- Email: [email protected]
The development of laterality in the forebrain projections of midline thalamic cell groups in the rat
Developmental Brain Research, 35 (1987) 275-282 Elsevier
275
BRD 50603
The development of laterality in the forebrain projections of midline thalamic cell groups in the rat M. Takada, Gord Fishell, Z.K. Li, Derek Van der Kooy and T. Hattori Department of Anatoray, Universityof Toronto, Toronto, Ont. (Canada) (Accepted 24 February 1987)
Key words: Midline thalamic nucleus; Forebrain; Laterality; Development; Cell migration; Retrograde double labeling; Rat
Bilateral forebrain (caudoputamen, nucleus accumbens and frontal cortical areas) injections of two different fluorescent retrograde tracers demonstrated that labeled cells situated in the midline nuclei of the thalamus and midbrain each project only unilaterally to the forebrain, regardless of the laterality of their perikarya. Thus, these intermingling midline perikarya send their axons primarily ipsilaterally and to a lesser degree contralaterally, but never bilaterally to the forebrain. At embryonic day 19, these midline nuclei exist as two bilaterally situated, independent structures, each projecting only ipsilaterally to the forebrain. By postnatal day 2, these perikarya fuse into a single mass on the midline. Upon fusion, many of the perikarya of the two developing subnuclei cross the midline, intermingle with each other, and thus some neurons come to have contralateral forebrain projections. These observations suggest that neurons are able to maintain their axonal projections while migrating short distances. INTRODUCTION Bilaterally situated cell groups, which lie on or across the midline of the brain, include the centromedial (Ce), rhomboid (Rh), reuniens (Re), paraventricular (Pv) and paratenial (Pt) nuclei of the thalamus and interfascicular (IF) and central linear (CL) nuclei of the midbrain. Neurons in all of these midline nuclei project to forebrain structures such as the caudoputamen (CPU) 1'2'4-6'10'15'18-20'22-24'27,nucleus accumbens (Acc) 1'2'4'5'12'13'16'21and frontal cortical areas (FC) 2'6'10'11'21. Retrograde tracing studies suggest that the unilateral forebrain injections result in neuronal labeling of the cell groups predominantly ipsilateral to the injections 1'4-6'10-13'15'16'18-24'27. However, a considerable n u m b e r of neurons on the midline and even contralateral to the forebrain injections can also be labeled 1'4'5'10'13'15'16'18-22'27. It seemed likely that single neurons located on the midline or contralateral to the injections might have bilateral projections to the forebrain. Contrary to this supposition, the present study using a fluorescent retrograde double-labeling tech-
nique has demonstrated that the adult axonal projections of these thalamic and midbrain cell groups to the forebrain are strictly unilateral, independent of the laterality of their cell bodies. Previous studies 9'25 on dorsal raphe projections to the forebrain came to a similar conclusion. The serotonin-containing axons of the dorsal raphe neurons are known to reach the forebrain very early in development (by embryonic day 15) 14. The dorsal raphe nucleus can be recognized as two bilaterally situated, independent structures early in development (embryonic day 18), which then fuse into a single mass on the midline (postnatal days 1-3) 7'8. Given a somewhat similar embryological maturation for the midline thalamic nuclei 17, we can ask about the developmental mechanisms that produce the bilateral organization of midline thalamic nuclei cells that have strictly unilateral forebrain axonal projections. Three developmental strategies seem possible. First, before midline fusion of the two subnuclei, there may be only ipsilaterally projecting cells. In this scenario, fusion involves the intermingling of the
Correspondence: M. Takada, Department of Anatomy, University of Toronto, Medical Sciences Building, Toronto, Ontario M5S 1A8, Canada. 0165-3806/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
276 projection cells across the midline, creating a new population of cells with only contralateral projections. Second, before midline fusion of the two subnuclei, many of the cells may have bilateral projections. In this case, after fusion there may be trophic support for only unilateral projections for each cell. The probability of retracting an ipsilateral axon collateral after fusion would decrease with distance from the midline. Third, before midline fusion of the two subnuclei, there may already be present in each subnucleus ipsilaterally and contralaterally projecting cells (or cells that are committed to these specific projections). In this case, fusion is nothing more than the two subnuclei meeting edge to edge on the midline. Employing the short- and long-term survival of rat pups injected at embryonic day 18 with different fluorescent retrograde tracers bilaterally into the forebrain, we now provide evidence supporting the first developmental strategy listed above. During subnuclei fusion, there is intermingling across the midline of some strictly ipsilaterally projecting perikarya, thus creating a new population of cells with only contralateral forebrain projections. MATERIALS AND METHODS
Adult injections Eleven male adult albino rats (Wistar, 250-300 g) were used for this study. Under anesthesia with sodium pentobarbital (65 mg/kg, i.p.), these rats were each injected with two fluorescent retrograde tracers: True blue (TB, 5% in distilled water) and diamidino yellow (DY, 3% in distilled water). Single TB injections were made stereotaxically into the CPU on one side of the brain, while single DY injections were made into the CPU on the opposite side of the brain (4 rats). A total volume of 0.2-0.8/A of each tracer was deposited slowly through a 1-ktl Hamilton microsyringe. Other rats received the same combinations of injections into the Acc (4 rats) and FC (3 rats) on both sides of the brain. After 2-4-day survival periods, the animals were reanesthetized deeply and perfused transcardially with 300 mi of 10% formalin in 0.1 M phosphate buffer (pH 7.4). The brains were removed immediately and saturated with 25% sucrose in the same buffer at 4 °C. Coronal frozen sections were serially cut at 40 ,um thickness on a cryostat, mounted onto clean slides, and then
observed using a Leitz fluorescence microscope. A filter providing excitation light of ca. 360 nm wave length was used to examine the blue-fluorescent TBcontaining cells and the yellow-fluorescent DY-containing cells.
Embryonic injections Injections into prenatal rats 3 were made using 5 embryonic day 18 (El8) timed pregnant rats. The pregnant rats were anesthetized and laparotomized. Each fetus was oriented inside the uterus and the dorsal calvarian suture lines were used to estimate the location of the forebrain. After puncturing the uterine wall with a small needle, a 1-/~1 Hamilton microsytinge was used to make single injections of 0.4/A of 5% TB and 3% DY into bilateral forebrain structures including the CPU, Acc and FC. The hole in the uterine wall was sealed with cyanoacrylic adhesive. The pregnant rats were sutured and allowed to recover. After a 24-h survival period (at E19), 2 of the 5 pregnant rats used were reanesthetized deeply and a caesarean section was performed. Previously injected fetal rat pups were removed and immediately perfused transcardially with 5 ml of 20% formalin in 0.1 M phosphate buffer (pH 7.4). The brains were removed, postfixed in the same fixative for 4 h, and then immersed in the same buffer containing 25% sucrose. Forty-ktm-thick frozen sections were serially cut in the coronal plane, mounted onto clean slides, and then observed using a Leitz fluorescence microscope under light at 360 nm excitation. One of the pregnant rats was again given a laparotomy under general anesthesia 24 h after the first operation. A unilateral hemitransection was performed using a small knife on each fetal brain which had previously received bilateral tracer injections into the forebrain. The knife cut was made caudal to the tracer injection site at the level of the rostral-most thalamus. The remaining 3 mothers were then allowed to give birth. Prenatally injected pups were anesthetized using cold temperature, sacrificed at postnatal day 2 (P2), and then processed in a manner identical to the methods used for prenatal fetuses. RESULTS
Adult injections Following bilateral injections into the CPU, retro-
277 gradely labeled cells were found in the midline thalamic regions including the Ce, Rh, Re and Pv, as well as in the intralaminar thalamic nuclei. In the case of the intralaminar thalamic nuclei, cell labeling was strictly ipsilateral to the injection site. Although labeled neurons in the midline thalamic nuclei were located predominantly ipsilaterally to each injection, there was also a substantial population distributed contralaterally. Thus, the distribution of TB-labeled cells from one CPU injection overlapped that of DYpositive neurons from the other CPU injection in the mediolateral plane throughout the midline thalamic nuclei. Neurons labeled with either TB or DY were not only intermingled with each other on the midline of the thalamus, but were also seen contralateral to each injection up to a maximum of 500/~m from the midline. Interestingly, double-labeled cells were never detected in the midline thalamic regions, in spite of the extensive overlapping of the distributions of the labeled cells (Figs. 1 and 2a). Thus, individual neurons in the midline thalamic areas project only unilaterally (either ipsilateraUy or contralateraUy), but not bilaterally to the CPU. The results after bilateral injections into the Acc were similar to those with the CPU injections described above. On the midline, a mediolateral overlapping of the distribution of TB- and DYlabeled cells was seen in the Pv and Pt of the thalamus (Fig. 2b). The Acc injections also resulted in the labeling of Ce, Rh and Re neurons, although labeled neurons in these nuclei were less in number than in the case of the CPU injections. Double-labeled cells were absent from the overlapping distribution areas in the midline nuclei. A number of neurons in the IF (Fig. 2c) and CL (Fig. 2d,e) of the midbrain were labeled from the Acc injections, in contrast with the case of the CPU injections which labeled few neurons here. There was almost a complete overlap in the IF and CL of the distributions of neurons labeled with either TB or DY. Nevertheless, TB-labeled cells again appeared totally separate from DY-positive cells. The data from injections into the bilateral FC were again similar to those from bilateral CPU and Acc injections described above. Neurons in the intralaminar thalamic nuclei were labeled strictly ipsilaterally. Conversely, throughout the midline thalamic (including the Ce, Rh, Re, Pv and Pt) and midbrain (in-
'~
MD
)
..,O,o.0oO\oo -
o~_~ o ..... •..-,o,~-,O;o - .• ,.~•eoili~'o • oo~:~,%OoO oo, o©o / •
~-
oO 0
oo • • , oo ~'~Rh ~
H
:o o°o '~ ~oOo
a
n /
.~
.~
l
"~,: I
o~
b
P~
Hb ,.,
MD
~
M
o•
C ._
. e o"
n
c
d
Fig. 1. Schematic drawings of the distribution of projection neurons in the midline thalamic nuclei following bilateral C P U injections. TB-positive cells from the left C P U are represented with solid circles, and DY-positive cells from the right CPU with open circles. Neurons in these areas are all single-labeled with either TB or DY regardless of their laterality. The drawings are arranged rostrocaudally from a to d. FR, fasciculus
retroflexus; Hb, habenula; MD, mediodorsal nucleus; MT, mammillothalamic tract; SM, stria medullaris; VM, ventromedial nucleus; 3V, third ventricle. cluding the IF and CL) nuclei, considerable overlapping of the distributions of TB- and DY-positive neurons was observed in the mediolateral plane (Fig. 3). Once again, the FC injections did not produce any double-labeling of these cells in the midline thalamic and midbrain nuclei.
Embryonic injections After bilateral injections into the forebrain of E18 prenatal rats, the distributions of labeled neurons in the thalamus and midbrain at E l 9 and P2 were examined. At El9, in the midline and intralaminar thalam-
278
Fig. 2. Photomicrographs of retrogradely labeled cells in the midline nuclei of the thalamus and midbrain after bilateral injections into the CPU or Acc. TB injections were located on the left side, and DY injections on the right side of the brain and of the photomicrographs. Small arrows indicate TB-positive cells, while arrowheads indicate DY-positive cells, a: neurons in the Ce single-labeled with either TB or DY following bilateral injections into the CPU. b: labeled neurons in the Pv following bilateral Acc injections. Large arrows in a and b point to the midline of the thalamus, c,d: the IF (c) and CL (d) of the midbrain following bilateral Acc injections, e: higher magnification of cells in the CL single-labeled with either TB or DY. CL, central linear nucleus; IF, interfascicular nucleus; IP, interpeduncular nucleus; VTA, ventral tegmental area. a,b,e, x 110; c,d, x40. ic regions all of the r e t r o g r a d e l y l a b e l e d p e r i k a r y a w e r e l o c a t e d strictly ipsilaterally to t h e i r i n j e c t i o n
and D Y - l a b e l i n g was v e r y sharply d e l i n e a t e d (Fig. 4a,b). T h e s e p a r a t i o n of t h e l a b e l e d cells in t h e s e nu-
sites. T h e b o u n d a r y on t h e m i d l i n e b e t w e e n the T B -
clei was m o r e d r a m a t i c at the c a u d a l t h a n at the ros-
279 several cells labeled with DY injected into the forebrain on the opposite side, and vice versa (Fig. 4d,e). Contralaterally situated neurons at P2 were neither as numerous nor situated as far contralaterally as their adult counterparts (a maximum of 200 # m from the midline at P2, compared to 500/~m from the midline in the adult). However, as in the adult, no double-labeled (bilaterally projecting) neurons were detected. A unilateral hemitransection between the injection site and the rostral-most thalamus at El9, produced results which were very similar to those from bilateral forebrain injections at E l 8 with no hemitransection. Thus, at P2, neurons single-labeled with either TB or DY were intermingled with each other across the midline of the thalamus and midbrain. Again, no double-labeling of cells was detectable. The prenatal hemitransection also did not affect the midline intermingling of forebrain projecting dorsal raphe perikarya 9'25, studied in the same rat pups at P2 (not depicted). DISCUSSION Fig. 3. Photomicrographs of retrogradely labeled neurons in the midline thalamic nuclei after bilateral injections into the FC. TB injections were made on the left side, and DY iniections on the right side of the brain and of the photomicrographs. Small arrows indicate TB-positive cells and arrowheads DYpositive cells, a: neurons in the Ce single-labeled with either TB or DY. b: labeled neurons in the Pv. Large arrows in a and b point to the midline of the thalamus, x 100. tral level of the thalamus. Indeed, at the level of the caudal thalamus at this age, a faintly autofluorescent structure was seen on the midline (Fig. 4a,b) separating the two populations of the labeled cells. In the rostral portions of the thalamus, cell bodies single labeled with either TB or DY did show some intermingling with each other on the midline at E l 9 (Fig. 4c). In contrast, the separation of the TB- and DY-labeling was definite and absolute on the midline of the caudal thalamus (Fig. 4a,b) and of the midbrain (in the IF and CL; not depicted). After sacrifice at P2, in comparison with E19, a relatively large overlapping of the two fluorescent-positive cell populations was found on the midline of the thalamus and midbrain. Thus, neurons labeled with either TB or DY were intermingled with each other on the midline. Furthermore, TB-positive cell areas labeled from the ipsilateral injections also contained
The present study has provided evidence that cell groups situated on the midline of the rat thalamus and midbrain each project only unilaterally to forebrain structures (including the CPU, Acc and FC), regardless of the laterality of their perikarya. Thus, they send their axons primarily ipsilaterally and to a lesser extent contralaterally, but never bilaterally to forebrain structures. Previous studies on the dorsal raphe projections of the adult rat 9'25 have provided support for the notion that midline-situated brain perikarya may generally have only unilateral ascending projections. The present data on thalamocortical projections of the adult rat are in disagreement with a recent report H, which has suggested that cortical projection neurons in the midline thalamic nuclei of the adult rat were located only ipsilaterally and never crossed the midline. However, intermingling of thalamic cells across the midline was observed in the newborn rat in this previous study 11. In the developmental portion of the present study, we have observed that the midline nuclei of the thalamus and midbrain exist at E l 9 as two bilaterally situated, independent structures each projecting only ipsilaterally to the forebrain. These nuclear groups
280
Fig. 4. Photomicrographs of retrogradely labeled neurons in the midline thalamic regions at E l 9 and P2 following bilateral forebrain injections at El8. TB injections were made on the left side, and D Y injections on the right side of the brain and of the photomicrographs. Small arrows indicate TB-positive cells and arrowheads DY-positive cells, a: the caudal thalamus at El9. A very sharply delineated border on the midline is seen between the TB- and DY-labeling. b: higher magnification of labeled thalamic cells in a. In a and b, a faintly autofluorescent structure is seen on the midline, c: the rostral thalamus at El9. TB- and DY-positive n e u r o n s are intermingled with each other on the midline, d: the caudal thalamus at P2. The degree of intermingling at this age is transitional between the fetus (Fig. 4a,b) and the adult (Figs. 2a and 3a). e: higher magnification of thalamic cells single labeled with either TB or D Y in d. Large arrows in a - e point to the midline of the thalamus, a,d, x 42: b,c,e, × 118.
281 fuse into a mass on the midline by P2. This time course is similar to the fusion time course reported in previous developmental work on the dorsal raphe nucleus7's. By P2, some midline thalamic, midbrain and raphe perikarya with initially ipsilateral forebrain projections have crossed the midline to intermingle with cells from the opposite side, and thus acquire contralateral axonal projections. The absence of contralaterally or bilaterally labeled cells at El9 would seem to rule out other possible developmental mechanisms (see Introduction) for organizing the midline nuclei and their projections. It remains possible that the contralaterally projecting cells do not send out their axons as early as the ipsilaterally projecting cells (which can be labeled by El9). This would imply a long-term, continual uptake from the injection site in animals injected at E l 8 and sacrificed at P2. However, our recent long-term survival experiments have provided evidence that the effective uptake time for the retrograde tracers may be less than 2 days 3. The prenatal hemitransection experiment reported here strongly suggests that perikarya with contralateral axonal projections postnatally already exist as ipsilaterally projecting cells at El9. Thus, a unilateral hemitransection at E l 9 (blocking further retrograde transport from E l 8 forebrain injection sites) did not prevent intermingling of labeled cells across the midline of the thalamus and midbrain at P2. The data propose that perikarya situated contralateral to their adult forebrain projections must migrate towards and across the midline during early development. Furthermore, there may be a rostrocaudal gradient in this developmental process. In the rostral portions of the thalamus, the contralaterally situated perikarya must have begun to migrate and intermingle on the midline even before El9. A similar gradient has been shown in the development of serotonergic raphe perikarya 7,8. This rostrocaudal cell migration gradient contrasts with the caudorostral gradient seen in other neural development processes such as closure of the neural tube and the growth of axons. The rostrocaudal migration of cells towards and across the midline of the thalamus that is reported here even contrasts with the midline fusion of the thalamic tissue itself, which occurs earlier in the embryonic period in a caudorostral gradient 17. The rostrocaudal course of midline nuclei cell migration might depend on the distance between the labeled
cells and forebrain. Thus, the rostral thalamic perikarya (closest to the forebrain) may be able to receive neurotrophic signals from the forebrain earlier than the caudal thalamic perikarya. However, our hemitransection studies suggest that if forebrain neurotrophic factors are important in the rostrocaudal midline cell migration gradient, then their signals must have reached the midline thalamic, midbrain and raphe nuclei before El9. Employing immunohistochemical staining against the S-100 protein, a transient glial structure has recently been described in the rat midline raphe during pre- and early postnatal development 26. This structure extends rostrally as far as the third ver]tricle26. It seems possible that the white (autofluorescent) linear structure seen on the midline of the E l 9 caudal thalamus (Fig. 4a,b), but not seen on the midline of the P2 caudal thalamus (Fig. 4d,e), is the rostralmost portion of this transient glial structure. The onset of cell migration towards and across the midline of the caudal thalamus seems to coincide with the disappearance of this autofluorescent structure (Fig. 4). The autofluorescent linear structure is not visible on the midline of the El9 rostral thalamus, nor can the transient glial structure be stained with S-100 antisera in the rostral thalamus 26. In the absence of this structure, intermingling of cells on the rostral thalamic midline has begun to occur by El9 (Fig. 4c). The perinatal presence of a transient glial barrier in the caudal thalamus may underlie the rostrocaudal midline cell migration gradient in the thalamus. Thus, there might be two consecutive events in the thalamic midline early in development. First, bilaterally situated, independent thalamic tissues fuse in a caudorostral gradient into a single mass on the midline. Second, a transient glial structure next forms a midline barrier within only the caudal thalamus, which then results in a rostrocaudal gradient of thalamic perikaryal migration towards and across the midline. In summary, achievement of the adult forebrain projection pattern involves a pre- and early postnatal perikaryal migration in cell groups situated on the midline of the rat thalamus and midbrain. The adult role of the unilateral (either ipsilateral or contralateral) forebrain projections of these midline situated neurons remains unknown. However, the midline intermingling of perikarya with a mixed laterality of
282 f o r e b r a i n p r o j e c t i o n s m a y allow for an i n f o r m a t i o n
ACKNOWLEDGEMENTS
p r o c e s s i n g c a p a c i t y which w o u l d be i m p o s s i b l e if the midline nuclei r e m a i n e d in their e m b r y o n i c f o r m as two bilaterally s i t u a t e d , i n d e p e n d e n t structures.
REFERENCES 1 Beckstead, R.M., The thalamostriatal projection in the cat, J. Comp. Neurol., 223 (1984) 313-346. 2 Cowan, W.M. and Powell, T.P.S., The projections of the midline and intralaminar nuclei of the thalamus of the rabbit, J. Neurol. Neurosurg. Psychiat., 18 (1955) 266-279. 3 Fishell, G. and Van der Kooy, D., Pattern formation in the striatum: developmental changes in the distribution of striatonigral neurons, J. Neurosci., in press. 4 Groenewegen, H.J., Becker, N.E.H.M. and Lohman, A.H.M., Subcortical afferents of the nucleus accumbens septi in the cat, studied with retrograde axonal transport of horseradish peroxidase and bisbenzimid, Neuroscience, 5 (1980) 1903-1916. 5 Jayaraman, A., Organization of thalamic projections in the nucleus accumbens and the caudate nucleus in cats and its relation with hippocampal and other subcortical afferents, J. Comp. Neurol., 231 (1985) 396-420. 6 Jones, E.G. and Leavitt, R.Y., Retrograde axonal transport and the demonstration of non-specific projections to the cerebral cortex and striatum from the thalamic intralaminar nuclei in the rat, cat and monkey, J. Comp. Neurol., 154 (1974) 349-378. 7 Levitt, P, and Moore, R.Y., Developmental organization of raphe serotonin neuron groups in the rat, Anat. Embryol., 154 (1978) 241-251. 8 Levitt, P. and Moore, R.Y., Developmental organization of serotonin-containing cell group within the brainstem raphe of the rat, Soc. Neurosci. Abstr., 4 (1978) 277. 9 Loughlin, S.E. and Fallon, J.H., Mesostriatal projections from ventral tegmentum and dorsal raphe: cells project ipsilaterally or contralaterally but not bilaterally, Neurosci. Lett., 32 (1982) 11-16. 10 Macchi, G., Bentivoglio, M., Molinari, M. and Minciacchi, D., The thalamo-caudate versus thalamo-cortical projections as studied in the cat with fluorescent retrograde double labeling, Exp. Brain Res., 54 (1984) 225-239. 11 Minciacchi, D., Granato, A., Molinari, M. and Macchi, G., Bilateral thalamo-cortical projections in the newborn rat, Neurosci. Lett. Suppl., 22 (1985) $39. 12 Nauta, W.J.H., Smith, G.P., Faull, R.L.M. and Domesick, V.B., Efferent connections and nigral afferents of the nucleus accumbens septi in the rat, Neuroscience. 3 (1978) 385-401. 13 Newman, R. and Winans, S.S., An experimental study of the ventral striatum of the golden hamster. I. Neuronal connections of the nucleus accumbens, J. Cornp. Neurol., 191 (1980) 167-192.
S u p p o r t e d by the M e d i c a l R e s e a r c h C o u n c i l of Canada. M . T . is an M R C P o s t p r o f e s s i o n a l Fellow.
14 Olson, L. and Seiger, ,&., Early prenatal ontogeny of central monoamine neurons in the rat: fluorescence histochemical observations, Z. Anat. Entwickl..Gesch., 137 (1972) 301-316. 15 Parent, A., Mackey, A. and De Bellefeuille, L., The subcortical afferents to caudate nucleus and putamen in primate. A fluorescence retrograde double labeling study, Neuroscience, 10 (1983) 1137-1150. 16 Phillipson, O.T. and Griffiths, A.C., The topographic order of inputs to nucleus accumbens in the rat, Neuroscience, 16 (1985) 275-296. 17 Rose, J.E., The ontogenetic development of the rabbit's diencephalon, J. Comp. Neurol., 77 (1942) 61-129. 18 Royce, G.J., Cells of origin of subcortical afferents to the caudate nucleus: a horseradish peroxidase study in the cat, Brain Res., 153 (1978) 465-475. 19 Sato, M., Itoh, K. and Mizuno, N., Distribution of thalamo-caudate neurons in the cat as demonstrated by horseradish peroxidase, Exp. Brain Res., 34 (1979) 143-153. 20 Smith, Y. and Parent, A., Differential connections of caudate nucleus and putamen in the squirrel monkey (Saimiri sciureus), Neuroscience, 18 (1986) 347-371. 21 Swanson, L.W., The projections of the ventral tegmental area and adjacent regions: a combined fluorescent retrograde tracer and immunofluorescence study in the rat, Brain Res. Bull., 9 (1982) 321-353. 22 Takada, M., Itoh, K., Sugimoto, T. and Mizuno, N., Topographical projections from the thalamus to the putamen in the cat, Neurosci. Lett., 54 (1985) 207-212. 23 Takada, M., Itoh, K., Yasui, Y., Sugimoto, T. and Mizuno, N., Topographical projections from the posterior thalamic regions to the striatum in the cat, with reference to possible tecto-thalamo-striatal connections, Exp. Brain Res., 60 (1985) 385-396. 24 Van der Kooy, D., The organization of the thalamic, nigral and raphe cells projecting to the medial vs lateral caudateputamen in the rat. A fluorescent retrograde double labeling study, Brain Res., 169 (1979) 381-387. 25 Van der Kooy, D. and Hattori, T., Bilaterally situated dorsal raphe cell bodies have only unilateral forebrain projections in rat, Brain Res., 192 (1980) 550-554. 26 Van Hartesveldt, C., Moore, B. and Hartman, B.K., Transient midline raphe glial structure in the developing rat, J. Comp. Neurol., 253 (1986) 175-184. 27 Veening, J.G., Cornelissen, F.M. and Lieven, P.A.J.M., The topical organization of the afferents to the caudatoputamen of the rat. A horseradish peroxidase study, Neuroscience, 5 (1980) 1253-1268.
275
BRD 50603
The development of laterality in the forebrain projections of midline thalamic cell groups in the rat M. Takada, Gord Fishell, Z.K. Li, Derek Van der Kooy and T. Hattori Department of Anatoray, Universityof Toronto, Toronto, Ont. (Canada) (Accepted 24 February 1987)
Key words: Midline thalamic nucleus; Forebrain; Laterality; Development; Cell migration; Retrograde double labeling; Rat
Bilateral forebrain (caudoputamen, nucleus accumbens and frontal cortical areas) injections of two different fluorescent retrograde tracers demonstrated that labeled cells situated in the midline nuclei of the thalamus and midbrain each project only unilaterally to the forebrain, regardless of the laterality of their perikarya. Thus, these intermingling midline perikarya send their axons primarily ipsilaterally and to a lesser degree contralaterally, but never bilaterally to the forebrain. At embryonic day 19, these midline nuclei exist as two bilaterally situated, independent structures, each projecting only ipsilaterally to the forebrain. By postnatal day 2, these perikarya fuse into a single mass on the midline. Upon fusion, many of the perikarya of the two developing subnuclei cross the midline, intermingle with each other, and thus some neurons come to have contralateral forebrain projections. These observations suggest that neurons are able to maintain their axonal projections while migrating short distances. INTRODUCTION Bilaterally situated cell groups, which lie on or across the midline of the brain, include the centromedial (Ce), rhomboid (Rh), reuniens (Re), paraventricular (Pv) and paratenial (Pt) nuclei of the thalamus and interfascicular (IF) and central linear (CL) nuclei of the midbrain. Neurons in all of these midline nuclei project to forebrain structures such as the caudoputamen (CPU) 1'2'4-6'10'15'18-20'22-24'27,nucleus accumbens (Acc) 1'2'4'5'12'13'16'21and frontal cortical areas (FC) 2'6'10'11'21. Retrograde tracing studies suggest that the unilateral forebrain injections result in neuronal labeling of the cell groups predominantly ipsilateral to the injections 1'4-6'10-13'15'16'18-24'27. However, a considerable n u m b e r of neurons on the midline and even contralateral to the forebrain injections can also be labeled 1'4'5'10'13'15'16'18-22'27. It seemed likely that single neurons located on the midline or contralateral to the injections might have bilateral projections to the forebrain. Contrary to this supposition, the present study using a fluorescent retrograde double-labeling tech-
nique has demonstrated that the adult axonal projections of these thalamic and midbrain cell groups to the forebrain are strictly unilateral, independent of the laterality of their cell bodies. Previous studies 9'25 on dorsal raphe projections to the forebrain came to a similar conclusion. The serotonin-containing axons of the dorsal raphe neurons are known to reach the forebrain very early in development (by embryonic day 15) 14. The dorsal raphe nucleus can be recognized as two bilaterally situated, independent structures early in development (embryonic day 18), which then fuse into a single mass on the midline (postnatal days 1-3) 7'8. Given a somewhat similar embryological maturation for the midline thalamic nuclei 17, we can ask about the developmental mechanisms that produce the bilateral organization of midline thalamic nuclei cells that have strictly unilateral forebrain axonal projections. Three developmental strategies seem possible. First, before midline fusion of the two subnuclei, there may be only ipsilaterally projecting cells. In this scenario, fusion involves the intermingling of the
Correspondence: M. Takada, Department of Anatomy, University of Toronto, Medical Sciences Building, Toronto, Ontario M5S 1A8, Canada. 0165-3806/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
276 projection cells across the midline, creating a new population of cells with only contralateral projections. Second, before midline fusion of the two subnuclei, many of the cells may have bilateral projections. In this case, after fusion there may be trophic support for only unilateral projections for each cell. The probability of retracting an ipsilateral axon collateral after fusion would decrease with distance from the midline. Third, before midline fusion of the two subnuclei, there may already be present in each subnucleus ipsilaterally and contralaterally projecting cells (or cells that are committed to these specific projections). In this case, fusion is nothing more than the two subnuclei meeting edge to edge on the midline. Employing the short- and long-term survival of rat pups injected at embryonic day 18 with different fluorescent retrograde tracers bilaterally into the forebrain, we now provide evidence supporting the first developmental strategy listed above. During subnuclei fusion, there is intermingling across the midline of some strictly ipsilaterally projecting perikarya, thus creating a new population of cells with only contralateral forebrain projections. MATERIALS AND METHODS
Adult injections Eleven male adult albino rats (Wistar, 250-300 g) were used for this study. Under anesthesia with sodium pentobarbital (65 mg/kg, i.p.), these rats were each injected with two fluorescent retrograde tracers: True blue (TB, 5% in distilled water) and diamidino yellow (DY, 3% in distilled water). Single TB injections were made stereotaxically into the CPU on one side of the brain, while single DY injections were made into the CPU on the opposite side of the brain (4 rats). A total volume of 0.2-0.8/A of each tracer was deposited slowly through a 1-ktl Hamilton microsyringe. Other rats received the same combinations of injections into the Acc (4 rats) and FC (3 rats) on both sides of the brain. After 2-4-day survival periods, the animals were reanesthetized deeply and perfused transcardially with 300 mi of 10% formalin in 0.1 M phosphate buffer (pH 7.4). The brains were removed immediately and saturated with 25% sucrose in the same buffer at 4 °C. Coronal frozen sections were serially cut at 40 ,um thickness on a cryostat, mounted onto clean slides, and then
observed using a Leitz fluorescence microscope. A filter providing excitation light of ca. 360 nm wave length was used to examine the blue-fluorescent TBcontaining cells and the yellow-fluorescent DY-containing cells.
Embryonic injections Injections into prenatal rats 3 were made using 5 embryonic day 18 (El8) timed pregnant rats. The pregnant rats were anesthetized and laparotomized. Each fetus was oriented inside the uterus and the dorsal calvarian suture lines were used to estimate the location of the forebrain. After puncturing the uterine wall with a small needle, a 1-/~1 Hamilton microsytinge was used to make single injections of 0.4/A of 5% TB and 3% DY into bilateral forebrain structures including the CPU, Acc and FC. The hole in the uterine wall was sealed with cyanoacrylic adhesive. The pregnant rats were sutured and allowed to recover. After a 24-h survival period (at E19), 2 of the 5 pregnant rats used were reanesthetized deeply and a caesarean section was performed. Previously injected fetal rat pups were removed and immediately perfused transcardially with 5 ml of 20% formalin in 0.1 M phosphate buffer (pH 7.4). The brains were removed, postfixed in the same fixative for 4 h, and then immersed in the same buffer containing 25% sucrose. Forty-ktm-thick frozen sections were serially cut in the coronal plane, mounted onto clean slides, and then observed using a Leitz fluorescence microscope under light at 360 nm excitation. One of the pregnant rats was again given a laparotomy under general anesthesia 24 h after the first operation. A unilateral hemitransection was performed using a small knife on each fetal brain which had previously received bilateral tracer injections into the forebrain. The knife cut was made caudal to the tracer injection site at the level of the rostral-most thalamus. The remaining 3 mothers were then allowed to give birth. Prenatally injected pups were anesthetized using cold temperature, sacrificed at postnatal day 2 (P2), and then processed in a manner identical to the methods used for prenatal fetuses. RESULTS
Adult injections Following bilateral injections into the CPU, retro-
277 gradely labeled cells were found in the midline thalamic regions including the Ce, Rh, Re and Pv, as well as in the intralaminar thalamic nuclei. In the case of the intralaminar thalamic nuclei, cell labeling was strictly ipsilateral to the injection site. Although labeled neurons in the midline thalamic nuclei were located predominantly ipsilaterally to each injection, there was also a substantial population distributed contralaterally. Thus, the distribution of TB-labeled cells from one CPU injection overlapped that of DYpositive neurons from the other CPU injection in the mediolateral plane throughout the midline thalamic nuclei. Neurons labeled with either TB or DY were not only intermingled with each other on the midline of the thalamus, but were also seen contralateral to each injection up to a maximum of 500/~m from the midline. Interestingly, double-labeled cells were never detected in the midline thalamic regions, in spite of the extensive overlapping of the distributions of the labeled cells (Figs. 1 and 2a). Thus, individual neurons in the midline thalamic areas project only unilaterally (either ipsilateraUy or contralateraUy), but not bilaterally to the CPU. The results after bilateral injections into the Acc were similar to those with the CPU injections described above. On the midline, a mediolateral overlapping of the distribution of TB- and DYlabeled cells was seen in the Pv and Pt of the thalamus (Fig. 2b). The Acc injections also resulted in the labeling of Ce, Rh and Re neurons, although labeled neurons in these nuclei were less in number than in the case of the CPU injections. Double-labeled cells were absent from the overlapping distribution areas in the midline nuclei. A number of neurons in the IF (Fig. 2c) and CL (Fig. 2d,e) of the midbrain were labeled from the Acc injections, in contrast with the case of the CPU injections which labeled few neurons here. There was almost a complete overlap in the IF and CL of the distributions of neurons labeled with either TB or DY. Nevertheless, TB-labeled cells again appeared totally separate from DY-positive cells. The data from injections into the bilateral FC were again similar to those from bilateral CPU and Acc injections described above. Neurons in the intralaminar thalamic nuclei were labeled strictly ipsilaterally. Conversely, throughout the midline thalamic (including the Ce, Rh, Re, Pv and Pt) and midbrain (in-
'~
MD
)
..,O,o.0oO\oo -
o~_~ o ..... •..-,o,~-,O;o - .• ,.~•eoili~'o • oo~:~,%OoO oo, o©o / •
~-
oO 0
oo • • , oo ~'~Rh ~
H
:o o°o '~ ~oOo
a
n /
.~
.~
l
"~,: I
o~
b
P~
Hb ,.,
MD
~
M
o•
C ._
. e o"
n
c
d
Fig. 1. Schematic drawings of the distribution of projection neurons in the midline thalamic nuclei following bilateral C P U injections. TB-positive cells from the left C P U are represented with solid circles, and DY-positive cells from the right CPU with open circles. Neurons in these areas are all single-labeled with either TB or DY regardless of their laterality. The drawings are arranged rostrocaudally from a to d. FR, fasciculus
retroflexus; Hb, habenula; MD, mediodorsal nucleus; MT, mammillothalamic tract; SM, stria medullaris; VM, ventromedial nucleus; 3V, third ventricle. cluding the IF and CL) nuclei, considerable overlapping of the distributions of TB- and DY-positive neurons was observed in the mediolateral plane (Fig. 3). Once again, the FC injections did not produce any double-labeling of these cells in the midline thalamic and midbrain nuclei.
Embryonic injections After bilateral injections into the forebrain of E18 prenatal rats, the distributions of labeled neurons in the thalamus and midbrain at E l 9 and P2 were examined. At El9, in the midline and intralaminar thalam-
278
Fig. 2. Photomicrographs of retrogradely labeled cells in the midline nuclei of the thalamus and midbrain after bilateral injections into the CPU or Acc. TB injections were located on the left side, and DY injections on the right side of the brain and of the photomicrographs. Small arrows indicate TB-positive cells, while arrowheads indicate DY-positive cells, a: neurons in the Ce single-labeled with either TB or DY following bilateral injections into the CPU. b: labeled neurons in the Pv following bilateral Acc injections. Large arrows in a and b point to the midline of the thalamus, c,d: the IF (c) and CL (d) of the midbrain following bilateral Acc injections, e: higher magnification of cells in the CL single-labeled with either TB or DY. CL, central linear nucleus; IF, interfascicular nucleus; IP, interpeduncular nucleus; VTA, ventral tegmental area. a,b,e, x 110; c,d, x40. ic regions all of the r e t r o g r a d e l y l a b e l e d p e r i k a r y a w e r e l o c a t e d strictly ipsilaterally to t h e i r i n j e c t i o n
and D Y - l a b e l i n g was v e r y sharply d e l i n e a t e d (Fig. 4a,b). T h e s e p a r a t i o n of t h e l a b e l e d cells in t h e s e nu-
sites. T h e b o u n d a r y on t h e m i d l i n e b e t w e e n the T B -
clei was m o r e d r a m a t i c at the c a u d a l t h a n at the ros-
279 several cells labeled with DY injected into the forebrain on the opposite side, and vice versa (Fig. 4d,e). Contralaterally situated neurons at P2 were neither as numerous nor situated as far contralaterally as their adult counterparts (a maximum of 200 # m from the midline at P2, compared to 500/~m from the midline in the adult). However, as in the adult, no double-labeled (bilaterally projecting) neurons were detected. A unilateral hemitransection between the injection site and the rostral-most thalamus at El9, produced results which were very similar to those from bilateral forebrain injections at E l 8 with no hemitransection. Thus, at P2, neurons single-labeled with either TB or DY were intermingled with each other across the midline of the thalamus and midbrain. Again, no double-labeling of cells was detectable. The prenatal hemitransection also did not affect the midline intermingling of forebrain projecting dorsal raphe perikarya 9'25, studied in the same rat pups at P2 (not depicted). DISCUSSION Fig. 3. Photomicrographs of retrogradely labeled neurons in the midline thalamic nuclei after bilateral injections into the FC. TB injections were made on the left side, and DY iniections on the right side of the brain and of the photomicrographs. Small arrows indicate TB-positive cells and arrowheads DYpositive cells, a: neurons in the Ce single-labeled with either TB or DY. b: labeled neurons in the Pv. Large arrows in a and b point to the midline of the thalamus, x 100. tral level of the thalamus. Indeed, at the level of the caudal thalamus at this age, a faintly autofluorescent structure was seen on the midline (Fig. 4a,b) separating the two populations of the labeled cells. In the rostral portions of the thalamus, cell bodies single labeled with either TB or DY did show some intermingling with each other on the midline at E l 9 (Fig. 4c). In contrast, the separation of the TB- and DY-labeling was definite and absolute on the midline of the caudal thalamus (Fig. 4a,b) and of the midbrain (in the IF and CL; not depicted). After sacrifice at P2, in comparison with E19, a relatively large overlapping of the two fluorescent-positive cell populations was found on the midline of the thalamus and midbrain. Thus, neurons labeled with either TB or DY were intermingled with each other on the midline. Furthermore, TB-positive cell areas labeled from the ipsilateral injections also contained
The present study has provided evidence that cell groups situated on the midline of the rat thalamus and midbrain each project only unilaterally to forebrain structures (including the CPU, Acc and FC), regardless of the laterality of their perikarya. Thus, they send their axons primarily ipsilaterally and to a lesser extent contralaterally, but never bilaterally to forebrain structures. Previous studies on the dorsal raphe projections of the adult rat 9'25 have provided support for the notion that midline-situated brain perikarya may generally have only unilateral ascending projections. The present data on thalamocortical projections of the adult rat are in disagreement with a recent report H, which has suggested that cortical projection neurons in the midline thalamic nuclei of the adult rat were located only ipsilaterally and never crossed the midline. However, intermingling of thalamic cells across the midline was observed in the newborn rat in this previous study 11. In the developmental portion of the present study, we have observed that the midline nuclei of the thalamus and midbrain exist at E l 9 as two bilaterally situated, independent structures each projecting only ipsilaterally to the forebrain. These nuclear groups
280
Fig. 4. Photomicrographs of retrogradely labeled neurons in the midline thalamic regions at E l 9 and P2 following bilateral forebrain injections at El8. TB injections were made on the left side, and D Y injections on the right side of the brain and of the photomicrographs. Small arrows indicate TB-positive cells and arrowheads DY-positive cells, a: the caudal thalamus at El9. A very sharply delineated border on the midline is seen between the TB- and DY-labeling. b: higher magnification of labeled thalamic cells in a. In a and b, a faintly autofluorescent structure is seen on the midline, c: the rostral thalamus at El9. TB- and DY-positive n e u r o n s are intermingled with each other on the midline, d: the caudal thalamus at P2. The degree of intermingling at this age is transitional between the fetus (Fig. 4a,b) and the adult (Figs. 2a and 3a). e: higher magnification of thalamic cells single labeled with either TB or D Y in d. Large arrows in a - e point to the midline of the thalamus, a,d, x 42: b,c,e, × 118.
281 fuse into a mass on the midline by P2. This time course is similar to the fusion time course reported in previous developmental work on the dorsal raphe nucleus7's. By P2, some midline thalamic, midbrain and raphe perikarya with initially ipsilateral forebrain projections have crossed the midline to intermingle with cells from the opposite side, and thus acquire contralateral axonal projections. The absence of contralaterally or bilaterally labeled cells at El9 would seem to rule out other possible developmental mechanisms (see Introduction) for organizing the midline nuclei and their projections. It remains possible that the contralaterally projecting cells do not send out their axons as early as the ipsilaterally projecting cells (which can be labeled by El9). This would imply a long-term, continual uptake from the injection site in animals injected at E l 8 and sacrificed at P2. However, our recent long-term survival experiments have provided evidence that the effective uptake time for the retrograde tracers may be less than 2 days 3. The prenatal hemitransection experiment reported here strongly suggests that perikarya with contralateral axonal projections postnatally already exist as ipsilaterally projecting cells at El9. Thus, a unilateral hemitransection at E l 9 (blocking further retrograde transport from E l 8 forebrain injection sites) did not prevent intermingling of labeled cells across the midline of the thalamus and midbrain at P2. The data propose that perikarya situated contralateral to their adult forebrain projections must migrate towards and across the midline during early development. Furthermore, there may be a rostrocaudal gradient in this developmental process. In the rostral portions of the thalamus, the contralaterally situated perikarya must have begun to migrate and intermingle on the midline even before El9. A similar gradient has been shown in the development of serotonergic raphe perikarya 7,8. This rostrocaudal cell migration gradient contrasts with the caudorostral gradient seen in other neural development processes such as closure of the neural tube and the growth of axons. The rostrocaudal migration of cells towards and across the midline of the thalamus that is reported here even contrasts with the midline fusion of the thalamic tissue itself, which occurs earlier in the embryonic period in a caudorostral gradient 17. The rostrocaudal course of midline nuclei cell migration might depend on the distance between the labeled
cells and forebrain. Thus, the rostral thalamic perikarya (closest to the forebrain) may be able to receive neurotrophic signals from the forebrain earlier than the caudal thalamic perikarya. However, our hemitransection studies suggest that if forebrain neurotrophic factors are important in the rostrocaudal midline cell migration gradient, then their signals must have reached the midline thalamic, midbrain and raphe nuclei before El9. Employing immunohistochemical staining against the S-100 protein, a transient glial structure has recently been described in the rat midline raphe during pre- and early postnatal development 26. This structure extends rostrally as far as the third ver]tricle26. It seems possible that the white (autofluorescent) linear structure seen on the midline of the E l 9 caudal thalamus (Fig. 4a,b), but not seen on the midline of the P2 caudal thalamus (Fig. 4d,e), is the rostralmost portion of this transient glial structure. The onset of cell migration towards and across the midline of the caudal thalamus seems to coincide with the disappearance of this autofluorescent structure (Fig. 4). The autofluorescent linear structure is not visible on the midline of the El9 rostral thalamus, nor can the transient glial structure be stained with S-100 antisera in the rostral thalamus 26. In the absence of this structure, intermingling of cells on the rostral thalamic midline has begun to occur by El9 (Fig. 4c). The perinatal presence of a transient glial barrier in the caudal thalamus may underlie the rostrocaudal midline cell migration gradient in the thalamus. Thus, there might be two consecutive events in the thalamic midline early in development. First, bilaterally situated, independent thalamic tissues fuse in a caudorostral gradient into a single mass on the midline. Second, a transient glial structure next forms a midline barrier within only the caudal thalamus, which then results in a rostrocaudal gradient of thalamic perikaryal migration towards and across the midline. In summary, achievement of the adult forebrain projection pattern involves a pre- and early postnatal perikaryal migration in cell groups situated on the midline of the rat thalamus and midbrain. The adult role of the unilateral (either ipsilateral or contralateral) forebrain projections of these midline situated neurons remains unknown. However, the midline intermingling of perikarya with a mixed laterality of
282 f o r e b r a i n p r o j e c t i o n s m a y allow for an i n f o r m a t i o n
ACKNOWLEDGEMENTS
p r o c e s s i n g c a p a c i t y which w o u l d be i m p o s s i b l e if the midline nuclei r e m a i n e d in their e m b r y o n i c f o r m as two bilaterally s i t u a t e d , i n d e p e n d e n t structures.
REFERENCES 1 Beckstead, R.M., The thalamostriatal projection in the cat, J. Comp. Neurol., 223 (1984) 313-346. 2 Cowan, W.M. and Powell, T.P.S., The projections of the midline and intralaminar nuclei of the thalamus of the rabbit, J. Neurol. Neurosurg. Psychiat., 18 (1955) 266-279. 3 Fishell, G. and Van der Kooy, D., Pattern formation in the striatum: developmental changes in the distribution of striatonigral neurons, J. Neurosci., in press. 4 Groenewegen, H.J., Becker, N.E.H.M. and Lohman, A.H.M., Subcortical afferents of the nucleus accumbens septi in the cat, studied with retrograde axonal transport of horseradish peroxidase and bisbenzimid, Neuroscience, 5 (1980) 1903-1916. 5 Jayaraman, A., Organization of thalamic projections in the nucleus accumbens and the caudate nucleus in cats and its relation with hippocampal and other subcortical afferents, J. Comp. Neurol., 231 (1985) 396-420. 6 Jones, E.G. and Leavitt, R.Y., Retrograde axonal transport and the demonstration of non-specific projections to the cerebral cortex and striatum from the thalamic intralaminar nuclei in the rat, cat and monkey, J. Comp. Neurol., 154 (1974) 349-378. 7 Levitt, P, and Moore, R.Y., Developmental organization of raphe serotonin neuron groups in the rat, Anat. Embryol., 154 (1978) 241-251. 8 Levitt, P. and Moore, R.Y., Developmental organization of serotonin-containing cell group within the brainstem raphe of the rat, Soc. Neurosci. Abstr., 4 (1978) 277. 9 Loughlin, S.E. and Fallon, J.H., Mesostriatal projections from ventral tegmentum and dorsal raphe: cells project ipsilaterally or contralaterally but not bilaterally, Neurosci. Lett., 32 (1982) 11-16. 10 Macchi, G., Bentivoglio, M., Molinari, M. and Minciacchi, D., The thalamo-caudate versus thalamo-cortical projections as studied in the cat with fluorescent retrograde double labeling, Exp. Brain Res., 54 (1984) 225-239. 11 Minciacchi, D., Granato, A., Molinari, M. and Macchi, G., Bilateral thalamo-cortical projections in the newborn rat, Neurosci. Lett. Suppl., 22 (1985) $39. 12 Nauta, W.J.H., Smith, G.P., Faull, R.L.M. and Domesick, V.B., Efferent connections and nigral afferents of the nucleus accumbens septi in the rat, Neuroscience. 3 (1978) 385-401. 13 Newman, R. and Winans, S.S., An experimental study of the ventral striatum of the golden hamster. I. Neuronal connections of the nucleus accumbens, J. Cornp. Neurol., 191 (1980) 167-192.
S u p p o r t e d by the M e d i c a l R e s e a r c h C o u n c i l of Canada. M . T . is an M R C P o s t p r o f e s s i o n a l Fellow.
14 Olson, L. and Seiger, ,&., Early prenatal ontogeny of central monoamine neurons in the rat: fluorescence histochemical observations, Z. Anat. Entwickl..Gesch., 137 (1972) 301-316. 15 Parent, A., Mackey, A. and De Bellefeuille, L., The subcortical afferents to caudate nucleus and putamen in primate. A fluorescence retrograde double labeling study, Neuroscience, 10 (1983) 1137-1150. 16 Phillipson, O.T. and Griffiths, A.C., The topographic order of inputs to nucleus accumbens in the rat, Neuroscience, 16 (1985) 275-296. 17 Rose, J.E., The ontogenetic development of the rabbit's diencephalon, J. Comp. Neurol., 77 (1942) 61-129. 18 Royce, G.J., Cells of origin of subcortical afferents to the caudate nucleus: a horseradish peroxidase study in the cat, Brain Res., 153 (1978) 465-475. 19 Sato, M., Itoh, K. and Mizuno, N., Distribution of thalamo-caudate neurons in the cat as demonstrated by horseradish peroxidase, Exp. Brain Res., 34 (1979) 143-153. 20 Smith, Y. and Parent, A., Differential connections of caudate nucleus and putamen in the squirrel monkey (Saimiri sciureus), Neuroscience, 18 (1986) 347-371. 21 Swanson, L.W., The projections of the ventral tegmental area and adjacent regions: a combined fluorescent retrograde tracer and immunofluorescence study in the rat, Brain Res. Bull., 9 (1982) 321-353. 22 Takada, M., Itoh, K., Sugimoto, T. and Mizuno, N., Topographical projections from the thalamus to the putamen in the cat, Neurosci. Lett., 54 (1985) 207-212. 23 Takada, M., Itoh, K., Yasui, Y., Sugimoto, T. and Mizuno, N., Topographical projections from the posterior thalamic regions to the striatum in the cat, with reference to possible tecto-thalamo-striatal connections, Exp. Brain Res., 60 (1985) 385-396. 24 Van der Kooy, D., The organization of the thalamic, nigral and raphe cells projecting to the medial vs lateral caudateputamen in the rat. A fluorescent retrograde double labeling study, Brain Res., 169 (1979) 381-387. 25 Van der Kooy, D. and Hattori, T., Bilaterally situated dorsal raphe cell bodies have only unilateral forebrain projections in rat, Brain Res., 192 (1980) 550-554. 26 Van Hartesveldt, C., Moore, B. and Hartman, B.K., Transient midline raphe glial structure in the developing rat, J. Comp. Neurol., 253 (1986) 175-184. 27 Veening, J.G., Cornelissen, F.M. and Lieven, P.A.J.M., The topical organization of the afferents to the caudatoputamen of the rat. A horseradish peroxidase study, Neuroscience, 5 (1980) 1253-1268.