve ~rminals within the injection spot. In control experiments HRP could be demonstrated in a~ area 800--800 gm in diameter 2 h ~£ter injection of 30--50 nl in the region of the ILC. Within this area terminals of the ILC, the dorsal part of the ventral horn, and the dorsal horn up to the level of ~arnina IV could have taken up HRP for
115 retrograde transport. The area of diffusion identified represents the maxim u m since, as demonstrated by Vanegas et al. [25], terminals which come into contact with HRP after more than 2 h do not take up HRP or show retrograde transport. It is obvious, from the fact that no labelled neurones were found in the red nucleus and that less than 1% of all labelled neurones were detected in other m o t o r nuclei of the brain stem, that although HRP had reached parts of the ventral horn by diffusion after more than 2 h this HRP was n o t retrogradely transported. It is conceded that an u n k n o w n n u m b e r of neurones which do not terminate in the ILC might be included in these results but those neurones which do terminate in this ILC region must ori~,inate within the described medullary areas. Neurones in the nuclei of the solitary tract with spinal projections have also been demonstrated by Kuypers and Maisky [18] and Loewy and Burton [20]. Loewy and Burton have shown that the descending axons terminate in the ILC, the region of the phrenic m o t o r neurones, and the thoracic ventral horn [20]. Using the fluorescence m e t h o d DahlstrSm and Fuxe [12] were able to localize noradrenergic neurones in the commissural nucleus. This area was called A2 by these authors. DahlstrSm and Fuxe provided evidence that neurones of the area A2 project to the ILC. the m e t h o d of anterograde transport of labelled amino acids has been used to show that neurones in the magnocellular tegmental field project to the ILC [ 5]. This ~wea partially overlaps with the ventrolateral region which contained HRP-stained neurones in our experiments. In the caudal prolongation of this area Dahlstr5m and Fuxe [12] have localized noradrenergic neurones with spinal projection. This area was called A1. Tryptaminergic neurones in the caudal raphe nuclei with descending fibres to the spinal cord have been demonstrated by several authors [11,12]. In the same area neurones with spinal projection have been localized by means of anterograde and retrograde axonal transport [5,7,18]. In agreement with our results Basbaum et al. have shown that part of these raphe neurones terminate at the ILC [ 5]. rTtt_ l z~e present results are also in agreement with the localization of buibospinal neurones by other authors. The existence of additional polysynaptic pathways from brain stem to ILC could n o t be investigated with this method. Future electrophysiological studies of bulbospinal neurones within these regions will be necessary to clarify their properties and functions relative to the regulation of the autonomic nervous system activity. ACKNOWLEDGEMENTS These studies were supported by the German Research Foundation within the SFB 90 "Cardiovascul~ires System". The technical assistance of Miss Helga Schmidt-Prestin is gratefully acknowledged. Thanks are due to Dr. D. Richter and Dr. M. Spyer for c o m m e n t s and criticism during the preparation of the manuscript.
116
REFERENCES 1 Alexander, R,S., Tonic and reflex function of medullary sympathetic: cardiovascular centers, J. Neurophysiol. 9 (1946) 205--217. 2 Amendt, K., Czachurski, J., Dembowsky, K. and Seller, H., Localization of brainstemneurons projecting to the intermediolateral nucleus in the cat, Pfliigers Arch. ges. Physiol., 373 (1978) R76. 3 Amendt, K., Czachurski, J., Dembowsky, K. and Seller, H., Neurones within the "chemosensitive area" on the ventral surface of the brainstem which project to the h~termediolateral column, Pfliigers Arch. ges. Physiol., 375 (1978) 289--292. 4 Bach, L.M.N., Relationships between bulbar respiratory, vasomotor and somatic facilitatory and inhibitory areas, Amer. J. Physiol., 171 (1952) 417--435. 5 Basbaum, A.I., Clanton, C.H. and Fields, H.L., Three bulbospinal pathways from the rostra| medulla of the cat: an autoradiographic study of pain modulating systems, J. comp. Neuro|., 178 (1978) 209-224. 6 Berman, A.L., The Brainstem of the Cat; A Cytoarchitectonic Atlas with Stereotaxic Coordinates, Univ. of Wisconsin Press, Madison, Wisc., 1968. 7 Bobillier, P., Seguin, S., Petiqean: F., SalverL D., Touret, M. and Jouvet, M., The raphe nuclei ,~f the cat brain stem: a topographical atlas of their efferent projections as revealed by autoradiography, Brain Res., 113 (1976) 449--486. 8 Brodal, A., The Reticular Formation of the Brainstem. Anatomical Aspects and Functional Correlations, Oliver and Boyd, Edinburgh-London, 1975. 9 Burton, H. and Leewy, A.D. Projections to the spinal cord from medullary somatosensory relay nuclei, J. comp. Neurol., 173 (1977) 773--792. 10 Calaresu, F.R., Faiers, A.A. and Mogenson, G.J., Central neural regulation of heart and blood vessels in mammals, Progr. Neurobiol., 5 (1975) 1--35. 11 Coote, J.H. and Macleod, V.H., The influence of bulbospinal monoaminergic pathways on sympathetic nerve activity, J. Physiol. (Lond.), 241 (1974) 453--475. 12 Dahlstr6m, A. and Fuxe, K., Evidence for the exisCence of monoamine neurons in the central nervo.u.ssystem, Acta physiol, scand., 64, Suppl. 247 (1965). 13 Dittmar, C., Uber die Lage des sogenannten Gefgsscentrums in der Medulla oblongata. Bet. Verh. siichs. Ges. Wiss. Leipzig. Math.-Ph:,.s. Ct., 25 (1873) 449--469. 14 Guertzenstein, P.G. and Silver, A., Fall in b|ood pres.~ure prod~leed from discrete regions of the ventrai surface of medutla by gtycine and lesions, J. Physiol. (Lond.), 242 (1974) 489--503. 15 Hedreen, J.C. and McGrath, S°, Observations on labeling of neuronal eel] bodies, axons, and terminals after injection of horseradish peroxid~e into rat brain, J. comp. Neurol.~ 176 (1977} 225--248. 1~ Henry., J.L. and Ca|aresu~ F.R., Excitatory and inhibitory inputs from medn|tary nuclei projecting to spina| cardioacceleratory neurons in the cat, Exp. Brain Res°, 20 (1974} 485--504. !7 Kim, C.G. and Strick~ P.L., Critical factors involved in the demonstration of horseradish peroxidase retrograde t~ansport, Brain Res., 103 (!976) 356--36~. 18 K~lypem~ H.G.J.M. and Maisky, V.A., Retrogrede axonal transport of h~rseradish peroxidize from spinal cord to brain s~em cell groups in the cat, Neu~osci. Lett., 1 (1975) 9--14. 19 Lebedev, V.P a~d MikhaHova, G.P.~ Stereotaxic investigation of the :~ateral horn of the cat spinal cord, Fiziok Zh. (Leningrad} 57 (197! } 677--683. 20 Loewy. A D. and Burton~ H°~ Nuclei of ~he soii~.ary trac~: Effe~en~ p~oiec~ions ~o the lower br~in stem and ~p}~a| co~d of ~.he ca~, Jo temp. Neurol., 181 (!978} 421--450. 21 Mes~lam~ M.M., ~f~he btae ~eac~Aon produc~ ~n horseradish peroxidase ne~rohistoci~emi~zy: i~c.~ba~icm pa~amete~-s and vi,~ihiii~y, J. H~s~ochem. Cy~;ochem., 24 (1976) 1273--1280.
117 22 Oldfield, B.J. and McLachlan, E.M., Uptake and retrograde transport of H R P by axons of intact and damaged periphera~ nerve trunks, Neurosci. Lett., 6 (1977) 135-141, 23 Owsjannikow, Ph., Die tonischen und reflektorischen Centren der Gef~issnerven. Ber. Verb. s//chs.Ges. Wiss. Leipzig. Math.-Phys. CI., 23 (1871) 135--147. 24 Price, P., Fisher, A.W.F. and Redstone, P., A simple apparatus for the injection of small (nanolitre)volumes of horseradish peroxidase, Neuroscl. Lett., 6 (1977) 21--25. 25 Vanegas, H.,, Holl~inder, H. and Distel, H., Early stages of uptake and transport of horseradish-peroxidase by cortical structures, and its use for the study of local neurons and their processes, J. comp. Neurol., 177 (1978) 193--212.
I03
© Elsevier/North-Holland Biomedical Press
BULBOSPINAL PROJECTIONS TO THE INTERMEDIOLATERAL CELL COLUMN; A NEUROANATOMICAL STUDY
K L A U S AMENDT, JI)RGEN CZACHURSKI, K L A U S D E M B O W S K Y and H O R S T SELLER I. Physiologisches lnstitut o f the University of Heidelberg, lm Neuenheimer F~ld 326. D-6900 Heidelberg 1 (G.F.R.)
(Received April 21st, 1979) (Accepted June 5th, 1979)
K e y w o r d s : Horseradish peroxidase -- intermediolateral column -- buibo-
spinal neurones
ABSTRACT
The location of those neurones in the brain stem that project to the i~termediolateral column (ILC) from which preganglionic sympathetic neurones have their origin was studied by the m e t h o d of retrograde Lransport of horseradish peroxidase (HRP). In cats 30--50 nl of a 30% :~RP solution was injected into the region of the ILC at T3 or L~ on one side. After a survival periox of 72 h the lower brain stem from CI to the inferior colliculi was sectioned and prepared for histological study under bright:'ield illumination. Ncurones stained with exogenous HRP were found in three regions: (a) in the ipsilateral, dorsomedial part of the nucleus of the solitary tract (NTS) (43% of all labelled neurones), in the cranial part of the NT:~, and also on the contralateral side (7%); (b) in the ventrolateral reticular for raation begiiming at the level of the obex up to 8 mm cranial to the obex (2,i% ipsilateral, 3% contralateral); and (c) in the ventral part of the raphe nuch,i (postpyramidal and inferior central nucleus) from 2 to 9 mm cranial to the c,bex (22%).
INTRODUCTION
The existence of a vasomotor centre in the lower brain stem has been investigated for more than a hundred years. In 1871 Ow;jannikow [23] showed that arterial blood pressure was unaffected by complete transection of the brain stem at the level of the inferior colliculi. Blood pressure, however, decreased following further transections at levels lower in the brain
104 stem. It fell to its lowest level, about 40 m m Hg~ after a transection 4 mm cranial to the obex. These results were later confirmed by Dittmar [13] in 1873. By dorsal hemisections Dittmar also found that the area which was essential for the maintenance of normal blood pressure was restricted to the ventral half of the medulla. From transection studies Alexander [ 1] showed that the lower brain stem was a source of the tonic activation of the sympathetic outflow. Since there is a partial recovery of sympathetic activity after a further transection at C2, Alexander postulated the existence of a depressor as well as a pressor area in the caudal brmn stem. Recent experiments with restricted bilateral lesions of the ventral surface of the medulla caudal to the trapezoid bodies have demonstrated more precisely that there is a circumscribed region important for the maintenance and regulation of normal blood pressure [14]. Electrical stimulation within the brain stem has not given consistent results relative to the location of pressor and depressor areas [4]. Electrical stimulation is, however, an inappropriate m e t h o d for the functional identification of brain stem areas [10] since there is always an uncertainty as to whether one stimulates a mixed population of antagonistic neurones or ascending and/or descending fibres within a particular region of the nervous system. Another method which has been used in attempts to identify bulbospinal neurones that terminate on spinal sympathetic neurones has been to antidromicaUy activate them by stimuli applied to their terminals in the intermediolateral column {ILC) [16]. Recordings of field potentials and single neuronal activity in the brain stem, however, have revealed a widespread dis. tribution of the cell bodies of bulbospinal neurones thus activated [16]. Again the precision effected by the methodology is questionable. Another method of greater promise, however, is now available. Neurones with axons descending to the spinal cord have been identified in the brain stem by means of retrograde cell degeneration [8] and by the retrograde axonal transport of horseradish peroxidase {HRP) [18]. The bulbospinal neurones thus identified have been located in the gigantocellular, magnocellular nuclei and lateral tegmental field, in the lateral reticular nucleus, in the nucleus of the solitm-y tract and in the nuclei of the raphe (nomenclature of medullary ~uclei according to Berman [ 6] ). These observations suggest that bulbospinal neurones whose axons project to the intermediolateral columns of the cord (ILC) are probably located somewhere within the limits of these medullary nuclei. It was the aim of the present investigation to identify these neurones by injection of small voltlmes of HRP into the region of the ILC. METHODS Twenty-three cats of either sex (body weight !.8--3.5 kg) were anaesthetized by intrapei~toneal injection of 35--40 mg/kg of pentobarbitone. A catheter was introduced in~o a cutaneous vein of the right foreleg. Cats were
105
intubated to permit artificial respiration. The left 3rd thoracic white ramus (WR-T3) was dissected and cut close to its junction with the sympathetic chain. A laminectomy was performed at the third thoracic segment. The tip of the cannula for the injection of HRP was positioned in the intermediolateral column by two different procedures: (a) the WR-T3 was stimulated and the antidromic field potential in the cord was recorded by means of a glass micropipette filled with KC1 solution (3 mol/litre). Then the recording electrode was removed and a glass pipette (tip diameter 30--40 pro) filled with HRP solution was inserted into the same position from which the identifying field potential had been recorded. (b) A glass pipette or a hypodermic needle (outer diameter 150 ~m) filled with HRP was positioned into the intermediolateral column (ILC) according to the stereotaxic coordinates elaborated by Lebedev and Mikhallova [19]. In each experiment three injections of 30--50 nl HRP solution were performed at a distance of i mm from each other in the cranio-caudal direction. The latter m e t h o d was als¢; used for injection of HRP into the ILC at the first lumbar segment. The localization and the e x t e n t of injected HRP was assessed histologically in each experimem;. Thirty t~ fifty nl of a solution of 30% HRP (Boehringer I) in Tris buffer (pH 7.6) was injected at each injection site by pressure according to the technique of Price et al. [24]. It has been shown that an injection of this volume of HRP solution is confined to an area of 200--500 ~m in diameter from which HRP is taken up for retrograde transport [25]. The evidence q m t e d was obtained from cerebral cortex injections but injections into cord ,;issue were similarly confined. After the injection the glass pipette or the hypodermic needle was l i f t in place for more than 5 min to allow the enzyme to diffuse from the tip before removing the pipette or needle and closing the wound. The animals were allowed to recover from anaesthesia and kept alive Zor a period of 72 h. They were then anaesthetized by pentobarbitone, a catheter was introduced into the abdominal aorta for perfusion and the inferior vena cava opened just caudal to the diaphragm. The cats were perfused first with 2 litres of Ringer solution then with 2 litres of a fixative solution which consisted of 1% paraformaldehyde, 2.5% glutaraldehyde and 1% sucrose in 0.1 mol/litre phosphate buffer at pH = 7.4, according to the method described by Kim and Strick [17]. The brain stem between C~ and the inferior colliculi and the spinal cord s,egments at T3 or L~ were removed immediately af'~er the perfusion. The tissue was placed in fresh fixative solution -- which was identical to that used in the perfusion -- for 12 h at 4°C and then transferred to a 20% sucrose solution for 24 h. The tissue was then cut into 40 tam thick sections with a freezing microtome. The histochemical identification of HRP was accomplished according to the m e t h o d of Mesulam [21] which consists principally of an incubation of the sections in a solution of benzidine dihydrochloride and sodium nitroferricyanide to which after 20 rain H202 is added.
106
T h e outer shapes of the slices and the exact position of each neurone stained by HRP were drawn b y means of a Leitz microscope drawing device. These drawings were transferred to the plates of a brain stem atlas [6]. If n o t stated differently the dots plotted in the figures represent all labelled cells seen on serial sections within half of the quoted distance between the next caudal and rostral plate. By comparing the distance between two points on the surface of the spinal cord before and after fixation, the longitudinal tissue shrinkage was estimated to be about 10%. RESULTS
H R P was injected into the intermediolateral column (ILC) at Ts in 18 experiments and at L~ in 5 experiments. In 15 experiments the injection pipette was inserted vertically into the ][LC at T3, while in a further 3 experiments ir, order to avoid retrograde transport by injured fibres in the dorsolateral funiculus or in the dorsal horn, the ILC at Ts wa~ approached at an
~
o~
-t8mm 2turn Fig. I. TwG, single charts of origh, al serial sections frc~m one e~periment at ~.0 mm di~° tance cranial and 1.8 ~ r caudal from the obe:r. Each d,)t represen~ oP.e labs~led neu~'one i~ these single sections. CI, central inferior m~c~eus; CUC, cuneate nucleus, caudal divisi~n; FTL, lateral tegmentai field; FTM, magaocetlular te~menta~ ~ie]d; GRR, ~vacile nucleus, rostral di~dsion; P, pyra~nidat tract; PPR, postpyramidat nucleus of the raphe; SM, medial nucleus of the solitary tract; 5ST, spinal trigen~ina] tract; 7L, facial nucleus, lateral division; 7M, facial nucleus, mediaJ division; 12~ hypoglossat ~l~cleu~.
~
, 4mm
,
~
)
,.~..,. ::. ~ ~
Injection at L1 (n-3}
.42~m
......
~4.gmm
......
.~ig, 2. Sequences of superimposed serial sections of the brain stem at different distances from the obex (+ is crapJa| and ~o-is caudal to Q ;he obex) from several different preparations. Each dot represents one neurone labelled w~th ~ x u ~ o u ~ ;~RP. -~
njeotion at T3 (n=5}
mm
,L
1.4
4.$
• "... ::
"- .!: ".:!-..
108 angle passing through the dorsal medical sulcus. The data presented are based on the results of 11 experiments (8 at T3, 3 at L~). In 3 of the remaining 12 experiments no HRP-stained neurone could be detected within the brain stem and 9 experiments were excluded from analysis because the neurones were not stained with sufficient intensity to permit an unequivocal identification. The main criterion used for identification was the presence of intracellular granulated reaction products of ?l~e enzyme. Histological examination of the location and the extent of spread of the injected HRP in the thoracic or lumbar spinal cord always produced evidence that the centre of the injection spot was within the ILC and largely confined there (Fig. 2). After injection of HRP into the ILC at T3 specifically stained neurones could be detected in three separate regions of the brain stem: a dorsomedial region, a ventrolateral region, and a ventromedial region (Fig. 1). These results were uniformly obtained and were independent from the various metb, ods of injection.
Dorsomedial region The dorsomedial regi,~m containing stained neurones e x t e n d e d from 4 m m caudal to 3.5 mm crani~fl to the obex (Figs. 3 and 4). In its caudal extension this region was close to and dorsolateral to the central canal; it deviated to the lateral side in its cranial prolongation. The neurones in the caudal part of this region, up to 1.5 m m cranial to the obex, were found ipsilateral to the site of the injection. Neurones in the cranial third of this region were located mainly contralateral to the site of the injection. The impregnated neurones in the dorsomedial region were mostly round with a diameter of 2(}--40 ~m (see Fig. 6C). According to tt~e atlas of Berman [6] most of the neurones were located in the medial nucleus of the solitary tract (Fig. 5). Less than 10% of all neurones were located in the adjacent gracile nucleus.
VentroIateral region Eighty percent of all labelled neurones in this region were located in an area extending from the obtx up to 8 m m in the cranial direction on the side ipsilateral to the injecti¢,,~ (Figs. 3 and 4). Only a few neurones were found caudal to the obex. Approximately 10% of the labelled neurones in this ventrolateral region were found contralateral to the site of the injection. The medio to lateral ,extent of this region was from 1.7 to 4.5 m m lateral to the nddline. Most of the neurones were located from 450 to 1150 pm beneath the ventral surface of the brain stem (Figs. 3, 5 mid 6). The shape of the neurones was mostly polygonal, the largest diameter being 40--70 pro. This region containing these HRP-identified neurones lies in the cranial prolongation of the lateral reticul~u¢ nucleus and is ventrotateral to ~ghemagnocellular tegmental field ( Fig. 5).
Veu tromedial re~,ion The neurones in this region were located in a narrow bead 500 um on
o
w
""
° "/ •
\
+4.9mm~
+5.6 mm
"
•
" ~
-r...:. !:.. o
~
+6.5m
+8.1mm
+8.9
,
10
~
mm
!
Fig. 3. Summary of the loea|ization of HRP-labelled neurones from 5 experiments after' injection at T3.
- ~.2 mm
= 1.9mm\~';-'"'~""'"/ ~~"
- 0.6mm • . -~--o - ' " ~ ' ) "~
+0.6mm
+I.4m
+29mm :
~
c
)
0
4
3ram
VENTROMED|AL REGION
3ram
t o
i
ee
.'.:-
VE~JTROLATERAL REGION
..
Fig. 4 L o c a l i z a t i o n o f H R P - | a b e | l e d ~ e u r o r l e s f r o m 5 e x p e r i m e n t s a f t e r i n j e c t i o n o f H R P a t T 3 in p r o j e c t i o n t o t h e d o r s a l s u r f a c e o f ~he l o w e r b r a i n stern.
DOR,~OM~D|AL ~I~/G|ON
°:°{
/
ti -i
III
A
~
/
4:,
~f i
q_
S
112
lO00pm
200~m
Fig. 6. T y p e s o f l a b e l l e d n e u r o n e s in t h e t h r e e d i f f e r e n t areas, A: v e n W o m e d i a l area; R : v e n t r o l a t e r a ] ; C: d o r s o m e d i a ~ r e g i o n .
three regions mentioned; a small number, however, (mean 20%) were difi~asely scattered throughout the brain stem. The concentration or dispersal of neurones thus identified within these regions, however, varied from experi-
;c
~mm 6
~mm
- - 1 9 mrrl
+ 4 , 9 mm
Fig. 7. D i s t r i b u t i o n o f l a b e l l e d n e u r o n e s a f t e r i n j e c t i o n o f t-IRP a t t h e I L C at T3 o n t h e left side. A - - D r e p r e s e n t 4 d i f f e r e n t e x p e r i m e n t s with the same amount of injected HRP and the same su~ival period.
\
: 7.0 mm
114 m e n t t o experiment (Fig. 7). For all experiments the distribution expressed as a percentage of the total number of neurones within file three regions was 50% in the dorsomedial, 28% in the ventrolateral, and 22% in the ventromedial region. Seventy-two hours after injection of HRP at L~ labelled neurones could be found in the three regions described above, but the total n u m b e r of labelled neurones per experiment was smaller than after injection at T3 (mean value 60 neurones) (Fig. 2). The distribution of neurones among the three regions was: dorsomedial region, 47%; ventrolateral region, 26%; ventromedial region, 27%. DISCUSSION The method of retrograde transport of HRP permitted identification of three brain stem areas which contain neurones with spinal descending projection to the ILC. This analysis contributes to our knowledge of autonomic: system control by at least identifying the pathways available for exerting excitatory and inhibitory tone and reflex control over preganglionic neurones. The method of retrograde labelling of neurones with HRP presents some general problems and limitations, therefore the interpretation of the results requires discussion. It has been demonstrated that retrograde axonal transport of HRP occurs if the HRP is taken up by nerve terminals or by injured axons [ 15,22]. HRP can ~,A~oenter axons en passant by diffusion but this HRP is not involved in the retrograde transport [15,22]. The neurones which have been identified in the brain stem in the present experiments by means of exogenous HRP therefore may have taken up the HRP through either injured ,axons or their nerve terminals. The injury of fibres could have been produced by the insertion of the injection cannula. This would have involved fibres which run through the dorsolateral funiculus and lateral parts of the dorsal horn. Therefore it must be considered possible that the population of bulbospina] neurones which has been demonstrated in these experiments might contain not only neurones which terminate within the ILC but also
some
,v,,,,.,,
~ , , , . , ~
in
~c
dorsal
H ~ U - -r- l l [~~ J'
. l? ~ .~AL'I,[I~L'[IIU[¢~ . . . . ~ . . . . . . . . L,I[1}.~5 ~ : . . . .~. o. D I ~ A D ' -. -
tion might include neurones which send axons through the dorsolateral funiculus and terminate in the ventral horn at the same or more caudal segmental level. However, experiments in which the insertion of the injection cannula was performed at an angle through the dor~aJ funiculus revealed the same findings as after vertical insertion making it likely that the presented observations are not due to uptake of HRP by damaged ~_xons in the injection canal but due to uptake by nerve ~rminals within the injection spot. In control experiments HRP could be demonstrated in a~ area 800--800 gm in diameter 2 h ~£ter injection of 30--50 nl in the region of the ILC. Within this area terminals of the ILC, the dorsal part of the ventral horn, and the dorsal horn up to the level of ~arnina IV could have taken up HRP for
115 retrograde transport. The area of diffusion identified represents the maxim u m since, as demonstrated by Vanegas et al. [25], terminals which come into contact with HRP after more than 2 h do not take up HRP or show retrograde transport. It is obvious, from the fact that no labelled neurones were found in the red nucleus and that less than 1% of all labelled neurones were detected in other m o t o r nuclei of the brain stem, that although HRP had reached parts of the ventral horn by diffusion after more than 2 h this HRP was n o t retrogradely transported. It is conceded that an u n k n o w n n u m b e r of neurones which do not terminate in the ILC might be included in these results but those neurones which do terminate in this ILC region must ori~,inate within the described medullary areas. Neurones in the nuclei of the solitary tract with spinal projections have also been demonstrated by Kuypers and Maisky [18] and Loewy and Burton [20]. Loewy and Burton have shown that the descending axons terminate in the ILC, the region of the phrenic m o t o r neurones, and the thoracic ventral horn [20]. Using the fluorescence m e t h o d DahlstrSm and Fuxe [12] were able to localize noradrenergic neurones in the commissural nucleus. This area was called A2 by these authors. DahlstrSm and Fuxe provided evidence that neurones of the area A2 project to the ILC. the m e t h o d of anterograde transport of labelled amino acids has been used to show that neurones in the magnocellular tegmental field project to the ILC [ 5]. This ~wea partially overlaps with the ventrolateral region which contained HRP-stained neurones in our experiments. In the caudal prolongation of this area Dahlstr5m and Fuxe [12] have localized noradrenergic neurones with spinal projection. This area was called A1. Tryptaminergic neurones in the caudal raphe nuclei with descending fibres to the spinal cord have been demonstrated by several authors [11,12]. In the same area neurones with spinal projection have been localized by means of anterograde and retrograde axonal transport [5,7,18]. In agreement with our results Basbaum et al. have shown that part of these raphe neurones terminate at the ILC [ 5]. rTtt_ l z~e present results are also in agreement with the localization of buibospinal neurones by other authors. The existence of additional polysynaptic pathways from brain stem to ILC could n o t be investigated with this method. Future electrophysiological studies of bulbospinal neurones within these regions will be necessary to clarify their properties and functions relative to the regulation of the autonomic nervous system activity. ACKNOWLEDGEMENTS These studies were supported by the German Research Foundation within the SFB 90 "Cardiovascul~ires System". The technical assistance of Miss Helga Schmidt-Prestin is gratefully acknowledged. Thanks are due to Dr. D. Richter and Dr. M. Spyer for c o m m e n t s and criticism during the preparation of the manuscript.
116
REFERENCES 1 Alexander, R,S., Tonic and reflex function of medullary sympathetic: cardiovascular centers, J. Neurophysiol. 9 (1946) 205--217. 2 Amendt, K., Czachurski, J., Dembowsky, K. and Seller, H., Localization of brainstemneurons projecting to the intermediolateral nucleus in the cat, Pfliigers Arch. ges. Physiol., 373 (1978) R76. 3 Amendt, K., Czachurski, J., Dembowsky, K. and Seller, H., Neurones within the "chemosensitive area" on the ventral surface of the brainstem which project to the h~termediolateral column, Pfliigers Arch. ges. Physiol., 375 (1978) 289--292. 4 Bach, L.M.N., Relationships between bulbar respiratory, vasomotor and somatic facilitatory and inhibitory areas, Amer. J. Physiol., 171 (1952) 417--435. 5 Basbaum, A.I., Clanton, C.H. and Fields, H.L., Three bulbospinal pathways from the rostra| medulla of the cat: an autoradiographic study of pain modulating systems, J. comp. Neuro|., 178 (1978) 209-224. 6 Berman, A.L., The Brainstem of the Cat; A Cytoarchitectonic Atlas with Stereotaxic Coordinates, Univ. of Wisconsin Press, Madison, Wisc., 1968. 7 Bobillier, P., Seguin, S., Petiqean: F., SalverL D., Touret, M. and Jouvet, M., The raphe nuclei ,~f the cat brain stem: a topographical atlas of their efferent projections as revealed by autoradiography, Brain Res., 113 (1976) 449--486. 8 Brodal, A., The Reticular Formation of the Brainstem. Anatomical Aspects and Functional Correlations, Oliver and Boyd, Edinburgh-London, 1975. 9 Burton, H. and Leewy, A.D. Projections to the spinal cord from medullary somatosensory relay nuclei, J. comp. Neurol., 173 (1977) 773--792. 10 Calaresu, F.R., Faiers, A.A. and Mogenson, G.J., Central neural regulation of heart and blood vessels in mammals, Progr. Neurobiol., 5 (1975) 1--35. 11 Coote, J.H. and Macleod, V.H., The influence of bulbospinal monoaminergic pathways on sympathetic nerve activity, J. Physiol. (Lond.), 241 (1974) 453--475. 12 Dahlstr6m, A. and Fuxe, K., Evidence for the exisCence of monoamine neurons in the central nervo.u.ssystem, Acta physiol, scand., 64, Suppl. 247 (1965). 13 Dittmar, C., Uber die Lage des sogenannten Gefgsscentrums in der Medulla oblongata. Bet. Verh. siichs. Ges. Wiss. Leipzig. Math.-Ph:,.s. Ct., 25 (1873) 449--469. 14 Guertzenstein, P.G. and Silver, A., Fall in b|ood pres.~ure prod~leed from discrete regions of the ventrai surface of medutla by gtycine and lesions, J. Physiol. (Lond.), 242 (1974) 489--503. 15 Hedreen, J.C. and McGrath, S°, Observations on labeling of neuronal eel] bodies, axons, and terminals after injection of horseradish peroxid~e into rat brain, J. comp. Neurol.~ 176 (1977} 225--248. 1~ Henry., J.L. and Ca|aresu~ F.R., Excitatory and inhibitory inputs from medn|tary nuclei projecting to spina| cardioacceleratory neurons in the cat, Exp. Brain Res°, 20 (1974} 485--504. !7 Kim, C.G. and Strick~ P.L., Critical factors involved in the demonstration of horseradish peroxidase retrograde t~ansport, Brain Res., 103 (!976) 356--36~. 18 K~lypem~ H.G.J.M. and Maisky, V.A., Retrogrede axonal transport of h~rseradish peroxidize from spinal cord to brain s~em cell groups in the cat, Neu~osci. Lett., 1 (1975) 9--14. 19 Lebedev, V.P a~d MikhaHova, G.P.~ Stereotaxic investigation of the :~ateral horn of the cat spinal cord, Fiziok Zh. (Leningrad} 57 (197! } 677--683. 20 Loewy. A D. and Burton~ H°~ Nuclei of ~he soii~.ary trac~: Effe~en~ p~oiec~ions ~o the lower br~in stem and ~p}~a| co~d of ~.he ca~, Jo temp. Neurol., 181 (!978} 421--450. 21 Mes~lam~ M.M., ~f~he btae ~eac~Aon produc~ ~n horseradish peroxidase ne~rohistoci~emi~zy: i~c.~ba~icm pa~amete~-s and vi,~ihiii~y, J. H~s~ochem. Cy~;ochem., 24 (1976) 1273--1280.
117 22 Oldfield, B.J. and McLachlan, E.M., Uptake and retrograde transport of H R P by axons of intact and damaged periphera~ nerve trunks, Neurosci. Lett., 6 (1977) 135-141, 23 Owsjannikow, Ph., Die tonischen und reflektorischen Centren der Gef~issnerven. Ber. Verb. s//chs.Ges. Wiss. Leipzig. Math.-Phys. CI., 23 (1871) 135--147. 24 Price, P., Fisher, A.W.F. and Redstone, P., A simple apparatus for the injection of small (nanolitre)volumes of horseradish peroxidase, Neuroscl. Lett., 6 (1977) 21--25. 25 Vanegas, H.,, Holl~inder, H. and Distel, H., Early stages of uptake and transport of horseradish-peroxidase by cortical structures, and its use for the study of local neurons and their processes, J. comp. Neurol., 177 (1978) 193--212.