Brain Research Bulletin, Vol. 1, pp. 443-451, 1976. Copyright 0 ANKHO All rights of reproduction in any form reserved. Printed in the U.S.A.
international
inc.
Horseradish Peroxidase Tracing of the Lateral Habenular-Midbrain Raphe Nuclei Connections in the Rat’ DANIEL A. PASQUIER, The Worcester
CLINT ANDERSON, Foundation
WILLIAM B. FORBES AND PETER J. MORGANE’
for Experimental
(Received
Biology, Shrewsbtq,
11 September
MA 01545
1976)
PASQUIER, D. A., C. ANDERSON, W. B. FORBES AND P. J. MORGANE. Horseradish peroxidase tracing of the lateral habenular-midbrain raphe nuclei connections in the rat. BRAIN RES. BULL. l(5) 443-451, 1976. - Connections of the habenular complex to the nuclei of the midline in the midbrain (interpeduncularis, medianus raphe, and dorsalis raphe) have been studied classically by anterograde degeneration in the monkey, the cat, and marsupials. Passing fibers from the medial septal nucleus and lateral preoptic area, however, have also been demonstrated which can complicate interpretation of these results. In this paper the habenular projections were studied in the rat by the retrograde axonal transport of horseradish peroxidase (HRP). After HRP injections in the medianus raphe nucleus labelied neurons appeared in the lateral habenular nucleus and parafascicular nucleus. Labelled neurons were also found in the lateral habenular nucleus after injections in either the dorsalis raphe nucleus or the caudal central gray substance. The habenular projections were always bilateral. There were no labelled neurons in the medial habenular nucleus after HRP injections in the medianus raphe nucleus, dorsalis raphe nucleus, or central gray. These data stress the lateral habenular influences upon the raphe nuclei, especially on the dorsalis raphe neurons which have usually been thought of as functionary related to other brainstem structures. The present results suggest also that in the rat the lateral habenular nucleus might be the link between basal forbrain inputs and the limbic midbrain area. Thus, the raphe nuclei of the midbrain appear to be crucial regions for integrating two descending circuits: first, a limbic (through septum) circuit, and, second, a basal forebrain (through lateral habenular-preoptic area) circuit.
Habenular nuclei Horseradish peroxidase Brainstem Dorsalis raphe afferents
Habenular-raphe relations
THE FIRST complete picture of the afferents to the dorsalis raphe nucleus (DR) have been carried out in the rat [22] by the retrograde axonal transport method using horseradish peroxidase (HRP). In that report it was suggested that a complex pattern of afferents carries inputs to the DR, especially from the lateral habenular nucleus and from various levels of the brainstem. By comparing these results with histofluorescence [ 3, 5, 251 and histochemical mappings [9, 10, 16, 17, 28, 291 it has been postulated also that the multiple inputs to DR might be mediated by various neurotransmitters. However, there is no method yet available to map histochemically many brain regions, especially in the forebrain. Moreover, in the brainstem the histofluorescence and histochemical methods in fact have only shown cholinergic, catecholaminergic and indolaminergic neuronal groups. Therefore, some of the brain regions projecting to the DR 1221 might be mediated by chemical circuitries utilizing other neurotransmitters. This might be especially the case of the dien~ephali~ descending projections to the DR. In our preliminary study numerous projections from the lateral habenular nucleus were found 1221. The role of the DR in several functional
Parafascicular nucleus
Sleep
phenomena,
such as sleep, thermoregulation, mood, and [6, 13, 141 emphasizes the interest in working out the connections of this nuclear formation. Projections from the habenular complex to the midbrain raphe have been demonstrated by silver techniques in the cat [l, 12, 15,30],rabbit [4] andmarsupials [30,31,32]. However, this connection has not received attention from the physiological point of view and its detailed morphology is not adequately known. For instance, the habenular efferents show many differences between species. It needs to be demonstrated which part of the habenular complex projects to the midbrain raphe and in which proportion they are distributed between the medianus and dorsaiis raphe nuclei. For example, degenerated fibers were described ambiguously in the “central gray caudal to the third cranial nerve”, in the “caudal central gray”, in the “entire extent of the central gray”, and “probably project to the dorsal tegmental nucleus of Gudden”. Finally, the elucidation of habenular-midbrain raphe connections need autoradiographic studies (injection of labelled substance in the habenular nuclei) or a retrograde labelling of the habenular neurons (injection of HRP in the midbrain raphe sexual
behavior
’ Supported by National Science Foundation Grant BNS 74-02620 (.P. J. M.) and by Cultural Agreement between Spain and U.S.A. (D. A. P.). *Reprint requests to: Dr. Peter J. Morgane, The Worcester Foundation for Experimental Biology, 222 Maple Avenue, Shrewsbury, MA 01545. 443
444
PASQUIER.
nuclei) because fibers of passage have been demonstrated to go through the habenular complex to the midbrain raphe. These passing fibers in the stria medullaris were traced from the entorhinal cortex [ 11, preoptic area [ 151, septum [4], and vertical limb of the diagonal band [ 26 ] In this paper we are reporting a conspicuous bilateral projection from the lateral habenular nucleus to the medianus and dorsalis raphe nuclei in the rat. This connection appears to reciprocate the direct neural circuit with the raphe nuclei (medianus and dorsalis) which might (complete) the raphe-lateral preoptic area integrate connection. METHOI)
Twenty albino rats (Charles River, 250-350 g weight) received a stereotaxic injection of 0.1 ~1, through a micropipette, of 30% HRP (Sigma type VI) solution in one of the following nuclei: dorsalis raphe or medianus raphe. Control injections were performed in the cerebellar cortex and oculomotor complex. After a survival period of 24 hr the animals were deeply anesthetized and perfused intracardially with a fixative solution (1% paraformaldehyde, 1% glutaradehyde in phosphate buffer pH = 7.4). The brains were removed immediately and cut in a freezing microtome at 40 p thick. The sections were processed for HRP reaction in the same day, as has been previously described [ 2 11. The reactions were studied under bright and dark field illumination.
1:‘7‘Al,.
RESULTS
Figure 1 shows the habenular complex and the symbols used in all subsequent figures. The control injections in the oculomotor complex did not show labelled neurons in the habenular complex but did in other brainstem regions as has been previously reported by us [22 I. After HRP injections in the cerebellar cortex labelled neurons appeared in the inferior olive and some reticular nuclear groups of the brainstem but none was found in the habenulal complex. HRP neurons in the lateral habenular nucleus were found after injections both in the dorsalis or medianus raphe nuclei. Also, labelled neurons appeared in the lateral habenular nucleus after injections place in the caudal central gray substance. Three animals will be described as examples of typical injections in these regions. ‘The rat RD 34 (Fig. 2) showed a large: HRP injection centered in the midbrain raphe area. There was no diffusion to the interpeduncular nucleus. Numerous labelled neurons were found in the lateral habenular nucleus (Fig. 2 a -e). They appear in the entire lateral habenular nucleus in such proportions that we were able to count 60 labelled neurons in one section 40 p thick. HRP neurons were not found in the medial habenular nucleus. Other labelled groups appeared surrounding Meynert‘s bundle (i’asciculus retrotlexus) and some among the fibers of this tract near to the habenular complex. The latter group of lahelled neurons seems to be placed primarily in the parafascicular nucleus (Fig. 7 a, b. c).
FIG. 1. Photograph of a frontal section of the normal rat brain showing the habenular complex. Abbreviations for all the figures: BC, brachium conjunctivum; CA, commissura anterior; CC, crus cerebri (cerebral peduncle): CI, colliculus inferior; CO, chiasma opticum; DR, n. dorsalis raphe; F, fornix; Fi, Fimbria; FMT, fasciculus mammillothalamicus; FR, fasciculus retroflexus (Meynert’s bundle); HaL, n. habenularis lateralis; HaM, n. habenularis medialis; Hi, hippocampus; MR, n. medianus raphe; Pf, n. parafascicularis; SM, stria medullaris; TCC, tractus corporis callosi; TO, tractus opticus; V3, ventriculus tertius; VII, genu septimi; SGV, substantia grisea ventralis; VL, ventriculus lateralis, Nissl stain.
HORSERADISW
PEROXIDASE
NEURONAL
TRACING
RD 34
a
FIG. 2. Diagrams of the rat RD 34. Sagittal sections. Riack area shows the center of the HRP injections (medianus rayhe); striped lines (in a, b, and c) around medianus raphe indicate diffusion area of the injection. Labelled neurons (asterisks) appear in the lateral habenuhr and parafasdcular nuclei. See abbreviations in Fig. 1.
445
446
FIG. 3. Phomgraph of the HKP injection in rat RD 29. E’rontai section, light Nissl stain. SW abbrevintions in Fig. 1.
b E‘IC. 4. Diagrams of the rat RD 29. Frontal sections. Labetied neurons {asterisks) appear bilaterally in the lateral ~abe~~~ar nuclens. See abbreviations in Fig. I.
44x
PASQIJIER,
I:‘?‘ .4L
FIG. 6. Photograph of the center of HRP injection in rat RD 47 Frontsf section. light Nissl stain. See ~~~re~~~t~~n~in Fig. 1. the brainstemt project to the dorsalis raphe nucleus [ZZ] , of these groups have not been related functionally to the dorsalis raphe nucleus. Nevertheless, the complexity of some functional states in which the dorsalis raphe seems to participate cannot be resolved at the brainstem level. For instance. in the sleep-waking cycle there occur many endocrine and metabolic changes which are not understood using madeis incorporating only lower brainstem structures. The cortical activity in this particular state is only one sign of what is happening during the cycle. Thus, the forebrain descending connections found aftzi HRP injections in the dorsalis raphe nucleus [X3], besides the habenular projections reported here, open the possibility of other interpretations of the function of the rnidb~a~n raphe. The dorsalis raphe afferents classically proposed are noradrenergic [ 11,271 . But, following the description of these afferents to the dorsalis raphe and surrounding central it appears that projections from serotonergio, gray, chohnergic and catecholaminergic brainstem groups are likely [?Z]. Very little information exists about the chemical pecularities of the neuronal groups at diencephalic and telencephalic regions [ 1O,lS] Therefore, the functional meaning of these babenular inputs to the dorsalis raphe remains as an open question. However, we can speculate that the functional effects of these habenular inputs on the dorsalis raphe shouid be conditioned by two main factors (1) the type of neurotransmitter used by these fibers and, (2) the chemical characteristic of the dorsalis raphe neurons on which the contacts are made, since not ail the neurons present in the dorsalis raphe are serotonergic in type. In this way, some topographical distribution of the neurans in the dorsalis raphe nucleus has been suggested in relation to the different areas of projection [ 19,21 1.
Yet, mast
The results reported here demonstrate a direct connection of .the lateral habenular nucleus with the medianus and dorsalis raphe nuclei in the rat. This habenular-midbrain raphe connection is restricted to projections from the lateral habenu~ar nucleus, as has been suggested in the cat using silver degeneration methods [ 15 f The llabenuiar projection is massive to both raphe nuclei. Taking in account that these nuclei are different in size, the habenular projection seems to be distributed almost proportionally. Evidence is also presented that these habenular efferents are not restricted to these raphe nuclei, but they reach the central gray areas bordering on the nucleus raphe dorsalis. The absence of labelled neurons in the medial habenuiar nucleus demonstrates indirectly that it projects only to the interpedu~cLllar nucleus and not to the other midline midbrain nuclei in the rat. These data also show that the lateral habenu~ar projection to the dorsalis raphe is bilateral and extremely abundant. Most labelled neurons were of medium size and were demonstrated everywhere in the nucleus. In spite of differences between injections these results agree with the opinion indicating that the number of labelled neurons is related to the place of the injection, i.e., depends on the proportion of terminal fibers in the injected region ]20]. The role played by the dorsalis raphe nucleus in different functional states has been questioned basically because very little is known about its afferent and efferent connections which are quite widespread, being distributed in the forebrain and brainstem regions. However, all these functional states may well need different structural subsets 1241. This speculation has been supported by the demonstration that diverse neuronal groups, at almost all levels of
HORSERADISH
PEROXIDASE
NEWRUNAL TRACING
RD 47
d
FIG. 7. Diagrams of frontal sections in rat RD 47. Labelled neurons (asterisks) were found bilaterally in the lateral habcnular nucleus. In ““d” some labelled neurons appear in the lateral border of the medial habenular nucleus. See abbreviations in Fig. I.
Thus, the raphe nuclei of the midbrain appear as a crucial region in which Ximbic inputs (seytal) complete a septal -- medial habenuiar - interpeduncular - raphc nuclei link with those inputs from the basal forebrain (lateral preoptic area> completing a preoptic - lateral habenular raphe nuclei circuit. In turn the medial and dorsalis raphe nuclei reciprocate the lateral habenuiar projection as well as those from the later-a1 preoptic and medial septal regions.
Thus, these results demonstrate strong forebrain influences an the midbrain raphe nuclei through the “dorsal projection system” (with a relay in the habenular complex)
in addition
tu the classically described “ventral”
pathway.
ACKNQWLEDGEMENT
The authors thank Drs. T. L. Kemper and V. Caveness for use of their special morpbolog~~l facilities and Mr, Morris Feinstein for his technical histoIogica1 help.
FIG. 8. M~cropbotogr~ph (montage) of the habenular complex in rat RD 47. Labelled neurons (arrows, dark field) appear in the lateral habenular nuckus. Originat magnification: 100 x. See abbreviations in Fig. 1.
REFERENCES 1,
2. 3.
4. 5.
Adey, W. R., N. C. R. Merrillees and S. Sun~~r~nd. The entorhinal area: behavioral, evoked potential and histologica studies of its interrelationships with brainstem regions. Brain 79: 4X4-439, L956. Akad, K. and E. W. Pow&. D~fere~t~a~ projections of hab&&ar nuclei. J. camp. Nellrod: 132: X3-274. j968. Bi&klund. A.. B. Faick and U. Stenevi. Classification of monoam&e nehrons in the rat mesencephalon: distribution of a new monoamine neuron system. Brain Rex 32: 269-285, 1971. Cragg, B. G. The connections of the habenuIa in the rabbit, f?xpxpE Neurof. 3: 388-409,196 I. Dahtstram, A. and K. Fuxe. Evidence for the existence of rno~oarn~e~ontai~~ neurons in the central nervous system. 1. Demonstration of monoamines in the cell bodies of brain stem neurons. Acta physioL stand. 62: Suppl. 232: l-80, 1964.
Fuxe, K., T. IItikfelt and U. Ungerstedt, MorphologicaT and functional aspects of central mani~am~e neurons. Irrt. RPE:. Rettrohiob. 13: 93-126, 1970. Carver, D. L. and J. R. Sladek. Mono~m~e dis~r~butjon in primate brain. f. Cate~holamine containing perikarya in the brainstem of Macaca speciosa. J. conap. Neuwl. 1.59: 289-304, 1975. Herkenham, M. A. Afferent connections of the h~be~ul~r nuclei in the rat. In: Proc. 5th Annual Meeting Society fiv Neuroscience, New York, p. 546, 191.5. 4. HLikfelt, ‘I’., K. Fuxe and M. Goldstein. ~mm~noh~stocbemica~ lo~al~ation of aromatic L-aminoacid decarboxylase (DOPA decarboxylase) in central dopamine and S-hydroxytry~tam~e nerve cell &dies of the rat. Brain. lies. 53: 175-180, 1373.
HOSRERADISH
PEROXIDASE
NEURONAL
10. Jacobowitz,
11. 12. 13.
14.
15.
16.
17.
18.
4.51
TRACING
D. M. and M. Palkovits. Topographical atlas of catecholamine and acetylcholinesterase-containing neurons in the rat brain. I. Forebrain (telencephalon, diencephalon). J. camp. Neural. 157: 13-28, 1974. Loizou, L. A. Projections of the nucleus locus coeruleus in the albino rat. Bruin Rex 15: 563-566, 1969. Mitchell, R. Connections of the habenula and of the interpeduncular nucleus in the cat. J. camp. Neurol. 121: 441-457, 1963. Morgane, P. J. and W. C. Stern. Relationship of sleep to ~euroanatomical circuits, biochemistry, and behavior. Ann. N. Y. Acud. Sci. 193: 95-111, 1972. Moruzzi, G. and M. Jouvet. Neurophysiology and neurochemistry of sleep and wakefulness. Ergeb. der Physiof. 64: l-342,1972. Nauta, W. J. H. Hippocampal projections and related neural pathways to the mid-brain in the cat. Bruin 81: 319-340, 19.58. Palkovits, M. and Jacobowitz, D. M. Topographic atlas of catecholamine and acetylcholinesterase-containing neurons in the rat brain. II. Hindbrain (mesencephalon, rhombencephalon). J. camp. Neural. 157: 29-42, 1974. Palkovits, M., M. Brownstein and J. M. Saavedra. Serotonin content of the brain stem nuclei in the rat. Brain Rex 80: 23?-249,1974. Parent, A. and L. L. Butcher. Acetylcholinesterase-con~in~g neurons in the dien~phalon of the rat. Anat. Rec. 184: 496, 1976.
19. Pasquier, D. A. A topometric distribution in the hippocampus of projections from the raphe brainstem and locus coeruleus in the cat. In: Proc. 10th Int. Gong. Anat. edited by E. Yamada, Tokyo: Science Council of Japan, p. 139, 1975. 20. Pasquier, D. A. and F. Reinoso-Sulirez. Direct projections from hypothalamus to hippocampus in the rat demonstrated by retrograde transport of horseradish peroxidase. Brain Res. 108: 16S-169,1976. 21. Pasquier, D. A. and F. Reinoso-Suirez. Differential efferent connections of the brainstem to the hippocampus in the cat. Brain Res. in press, 1976.
22. Pasquier, D. A., W. B. Forbes and P. J. Morgane. Afferent connections of the nucleus raphe dorsalis and the oculomotor complex in the rat. In: Proc. 6th Annual Meeting Society for Neuroscience, Toronto, Vol. 2, p. 470, 1976. 23. Pasquier, D. A., W. B. Forbes and P. J. Morgane. Afferent connections of the nucleus raphe dorsalis. A retrograde axonal transport study in the rat, in preparation. 24. Pasquier, D. A., F. Reinoso-Suirez and P. J. Moraane. Effect of raphe lesions on brain serotonin in the cat. Brai;? Rex. Bull, 1: 279-283.1976. 2s. Pin, C., B. Jones and M. Jouvet. Topographie des neurones monoaminergiques du tronc cerebral du chat: etude par histoflurescence. C. r. Sot. Biol. (Par&] 162: 2136-2140, 1968.
26. Powell, E. W. Septohabenular connections in the rat, cat and monkey. J. camp. Neural. 134: 145-150,1968. 27. Roizen, M. F. and D. M. Jacobowitz. Studies of the origin of innervation of the noradrenergic area bordering on the nucleus raphe dorsalis. Bruin Rex 101: 561-568, 1975. 28. Saavedra, J. M., M. Brownstein and M. Palkovits. Serotonin content of the limbic system in rat. Bruin Rex 79: 437-441, 1974. 29. Saavedra, J. M., M. Palkovits, M. Brownstein and 3. Axelrod. Serotonin distribution in the nuclei of the rat hypothalamus and preoptic region. Brain Res. 77: 157-165, 1974. 30. Smaha, L. A. and W. W. Kaelber. Efferent fiber projections of the habenula and the interpeduncular nucleus. An experimental study in the opposum and cat. Ejrpt Brain Res. 16: 291-308,1973. 31. Way, J. S. A degeneration study of some habenular efferents to the midbrain in a Wallaby. Am. J. Anat. 142: l-14,1975. 32. Way, J. S. and W. W. Kaelber. A degeneration study of efferent connections of the habenular complex in the opposum. Am. J. Anat. 124: 31-46.1969.
international
inc.
Horseradish Peroxidase Tracing of the Lateral Habenular-Midbrain Raphe Nuclei Connections in the Rat’ DANIEL A. PASQUIER, The Worcester
CLINT ANDERSON, Foundation
WILLIAM B. FORBES AND PETER J. MORGANE’
for Experimental
(Received
Biology, Shrewsbtq,
11 September
MA 01545
1976)
PASQUIER, D. A., C. ANDERSON, W. B. FORBES AND P. J. MORGANE. Horseradish peroxidase tracing of the lateral habenular-midbrain raphe nuclei connections in the rat. BRAIN RES. BULL. l(5) 443-451, 1976. - Connections of the habenular complex to the nuclei of the midline in the midbrain (interpeduncularis, medianus raphe, and dorsalis raphe) have been studied classically by anterograde degeneration in the monkey, the cat, and marsupials. Passing fibers from the medial septal nucleus and lateral preoptic area, however, have also been demonstrated which can complicate interpretation of these results. In this paper the habenular projections were studied in the rat by the retrograde axonal transport of horseradish peroxidase (HRP). After HRP injections in the medianus raphe nucleus labelied neurons appeared in the lateral habenular nucleus and parafascicular nucleus. Labelled neurons were also found in the lateral habenular nucleus after injections in either the dorsalis raphe nucleus or the caudal central gray substance. The habenular projections were always bilateral. There were no labelled neurons in the medial habenular nucleus after HRP injections in the medianus raphe nucleus, dorsalis raphe nucleus, or central gray. These data stress the lateral habenular influences upon the raphe nuclei, especially on the dorsalis raphe neurons which have usually been thought of as functionary related to other brainstem structures. The present results suggest also that in the rat the lateral habenular nucleus might be the link between basal forbrain inputs and the limbic midbrain area. Thus, the raphe nuclei of the midbrain appear to be crucial regions for integrating two descending circuits: first, a limbic (through septum) circuit, and, second, a basal forebrain (through lateral habenular-preoptic area) circuit.
Habenular nuclei Horseradish peroxidase Brainstem Dorsalis raphe afferents
Habenular-raphe relations
THE FIRST complete picture of the afferents to the dorsalis raphe nucleus (DR) have been carried out in the rat [22] by the retrograde axonal transport method using horseradish peroxidase (HRP). In that report it was suggested that a complex pattern of afferents carries inputs to the DR, especially from the lateral habenular nucleus and from various levels of the brainstem. By comparing these results with histofluorescence [ 3, 5, 251 and histochemical mappings [9, 10, 16, 17, 28, 291 it has been postulated also that the multiple inputs to DR might be mediated by various neurotransmitters. However, there is no method yet available to map histochemically many brain regions, especially in the forebrain. Moreover, in the brainstem the histofluorescence and histochemical methods in fact have only shown cholinergic, catecholaminergic and indolaminergic neuronal groups. Therefore, some of the brain regions projecting to the DR 1221 might be mediated by chemical circuitries utilizing other neurotransmitters. This might be especially the case of the dien~ephali~ descending projections to the DR. In our preliminary study numerous projections from the lateral habenular nucleus were found 1221. The role of the DR in several functional
Parafascicular nucleus
Sleep
phenomena,
such as sleep, thermoregulation, mood, and [6, 13, 141 emphasizes the interest in working out the connections of this nuclear formation. Projections from the habenular complex to the midbrain raphe have been demonstrated by silver techniques in the cat [l, 12, 15,30],rabbit [4] andmarsupials [30,31,32]. However, this connection has not received attention from the physiological point of view and its detailed morphology is not adequately known. For instance, the habenular efferents show many differences between species. It needs to be demonstrated which part of the habenular complex projects to the midbrain raphe and in which proportion they are distributed between the medianus and dorsaiis raphe nuclei. For example, degenerated fibers were described ambiguously in the “central gray caudal to the third cranial nerve”, in the “caudal central gray”, in the “entire extent of the central gray”, and “probably project to the dorsal tegmental nucleus of Gudden”. Finally, the elucidation of habenular-midbrain raphe connections need autoradiographic studies (injection of labelled substance in the habenular nuclei) or a retrograde labelling of the habenular neurons (injection of HRP in the midbrain raphe sexual
behavior
’ Supported by National Science Foundation Grant BNS 74-02620 (.P. J. M.) and by Cultural Agreement between Spain and U.S.A. (D. A. P.). *Reprint requests to: Dr. Peter J. Morgane, The Worcester Foundation for Experimental Biology, 222 Maple Avenue, Shrewsbury, MA 01545. 443
444
PASQUIER.
nuclei) because fibers of passage have been demonstrated to go through the habenular complex to the midbrain raphe. These passing fibers in the stria medullaris were traced from the entorhinal cortex [ 11, preoptic area [ 151, septum [4], and vertical limb of the diagonal band [ 26 ] In this paper we are reporting a conspicuous bilateral projection from the lateral habenular nucleus to the medianus and dorsalis raphe nuclei in the rat. This connection appears to reciprocate the direct neural circuit with the raphe nuclei (medianus and dorsalis) which might (complete) the raphe-lateral preoptic area integrate connection. METHOI)
Twenty albino rats (Charles River, 250-350 g weight) received a stereotaxic injection of 0.1 ~1, through a micropipette, of 30% HRP (Sigma type VI) solution in one of the following nuclei: dorsalis raphe or medianus raphe. Control injections were performed in the cerebellar cortex and oculomotor complex. After a survival period of 24 hr the animals were deeply anesthetized and perfused intracardially with a fixative solution (1% paraformaldehyde, 1% glutaradehyde in phosphate buffer pH = 7.4). The brains were removed immediately and cut in a freezing microtome at 40 p thick. The sections were processed for HRP reaction in the same day, as has been previously described [ 2 11. The reactions were studied under bright and dark field illumination.
1:‘7‘Al,.
RESULTS
Figure 1 shows the habenular complex and the symbols used in all subsequent figures. The control injections in the oculomotor complex did not show labelled neurons in the habenular complex but did in other brainstem regions as has been previously reported by us [22 I. After HRP injections in the cerebellar cortex labelled neurons appeared in the inferior olive and some reticular nuclear groups of the brainstem but none was found in the habenulal complex. HRP neurons in the lateral habenular nucleus were found after injections both in the dorsalis or medianus raphe nuclei. Also, labelled neurons appeared in the lateral habenular nucleus after injections place in the caudal central gray substance. Three animals will be described as examples of typical injections in these regions. ‘The rat RD 34 (Fig. 2) showed a large: HRP injection centered in the midbrain raphe area. There was no diffusion to the interpeduncular nucleus. Numerous labelled neurons were found in the lateral habenular nucleus (Fig. 2 a -e). They appear in the entire lateral habenular nucleus in such proportions that we were able to count 60 labelled neurons in one section 40 p thick. HRP neurons were not found in the medial habenular nucleus. Other labelled groups appeared surrounding Meynert‘s bundle (i’asciculus retrotlexus) and some among the fibers of this tract near to the habenular complex. The latter group of lahelled neurons seems to be placed primarily in the parafascicular nucleus (Fig. 7 a, b. c).
FIG. 1. Photograph of a frontal section of the normal rat brain showing the habenular complex. Abbreviations for all the figures: BC, brachium conjunctivum; CA, commissura anterior; CC, crus cerebri (cerebral peduncle): CI, colliculus inferior; CO, chiasma opticum; DR, n. dorsalis raphe; F, fornix; Fi, Fimbria; FMT, fasciculus mammillothalamicus; FR, fasciculus retroflexus (Meynert’s bundle); HaL, n. habenularis lateralis; HaM, n. habenularis medialis; Hi, hippocampus; MR, n. medianus raphe; Pf, n. parafascicularis; SM, stria medullaris; TCC, tractus corporis callosi; TO, tractus opticus; V3, ventriculus tertius; VII, genu septimi; SGV, substantia grisea ventralis; VL, ventriculus lateralis, Nissl stain.
HORSERADISW
PEROXIDASE
NEURONAL
TRACING
RD 34
a
FIG. 2. Diagrams of the rat RD 34. Sagittal sections. Riack area shows the center of the HRP injections (medianus rayhe); striped lines (in a, b, and c) around medianus raphe indicate diffusion area of the injection. Labelled neurons (asterisks) appear in the lateral habenuhr and parafasdcular nuclei. See abbreviations in Fig. 1.
445
446
FIG. 3. Phomgraph of the HKP injection in rat RD 29. E’rontai section, light Nissl stain. SW abbrevintions in Fig. 1.
b E‘IC. 4. Diagrams of the rat RD 29. Frontal sections. Labetied neurons {asterisks) appear bilaterally in the lateral ~abe~~~ar nuclens. See abbreviations in Fig. I.
44x
PASQIJIER,
I:‘?‘ .4L
FIG. 6. Photograph of the center of HRP injection in rat RD 47 Frontsf section. light Nissl stain. See ~~~re~~~t~~n~in Fig. 1. the brainstemt project to the dorsalis raphe nucleus [ZZ] , of these groups have not been related functionally to the dorsalis raphe nucleus. Nevertheless, the complexity of some functional states in which the dorsalis raphe seems to participate cannot be resolved at the brainstem level. For instance. in the sleep-waking cycle there occur many endocrine and metabolic changes which are not understood using madeis incorporating only lower brainstem structures. The cortical activity in this particular state is only one sign of what is happening during the cycle. Thus, the forebrain descending connections found aftzi HRP injections in the dorsalis raphe nucleus [X3], besides the habenular projections reported here, open the possibility of other interpretations of the function of the rnidb~a~n raphe. The dorsalis raphe afferents classically proposed are noradrenergic [ 11,271 . But, following the description of these afferents to the dorsalis raphe and surrounding central it appears that projections from serotonergio, gray, chohnergic and catecholaminergic brainstem groups are likely [?Z]. Very little information exists about the chemical pecularities of the neuronal groups at diencephalic and telencephalic regions [ 1O,lS] Therefore, the functional meaning of these babenular inputs to the dorsalis raphe remains as an open question. However, we can speculate that the functional effects of these habenular inputs on the dorsalis raphe shouid be conditioned by two main factors (1) the type of neurotransmitter used by these fibers and, (2) the chemical characteristic of the dorsalis raphe neurons on which the contacts are made, since not ail the neurons present in the dorsalis raphe are serotonergic in type. In this way, some topographical distribution of the neurans in the dorsalis raphe nucleus has been suggested in relation to the different areas of projection [ 19,21 1.
Yet, mast
The results reported here demonstrate a direct connection of .the lateral habenular nucleus with the medianus and dorsalis raphe nuclei in the rat. This habenular-midbrain raphe connection is restricted to projections from the lateral habenu~ar nucleus, as has been suggested in the cat using silver degeneration methods [ 15 f The llabenuiar projection is massive to both raphe nuclei. Taking in account that these nuclei are different in size, the habenular projection seems to be distributed almost proportionally. Evidence is also presented that these habenular efferents are not restricted to these raphe nuclei, but they reach the central gray areas bordering on the nucleus raphe dorsalis. The absence of labelled neurons in the medial habenuiar nucleus demonstrates indirectly that it projects only to the interpedu~cLllar nucleus and not to the other midline midbrain nuclei in the rat. These data also show that the lateral habenu~ar projection to the dorsalis raphe is bilateral and extremely abundant. Most labelled neurons were of medium size and were demonstrated everywhere in the nucleus. In spite of differences between injections these results agree with the opinion indicating that the number of labelled neurons is related to the place of the injection, i.e., depends on the proportion of terminal fibers in the injected region ]20]. The role played by the dorsalis raphe nucleus in different functional states has been questioned basically because very little is known about its afferent and efferent connections which are quite widespread, being distributed in the forebrain and brainstem regions. However, all these functional states may well need different structural subsets 1241. This speculation has been supported by the demonstration that diverse neuronal groups, at almost all levels of
HORSERADISH
PEROXIDASE
NEWRUNAL TRACING
RD 47
d
FIG. 7. Diagrams of frontal sections in rat RD 47. Labelled neurons (asterisks) were found bilaterally in the lateral habcnular nucleus. In ““d” some labelled neurons appear in the lateral border of the medial habenular nucleus. See abbreviations in Fig. I.
Thus, the raphe nuclei of the midbrain appear as a crucial region in which Ximbic inputs (seytal) complete a septal -- medial habenuiar - interpeduncular - raphc nuclei link with those inputs from the basal forebrain (lateral preoptic area> completing a preoptic - lateral habenular raphe nuclei circuit. In turn the medial and dorsalis raphe nuclei reciprocate the lateral habenuiar projection as well as those from the later-a1 preoptic and medial septal regions.
Thus, these results demonstrate strong forebrain influences an the midbrain raphe nuclei through the “dorsal projection system” (with a relay in the habenular complex)
in addition
tu the classically described “ventral”
pathway.
ACKNQWLEDGEMENT
The authors thank Drs. T. L. Kemper and V. Caveness for use of their special morpbolog~~l facilities and Mr, Morris Feinstein for his technical histoIogica1 help.
FIG. 8. M~cropbotogr~ph (montage) of the habenular complex in rat RD 47. Labelled neurons (arrows, dark field) appear in the lateral habenular nuckus. Originat magnification: 100 x. See abbreviations in Fig. 1.
REFERENCES 1,
2. 3.
4. 5.
Adey, W. R., N. C. R. Merrillees and S. Sun~~r~nd. The entorhinal area: behavioral, evoked potential and histologica studies of its interrelationships with brainstem regions. Brain 79: 4X4-439, L956. Akad, K. and E. W. Pow&. D~fere~t~a~ projections of hab&&ar nuclei. J. camp. Nellrod: 132: X3-274. j968. Bi&klund. A.. B. Faick and U. Stenevi. Classification of monoam&e nehrons in the rat mesencephalon: distribution of a new monoamine neuron system. Brain Rex 32: 269-285, 1971. Cragg, B. G. The connections of the habenuIa in the rabbit, f?xpxpE Neurof. 3: 388-409,196 I. Dahtstram, A. and K. Fuxe. Evidence for the existence of rno~oarn~e~ontai~~ neurons in the central nervous system. 1. Demonstration of monoamines in the cell bodies of brain stem neurons. Acta physioL stand. 62: Suppl. 232: l-80, 1964.
Fuxe, K., T. IItikfelt and U. Ungerstedt, MorphologicaT and functional aspects of central mani~am~e neurons. Irrt. RPE:. Rettrohiob. 13: 93-126, 1970. Carver, D. L. and J. R. Sladek. Mono~m~e dis~r~butjon in primate brain. f. Cate~holamine containing perikarya in the brainstem of Macaca speciosa. J. conap. Neuwl. 1.59: 289-304, 1975. Herkenham, M. A. Afferent connections of the h~be~ul~r nuclei in the rat. In: Proc. 5th Annual Meeting Society fiv Neuroscience, New York, p. 546, 191.5. 4. HLikfelt, ‘I’., K. Fuxe and M. Goldstein. ~mm~noh~stocbemica~ lo~al~ation of aromatic L-aminoacid decarboxylase (DOPA decarboxylase) in central dopamine and S-hydroxytry~tam~e nerve cell &dies of the rat. Brain. lies. 53: 175-180, 1373.
HOSRERADISH
PEROXIDASE
NEURONAL
10. Jacobowitz,
11. 12. 13.
14.
15.
16.
17.
18.
4.51
TRACING
D. M. and M. Palkovits. Topographical atlas of catecholamine and acetylcholinesterase-containing neurons in the rat brain. I. Forebrain (telencephalon, diencephalon). J. camp. Neural. 157: 13-28, 1974. Loizou, L. A. Projections of the nucleus locus coeruleus in the albino rat. Bruin Rex 15: 563-566, 1969. Mitchell, R. Connections of the habenula and of the interpeduncular nucleus in the cat. J. camp. Neurol. 121: 441-457, 1963. Morgane, P. J. and W. C. Stern. Relationship of sleep to ~euroanatomical circuits, biochemistry, and behavior. Ann. N. Y. Acud. Sci. 193: 95-111, 1972. Moruzzi, G. and M. Jouvet. Neurophysiology and neurochemistry of sleep and wakefulness. Ergeb. der Physiof. 64: l-342,1972. Nauta, W. J. H. Hippocampal projections and related neural pathways to the mid-brain in the cat. Bruin 81: 319-340, 19.58. Palkovits, M. and Jacobowitz, D. M. Topographic atlas of catecholamine and acetylcholinesterase-containing neurons in the rat brain. II. Hindbrain (mesencephalon, rhombencephalon). J. camp. Neural. 157: 29-42, 1974. Palkovits, M., M. Brownstein and J. M. Saavedra. Serotonin content of the brain stem nuclei in the rat. Brain Rex 80: 23?-249,1974. Parent, A. and L. L. Butcher. Acetylcholinesterase-con~in~g neurons in the dien~phalon of the rat. Anat. Rec. 184: 496, 1976.
19. Pasquier, D. A. A topometric distribution in the hippocampus of projections from the raphe brainstem and locus coeruleus in the cat. In: Proc. 10th Int. Gong. Anat. edited by E. Yamada, Tokyo: Science Council of Japan, p. 139, 1975. 20. Pasquier, D. A. and F. Reinoso-Sulirez. Direct projections from hypothalamus to hippocampus in the rat demonstrated by retrograde transport of horseradish peroxidase. Brain Res. 108: 16S-169,1976. 21. Pasquier, D. A. and F. Reinoso-Suirez. Differential efferent connections of the brainstem to the hippocampus in the cat. Brain Res. in press, 1976.
22. Pasquier, D. A., W. B. Forbes and P. J. Morgane. Afferent connections of the nucleus raphe dorsalis and the oculomotor complex in the rat. In: Proc. 6th Annual Meeting Society for Neuroscience, Toronto, Vol. 2, p. 470, 1976. 23. Pasquier, D. A., W. B. Forbes and P. J. Morgane. Afferent connections of the nucleus raphe dorsalis. A retrograde axonal transport study in the rat, in preparation. 24. Pasquier, D. A., F. Reinoso-Suirez and P. J. Moraane. Effect of raphe lesions on brain serotonin in the cat. Brai;? Rex. Bull, 1: 279-283.1976. 2s. Pin, C., B. Jones and M. Jouvet. Topographie des neurones monoaminergiques du tronc cerebral du chat: etude par histoflurescence. C. r. Sot. Biol. (Par&] 162: 2136-2140, 1968.
26. Powell, E. W. Septohabenular connections in the rat, cat and monkey. J. camp. Neural. 134: 145-150,1968. 27. Roizen, M. F. and D. M. Jacobowitz. Studies of the origin of innervation of the noradrenergic area bordering on the nucleus raphe dorsalis. Bruin Rex 101: 561-568, 1975. 28. Saavedra, J. M., M. Brownstein and M. Palkovits. Serotonin content of the limbic system in rat. Bruin Rex 79: 437-441, 1974. 29. Saavedra, J. M., M. Palkovits, M. Brownstein and 3. Axelrod. Serotonin distribution in the nuclei of the rat hypothalamus and preoptic region. Brain Res. 77: 157-165, 1974. 30. Smaha, L. A. and W. W. Kaelber. Efferent fiber projections of the habenula and the interpeduncular nucleus. An experimental study in the opposum and cat. Ejrpt Brain Res. 16: 291-308,1973. 31. Way, J. S. A degeneration study of some habenular efferents to the midbrain in a Wallaby. Am. J. Anat. 142: l-14,1975. 32. Way, J. S. and W. W. Kaelber. A degeneration study of efferent connections of the habenular complex in the opposum. Am. J. Anat. 124: 31-46.1969.
