119
Neuroscience Research, 13 (1992) 119-137 © 1992 Elsevier Scientific Publishers Ireland, Ltd. 0168-0102/92/$05.00 NEURES 00523
Research Reports
An autoradiographic study of cortical projections from motor thalamic nuclei in the macaque monkey Katsuma Nakano t, Akinori Tokushige
2 Masako
Kohno z, Yasuo Hasegawa Tetsuro Kayahara i and Kazuo Sasaki 3
l
l Department of.4natomy, School of Medicine, Mie University, Tsu, Mie, 2 Department of Anatomy, Faculty of Medicine, Kagoshima Unicersity, Kagoshima and 3 Department of Integrative Brain Science, Faculty of Medicine, Kyoto UniL,ersity, Kyoto (Japan) (Received 26 July 1991; Revised version received 22 October 1991; Accepted 23 October 1991)
Key words." Motor cortex; Premotor cortex; Thalamic nuclei; Cerebellar nuclei; Basal ganglia; Monkey
SUMMARY The special areal and laminar distributions of cortical afferent connections from various thalamic nuclei in the monkey (Macaca fuscata) were studied by using the anterograde axonal transport technique of autoradiography. The following findings were obtained. The superficial thalamocortical (T-C) projections, terminating in the (superficial half of) cortical layer I, arise mainly from the nucleus ventralis anterior, pars principalis (VApc) and nucleus ventralis lateralis, pars oralis (VLo), and possibly from the nucleus ventralis lateralis, pars medialis (VLm) and nucleus ventralis anterior, pars magnocellularis (VAmc). The VApc gives rise to the superficial T-C and deep T-C projections onto the postarcuate premotor area around the arcuate genu and spur, and onto the dorsomedial part of the caudal premotor area as well as the supplementary motor area (SMA). The VApc also gives rise to only deep T-C projections onto the remaining premotor area and onto the rostral bank of the arcuate sulcus as well as the ventral bank of the cingulate sulcus at the level of the premotor area. The VLo gives rise to the superficial T-C projections onto the ventrolateral part of the motor area (mainly to the forelimb motor area) and onto the dorsomedial part to the mesial cortex at the rostral level of the motor area. The VAmc gives rise to the superficial T-C projections onto the banks of the arcuate genu and adjacent region of area 8. Area X, the nucleus ventralis posterolateralis, pars oralis (VPLo), nucleus ventralis posterolateralis, pars caudalis (VPLc), nucleus ventralis posteromedialis (VPM) and possibly the nucleus ventralis lateralis, pars caudalis (VLc) send only deep T-C projections. The dorsal and medial parts of the VLc project onto the premotor area, the rostral part of the motor area and the SMA, and also the ventral bank of the cingulate sulcus. Area X projects onto the premotor area, the SMA, and the caudal part of area 8. The thalamic relay nuclei projecting onto the frontal association cortex were found to be the VAmc, medial VLc and area X.
INTRODUCTION
Knowledge of thalamic projection onto the cerebral cortex is important in understanding the functional properties of cortical neurons. Various thalamic nuclei project onto the cytoarchitectural boundaries of the cerebral cortex with a variety of laminar patterns 18. Recent studies of the thalamocortical connections have emphasized lamina I Correspondence: Prof. Katsuma Nakano, M.D., Department of Anatomy, School of Medicine, Mie University, Tsu, Mie 514, Japan.
120
projection from various thalamic nuclei (see Jones 2o for a review). This projection m the frontal cortex has also been reported in the monkey ~2,~.25.5~,,5,~.~,~ Following a thalamic lesion, terminal degeneration in lamina I was demonstrated by an electron-microscopic study 59.~0. Slight lamina 1 projection from the nucleus ventralis lateralis (VI.) was observed in the motor cortex J~, and substantial lamina I projection from the nucleus ventralis anterior (VA) was indicated in the premotor cortex 25 Distinct laminar differentiation was proposed as the superficial and deep thalamocortical (T-C) projections after electrophysiological studies in cats 52. Further experiments ~ in monkeys revealed laminar differentiation as follows: the fastigial cerebellar nucleus projects mainly onto the medial part of the motor cortex and the parietal association cortex via deep T-C projection neurons terminating in the deep layers of the cortex; the interpositus nucleus projects to the intermediate part of the motor cortex and the premotor cortex via superficial T-C projection neurons terminating in the superficial layer of the cortex; and the dentate nucleus projects onto the lateral part of the motor cortex and the premotor cortex via superficial T-C projection neurons, and also onto area 9 via deep T-C projection neurons 51 Sasaki et al. 55 suggested that the caudomedial part of the nucleus ventralis anterior and nucleus ventralis lateralis complex (VA-VL complex) is the relay part of the deep T-C projection conveying the fastigial input, and the rostrolateral part of the nuclear complex is the relay part of the superficial T-C projection conveying input from the dentate and interpositus cerebellar nuclei. Some electrophysiological studies combined with the horseradish peroxidase (HRP) method also suggest these thalamic relay nuclei 45,49,70. However, the locations of these relay nuclei are still not well defined. In the macaque monkey, the VA-VL complex was divided into subnuclei z0.4~ and the segregated projections of the cerebello-thalamo-cortical and pallido-thalamo-cortical connections were reported on the basis of findings from modern tracing techniques 2,20,57,66.However, these cerebral connections of motor thalamic nuclei have not been in full agreement with the findings of other researchers 25.30.36.37,~,4. Systematic studies concerning the areal and laminar projections from individual subnuclei of the motor thalamic nuclei in monkeys are required. No experimental studies, however, have been conducted on these projections using anterograde axonal tracing methods. In the present study, the thalamocortical projections from distinct subnuclei of the motor thalamic nuclei were studied by the autoradiographic technique. A preliminary report of this work has appeared in part elsewhere 43 MATERIALS AND METHODS
Experiments were performed on 26 Japanese monkeys (Macaca fuscata) ranging in weight from 2.8 to 12 kg. Tritiated amino acid was evaporated under nitrogen gas supply and then reconstituted in the following concentrations with sterile saline or deionized, distilled water: 20-100 p~Ci//xl of 3H-leucine (L-4,5-3H-leucine, specific activity 57 Ci/mmol; Radiochemical Centre, Amersham, U.K.), or an equal mixture of 3H-leucine and 3H-proline (L-5-3H-proline, specific activity 22 Ci/mmol; Radiochemical Centre, Amersham, U.K.). The animals were held in a stereotaxic apparatus (David KopD and craniotomies were performed in aseptic conditions to expose the appropriate cortices. They were anesthetized by an intramuscular injection of 7-8 mg/kg of ketamine hydrochloride (Ketalar, 50 mg/ml) followed by intraperitoneal injection of 15-30 mg/kg of sodium pentobarbital. A single stereotaxic injection of the concentrated isotope was made in
121 different parts of the ventral thalamic nuclei of both cerebral hemispheres in each animal. The injection was achieved through a 31-gauge steel needle attached to a 1-~1 Hamilton syringe with the total injected volume varying in different experiments from 0.3 to 0.5/.d. These injections were made over a period of 10 min and the needle was left in place for 10 min after the injection. These animals were deeply re-anesthetized 5-9 days after the injection of the tracer, and perfused through the left ventricle by 2000 ml of 10% buffered neutral formalin. The brains were blocked, then removed immediately and kept in the same fixative with 10% sucrose for more than 1 week. Coronal sections 7 p~m thick were cut on a paraffin microtome. For autoradiography, selected series of sections mounted on slides were processed according to the procedure described by Cowan et al. 8. Alternate series were dipped in Sakura NR-M2 emulsion and exposed for 4 or 8-28 weeks. After developing in Rendol for 8 min, the sections were counterstained with cresyl violet. These sections were examined under dark- and bright-field microscopy using a Nikon phase-contrast condenser, which is able to switch dark and bright fields; silver grains indicative of terminal labeling and labeled fibers were charted on the representative projection drawings. Serial sections through the whole brain of monkey were made 40 tzm in thickness after embedding in celloidin, and stained with toluidine blue for observation of the normal cytoarchitecture of the thalamus and some cortical areas. The terminology used for the diencephalon essentially follows that of Olszewski 46 and for the cerebral cortical areas follows that of Barbas and Pandya 3 and Brodmann 5. RESULTS Animals were divided into 7 groups based on the injection sites. Only one representative case of each group will be described in detail.
Nucleus ventralis anterior, pars principalis (VApc) injection The injection site was confined to the ventral part of the VApc in M 1R, the dorsal part in M 23R, and the central part of VApc in M 28R (Fig. 1A). In M 28R, a reasonably well-confined deposit of isotope was placed in the central part of the VApc with little involvement of the lateral part of the nucleus ventralis anterior pars magnocellularis (VAmc), and only gliosis was observed in its needle tip and track (Fig. 1A, level 3). This injection site was extended caudally to the rostrodorsal part of area X, which was faintly labeled (Fig. 1A, level 1). In this case (Fig. 1B), the labeled fibers emerged from the injection site and ran dorsolaterally through the dorsomedial part of the anterior limb of the internal capsule to reach the subcortical white matter of the frontal cortex. Terminal labelings in the superficial cortical layer (superficial half of lamina I) and the deep cortical layer (laminae III-VI) were observed in the postarcuate premotor area (Figs. 1B and 2A-C) including the rostral bank of the arcuate genu, and in dorsal caudal area 6 (6 DC as defined by Barbas and Pandya 3) as well as the mesial cortex, facing the falx cerebri, rostral to the motor area (supplementary motor area, SMA). Only deep cortical terminations were found in both banks of the rostral part of the inferior limb of the arcuate sulcus (inferior arcuate sulcus, Ai), and in the lower part of ventral area 6 (6 Vb as defined by Barbas and Pandya 3). The terminal labeling in the superficial cortical layer was heaviest in the spur and genu of the arcuate sulcus (Figs. 1B, levels 3-6, and 2A-C). Some of the labeled fibers were observed running ventrolaterally or ventrally through the external capsule and lateral medullary
122
~
d
2
r
~_
"
/ ~ . / /
}fll-J'//l
I¢h
5
/
,,~ . c~
.. ::)
..-- .( :,fillip3
k
4
,---I)
~,--#
f
.! "~
,
.
1~{7/J
,~,)r
I
.....
~ ."vA x. ,.!\!
t)~
Neuroscience Research, 13 (1992) 119-137 © 1992 Elsevier Scientific Publishers Ireland, Ltd. 0168-0102/92/$05.00 NEURES 00523
Research Reports
An autoradiographic study of cortical projections from motor thalamic nuclei in the macaque monkey Katsuma Nakano t, Akinori Tokushige
2 Masako
Kohno z, Yasuo Hasegawa Tetsuro Kayahara i and Kazuo Sasaki 3
l
l Department of.4natomy, School of Medicine, Mie University, Tsu, Mie, 2 Department of Anatomy, Faculty of Medicine, Kagoshima Unicersity, Kagoshima and 3 Department of Integrative Brain Science, Faculty of Medicine, Kyoto UniL,ersity, Kyoto (Japan) (Received 26 July 1991; Revised version received 22 October 1991; Accepted 23 October 1991)
Key words." Motor cortex; Premotor cortex; Thalamic nuclei; Cerebellar nuclei; Basal ganglia; Monkey
SUMMARY The special areal and laminar distributions of cortical afferent connections from various thalamic nuclei in the monkey (Macaca fuscata) were studied by using the anterograde axonal transport technique of autoradiography. The following findings were obtained. The superficial thalamocortical (T-C) projections, terminating in the (superficial half of) cortical layer I, arise mainly from the nucleus ventralis anterior, pars principalis (VApc) and nucleus ventralis lateralis, pars oralis (VLo), and possibly from the nucleus ventralis lateralis, pars medialis (VLm) and nucleus ventralis anterior, pars magnocellularis (VAmc). The VApc gives rise to the superficial T-C and deep T-C projections onto the postarcuate premotor area around the arcuate genu and spur, and onto the dorsomedial part of the caudal premotor area as well as the supplementary motor area (SMA). The VApc also gives rise to only deep T-C projections onto the remaining premotor area and onto the rostral bank of the arcuate sulcus as well as the ventral bank of the cingulate sulcus at the level of the premotor area. The VLo gives rise to the superficial T-C projections onto the ventrolateral part of the motor area (mainly to the forelimb motor area) and onto the dorsomedial part to the mesial cortex at the rostral level of the motor area. The VAmc gives rise to the superficial T-C projections onto the banks of the arcuate genu and adjacent region of area 8. Area X, the nucleus ventralis posterolateralis, pars oralis (VPLo), nucleus ventralis posterolateralis, pars caudalis (VPLc), nucleus ventralis posteromedialis (VPM) and possibly the nucleus ventralis lateralis, pars caudalis (VLc) send only deep T-C projections. The dorsal and medial parts of the VLc project onto the premotor area, the rostral part of the motor area and the SMA, and also the ventral bank of the cingulate sulcus. Area X projects onto the premotor area, the SMA, and the caudal part of area 8. The thalamic relay nuclei projecting onto the frontal association cortex were found to be the VAmc, medial VLc and area X.
INTRODUCTION
Knowledge of thalamic projection onto the cerebral cortex is important in understanding the functional properties of cortical neurons. Various thalamic nuclei project onto the cytoarchitectural boundaries of the cerebral cortex with a variety of laminar patterns 18. Recent studies of the thalamocortical connections have emphasized lamina I Correspondence: Prof. Katsuma Nakano, M.D., Department of Anatomy, School of Medicine, Mie University, Tsu, Mie 514, Japan.
120
projection from various thalamic nuclei (see Jones 2o for a review). This projection m the frontal cortex has also been reported in the monkey ~2,~.25.5~,,5,~.~,~ Following a thalamic lesion, terminal degeneration in lamina I was demonstrated by an electron-microscopic study 59.~0. Slight lamina 1 projection from the nucleus ventralis lateralis (VI.) was observed in the motor cortex J~, and substantial lamina I projection from the nucleus ventralis anterior (VA) was indicated in the premotor cortex 25 Distinct laminar differentiation was proposed as the superficial and deep thalamocortical (T-C) projections after electrophysiological studies in cats 52. Further experiments ~ in monkeys revealed laminar differentiation as follows: the fastigial cerebellar nucleus projects mainly onto the medial part of the motor cortex and the parietal association cortex via deep T-C projection neurons terminating in the deep layers of the cortex; the interpositus nucleus projects to the intermediate part of the motor cortex and the premotor cortex via superficial T-C projection neurons terminating in the superficial layer of the cortex; and the dentate nucleus projects onto the lateral part of the motor cortex and the premotor cortex via superficial T-C projection neurons, and also onto area 9 via deep T-C projection neurons 51 Sasaki et al. 55 suggested that the caudomedial part of the nucleus ventralis anterior and nucleus ventralis lateralis complex (VA-VL complex) is the relay part of the deep T-C projection conveying the fastigial input, and the rostrolateral part of the nuclear complex is the relay part of the superficial T-C projection conveying input from the dentate and interpositus cerebellar nuclei. Some electrophysiological studies combined with the horseradish peroxidase (HRP) method also suggest these thalamic relay nuclei 45,49,70. However, the locations of these relay nuclei are still not well defined. In the macaque monkey, the VA-VL complex was divided into subnuclei z0.4~ and the segregated projections of the cerebello-thalamo-cortical and pallido-thalamo-cortical connections were reported on the basis of findings from modern tracing techniques 2,20,57,66.However, these cerebral connections of motor thalamic nuclei have not been in full agreement with the findings of other researchers 25.30.36.37,~,4. Systematic studies concerning the areal and laminar projections from individual subnuclei of the motor thalamic nuclei in monkeys are required. No experimental studies, however, have been conducted on these projections using anterograde axonal tracing methods. In the present study, the thalamocortical projections from distinct subnuclei of the motor thalamic nuclei were studied by the autoradiographic technique. A preliminary report of this work has appeared in part elsewhere 43 MATERIALS AND METHODS
Experiments were performed on 26 Japanese monkeys (Macaca fuscata) ranging in weight from 2.8 to 12 kg. Tritiated amino acid was evaporated under nitrogen gas supply and then reconstituted in the following concentrations with sterile saline or deionized, distilled water: 20-100 p~Ci//xl of 3H-leucine (L-4,5-3H-leucine, specific activity 57 Ci/mmol; Radiochemical Centre, Amersham, U.K.), or an equal mixture of 3H-leucine and 3H-proline (L-5-3H-proline, specific activity 22 Ci/mmol; Radiochemical Centre, Amersham, U.K.). The animals were held in a stereotaxic apparatus (David KopD and craniotomies were performed in aseptic conditions to expose the appropriate cortices. They were anesthetized by an intramuscular injection of 7-8 mg/kg of ketamine hydrochloride (Ketalar, 50 mg/ml) followed by intraperitoneal injection of 15-30 mg/kg of sodium pentobarbital. A single stereotaxic injection of the concentrated isotope was made in
121 different parts of the ventral thalamic nuclei of both cerebral hemispheres in each animal. The injection was achieved through a 31-gauge steel needle attached to a 1-~1 Hamilton syringe with the total injected volume varying in different experiments from 0.3 to 0.5/.d. These injections were made over a period of 10 min and the needle was left in place for 10 min after the injection. These animals were deeply re-anesthetized 5-9 days after the injection of the tracer, and perfused through the left ventricle by 2000 ml of 10% buffered neutral formalin. The brains were blocked, then removed immediately and kept in the same fixative with 10% sucrose for more than 1 week. Coronal sections 7 p~m thick were cut on a paraffin microtome. For autoradiography, selected series of sections mounted on slides were processed according to the procedure described by Cowan et al. 8. Alternate series were dipped in Sakura NR-M2 emulsion and exposed for 4 or 8-28 weeks. After developing in Rendol for 8 min, the sections were counterstained with cresyl violet. These sections were examined under dark- and bright-field microscopy using a Nikon phase-contrast condenser, which is able to switch dark and bright fields; silver grains indicative of terminal labeling and labeled fibers were charted on the representative projection drawings. Serial sections through the whole brain of monkey were made 40 tzm in thickness after embedding in celloidin, and stained with toluidine blue for observation of the normal cytoarchitecture of the thalamus and some cortical areas. The terminology used for the diencephalon essentially follows that of Olszewski 46 and for the cerebral cortical areas follows that of Barbas and Pandya 3 and Brodmann 5. RESULTS Animals were divided into 7 groups based on the injection sites. Only one representative case of each group will be described in detail.
Nucleus ventralis anterior, pars principalis (VApc) injection The injection site was confined to the ventral part of the VApc in M 1R, the dorsal part in M 23R, and the central part of VApc in M 28R (Fig. 1A). In M 28R, a reasonably well-confined deposit of isotope was placed in the central part of the VApc with little involvement of the lateral part of the nucleus ventralis anterior pars magnocellularis (VAmc), and only gliosis was observed in its needle tip and track (Fig. 1A, level 3). This injection site was extended caudally to the rostrodorsal part of area X, which was faintly labeled (Fig. 1A, level 1). In this case (Fig. 1B), the labeled fibers emerged from the injection site and ran dorsolaterally through the dorsomedial part of the anterior limb of the internal capsule to reach the subcortical white matter of the frontal cortex. Terminal labelings in the superficial cortical layer (superficial half of lamina I) and the deep cortical layer (laminae III-VI) were observed in the postarcuate premotor area (Figs. 1B and 2A-C) including the rostral bank of the arcuate genu, and in dorsal caudal area 6 (6 DC as defined by Barbas and Pandya 3) as well as the mesial cortex, facing the falx cerebri, rostral to the motor area (supplementary motor area, SMA). Only deep cortical terminations were found in both banks of the rostral part of the inferior limb of the arcuate sulcus (inferior arcuate sulcus, Ai), and in the lower part of ventral area 6 (6 Vb as defined by Barbas and Pandya 3). The terminal labeling in the superficial cortical layer was heaviest in the spur and genu of the arcuate sulcus (Figs. 1B, levels 3-6, and 2A-C). Some of the labeled fibers were observed running ventrolaterally or ventrally through the external capsule and lateral medullary
122
~
d
2
r
~_
"
/ ~ . / /
}fll-J'//l
I¢h
5
/
,,~ . c~
.. ::)
..-- .( :,fillip3
k
4
,---I)
~,--#
f
.! "~
,
.
1~{7/J
,~,)r
I
.....
~ ."vA x. ,.!\!
t)~
