- Email: [email protected]
Substance P containing and cholinergic projections from the habenula
Brain Research, 149 (1978) 413 429
413
© Elsevier/North-Holland Biomedical Press
SUBSTANCE P C O N T A I N I N G A N D C H O L I N E R G I C PROJECTIONS F R O M THE H A B E N U L A
A. CLAUD[O CUELLO*, PIERS C. EMSON, GEORGE PAXINOSt and THOMAS JESSELL~¢ M RC Neurochemical Pharmacology Unit, Dept. of Pharmacology, Medical School Cambridge (Great Britain )
(Accepted October 27th, 1977)
SUMMARY Electrolytic lesions and surgical transection of the habenulo-interpeduncularventrotegmental tract have established the existence of separate habenulo-interpeduncular-ventrotegmental substance P and cholinergic projections. Micro-knife lesions separating the habenula nuclei showed the medial habenular nucleus to be the source of substance P fibres running via the fasciculus retroflexus to the ventral tegmental area. The lateral habenular nucleus receives a substance P projection from the medial habenular nucleus and is the source of cholinergic projection to the interpeduncular nucleus and to the medial habenular nucleus. Lesions of the ventrotegmental-interpeduncular area did not modify the levels of substance P and choline acetyltransferase in the habenula. These observations suggested that there are no substance P or ACh containing afferents to the habenula from the ventrotegmental-interpeduncular area and the accumulation of substance P and AChE proximal to but not caudal to transections of the fasciculus retroflexus confirmed this view.
INTRODUCTION The identification of substance p4,5 as an undecapeptide was accomplished many years after the first report by von Euler and Gaddum 16 in 1931 of the existence of a vasoactive peptide in the brain and gut. The availability of pure substance P has now allowed the development of the appropriate immunochemical techniques which have clarified the role of substance P in the mammalian nervous system. Thus,
* To who correspondence and reprint requests should be sent. Present addresses: ¢ School of Psychology,The University of New South Wales, P.O. Box B, Kensington, New South Wales, Australia 2033. t¢ Dept. of Pharmacology, Harvard Medical School, 25 Shattuck Street, Boston, Mass. 02115, U.S.A.
414 immunoreactive substance P has been detected in the cell bodies and nerve terminals in specific regions of the rat, cat and human nervous systemll,12,1a,34. Sensitive radioimmunoassay (RIA) methods have demonstrated that substance P is unevenly distributed in the mouse and rat central nervous system1, ~z and a calcium-dependent release of this peptide has been demonstrated in vitro from the spinal cord 3~, hypothalamus 22, substantia nigra ~t and from synaptosomal preparations 39. More recently a vesicular localization of substance P has been demonstratedl0, zs. These observations have led to the suggestion that this peptide is a neurotransmitter (or modulator) in a number of well defined neuroanatomical pathways (see reviewsa,28.:~:'). There have been reports claiming a joint substance P/cholinergic projection from the habenula nuclei to the interpeduncular nucleus area~a,20,z3 although the precise sites of origin and termination of these projections remained uncertain. In some of these reports20, 33 the possible coexistence of substance P and acetylcholine in the same neurones has been discussed. The present investigation which employed discrete lesions to the habenula nuclei and pathway, allows us to dissociatc the two systems. MATERIALS AND METHODS
Microdissection Under the stereomicroscope, discrete nuclear areas were dissected at 4 °C from 0.6-1.0 mm thick coronal sections of fresh, unperfused, unfrozen rat brain. Coronal sections of the brain were prepared by hand using a series of parallel razor blades with a fixed inter-blade distance.
Radioimmunoassay of substance P The tissue pieces obtained were frozen on dry ice and weighed. Substance P was extracted initially with 50/~1 acetone/1 N HC1 (100:3, v/v) and then with 50/~1 acetone/0.01 N HCI (80:20, v/v) according to the method described by Chang and Leeman4, with the exception that the filtration procedure was replaced by centrifugation. Combined supernatant fractions were washed three times with pretroleum ether and lyophilized substance P was measured by radioimmunoassay using a method previously described by Powell et al. 4°. The specificity of the substance P-antisera prepared in guinea pig was checked with the related peptide eledoisin, and with somatostatin, met-enkephalin and fl-endorphin, all of which failed to cross-react with substance P when present in a 1,000 fold molar excess. The 9 and 10 amino acid fragments of the C-terminal sequence of substance P were found to possess immunoreactivity equal to that of the complete peptide. The sensitivity of the RIA allows accurate measurements of as little of 8 fmol substance P using antiserum dilutions in the range of 1:50,000-1:250,000. The post-mortem stability of substance P in brain tissue has been discussed in a previous publication23.
Histochemistry Experimental and control rats were anaesthetized with equithesin and perfused intra-
415 cardially with cold 4 ~ paraformaldehyde dissolved in 0.1 M phosphate buffer pH 7.4. Brains were removed and kept in the same fixative solution for 2-4 h. The fixed samples were transferred to a cold (4 °C) 0. l M phosphate buffer solution containing 5 % sucrose for up to 3 days. Thick sections (10 ttm) were cut on a Brights cryostat at 20 °C. Alternate sections were processed by immunocytochemistry, acetylcholinesterase or Nissl staining. Immunocytochemical investigation of substance P was performed by applying an indirect immunofluorescent method 6. Firstly antiserum (anti-substance P) raised in the guinea pig was applied in dilutions ranging from l :100 up to 1:450 (v/v). The second antiserum, anti-guinea pig F | T C conjugated lgG (Wellcome Foundation, Great Britain) was used in a 1:6 (v/v) dilution. Controls were performed using non-immune guinea pig serum or substance P-antiserum absorbed with 100 or 200 #g/ml of substance P. The tissue preparations were observed in a Leitz S M - L U X microscope equipped with epifluorescence optics. Acetylcholinesterase activity was demonstrated as described by Koelle and Friedenwald 25.
Biochemical determinations Choline acetyltransferase activity (CHAT) was determined as described by Fonnum tT. Protein concentrations were determined according to the method of Lowry et al. 3°.
Operative procedures Male 300 g Wistar rats were anaesthetized with equithesin (Jensen-Salsberg Labs. 4.0 ml/kg) and placed in a K o p f stereotaxic frame. Stereotaxic coordinates for radio frequency lesions were based on the stereotaxic atlas of Pellegrino and Cushman 3s. In the habenula nuclei (lateral 0.7 mm, posterior --2.0 mm and vertical 5.0 mm using the bregma as rostrocaudal zero) and ventral tegmental area (lateral 0.0 ram, posterior - - 3 . 0 m m and vertical 9.0 mm) lesions were made using a Radionic radio frequency lesion maker (model RGF-4). The temperature at the electrode tip was 55 °C during habenular lesions and 60 °C for ventral tegmental area lesions and were maintained for one minute. For micro-knife lesions the atlas served only as a guide and a series of rats were lesioned at different coordinates. Unilateral micro-knife lesions separating the lateral and medial habenula nuclei, transecting the fasciculus retroflexus or cutting the stria medullaris were made using a micro-knife similar to that described by Sclafani and Grossman 41. For the fasciculus retroflexus cuts the incisor bar was lowered to 7 m m below horizontal and the guide cannula was inserted in the brain at a 30 ° angle to the vertical so that the knife would intercept the fasciculus retroflexus approximately perpendicularly. The guide cannula was inserted into the brain 3.0 mm lateral to the midline and was lowered until the bottom of the wire would, if extended, reach 0.5 mm posterior to the anterior-posterior stereotaxic zero. The wire was then forced through the cannula and out the curved end in a coronal plane to a length of 3.0 m m from the cannula tip. The actual knife cut was made by lowering the cannula with the extended wire a distance of 5.7 mm. For the cuts that separated the medial
416
Fig. 1. Pattern of innervation by substance P-immunoreactive fibres of the rostral (A) and Caudal (B) aspects of the ventrotegmental area in normal rats. NIP, interpeduncular nucleus. :~: 65.
417 from the lateral habenula a 4.0 mm knife was positioned on a parasagittal plane 0.4 m m lateral to the midline and with its anterior tip 2.4 m m posterior to bregma. The guide cannula was lowered until the bottom of the wire would, if extended, reach 6.1 m m below the skull surface. The extended wire and cannula were raised 3.5 mm. The stria medullaris was transected 4.0 mm rostral to the habenula by a knife 6.0 mm wide, lowered to a depth of 6.0 m m below the dura. Animals were sacrificed between 8-10 days after the lesions. Correct placement of the lesions was verified by direct observation of thick tissue sections in material used for R I A or biochemical assay. Animals with misplaced lesions, infections or substantial tissue necrosis were rejected. RESULTS As already reportedtl,19, a4 dense groups of substance P immunofluorescent fibres were observed both in the ventrotegmental area (Fig. 1) and in the lateral habenular nucleus and positive cell bodies in the medial habenular nucleus. Bilateral electrolytic lesions of the habenula nuclei resulted in a fall in both ChAT and substance P content in the interpeduncular nuclear area (Table 1). Similar lesions produced a marked depletion of substance P-immunofluorescence in the ventral tegmental area but no change was observed in the substance P-rich substantia nigra present in the same sections of the mesencephalic region (Fig. 2). Following the transections of the stria medullaris proximal to the habenula the substance P immunofluorescence and the pattern of the AChE staining in the habenula was unaltered. Lesions of the ventral tegmental area (including the interpeduncular nucleus) did not alter the habenula content of immunoreactive substance P or ChAT (Table I1). In addition no change was seen in the density of immunofluorescent fibres normally present in the lateral habenular nucleusll,~9, a4 (Fig. 3). Unilateral electrothermic lesions of the habenula (Fig. 11, lesion 3) resulted in an ipsilateral depletion of substance P-immunoreactive fibres in the ventral tegmental area. Micro-knife cuts (Fig. 11, lesion 2) separating the medial from the lateral habenula without damaging the fasciculus retroflexus produced a total ablation of the substance P-immunoreactive fibres in the ipsilateral lateral habenula (Figs. 6, 7). TABLE 1 Effect o f bilateral electrolytic lesion o f the habenular nuclei on the substance P content and ChAT activity o f the nucleus interpeduncular-ventrotegmental area
Values represent the mean ± S.E.M. of 6 animals
Control Habenularlesion * P < 0.05.
** P <: 0.01 paired t-test.
ChAT (itmole/g protein~h)
SP (pmole/g)
930.0 ± 50 88.5 ~ 24**
1,126 ± 61 * 323 ± 69*
418
Fig. 2. Depletion of substance P-immunoreactivity in the ventrotegmental area (*) following bilateral electrothermic lesions of the habenular nuclei. Note that substance P immunofluorescent fibres in the substantia nigra ( a r r o w s ) r e m a i n unchanged. NIP, interpeduncular nucleus, ML. medial lemniscus. × 65.
The same knife cut abolished the AChE positive material in the medial habenular nucleus (Fig. 9). These effects were confirmed by showing that these knife cuts caused a significant fall in the substance P content of the lateral habenular nucleus but no change in substance P content of medial habenular nucleus (Table 1II). In agreement with the results obtained by staining for AChE a depletion of ChAT activity was found in the medial but not in the lateral habenula (Table I11). In the ventral tegmental area this cut (Fig. I l, lesion 2) produced an ipsilateral depletion of substance P immunofluorescent fibres similar to that seen after unilateral electrothermic lesions (Fig. 4). Nevertheless, at caudal levels of the ventrotegmental area some immunoreactive fibres survived both procedures. Similarly these cuts produced a significant depletion in the substance P content of the nucleus interpeduncular-ventrotegmental area, but no change in ChAT content (Table !1I). TABLE 11
Effect of electrolytic lesions o['the nucleus interpeduncular-ventrotegmental P content and ChAT activity Values represent the mean 5_ S.E.M. of 6 animals
Control Lesion
ChAT (r~rnole/g protein/h
SP (pmole/g)
521 2~ 45 475 3_ 75 : NS
358 ~- 44 319 -~-:101 : NS
NS, no significant difference from control, paired t-test.
area on habenular substance
419
Fig. 3. Persistence of s u b s t a n c e P - i m m u n o r e a c t i v e fibres in the lateral h a b e n u l a r nucleus (LH) after lesion of the i n t e r p e d u n c u l a r nucleus a n d ventrotegmental area. Note positive cell bodies in the medial h a b e n u l a r nucleus (arrow) (MH). >~ 65.
T A B L E Ill
Eff'ect o f the unilateral transection o f the babenMar nuclei on the substance P content and ChAT activio, 0/' the individual habenular nuclei and the ventrotegmental area Values represent the m e a n of 5 a n i m a l s -- S.E.M.
Medial h a b e n u l a r nucleus Lateral h a b e n u l a r nucleus N u c l e u s interpeduncularventrotegmental area
Control Lesion Control Lesion Control Lesion
* P < 0.05. ** P < 0.02. NS, n o t significant f r o m control, paired t-test.
ChA T (/¢mole/g protein/h)
Substance P (pmole/g wet wt.)
1023 238 236 255 1100 1290
412 434 271 155 733 352
~ 107 ~ 144'* ~ 82 -F 68 : NS =k 175 ± 240 : NS
~ 100 ~c 179 : N S J- 17 z~ 18" :F 69 ~- 45**
420
Fig. 4. (A) Depletion of substance P-immunoreactive fibres in the ipsilaterat ventrotegmental area after a knife cut lesion separating the medial habenular nucleus. (B) Contralaterai untreated side of the same section. NIP, interpeduncular nucleus. :,~ 160.
Fig. 5. Almost total loss of substance P-immunoreactive fibres ( ) from the ipsilateral side of the ventrotegmental area following a knife-cut transection of the fasciculus retroflexus. ,~ 74.
421 T A B L E IV
E/l~'ct o/'unilateral transection o f the fasciculus retroflexus on the substance P content and ChAT activity q[ the habenular Jasciculus retroflexus (proximal to habenula) and the ventrotegmental area Each value represent the m e a n ± S.E.M. of five animals.
Habenula Fasciculus retroflexus Nucleus interpeduncularventrotegmental area
Controls±de Lesion side Controls±de Lesion side Control side Lesion side
ChAT (/tmole/g protein/h)
Substance P (pmole/g)
256 290 178 331 824 343
382 420 134 304 757 414
:~ ± L ± ± :)
121 154 83 61" 251 134"**
± 42 ± 73 _i: 61 ± 49* ± 45 ~ 95**
* NSatP~ 0.1. ** P -< 0.05. *** P - 0,01, paired t-test.
Transections of the fasciculus retroflexus (Fig. 11, lesion 3) resulted in an ±psilateral loss of substance P-immunofluorescence from the ventrotegmental area (Fig. 5) and an accumulation of substance P immunofluorescence (Fig. 10) and AChE (Fig. 8) reaction in fibres immediately proximal to the knife cut. Eight days after surgery biochemical measurements in microdissected fasciculus retroflexus tissue proximal to the
B 't
V
V
O !i
7¸
Med
H
Med #
L~t
iI
H
Lat
H
' !
Fig. 6. A : low magnification view o f the transected h a b e n u l a r nuclei s h o w i n g the loss of substance P - i m m u n o r e a c t i v e fibres (*) f r o m the lateral h a b e n u l a r nucleus (Lat. H.). Some unspecific fluorescence is observed lying in the ventricle (V). B: contralateral untreated side. Med. H., medial h a b e n u l a r nucleus. D a s h e d line represents knife tract. "< 95.
422
Fig. 7. Higher magnification view of an analogous preparation to Figl 6 (A) untreated side, (B) bisected habenula (contralateral side). Note the complete depletion of substance P-immunoreactive fibres as compared with (A). Some unspecific fluorescence can be noticed in the knife tract (dashed line), x 380.
423
Fig. 8. Acetylcholinesterase staining in the habenular complex following transection of the fasciculus retroflexus (dashed lines). Note accumulation of AChE reaction products proximal to the cut and the depletion of the distal part of the fasciculus retroflexus (arrow). MH, medial habenular nucleus, LH, lateral habenular nucleus, x 84.
Fig. 9. Loss of acetylcholinesterase staining in the medial habenular nucleus (*) in the bisected habenula. Compare with untreated side (left). MH, medial habenular nucleus, LH, lateral habenular nucleus. Dashed lines represent course of micro-knife, x 120.
424
Fig. 10. Pile up of substance P-immunoreactivity in the proximal stump of the transected fasciculus retroflexus, x 650.
425
® J
,I LI
Hob.
FQSC. Retr
ACh
~
/I I
i-
(~
/
Fig. I I. Schematic representation of the various lesions performed in this study and reconstruction of the substance P and acetylcholine containing neurones originating from the habenula. (1) Electrothermic lesion of the habenula nuclei, (2) micro-knife transections of the habenula and (3) transection of the fasciculus retroflexus and proximal pile up of substance P and ACHE. This scheme is intended only to represent the sites of origin and termination of the substance P and cholinergic neurones in the habenulo-mesencephalic complex and not the cellular features of these neurones. L, lateral habenula, M, medial habenula, VTA, ventrotegmental area, NIP, nucleus interpeduncularis.
transection revealed an increase in substance P c o n t e n t a n d C h A T activity in this area (Table IV). Analysis of the i n t e r p e d u n c u l a r area revealed a significant depletion of both substance P a n d C h A T c o n t e n t in these animals (Table IV). DISCUSSION
The reported existence of substance P - c o n t a i n i n g elements in the h a b e n u l a a n d the nucleus i n t e r p e d u n c u l a r - v e n t r o t e g m e n t a l areal1,19, a4 has been confirmed b o t h by
426 radioimmunoassay and immunocytochemistry. The disappearance of substance P following bilateral lesions of the habenular nuclei suggest that these substance Pimmunoreactive elements are part of a neuronal pathway originating in these nuclei TM 19,33. The results of the present study indicate that this substance P pathway originates from immunoreactive cell bodies located in the medial habenular nucleus and terminating almost exclusively in the ventrotegmental area rather than in the interpeduncular nucleus. The latter structure seems to receive only a very limited number of substance P containing fibres. A close correlation was observed between the fall of substance P content in the interpeduncular nucteus-ventrotegmental area (Tables I-Ill) and disappearance of immunofluorescent fibres. The fact that these fibres originate in the medial habenular nucleus is confirmed by the fall in substance P which followed the disconnection of the nucleus from more laterally located structures (Table 111). This would indicate that the substance P-fibres originating from the medial habenular nucleus run laterally to join the fasciculus retroflexus, an axonal bundle which connects the habenula with the ventral mesencephalon and interpeduncular nucleus. That this is the case is clearly demonstrated by the unilateral depletion of immunofluorescent fibres in the ventrotegmental area and the interpeduncular nucleus and the fall of substance P content in the same regions (Table II). Furthermore, a 'pile up' of substance P (Fig. 10) was observed in the fasciculus retroflexus proximal (dorsal) to a knife cut lesion which severed the axons of this fibre bundle. The characteristics of this axonal transport of substance P will be described in more detail in a separate publication 37. In parallel with the immunofluorescent pile up of substance P an increase in the substance P content of the fasciculus retroflexus was also observed (Table IV). • 30/o) °j in The persistence of a small but significant amount of substance P (10-. the ventral tegmental area - - interpeduncular nucleus following bilateral electrolytic lesions of the habenula nuclei may indicate the existence of a separate population of substance P containing fibres or cells. ]mmunohistochemical findings indicate that substance P positive fibres persist in the caudal portion of the ventral tegmental area even after complete habenular lesions. The origin of this separate substance P projection to the caudal ventral tegmental area remains to be elucidated. The distribution of substance P-containing nerve terminals in the ventrotegmental area corresponds closely to that of the A-10 group of dopaminergic neurones as described by a catecholamine histofluorescence technique ~:~, Experimental studies have indicated that these cells represent the origin of the so-called mesolimbic and mesocortical dopaminergic systems 15,')6,29. From these observations an interaction between the substance P neurones of the habenulo-ventrotegmental pathway and the dopaminergic mesocortical and mesolimbic systems may be predicted. Substance P-containing fibres are also present in the lateral habenular nucleus. Since substance P-positive cell bodies are present in the interpeduncular nucleus the possibility of a reciprocal peptidergic innervation between the habenula and the interpeduncular nucleus was examined. Classical neuroanatomical studies in several species describe the existence in the habenula of interpeduncular nucleus afferents 8,1s.
427 31,3'~,43. However, in our study large lesions of the interpeduncular nucleus-ventrotegmental area failed to produce any change in the pattern or intensity of substance P immunofluorescence or in ChAT or substance P content in the habenula (Table II). Therefore, the existence of ascending cholinergic or substance P-containing fibres innervating the lateral or medial habenula nuclei from the interpeduncular nucleus region can be ruled out. The lack of change in the immunofluorescent pattern or cholinesterase staining in the habenula after the stria medullaris le~ion would indicate that the majority of substance P-containing or cholinesterase containing fibres of the habenula do not originate from forebrain limbic areas. Some evidence obtained in invertebrate species indicate that transmitter substances may coexist in a single neurone 2,a,7. However there is no unequivocal evidence for such a dual storage of neurotransmitters in the vertebrate nervous system. Lesions which involved both habenula nuclei or the fibre tract connecting the habenula with the interpeduncular nucleus-ventrotegmental area produced a fall in both substance P and ChAT (Tables I, IV). This type of finding has prompted some workers ~°,33 to postulate the possible coexistence of substance P with acetylcholine in the same neurones of the habenulo-interpeduncular tract. However, when the nuclei of the habenula were divided by a stereotaxically guided micro-knife the projections from the two nuclei could be dissociated. Thus it was observed that the cholinergic projection to the interpeduncular nucleus-ventrotegmental area was preserved whereas the substance P projection to the ventrotegmental area was markedly depleted (Table I11). This procedure also revealed that substance P-immunoreactivity in the lateral habenular nucleus was lost (Figs. 6, 7, Table i|1) indicating the existence of a short intrahabenular substance P projection. The same procedure also showed that the ChAT activity of the medial habenular nucleus was diminished (Table lIl) as was the AChE staining (Fig. 9). These findings indicate that a reciprocal peptidergic and cholinergic innervation exists between the habenula nuclei. The absence of any ChAT depletion following the surgical disconnection of medial habenular nucleus (Table Ill) from the nucleus interpeduncular-ventrotegmentM area indicates that the site of origin of the cholinergic habenulo-interpeduncular pathway is primarily in the lateral habenular nucleus. The organisation of the substance P and acetylcholine containing pathways revealed by these experiments in the habenula-interpeduncular nucleus-ventrotegmental area tract is illustrated schematically in Fig. 11. The physiological implications of these two pathways in the habenular complex remain to be established. ACKNOWLEDGEMENTS The authors wish to thank Drs. L. L. Iversen and J. S. Kelly for revising the manuscript and to Mrs L. Mercer, A. Nelson and J. Reed for technical assistance. T. M. J. was an MRC scholar. We gratefully acknowledge the gifts of substance P from Dr. R. Hirschmann Merk, Sharpe and Dohme.
428 REFERENCES 1 Brownstein, M. J., Mroz, E. A., Kizer, J. S., Palkovits, M. and Leeman, S. E., Regional distribution of substance P in the brain of the rat, Brain Research, 116 (1976) 299-305. 2 Brownstein, M. J., Saavedra, J. M., Axelrod, J., Zeman, G. H. and Carpenter, D. O., Coexistence of several putative neurotransmitters in single identified neurons of Aplysia, Proe. nat. Acad. Sci., (Wash.), 71 (1974) 4662-4668. 3 Burnstock, G., Do some nerve cells release more than one transmitter?, Neuroscience, I (1976) 239-278. 4 Chang, M. M. and Leeman, S. E., Isolation of a sialogogic peptide from bovine hypothalamic tissue and its characterization as substance P, J. biol. Chem., 245 (1970) 4784-4790. 5 Chang, M. M., Leeman, S. E. and Niall, H. D., Amino-acid sequence of substance P, Nature New Biology, 232 (1971) 86-87. 6 Coons, A. H. and Kaplan, M. H., Localization of antigens in tissue cells. II. Improvements in a method for the detection of antigen by means of fluorescent antibody, J. exp. Med., 91 (1950) I-9. 7 Cottrell, G. A., Identified amine-containing neurones and their synaptic connexions, Neuroscience, 2 (1977) 1-18. 8 Cragg, B. G., The connections of the habenula in the rabbit, Exp. Neurol., 3 (1961) 388-409. 9 Cuello, A. C., Emson, P. del Fiacco, M., Gale, J., Iversen, L. L., Jessell, T. M., Kanazawa, I., Paxinos, G. and Quik, M., Distribution and release of substance P in the central nervous system. In J. Hughes (Ed.), Centrally Acting Peptides, Biological Council Symposium, Macmillan, London (in press). l0 Cuello, A. C., Jessell, T., Kanazawa, 1. and lversen, L. L., Substance P: localization in synaptic vesicles in rat central nervous system, J. Neurochem., 29 (1977) 747-751. II Cuello, A. C. and Kanazawa, I., The distribution of substance P immunoreactive fibres in rat central nervous system, J. comp. NeuroL (in press). 12 Cuello, A. C., Polak, J. and Pearse, A. E. G., Substance P: a naturally occurring transmitter in human spinal cord, Lancet, II (1976) 1054-1056. 13 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine-containing neurones in the central nervous system. I Demonstration of monoamines in the cell bodies of brain stem neurons, Acta physiol, scand., 62 (Suppl. 232) 0964) 1-80. 14 Emson, P. C., Kanazawa, I., Cuello, A. C. and Jessel, T., Substance P pathways in rat brain, Biochem. Soc. Trans., 42 (1976) 187-189. 15 Emson, P. C. and Koob, G., The origin and distribution of dopamine-containing afferents to the rat frontal cortex, Brain Research, (in press). 16 Euler, U.S. von and Gaddum, J . H . , An unidentified depressor substance in certain tissue extracts, J. Physiol. (Lond.), 72 (1931) 74-87. 17 Fonnum, F., Radiochemical microassays for the determination of choline acetyltransferase and acetylcholinesterase activity, Biochem. J., l 15 (t969) 465-472. 18 Herkenharm, M. and Nauta, W. J. H., Afferent connections of the habenula nuclei in the rat. A horseradish peroxidase study, with a note on the fibre-of passage problem, J. comp. Neurol., 173 (t977) 123-146. 19 H6kfelt, T., Kellerth, J. O., Nilsson, G. and Pernow, B:, Substance P: localization in the central nervous system and in some primary sensory neurons, Science, 190 (1975) 889-890. 20 Hong, J. S., Costa, E. and Yang, H.-Y., T., Effects of habenular lesions on substance P content of various brain regions, Brain Research, 118 (1976) 523-525. 21 Jessell, T. and Iversen, L. L., Release of substance P from the rat substantia nigra, in vitro, Brain Research, (in press). 22 Jessell, T., Iversen, L. L. and Kanazawa, !., Release and metabolism of substance P in rat hypothalamus, Nature (Lond.), 264 (1976) 81-83. 23 Kanazawa, I. and Jessell, T., Post-mortem changes and regional distribution of substance P in the rat and mouse nervous system, Brain Research, 117 (1976) 362-367. 24 Kataoka, K., Nakamura, Y. and Hassler, R., Habenulo-interpeduncular tract: a possible cholinergic neuron in rat brain, Brain Research, 62 (1973) 264-267. 25 Koelle, G. G. and Friedenwald, J. S., A histochemical method for localizing cholinesterase activity, Proc. Soc. exp. Biol. (N. Y.), 70 (1949) 617-622. 26 Koob, G. F., Balcom, G. T. and Meyerhoff, J., Dopamine and norepinephrine levels in the nucleus
429
27
28 29 30 31 32 33 34
35 36 37 38 39 40 41 42 43
accumbens, olfactory tubercle and corpus striatum following lesions in the ventral tegmental are~, Brain Research, 94 (1975) 45-55. Leranth, C. S., Brownstein, M. J., Zaborsky, L., Jaranyi, Z. S. and Palkovits, M., Morphological and biochemical changes in the rat interpeduncular nucleus following the transection of the habenulo-interpeduncular tract, Brain Research, 99 (1975) 127 128. Leeman, S. E. and Mroz, E. A., Minireview: substance P, LiJ~" Sci., 15 (1974) 2033 2041. Lindvall, O., Bj6rklund, A., Moore, E. Y. and Stenevi, U., Mesencephalic dopamine neurons projecting to neocortex, Brain Research, 81 (1974) 325-331. Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the folin phenol reagent, J. biol. Chem., 193 (1951) 265 275. Massopust, L. C. Jr. and Thompson, R., A new interpedunculodiencephalic pathway in rats and cats, J. comp. Neurol., 118 (1962) 97 105. Mitchell, R., Connections of the habenula and the interpeduncular nucleus in the cat, J. comp. Neurol., 121 (1963) 441-457. Mroz, E. A., Brownstein, M.J. and Leeman, S. E., Evidence for substance P in the habenulointerpeduncular tract, Brain Research, I 13 (1976) 597-599. Nilsson, G., H6kfelt, T. and Pernow, B, Distribution of substance P-like immunoreactivity in the rat central nervous system as revealed by immunohistocbemistry, Med. Biol., 52 0974) 424427. Otsuka, M. and Konishi, S., Substance P an excitatory transmitter of primary sensory neurons, Cold Spr. Harb. Symp. quant. Biol., Vol. 15 (1976) 135 143. Otsuka, M. and Konishi, S., Release of substance P-like immunoreactivity from isolated spinal cord of newborn rat, Nature (Lond.), 264 (1976) 83-84. Paxinos, G. and Cuello, A. C., Axonal transport of substance P in the central nervous system, in preparation. Pellegrino, L. and Cushman, A., A Stereotaxic Atlas of the Rat Brain, Appleton Century Crofts, New York, 1967. Pickel, V. M., Reis, D. J. and Leeman, S. E., Ultrastructural localization of substance P in neutones of rat spinal cord, Brain Research, 122 (1977) 534-540. Powell, D., Leeman, S. E., Tregear, G. W., Niall, H. D. and Potts, J. T. Jr., Radioimmunoassay for substance P, Nature (Lond.), 241 (1973) 252 254. Sclafani, A. and Grossman, S. P., Hyperphagia produced by knife cuts between the medial and lateral hypothalamus in the rat. Physiol. Behav., 4 (1969) 533-538. Shuster, C., Mroz, E. A. and Leeman, S. E., Release of substance P from isolated nerve endings, Nature (Lond.), 264 (1976) 790-792. Smaha, L. A. and Kaelber, W. W., Efferent fibre projectory of the babenula and the interpeduncular nucleus. An experimental study in the oppossum and cat, Exp. Brain Res., 16 (1973) 291-308.
413
© Elsevier/North-Holland Biomedical Press
SUBSTANCE P C O N T A I N I N G A N D C H O L I N E R G I C PROJECTIONS F R O M THE H A B E N U L A
A. CLAUD[O CUELLO*, PIERS C. EMSON, GEORGE PAXINOSt and THOMAS JESSELL~¢ M RC Neurochemical Pharmacology Unit, Dept. of Pharmacology, Medical School Cambridge (Great Britain )
(Accepted October 27th, 1977)
SUMMARY Electrolytic lesions and surgical transection of the habenulo-interpeduncularventrotegmental tract have established the existence of separate habenulo-interpeduncular-ventrotegmental substance P and cholinergic projections. Micro-knife lesions separating the habenula nuclei showed the medial habenular nucleus to be the source of substance P fibres running via the fasciculus retroflexus to the ventral tegmental area. The lateral habenular nucleus receives a substance P projection from the medial habenular nucleus and is the source of cholinergic projection to the interpeduncular nucleus and to the medial habenular nucleus. Lesions of the ventrotegmental-interpeduncular area did not modify the levels of substance P and choline acetyltransferase in the habenula. These observations suggested that there are no substance P or ACh containing afferents to the habenula from the ventrotegmental-interpeduncular area and the accumulation of substance P and AChE proximal to but not caudal to transections of the fasciculus retroflexus confirmed this view.
INTRODUCTION The identification of substance p4,5 as an undecapeptide was accomplished many years after the first report by von Euler and Gaddum 16 in 1931 of the existence of a vasoactive peptide in the brain and gut. The availability of pure substance P has now allowed the development of the appropriate immunochemical techniques which have clarified the role of substance P in the mammalian nervous system. Thus,
* To who correspondence and reprint requests should be sent. Present addresses: ¢ School of Psychology,The University of New South Wales, P.O. Box B, Kensington, New South Wales, Australia 2033. t¢ Dept. of Pharmacology, Harvard Medical School, 25 Shattuck Street, Boston, Mass. 02115, U.S.A.
414 immunoreactive substance P has been detected in the cell bodies and nerve terminals in specific regions of the rat, cat and human nervous systemll,12,1a,34. Sensitive radioimmunoassay (RIA) methods have demonstrated that substance P is unevenly distributed in the mouse and rat central nervous system1, ~z and a calcium-dependent release of this peptide has been demonstrated in vitro from the spinal cord 3~, hypothalamus 22, substantia nigra ~t and from synaptosomal preparations 39. More recently a vesicular localization of substance P has been demonstratedl0, zs. These observations have led to the suggestion that this peptide is a neurotransmitter (or modulator) in a number of well defined neuroanatomical pathways (see reviewsa,28.:~:'). There have been reports claiming a joint substance P/cholinergic projection from the habenula nuclei to the interpeduncular nucleus area~a,20,z3 although the precise sites of origin and termination of these projections remained uncertain. In some of these reports20, 33 the possible coexistence of substance P and acetylcholine in the same neurones has been discussed. The present investigation which employed discrete lesions to the habenula nuclei and pathway, allows us to dissociatc the two systems. MATERIALS AND METHODS
Microdissection Under the stereomicroscope, discrete nuclear areas were dissected at 4 °C from 0.6-1.0 mm thick coronal sections of fresh, unperfused, unfrozen rat brain. Coronal sections of the brain were prepared by hand using a series of parallel razor blades with a fixed inter-blade distance.
Radioimmunoassay of substance P The tissue pieces obtained were frozen on dry ice and weighed. Substance P was extracted initially with 50/~1 acetone/1 N HC1 (100:3, v/v) and then with 50/~1 acetone/0.01 N HCI (80:20, v/v) according to the method described by Chang and Leeman4, with the exception that the filtration procedure was replaced by centrifugation. Combined supernatant fractions were washed three times with pretroleum ether and lyophilized substance P was measured by radioimmunoassay using a method previously described by Powell et al. 4°. The specificity of the substance P-antisera prepared in guinea pig was checked with the related peptide eledoisin, and with somatostatin, met-enkephalin and fl-endorphin, all of which failed to cross-react with substance P when present in a 1,000 fold molar excess. The 9 and 10 amino acid fragments of the C-terminal sequence of substance P were found to possess immunoreactivity equal to that of the complete peptide. The sensitivity of the RIA allows accurate measurements of as little of 8 fmol substance P using antiserum dilutions in the range of 1:50,000-1:250,000. The post-mortem stability of substance P in brain tissue has been discussed in a previous publication23.
Histochemistry Experimental and control rats were anaesthetized with equithesin and perfused intra-
415 cardially with cold 4 ~ paraformaldehyde dissolved in 0.1 M phosphate buffer pH 7.4. Brains were removed and kept in the same fixative solution for 2-4 h. The fixed samples were transferred to a cold (4 °C) 0. l M phosphate buffer solution containing 5 % sucrose for up to 3 days. Thick sections (10 ttm) were cut on a Brights cryostat at 20 °C. Alternate sections were processed by immunocytochemistry, acetylcholinesterase or Nissl staining. Immunocytochemical investigation of substance P was performed by applying an indirect immunofluorescent method 6. Firstly antiserum (anti-substance P) raised in the guinea pig was applied in dilutions ranging from l :100 up to 1:450 (v/v). The second antiserum, anti-guinea pig F | T C conjugated lgG (Wellcome Foundation, Great Britain) was used in a 1:6 (v/v) dilution. Controls were performed using non-immune guinea pig serum or substance P-antiserum absorbed with 100 or 200 #g/ml of substance P. The tissue preparations were observed in a Leitz S M - L U X microscope equipped with epifluorescence optics. Acetylcholinesterase activity was demonstrated as described by Koelle and Friedenwald 25.
Biochemical determinations Choline acetyltransferase activity (CHAT) was determined as described by Fonnum tT. Protein concentrations were determined according to the method of Lowry et al. 3°.
Operative procedures Male 300 g Wistar rats were anaesthetized with equithesin (Jensen-Salsberg Labs. 4.0 ml/kg) and placed in a K o p f stereotaxic frame. Stereotaxic coordinates for radio frequency lesions were based on the stereotaxic atlas of Pellegrino and Cushman 3s. In the habenula nuclei (lateral 0.7 mm, posterior --2.0 mm and vertical 5.0 mm using the bregma as rostrocaudal zero) and ventral tegmental area (lateral 0.0 ram, posterior - - 3 . 0 m m and vertical 9.0 mm) lesions were made using a Radionic radio frequency lesion maker (model RGF-4). The temperature at the electrode tip was 55 °C during habenular lesions and 60 °C for ventral tegmental area lesions and were maintained for one minute. For micro-knife lesions the atlas served only as a guide and a series of rats were lesioned at different coordinates. Unilateral micro-knife lesions separating the lateral and medial habenula nuclei, transecting the fasciculus retroflexus or cutting the stria medullaris were made using a micro-knife similar to that described by Sclafani and Grossman 41. For the fasciculus retroflexus cuts the incisor bar was lowered to 7 m m below horizontal and the guide cannula was inserted in the brain at a 30 ° angle to the vertical so that the knife would intercept the fasciculus retroflexus approximately perpendicularly. The guide cannula was inserted into the brain 3.0 mm lateral to the midline and was lowered until the bottom of the wire would, if extended, reach 0.5 mm posterior to the anterior-posterior stereotaxic zero. The wire was then forced through the cannula and out the curved end in a coronal plane to a length of 3.0 m m from the cannula tip. The actual knife cut was made by lowering the cannula with the extended wire a distance of 5.7 mm. For the cuts that separated the medial
416
Fig. 1. Pattern of innervation by substance P-immunoreactive fibres of the rostral (A) and Caudal (B) aspects of the ventrotegmental area in normal rats. NIP, interpeduncular nucleus. :~: 65.
417 from the lateral habenula a 4.0 mm knife was positioned on a parasagittal plane 0.4 m m lateral to the midline and with its anterior tip 2.4 m m posterior to bregma. The guide cannula was lowered until the bottom of the wire would, if extended, reach 6.1 m m below the skull surface. The extended wire and cannula were raised 3.5 mm. The stria medullaris was transected 4.0 mm rostral to the habenula by a knife 6.0 mm wide, lowered to a depth of 6.0 m m below the dura. Animals were sacrificed between 8-10 days after the lesions. Correct placement of the lesions was verified by direct observation of thick tissue sections in material used for R I A or biochemical assay. Animals with misplaced lesions, infections or substantial tissue necrosis were rejected. RESULTS As already reportedtl,19, a4 dense groups of substance P immunofluorescent fibres were observed both in the ventrotegmental area (Fig. 1) and in the lateral habenular nucleus and positive cell bodies in the medial habenular nucleus. Bilateral electrolytic lesions of the habenula nuclei resulted in a fall in both ChAT and substance P content in the interpeduncular nuclear area (Table 1). Similar lesions produced a marked depletion of substance P-immunofluorescence in the ventral tegmental area but no change was observed in the substance P-rich substantia nigra present in the same sections of the mesencephalic region (Fig. 2). Following the transections of the stria medullaris proximal to the habenula the substance P immunofluorescence and the pattern of the AChE staining in the habenula was unaltered. Lesions of the ventral tegmental area (including the interpeduncular nucleus) did not alter the habenula content of immunoreactive substance P or ChAT (Table I1). In addition no change was seen in the density of immunofluorescent fibres normally present in the lateral habenular nucleusll,~9, a4 (Fig. 3). Unilateral electrothermic lesions of the habenula (Fig. 11, lesion 3) resulted in an ipsilateral depletion of substance P-immunoreactive fibres in the ventral tegmental area. Micro-knife cuts (Fig. 11, lesion 2) separating the medial from the lateral habenula without damaging the fasciculus retroflexus produced a total ablation of the substance P-immunoreactive fibres in the ipsilateral lateral habenula (Figs. 6, 7). TABLE 1 Effect o f bilateral electrolytic lesion o f the habenular nuclei on the substance P content and ChAT activity o f the nucleus interpeduncular-ventrotegmental area
Values represent the mean ± S.E.M. of 6 animals
Control Habenularlesion * P < 0.05.
** P <: 0.01 paired t-test.
ChAT (itmole/g protein~h)
SP (pmole/g)
930.0 ± 50 88.5 ~ 24**
1,126 ± 61 * 323 ± 69*
418
Fig. 2. Depletion of substance P-immunoreactivity in the ventrotegmental area (*) following bilateral electrothermic lesions of the habenular nuclei. Note that substance P immunofluorescent fibres in the substantia nigra ( a r r o w s ) r e m a i n unchanged. NIP, interpeduncular nucleus, ML. medial lemniscus. × 65.
The same knife cut abolished the AChE positive material in the medial habenular nucleus (Fig. 9). These effects were confirmed by showing that these knife cuts caused a significant fall in the substance P content of the lateral habenular nucleus but no change in substance P content of medial habenular nucleus (Table 1II). In agreement with the results obtained by staining for AChE a depletion of ChAT activity was found in the medial but not in the lateral habenula (Table I11). In the ventral tegmental area this cut (Fig. I l, lesion 2) produced an ipsilateral depletion of substance P immunofluorescent fibres similar to that seen after unilateral electrothermic lesions (Fig. 4). Nevertheless, at caudal levels of the ventrotegmental area some immunoreactive fibres survived both procedures. Similarly these cuts produced a significant depletion in the substance P content of the nucleus interpeduncular-ventrotegmental area, but no change in ChAT content (Table !1I). TABLE 11
Effect of electrolytic lesions o['the nucleus interpeduncular-ventrotegmental P content and ChAT activity Values represent the mean 5_ S.E.M. of 6 animals
Control Lesion
ChAT (r~rnole/g protein/h
SP (pmole/g)
521 2~ 45 475 3_ 75 : NS
358 ~- 44 319 -~-:101 : NS
NS, no significant difference from control, paired t-test.
area on habenular substance
419
Fig. 3. Persistence of s u b s t a n c e P - i m m u n o r e a c t i v e fibres in the lateral h a b e n u l a r nucleus (LH) after lesion of the i n t e r p e d u n c u l a r nucleus a n d ventrotegmental area. Note positive cell bodies in the medial h a b e n u l a r nucleus (arrow) (MH). >~ 65.
T A B L E Ill
Eff'ect o f the unilateral transection o f the babenMar nuclei on the substance P content and ChAT activio, 0/' the individual habenular nuclei and the ventrotegmental area Values represent the m e a n of 5 a n i m a l s -- S.E.M.
Medial h a b e n u l a r nucleus Lateral h a b e n u l a r nucleus N u c l e u s interpeduncularventrotegmental area
Control Lesion Control Lesion Control Lesion
* P < 0.05. ** P < 0.02. NS, n o t significant f r o m control, paired t-test.
ChA T (/¢mole/g protein/h)
Substance P (pmole/g wet wt.)
1023 238 236 255 1100 1290
412 434 271 155 733 352
~ 107 ~ 144'* ~ 82 -F 68 : NS =k 175 ± 240 : NS
~ 100 ~c 179 : N S J- 17 z~ 18" :F 69 ~- 45**
420
Fig. 4. (A) Depletion of substance P-immunoreactive fibres in the ipsilaterat ventrotegmental area after a knife cut lesion separating the medial habenular nucleus. (B) Contralaterai untreated side of the same section. NIP, interpeduncular nucleus. :,~ 160.
Fig. 5. Almost total loss of substance P-immunoreactive fibres ( ) from the ipsilateral side of the ventrotegmental area following a knife-cut transection of the fasciculus retroflexus. ,~ 74.
421 T A B L E IV
E/l~'ct o/'unilateral transection o f the fasciculus retroflexus on the substance P content and ChAT activity q[ the habenular Jasciculus retroflexus (proximal to habenula) and the ventrotegmental area Each value represent the m e a n ± S.E.M. of five animals.
Habenula Fasciculus retroflexus Nucleus interpeduncularventrotegmental area
Controls±de Lesion side Controls±de Lesion side Control side Lesion side
ChAT (/tmole/g protein/h)
Substance P (pmole/g)
256 290 178 331 824 343
382 420 134 304 757 414
:~ ± L ± ± :)
121 154 83 61" 251 134"**
± 42 ± 73 _i: 61 ± 49* ± 45 ~ 95**
* NSatP~ 0.1. ** P -< 0.05. *** P - 0,01, paired t-test.
Transections of the fasciculus retroflexus (Fig. 11, lesion 3) resulted in an ±psilateral loss of substance P-immunofluorescence from the ventrotegmental area (Fig. 5) and an accumulation of substance P immunofluorescence (Fig. 10) and AChE (Fig. 8) reaction in fibres immediately proximal to the knife cut. Eight days after surgery biochemical measurements in microdissected fasciculus retroflexus tissue proximal to the
B 't
V
V
O !i
7¸
Med
H
Med #
L~t
iI
H
Lat
H
' !
Fig. 6. A : low magnification view o f the transected h a b e n u l a r nuclei s h o w i n g the loss of substance P - i m m u n o r e a c t i v e fibres (*) f r o m the lateral h a b e n u l a r nucleus (Lat. H.). Some unspecific fluorescence is observed lying in the ventricle (V). B: contralateral untreated side. Med. H., medial h a b e n u l a r nucleus. D a s h e d line represents knife tract. "< 95.
422
Fig. 7. Higher magnification view of an analogous preparation to Figl 6 (A) untreated side, (B) bisected habenula (contralateral side). Note the complete depletion of substance P-immunoreactive fibres as compared with (A). Some unspecific fluorescence can be noticed in the knife tract (dashed line), x 380.
423
Fig. 8. Acetylcholinesterase staining in the habenular complex following transection of the fasciculus retroflexus (dashed lines). Note accumulation of AChE reaction products proximal to the cut and the depletion of the distal part of the fasciculus retroflexus (arrow). MH, medial habenular nucleus, LH, lateral habenular nucleus, x 84.
Fig. 9. Loss of acetylcholinesterase staining in the medial habenular nucleus (*) in the bisected habenula. Compare with untreated side (left). MH, medial habenular nucleus, LH, lateral habenular nucleus. Dashed lines represent course of micro-knife, x 120.
424
Fig. 10. Pile up of substance P-immunoreactivity in the proximal stump of the transected fasciculus retroflexus, x 650.
425
® J
,I LI
Hob.
FQSC. Retr
ACh
~
/I I
i-
(~
/
Fig. I I. Schematic representation of the various lesions performed in this study and reconstruction of the substance P and acetylcholine containing neurones originating from the habenula. (1) Electrothermic lesion of the habenula nuclei, (2) micro-knife transections of the habenula and (3) transection of the fasciculus retroflexus and proximal pile up of substance P and ACHE. This scheme is intended only to represent the sites of origin and termination of the substance P and cholinergic neurones in the habenulo-mesencephalic complex and not the cellular features of these neurones. L, lateral habenula, M, medial habenula, VTA, ventrotegmental area, NIP, nucleus interpeduncularis.
transection revealed an increase in substance P c o n t e n t a n d C h A T activity in this area (Table IV). Analysis of the i n t e r p e d u n c u l a r area revealed a significant depletion of both substance P a n d C h A T c o n t e n t in these animals (Table IV). DISCUSSION
The reported existence of substance P - c o n t a i n i n g elements in the h a b e n u l a a n d the nucleus i n t e r p e d u n c u l a r - v e n t r o t e g m e n t a l areal1,19, a4 has been confirmed b o t h by
426 radioimmunoassay and immunocytochemistry. The disappearance of substance P following bilateral lesions of the habenular nuclei suggest that these substance Pimmunoreactive elements are part of a neuronal pathway originating in these nuclei TM 19,33. The results of the present study indicate that this substance P pathway originates from immunoreactive cell bodies located in the medial habenular nucleus and terminating almost exclusively in the ventrotegmental area rather than in the interpeduncular nucleus. The latter structure seems to receive only a very limited number of substance P containing fibres. A close correlation was observed between the fall of substance P content in the interpeduncular nucteus-ventrotegmental area (Tables I-Ill) and disappearance of immunofluorescent fibres. The fact that these fibres originate in the medial habenular nucleus is confirmed by the fall in substance P which followed the disconnection of the nucleus from more laterally located structures (Table 111). This would indicate that the substance P-fibres originating from the medial habenular nucleus run laterally to join the fasciculus retroflexus, an axonal bundle which connects the habenula with the ventral mesencephalon and interpeduncular nucleus. That this is the case is clearly demonstrated by the unilateral depletion of immunofluorescent fibres in the ventrotegmental area and the interpeduncular nucleus and the fall of substance P content in the same regions (Table II). Furthermore, a 'pile up' of substance P (Fig. 10) was observed in the fasciculus retroflexus proximal (dorsal) to a knife cut lesion which severed the axons of this fibre bundle. The characteristics of this axonal transport of substance P will be described in more detail in a separate publication 37. In parallel with the immunofluorescent pile up of substance P an increase in the substance P content of the fasciculus retroflexus was also observed (Table IV). • 30/o) °j in The persistence of a small but significant amount of substance P (10-. the ventral tegmental area - - interpeduncular nucleus following bilateral electrolytic lesions of the habenula nuclei may indicate the existence of a separate population of substance P containing fibres or cells. ]mmunohistochemical findings indicate that substance P positive fibres persist in the caudal portion of the ventral tegmental area even after complete habenular lesions. The origin of this separate substance P projection to the caudal ventral tegmental area remains to be elucidated. The distribution of substance P-containing nerve terminals in the ventrotegmental area corresponds closely to that of the A-10 group of dopaminergic neurones as described by a catecholamine histofluorescence technique ~:~, Experimental studies have indicated that these cells represent the origin of the so-called mesolimbic and mesocortical dopaminergic systems 15,')6,29. From these observations an interaction between the substance P neurones of the habenulo-ventrotegmental pathway and the dopaminergic mesocortical and mesolimbic systems may be predicted. Substance P-containing fibres are also present in the lateral habenular nucleus. Since substance P-positive cell bodies are present in the interpeduncular nucleus the possibility of a reciprocal peptidergic innervation between the habenula and the interpeduncular nucleus was examined. Classical neuroanatomical studies in several species describe the existence in the habenula of interpeduncular nucleus afferents 8,1s.
427 31,3'~,43. However, in our study large lesions of the interpeduncular nucleus-ventrotegmental area failed to produce any change in the pattern or intensity of substance P immunofluorescence or in ChAT or substance P content in the habenula (Table II). Therefore, the existence of ascending cholinergic or substance P-containing fibres innervating the lateral or medial habenula nuclei from the interpeduncular nucleus region can be ruled out. The lack of change in the immunofluorescent pattern or cholinesterase staining in the habenula after the stria medullaris le~ion would indicate that the majority of substance P-containing or cholinesterase containing fibres of the habenula do not originate from forebrain limbic areas. Some evidence obtained in invertebrate species indicate that transmitter substances may coexist in a single neurone 2,a,7. However there is no unequivocal evidence for such a dual storage of neurotransmitters in the vertebrate nervous system. Lesions which involved both habenula nuclei or the fibre tract connecting the habenula with the interpeduncular nucleus-ventrotegmental area produced a fall in both substance P and ChAT (Tables I, IV). This type of finding has prompted some workers ~°,33 to postulate the possible coexistence of substance P with acetylcholine in the same neurones of the habenulo-interpeduncular tract. However, when the nuclei of the habenula were divided by a stereotaxically guided micro-knife the projections from the two nuclei could be dissociated. Thus it was observed that the cholinergic projection to the interpeduncular nucleus-ventrotegmental area was preserved whereas the substance P projection to the ventrotegmental area was markedly depleted (Table I11). This procedure also revealed that substance P-immunoreactivity in the lateral habenular nucleus was lost (Figs. 6, 7, Table i|1) indicating the existence of a short intrahabenular substance P projection. The same procedure also showed that the ChAT activity of the medial habenular nucleus was diminished (Table lIl) as was the AChE staining (Fig. 9). These findings indicate that a reciprocal peptidergic and cholinergic innervation exists between the habenula nuclei. The absence of any ChAT depletion following the surgical disconnection of medial habenular nucleus (Table Ill) from the nucleus interpeduncular-ventrotegmentM area indicates that the site of origin of the cholinergic habenulo-interpeduncular pathway is primarily in the lateral habenular nucleus. The organisation of the substance P and acetylcholine containing pathways revealed by these experiments in the habenula-interpeduncular nucleus-ventrotegmental area tract is illustrated schematically in Fig. 11. The physiological implications of these two pathways in the habenular complex remain to be established. ACKNOWLEDGEMENTS The authors wish to thank Drs. L. L. Iversen and J. S. Kelly for revising the manuscript and to Mrs L. Mercer, A. Nelson and J. Reed for technical assistance. T. M. J. was an MRC scholar. We gratefully acknowledge the gifts of substance P from Dr. R. Hirschmann Merk, Sharpe and Dohme.
428 REFERENCES 1 Brownstein, M. J., Mroz, E. A., Kizer, J. S., Palkovits, M. and Leeman, S. E., Regional distribution of substance P in the brain of the rat, Brain Research, 116 (1976) 299-305. 2 Brownstein, M. J., Saavedra, J. M., Axelrod, J., Zeman, G. H. and Carpenter, D. O., Coexistence of several putative neurotransmitters in single identified neurons of Aplysia, Proe. nat. Acad. Sci., (Wash.), 71 (1974) 4662-4668. 3 Burnstock, G., Do some nerve cells release more than one transmitter?, Neuroscience, I (1976) 239-278. 4 Chang, M. M. and Leeman, S. E., Isolation of a sialogogic peptide from bovine hypothalamic tissue and its characterization as substance P, J. biol. Chem., 245 (1970) 4784-4790. 5 Chang, M. M., Leeman, S. E. and Niall, H. D., Amino-acid sequence of substance P, Nature New Biology, 232 (1971) 86-87. 6 Coons, A. H. and Kaplan, M. H., Localization of antigens in tissue cells. II. Improvements in a method for the detection of antigen by means of fluorescent antibody, J. exp. Med., 91 (1950) I-9. 7 Cottrell, G. A., Identified amine-containing neurones and their synaptic connexions, Neuroscience, 2 (1977) 1-18. 8 Cragg, B. G., The connections of the habenula in the rabbit, Exp. Neurol., 3 (1961) 388-409. 9 Cuello, A. C., Emson, P. del Fiacco, M., Gale, J., Iversen, L. L., Jessell, T. M., Kanazawa, I., Paxinos, G. and Quik, M., Distribution and release of substance P in the central nervous system. In J. Hughes (Ed.), Centrally Acting Peptides, Biological Council Symposium, Macmillan, London (in press). l0 Cuello, A. C., Jessell, T., Kanazawa, 1. and lversen, L. L., Substance P: localization in synaptic vesicles in rat central nervous system, J. Neurochem., 29 (1977) 747-751. II Cuello, A. C. and Kanazawa, I., The distribution of substance P immunoreactive fibres in rat central nervous system, J. comp. NeuroL (in press). 12 Cuello, A. C., Polak, J. and Pearse, A. E. G., Substance P: a naturally occurring transmitter in human spinal cord, Lancet, II (1976) 1054-1056. 13 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine-containing neurones in the central nervous system. I Demonstration of monoamines in the cell bodies of brain stem neurons, Acta physiol, scand., 62 (Suppl. 232) 0964) 1-80. 14 Emson, P. C., Kanazawa, I., Cuello, A. C. and Jessel, T., Substance P pathways in rat brain, Biochem. Soc. Trans., 42 (1976) 187-189. 15 Emson, P. C. and Koob, G., The origin and distribution of dopamine-containing afferents to the rat frontal cortex, Brain Research, (in press). 16 Euler, U.S. von and Gaddum, J . H . , An unidentified depressor substance in certain tissue extracts, J. Physiol. (Lond.), 72 (1931) 74-87. 17 Fonnum, F., Radiochemical microassays for the determination of choline acetyltransferase and acetylcholinesterase activity, Biochem. J., l 15 (t969) 465-472. 18 Herkenharm, M. and Nauta, W. J. H., Afferent connections of the habenula nuclei in the rat. A horseradish peroxidase study, with a note on the fibre-of passage problem, J. comp. Neurol., 173 (t977) 123-146. 19 H6kfelt, T., Kellerth, J. O., Nilsson, G. and Pernow, B:, Substance P: localization in the central nervous system and in some primary sensory neurons, Science, 190 (1975) 889-890. 20 Hong, J. S., Costa, E. and Yang, H.-Y., T., Effects of habenular lesions on substance P content of various brain regions, Brain Research, 118 (1976) 523-525. 21 Jessell, T. and Iversen, L. L., Release of substance P from the rat substantia nigra, in vitro, Brain Research, (in press). 22 Jessell, T., Iversen, L. L. and Kanazawa, !., Release and metabolism of substance P in rat hypothalamus, Nature (Lond.), 264 (1976) 81-83. 23 Kanazawa, I. and Jessell, T., Post-mortem changes and regional distribution of substance P in the rat and mouse nervous system, Brain Research, 117 (1976) 362-367. 24 Kataoka, K., Nakamura, Y. and Hassler, R., Habenulo-interpeduncular tract: a possible cholinergic neuron in rat brain, Brain Research, 62 (1973) 264-267. 25 Koelle, G. G. and Friedenwald, J. S., A histochemical method for localizing cholinesterase activity, Proc. Soc. exp. Biol. (N. Y.), 70 (1949) 617-622. 26 Koob, G. F., Balcom, G. T. and Meyerhoff, J., Dopamine and norepinephrine levels in the nucleus
429
27
28 29 30 31 32 33 34
35 36 37 38 39 40 41 42 43
accumbens, olfactory tubercle and corpus striatum following lesions in the ventral tegmental are~, Brain Research, 94 (1975) 45-55. Leranth, C. S., Brownstein, M. J., Zaborsky, L., Jaranyi, Z. S. and Palkovits, M., Morphological and biochemical changes in the rat interpeduncular nucleus following the transection of the habenulo-interpeduncular tract, Brain Research, 99 (1975) 127 128. Leeman, S. E. and Mroz, E. A., Minireview: substance P, LiJ~" Sci., 15 (1974) 2033 2041. Lindvall, O., Bj6rklund, A., Moore, E. Y. and Stenevi, U., Mesencephalic dopamine neurons projecting to neocortex, Brain Research, 81 (1974) 325-331. Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the folin phenol reagent, J. biol. Chem., 193 (1951) 265 275. Massopust, L. C. Jr. and Thompson, R., A new interpedunculodiencephalic pathway in rats and cats, J. comp. Neurol., 118 (1962) 97 105. Mitchell, R., Connections of the habenula and the interpeduncular nucleus in the cat, J. comp. Neurol., 121 (1963) 441-457. Mroz, E. A., Brownstein, M.J. and Leeman, S. E., Evidence for substance P in the habenulointerpeduncular tract, Brain Research, I 13 (1976) 597-599. Nilsson, G., H6kfelt, T. and Pernow, B, Distribution of substance P-like immunoreactivity in the rat central nervous system as revealed by immunohistocbemistry, Med. Biol., 52 0974) 424427. Otsuka, M. and Konishi, S., Substance P an excitatory transmitter of primary sensory neurons, Cold Spr. Harb. Symp. quant. Biol., Vol. 15 (1976) 135 143. Otsuka, M. and Konishi, S., Release of substance P-like immunoreactivity from isolated spinal cord of newborn rat, Nature (Lond.), 264 (1976) 83-84. Paxinos, G. and Cuello, A. C., Axonal transport of substance P in the central nervous system, in preparation. Pellegrino, L. and Cushman, A., A Stereotaxic Atlas of the Rat Brain, Appleton Century Crofts, New York, 1967. Pickel, V. M., Reis, D. J. and Leeman, S. E., Ultrastructural localization of substance P in neutones of rat spinal cord, Brain Research, 122 (1977) 534-540. Powell, D., Leeman, S. E., Tregear, G. W., Niall, H. D. and Potts, J. T. Jr., Radioimmunoassay for substance P, Nature (Lond.), 241 (1973) 252 254. Sclafani, A. and Grossman, S. P., Hyperphagia produced by knife cuts between the medial and lateral hypothalamus in the rat. Physiol. Behav., 4 (1969) 533-538. Shuster, C., Mroz, E. A. and Leeman, S. E., Release of substance P from isolated nerve endings, Nature (Lond.), 264 (1976) 790-792. Smaha, L. A. and Kaelber, W. W., Efferent fibre projectory of the babenula and the interpeduncular nucleus. An experimental study in the oppossum and cat, Exp. Brain Res., 16 (1973) 291-308.