Somatotopically arranged inputs from putamen and subthalamic nucleus to primary motor cortex.
ABSTRACT Employing retrograde transsynaptic transport of rabies virus, we investigated the organization of basal ganglia inputs to hindlimb, proximal and distal forelimb, and orofacial representations of the macaque primary motor cortex (MI). Four days after rabies injections into these MI regions, neuronal labeling occurred in the striatum and the subthalamic nucleus (STN) through the cortico-basal ganglia loop circuits. In the striatum, two distinct sets of the labeling were observed: one in the dorsal putamen, and the other in the ventral striatum (ventromedial putamen and nucleus accumbens). The dorsal striatal labeling was somatotopically arranged and its distribution pattern was in good accordance with that of the corticostriatal inputs, such that the hindlimb, orofacial, or forelimb area was located in the dorsal, ventral, or intermediate zone of the putamen, respectively. The distribution pattern of the ventral striatal labeling was essentially the same in all cases. In the STN, the somatotopic arrangement of labeled neurons was in register with that of corticosubthalamic inputs. The present results suggest that the cortico-basal ganglia motor circuits involving the dorsal putamen and the STN may constitute separate closed loops based on the somatotopy, while the ventral striatum provides common multisynaptic projections to all body-part representations in the MI.
- [show abstract] [hide abstract]
ABSTRACT: The neural circuits that link the basal ganglia with the cerebral cortex are critically involved in the generation and control of voluntary movement. Retrograde transneuronal transport of herpes simplex virus type 1 was used to examine the organization of connections in the cebus monkey between an output nucleus of the basal ganglia, the internal segment of the globus pallidus (GPi), and three cortical areas: the primary motor cortex, the supplementary motor ara, and the ventral premotor area. Spatially separate regions of the GPi were labeled after virus injections into each cortical area. The GPi projects to multiple cortical motor areas, and this pallidal output is organized into discrete channels. This information provides a new anatomical framework for examining the function of the basal ganglia in skeletomotor control.Science 03/1993; 259(5096):819-21. · 31.03 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: How the motor-related cortical areas modulate the activity of the output nuclei of the basal ganglia is an important issue for understanding the mechanisms of motor control by the basal ganglia. In the present study, by using awake monkeys, the polysynaptic effects of electrical stimulation in the forelimb regions of the primary motor and primary somatosensory cortices on the activity of globus pallidus (GP) neurons, especially mediated by the subthalamic nucleus (STN), have been characterized. Cortical stimulation induced an early, short-latency excitation followed by an inhibition and a late excitation in neurons of both the external and internal segments of the GP. It also induced an early, short-latency excitation followed by a late excitation and an inhibition in STN neurons. The early excitation in STN neurons preceded that in GP neurons. Blockade of STN neuronal activity by muscimol (GABA(A) receptor agonist) injection resulted in abolishment of both the early and late excitations evoked in GP neurons by cortical stimulation. At the same time, the spontaneous discharge rate of GP neurons decreased, pauses between the groups of spikes of GP neurons became prominent, and the firing pattern became regular. Injection of (+/-)-3-(2-carboxypiperazin-4-yl)-propyl-1-phosphonic acid (CPP) [N-methyl-D-aspartate (NMDA) receptor antagonist], but not 1,2,3, 4-tetrahydro-6-nitro-2,3-dioxo-benzo[f]quinoxaline-7-sulfonamide disodium [NBQX (non-NMDA receptor antagonist)], into the STN attenuated the early and late excitations in GP neurons, suggesting that cortico-subthalamic transmission is mediated mainly by NMDA receptors. Interference with the pallido-subthalamic transmission by bicuculline (GABA(A) receptor antagonist) injection into the STN made the inhibition distinct without affecting the early excitation. The present results indicate that the cortico-subthalamo-pallidal pathway conveys powerful excitatory effects from the motor-related cortical areas to the GP with shorter conduction time than the effects conveyed through the striatum.Journal of Neurophysiology 08/2000; 84(1):289-300. · 3.30 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The basal ganglia receive massive inputs from the neocortex and send outputs that exert both inhibitory and disinhibitory control over parts of the frontal cortex and brainstem. Between these basal ganglia inputs and outputs lies the striatum, which receives most of the cortical afferents and projects to the basal ganglia output nuclei--the globus pallidus and substantia nigra. To analyze this system we conjointly labeled, in squirrel monkeys, sensorimotor cortical inputs to the striatum and striatal outputs to the globus pallidus. Anterograde tracers were injected into the motor (MI) and somatosensory (SI) cortical body maps, at sites determined by electrophysiological stimulation and recording. Retrograde tracers were stereotaxically injected into the external and internal pallidal segments (GPe and GPi). We found that multiple dispersed modules ("matrisomes") in the putamen that all received inputs from single body-part representations in sensorimotor cortex could, in turn, send convergent outputs to single sites in the pallidum. This divergence-reconvergence pattern was found for both GPe and GPi sites, and for inputs from both SI and MI cortex. Thus, information from a single functional region in the cortex can be split up at the striatal stage only to be brought back together in the pallidum. The temporary divergence may increase lateral interactions between sensorimotor matrisomes, as well as between matrisomes and striosomes. One function of striatal modularity may thus be to set up an associative network in the striatum, which might contribute to sensorimotor learning. We also found that some sets of matrisomes did not receive strong sensorimotor inputs, even though they projected to regions of GPe and GPi that are near the sensorimotor-recipient zones described above. Thus, the matrisomal system may sort MI/SI inputs and other inputs before transfer to paired regions of GPe and GPi.Journal of Neuroscience 03/1994; 14(2):599-610. · 6.91 Impact Factor
Somatotopically arranged inputs from putamen and
subthalamic nucleus to primary motor cortex
Shigehiro Miyachia,b,*, Xiaofeng Luc, Michiko Imanishia,
Kaori Sawadaa,d, Atsushi Nambue, Masahiko Takadaa,b
aDepartment of System Neuroscience, Tokyo Metropolitan Institute for Neuroscience,
Tokyo Metropolitan Organization for Medical Research, Fuchu, Tokyo 183-8526, Japan
bCREST, Japan Science and Technology Agency, Kawaguchi, Saitama 332-0012, Japan
cDepartment of Physiology, Juntendo University School of Medicine, Tokyo 113-8421, Japan
dDivision of Applied System Neuroscience, Nihon University School of Medicine, Tokyo 173-8610, Japan
eDivision of System Neurophysiology, National Institute for Physiological Sciences, Okazaki, Aichi 444-8585, Japan
Received 1 May 2006; accepted 27 July 2006
Available online 12 September 2006
distal forelimb, and orofacial representations of the macaque primary motor cortex (MI). Four days after rabies injections into these MI regions,
neuronal labeling occurred in the striatum and the subthalamic nucleus (STN) through the cortico-basal ganglia loop circuits. In the striatum, two
distinct sets of the labeling were observed: one in the dorsal putamen, and the other in the ventral striatum (ventromedial putamen and nucleus
accumbens). The dorsal striatal labeling was somatotopically arranged and its distribution pattern was in good accordance with that of the
corticostriatal inputs, such that the hindlimb, orofacial, or forelimb area was located in the dorsal, ventral, or intermediate zone of the putamen,
respectively. The distribution pattern of the ventral striatal labeling was essentially the same in all cases. In the STN, the somatotopic arrangement
of labeled neurons was in register with that of corticosubthalamic inputs. The present results suggest that the cortico-basal ganglia motor circuits
involving the dorsal putamen and the STN may constitute separate closed loops based on the somatotopy, while the ventral striatum provides
common multisynaptic projections to all body-part representations in the MI.
# 2006 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved.
Keywords: Basal ganglia; Monkey; Rabies virus; Transneuronal labeling; Ventral striatum
The basal ganglia constitute loop circuits with the frontal
lobe through the thalamus. The striatum and the subthalamic
nucleus (STN) receive inputs directly from the cortex and send
outputs to the internal segment of the globus pallidus (GPi;
Hedreen and DeLong, 1991; Flaherty and Graybiel, 1994;
along the cortico-basal ganglia loops. The results of anatomical
studies in the last two decades suggest that the cortico-basal
ganglia loop circuit consists of multiple parallel loops. The
frontal association, motor, and limbic cortical areas project to
distinct territories in the striatum (caudate nucleus/rostral
putamen, dorsal putamen, and ventral putamen/nucleus
accumbens, respectively). Each striatal territory projects back
to the same cortical areas via the GPi or the substantia nigra
pars reticulata, and then the thalamus (Alexander et al., 1990;
Parent and Hazrati, 1995).
The corticostriatal projections from the primary motor
cortex (MI) are classified in a somatotopic manner: the
hindlimb, orofacial, and forelimb regions of the MI project to
the dorsal, ventral, and intermediate sectors of the caudal
putamen, respectively (Alexander and DeLong, 1985; Flaherty
and Graybiel, 1993; Takada et al., 1998). Within the forelimb
region of the putamen, the distal part is represented more
laterally, as compared to the proximal counterpart (Tokuno
Neuroscience Research 56 (2006) 300–308
Sciences, Primate Research Institute, Kyoto University, Inuyama, Aichi 484-
8506, Japan. Tel.: +81 568 63 0559; fax: +81 568 63 0563.
E-mail address: firstname.lastname@example.org (S. Miyachi).
0168-0102/$ – see front matter # 2006 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved.
et al., 1999). However, whether such a somatotopy is main-
tained throughout the cortico-basal ganglia motor loop has not
been determined yet.
Using retrograde transsynaptic transport of herpes simplex
virus, Hoover and Strick (1993, 1999) have demonstrated that
the GPi contains somatotopically organized output channels to
the MI via the motor thalamus. Moreover, the application of
another neurotropic virus, rabies virus, by Kelly and Strick
(2004) has provided evidence that the distribution area in the
putamen of striatal neurons projecting multisynaptically to the
MI forelimb representation is similar to that of corticostriatal
terminals arising from the same MI region. They have also
shown that retrograde transneuronal labeling in the STN
occurrs in the region that receives corticosubthalamic fibers
from the MI forelimb representation. These data suggest that
the cortico-basal ganglia motor loop may consist of multiple
closed subloops representing different body parts.
Toclarifytheentire organization ofthecortico-basal ganglia
motor loop, we injected the CVS-11 strain of rabies virus into
the four different body-part representations of the monkey MI:
the hindlimb, proximal forelimb, distal forelimb, and orofacial
regions. This virus has been shown to infect neurons selectively
and moves retrogradely through synapses in a time-dependent
manner (Kelly and Strick, 2000, 2004; Graf et al., 2002;
Moschovakis et al., 2004; Miyachi et al., 2005). We examined
the distribution patterns of labeled neurons in the striatum and
STN at a 4-day postinjection period when the viral infection
was expected to reach third-order neurons by way of the motor
thalamus and GPi (Fig. 1).
2. Materials and methods
Seven macaque monkeys (three rhesus, three Japanese, and one crab-eating
monkeys) of either sex weighing 3.7–11 kg were used in this study (Table 1).
Part of data obtained in monkeys Vi and Ri was utilized elsewhere for analysis
of frontal cortical labeling (Miyachi et al., 2005). The experimental protocol
was approved by the Animal Care and Use Committee of the Tokyo Metro-
politan Institute for Neuroscience, and all experiments were done in line with
2.2. Surgical procedures and electrophysiological mapping
First of all, a head holder was surgically attached to the skull in aseptic
conditions. Under general anesthesia with ketamine hydrochloride (5 mg/kg,
i.m.) and sodium pentobarbital (20 mg/kg, i.v.), each monkey was positioned
in a stereotaxic apparatus, and the head holder was fixed on the skull with
anchor screws and dental acrylic resin. Several days later, body part repre-
sentations in the MI were mapped by means of intracortical microstimulation.
The monkeys were anesthetized with a combination of ketamine hydrochlor-
ide (5–10 mg/kg, i.m.) and xylazine hydrochloride (0.5–1 mg/kg, i.m.), and
coated Elgiloy-alloy microelectrode (0.5–1.5 MV at 1 kHz) was introduced
perpendicular to the dural surface. When trains of 12 cathodal pulses (200 ms
duration at 333 Hz, currents of less than 50 mA) were delivered, evoked
movements of various body parts were carefully monitored by visual inspec-
tion. To preserve the exposed dural surface, a rectangular chamber was
attached to the skull with acrylic resin.
2.3. Viral injections
Based on the somatotopic map of the MI (viewed from the cortical surface),
the CVS-11 strain of rabies virus was injected into its proximal forelimb, distal
forelimb, hindlimb, and orofacial regions. The virus was derived from the
Centerfor Disease Controland Prevention (Atlanta, GA,USA) and amplified at
the National Institute of Infectious Diseases (Tokyo, Japan). The titer of a stock
virus suspension was 1.4 ? 108focus-forming units/ml. After the anesthesia
with ketamine hydrochloride (5–10 mg/kg, i.m.) and xylazine hydrochloride
S. Miyachi et al./Neuroscience Research 56 (2006) 300–308301
Fig. 1. Order of retrograde viral labeling in the cortico-basal ganglia loop
the direction of viral transport. Medium-sized spiny projection neurons in the
putamen (MS) and subthalamic nucleus neurons (STN) consist of third-order
neurons, while acetylcholine-containing (Ach) and parvalbumin-containing
(PV) interneurons in the putamen consist of fourth-order neurons.
Summary of experiments
C: Crab-eating monkey (Macaca fascicularis); Dist: distal forelimb region; F:
female; HL: hindlimb region; J: Japanese monkey (Macaca fuscata); M: male;
OF: orofacial region; Prox: proximal forelimb region; R: rhesus monkey
(Macaca mulatta). For the cases of proximal and distal forelimb injections,
the volume of the rabies virus suspension injected into the precentral bank area
is indicated in parentheses.
(0.5–1 mg/kg, i.m.), each monkey received injections of the viral suspension
into one of the electrophysiologically identified MI regions through a 10-ml
2.4. Histological procedures
After a survival period of4 days, the monkeys wereanesthetized deeply with
sodium pentobarbital (50 mg/kg) and perfused transcardially with 0.1 M phos-
phate-buffered saline (PBS, pH 7.4), followed by a mixture of 10% formalin and
15% saturated picric acid in 0.1 M phosphate buffer (pH 7.4). The brains were
removed from the skull, postfixed in the same fresh fixative overnight at 4 8C,
saturated with 30% sucrose at 4 8C, and then cut serially into coronal sections
60 mm thick on a freezing microtome. Every sixth section was immunohisto-
means of the standard avidin-biotin-peroxidase complex (ABC) method.
The sections were briefly washedin PBS,soaked with 1% skim milkin PBS
for 2 h, and then incubated overnight with the primary antibody (1:10,000
dilution) in PBS containing 0.1% Triton X-100 and 1% normal goat serum.
Subsequently, the sections were immersed for 2 h in the same fresh incubation
medium containing biotinylated goat anti-rabbit IgG antibody (1:200 dilution;
Vector Laboratories, Burlingame, CA, USA) and processed with the ABC kit
(ABC Elite; Vector Laboratories). For visualization of the antigen, the sections
were reacted in 0.05 M Tris–HCl buffer (pH 7.6) containing 0.04% diamino-
in PBS, the sections were mounted onto gelatin-coated glass slides and cover-
slipped. An adjacent group of the sections (60 mm apart) were Nissl-stained
with 1% Neutral Red or Cresyl Violet.
Some groups of the sections through the striatum were immunostained for a
combination of rabies virus and calbindin D28k, rabies virus and choline
acetyltransferase (ChAT), or rabies virus and parvalbumin (PV).
Block Ace (Yukijirushi Nyugyo, Sapporo, Japan) for 1 h, and then incubated
overnight with a mixture of primary antibodies dissolved in PBS containing 1%
normal horse or goat serum and 0.1% Triton X-100. The primary antibodies
(Swant, Bellinzona, Switzerland) antibodies, at 1:500 for goat anti-ChAT
antibody (Chemicon International, Temecla, CA), and at 1:2000 for mouse
anti-PVantibody (Chemicon International). After washes in PBS, the sections
were reactedfor2 hwithbiotinylateddonkeyanti-goatIgGor horseanti-mouse
IgG antibody (1:200 dilution, Vector Laboratories), followed by a mixture of
donkey Cy3-conjugated anti-rabbit IgG antybody (1:200 dilution, Chemicon
International) and FITC-conjugated streptavidin (1:100 dilution, Chemicon
International) dissolved in the same fresh medium. The sections were washed
in PBS, mounted onto gelatin-coated glass slides, and then coverslipped.
2.5. Data analysis
2.5.1. Cell density analysis
Projection drawings of representative coronal sections through the striatum
and STN were prepared in five secitons for the striatum (2880 mm apart) and in
were devided into 500- and 250-mm bins, respectively. The number of labeled
neurons in each bin was counted and color-coded.
2.5.2. Cell counts
into the MI distal forelimb region produced a much larger number of labeled
neurons as compared to the proximal injections. Thus, the actual number of
labeled neurons in the putamen and STN was counted in the cases of proximal
(monkeys Lo and X) and distal forelimb (monkeys Su and Si) injections. In the
putamen, cell counts were carried out within the region dorsal to theventral end
of the globus pallidus between the rostrocaudal levels of the anterior commis-
sure and the caudal end of the globus pallidus, which corresponds nearly to the
motor territory of the putamen (8 equidistant sections 1440 mm apart). In the
STN, cell counts were done in a series of sections (10 or 11 equidistant sections
360 mm apart).
2.6. Safety issues
The entire experiments were carried out in a special primate laboratory
(biosafety level 2) designated for in vivo virus experiments. Throughout the
experiments, the monkeys were kept in individual cages which were placed
inside a special safety cabinet. To avoid accidental infection with the virus, all
investigators received immunization beforehand and wore protective clothing
S. Miyachi et al./Neuroscience Research 56 (2006) 300–308302
Fig. 2. Results of intracortical microstimulation mapping and sites of rabies injections into hindlimb (monkey Vi), proximal forelimb (monkey Lo), distal forelimb
(monkey Su), and orofacial (monkey Ch) representations of MI. Each letter on the cortical surface map denotesthe site of electrode penetration, stimulation of which
elicited movement predominantly from one of the following body parts: A, ankle; D, digit; E, elbow; H, hip; J, jaw; Kn, knee; L, lip; S, shoulder; Tg, tongue; Tl, tail;
To, toe; Tr, trunk; W, wrist. Each circle indicates the approximate extent of the injection site. CS, central sulcus.
during the experimental sessions. Equipment was disinfected with 70% ethanol
after each experimental session, and waste was autoclaved prior to disposal.
3.1. Injection sites
In seven monkeys, multiple injections of rabies virus were
made into the forelimb, hindlimb, and orofacial regions of the
MI (Fig. 2, Table 1). The injection sites in the forelimb
representation were restricted to its proximal or distal region,
respectively, in monkeys X and Lo or monkeys Su and Si. On
the other hand, the injection sites in the hindlimb or orofacial
representation were located through its large extent in monkeys
Ri and Vi or monkey Ch, respectively.
3.2. Distribution of striatal neuron labeling
In all monkeys, labeled neurons formed two distinct clusters:
one in the dorsal putamen and the other in the ventral striatum
including the ventromedial portion of the putamen and the
nucleus accumbens (Fig. 3, upper panel). The distribution
patterns of neuronal labeling in the dorsal putamen were
dependent on the injection sites of rabies virus. After the
injections into the distal (monkeys Su and Si) and proximal
(monkeys Lo and X) forelimb representations of the MI, labeled
sector. As compared to such labeling from the forelimb regions,
striatal neurons labeled from the hindlimb representation
(monkeys Vi and Ri) were clustered more dorsally. Dense
labeling from the orofacial region was located in a more ventral
zone, which seemed continuous to the ventral striatum as
When the proximal and distal forelimb injection cases were
laterally than that from the proximal injection. The number of
was much greater than that of striatal neurons labeled from the
proximal forelimb region. The mean (?S.D.) number of the
S. Miyachi et al./Neuroscience Research 56 (2006) 300–308 303
Fig. 3. (Upper panel) Distribution patterns of neuronal labeling in the striatum
(monkey Ch; bottom) representations of the MI. Five equidistant coronal
sections (2880 mm apart) are arranged rostrocaudally in A–E, A0–E0, A00–E00,
or A000–E000. The rostrocaudal level of sections B, B0, B00, and B000corresponds to
that of the anterior commissure. Each striatal section was divided into
0.5 mm ? 0.5 mmbins.Thenumberoflabeledneuronsineachbinwascounted
and color-coded as follows: white, bins with the highest cell density (approxi-
mately top 5% of all bins); yellow, next 5%; red, next 10%; blue, next 30%. For
monkey Vi, blue, red, yellow, and white bins contain 2–4, 5–6, 7–11, and more
than 12 cells, respectively. For the other monkeys, these bins contain 3–4, 5–6,
7–10, and more than 11 (monkey Lo), 3–6, 7–9, 10–13, and more than 14
(monkey Su), or 16–48, 49–67, 68–83, and more than 84 (monkey Ch),
respectively. The areas of neuronal labeling in the dorsal putamen are soma-
totopically organized, whereas those in the ventral striatum (ventromedial
putamen and nucleus accumbens) are quite similar regardless of the injection
sites. Acc, nucleus accumbens; Cd, caudate nucleus; Put, putamen. (Lower
panel) Distribution patterns of neuronal labeling in the STN of the same
monkeys as in the upper panel. Four equidistant coronal sections (720 mm
apart) are arranged rostrocaudally in A–D, A0–D0, A00–D00, or A000–D000. The
number of labeled neurons in blue, red, yellow, and white bins is 1, 2, 3, and
8, 9–10, 11–12, and more than 13 (monkey Ch), respectively. Note that the
major zone representing the hindlimb or orofacial part resides in the dorsome-
dial or ventrolateral portion in the lateral aspect of the STN, respectively. The
proximal and distal forelimb representations are located in between, with the
distal forelimb represented more laterally.
injection case (monkeys Lo and X) and 6036 (?3891) for the
distal injection case (monkeys Su and Si) (Fig. 4).
In contrast with the somatotopically arranged labeling in the
dorsal putamen, the labeling pattern in the ventral striatum was
rather consistent in all injection cases. All injections yielded
dense labeling in the ventromedial portion of the putamen and
the nucleus accumbens.
3.3. Labeling of striatal interneurons
To examine whether or not the viral labeling might extend to
interneuron groups in the striatum (fourth-order neurons),
representative sections were immunostained for a combination
of rabies virus and ChAT, a marker for striatal chlolinergic
interneurons, or rabies virus and PV, a marker for a group of
striatal GABAergic interneurons (see Fig. 1). In the putamen of
monkeys Vi (hindlimb), Lo (proximal forelimb), and Su (distal
forelimb), no double-labeled neurons were observed even in the
densely rabies-labeled zones, indicating that the viral labeling
double-labeled neurons were found in the zones of dense rabies
in this monkey involved at least partly fourth-order neurons.
3.4. Relationship of rabies labeling to striatal patch/matrix
Some striatal sections obtained from monkeys Su, Lo, and Vi
were immunostained for a combination of rabies virus and
calbindin D28k, a marker forthe striatal matrix compartment. As
shown in Fig. 5D–D00and E–E00, rabies-labeled neurons were
distributed not only in the calbindin-positive matrix compart-
ment, but also in the calbindin-negative patch compartment.
3.5. Distribution of subthalamic neuron labeling
A largenumber of labeled neurons were found in the STN as
well as in the striatum (Fig. 3, lower panel). The densest
labeling zone from the hindlimb or orofacial representation of
the MI was located in the dorsomedial or ventrolateral portion,
respectively. Neuronal labeling from the proximal and distal
forelimb regions of the MI seemed to occur in a zone between
the labeling zones from the hindlimb and orofacial representa-
tions (Fig. 3, lower panel). Again, many more neurons were
labeled from the distal forelimb region than from the proximal
forelimb region. The mean (?S.D.) number of labeled neurons
was 313 (?212) for the proximal injection case (monkeys Lo
and X) and 1458 (?410) for the distal injection case (monkeys
Su and Si) (Fig. 4). The dense labeling zones from the distal
forelimb region were located more laterally than those from the
proximal forelimb region.
Using retrograde transsynaptic transport of rabies virus, we
examined the organization of multisynaptic projections from
the striatum and STN to the MI. Both the striatum and the STN
receive direct inputs from the MI and send outputs back to the
MI via the GPi and thalamic nuclei (Smith et al., 1990; Flaherty
and Graybiel, 1994; Nambu et al., 1996; Sidibe et al., 1997;
Kelly and Strick, 2004). In the thalamus, the nucleus ventralis
lateralis pars oralis (VLo) that receives basal ganglia inputs is
et al., 1990; Darian-Smith et al., 1990). All monkeys tested in
the present study, in fact, had very dense neuronal labeling in
the VLo and the nucleus ventralis posterolateralis pars oralis
where the orofacial part, forelimb, and hindlimb were
represented mediolaterally in this order (data not shown).
Previous studies have shown that rabies virus labels third-order
neurons in approximately 4 days after the injection (Kelly and
Strick, 2000, 2004; Miyachi et al., 2005). Therefore, we
investigated the distribution patterns of retrograde transneur-
onal labeling in the striatum and STN 4 days after the viral
injectionsintothe hindlimb,proximal forelimb,distal forelimb,
and orofacial representations of the MI. In the striatum, two
distinct groups of neurons were labeled from each body-part
representation in the MI: one in the dorsal putamen (i.e., the
motor striatum) and the other in the ventral striatum, including
the ventromedial putamen and the nucleus accumbens (i.e., the
limbic striatum). The distribution pattern of neuronal labeling
in the dorsal putamen is in full agreement with the topography
of corticostriatal terminals from the MI. According to
anatomical (Flaherty and Graybiel, 1993; Takada et al.,
1998; Miyachi et al., 2005) and electrophysiological (Alex-
ander and DeLong, 1985; Nambu et al., 2002b) studies, the
hindlimb, forelimb, and orofacial representations are arranged
dorsoventrally in the putamen. In favor of such previous data,
our results obtained with retrograde transneuronal labeling
from the MI indicate that the hindlimb, forelimb, and orofacial
part are represented sequentially from dorsal to ventral in the
putamen (see also Kelly and Strick, 2004; Miyachi etal., 2005).
S. Miyachi et al./Neuroscience Research 56 (2006) 300–308304
labeled from the proximal vs. distal forelimb representations of the MI. Cell
counts in the putamen were performed in eight equidistant sections (1440 mm
end of the globus pallidus, while those in the STN were performed in 10 or 11
equidistant sections (360 mm apart).
S. Miyachi et al./Neuroscience Research 56 (2006) 300–308 305
Fig. 5. Double immunofluorescence histochemistry for rabies virus and ChAT, PV, or calbindin D28k. In the putamen of monkey Su (A–A00), no ChAT-positive
In this monkey, PVinterneurons (C–C00) are also double-labeled with rabies. In the putamen of monkey Su, rabies labeling is observed in both the calbindin-negative
patch (D–D00) and the calbindin-positive matrix (E–E00) compartments. Scale bars, 100 mm.
It has also been revealed in the present study that within the
forelimb territory of the putamen, the distal representation is
located more laterally than the proximal representation
(Fig. 6A). This is well consistent with the reported distribution
patterns of corticostriatal terminals from the corresponding
regions of the MI (Tokuno et al., 1999) and of striatal loci
responding to MI stimulation (Nambu et al., 2002b). Thus, the
present data combined with the previous findings clearly
indicate that the cortico-basal ganglia motor loop involving the
MI and the putamen consists of multiple closed subloops
representing different body parts.
In our experiments, the rabies injection into the distal
forelimb region of the MI produced much denser neuronal
labeling in the striatum (and also in the STN) than the injection
into the proximal forelimb region. This does not seem
technical, because the same results were repeatedly obtained
in four monkeys (two monkeys in each group), in which the
total volume of the viral suspension injected into the proximal
region was almost identical to or even larger than that injected
into the distal region. Moreover, in the proximal forelimb-
injection cases, the volume injected into the precentral bank
area of the MI was the same as or not much different from that
in the distal forelimb-injection cases, and constituted about
70% of the total volume injected into the proximal region (see
Table1;4of5.5 mlinmonkeyLo and3of4.5 mlinmonkeyX).
Thus, it can be concluded that the distal forelimb is represented
more prominently in the basal ganglia than the proximal
forelimb. In this context, it should be noted here that
stimulation of the distal forelimb region in the putamen readily
evoked wrist or digit movements, whereas stimulation of the
proximal forelimb region hardly elicited movements of either
the shoulder or the elbow (Nambu et al., 2002b).
Our double immunofluorescence histochemistry for rabies
virus and ChAT or PV has confirmed that striatal interneurons
containing acetylcholine and PV, relevant to fourth-order
neurons, are labeled transsynaptically from medium-sized
spiny projection neurons in the case where the viral infection
or PV immunoreactivity in the striatum was largely dampened
S. Miyachi et al./Neuroscience Research 56 (2006) 300–308306
Fig. 6. (A and B) Summary diagrams showing the arrangement of somatotopic representations in the putamen (A) and STN (B), as revealed by the distribution
patterns of retrograde transneuronal labeling from the MI. (C) Summary diagram showing the two somatotopically arranged closed motor loops in the cortico-basal
ganglia circuit through the direct and hyperdirect pathways.
(data not shown). This implies that rabies virus may probably
In addition to the somatotopically arranged labeling in the
‘motor striatum’, many neurons in the ‘limbic striatum’
including the ventromedial putamen and the nucleus accum-
bens were labeled regardless of the injection sites of rabies
virus. Kelly and Strick (2004) reported a similar pattern of
labeling in theventral striatum after rabies injection into the MI
arm representation. These data indicate that the ventral striatal
territory projects diffusely to the large area of the MI involving
the hindlimb, forelimb, and orofacial representations. In our
neurons were labeled in the ventral striatum as early as 3 days
point as GPi neurons were labeled). We assumed that this
labeling in the ventral striatum might be the second-order
labeling via the basal forebrain which receives strong input
from the ventral striatum and sends widespread output to the
cerebral cortex (Kievit and Kuypers, 1975; Mesulam et al.,
1983; Haber et al., 1990). The dense ventral labeling in the
present experiments can also be considered such second-order
labeling (and may perhaps include the third-order labeling
through the cortical areas projecting directly to the MI). It has
been well documented by Gerfen (1992) that not only the
ventral striatum, but also the patch compartment in the dorsal
striatum receives input from the limbic cortical areas (Eblen
and Graybiel, 1995; Ferry et al., 2000) and provides
‘nonspecific’ feedback to the cortex via cholinergic neurons
in the basal forebrain and dopaminergic neurons in the
midbrain, whereas the matrix compartment in the dorsal
striatum receives input from the sensorimotor cortical areas
(Flaherty and Graybiel, 1993) and provides more ‘specific’ or
topographically organized feedback to the motor cortex by way
of the basal ganglia-thalamo-cortical circuit (see also Gerfen,
1984, 1985; Grove et al., 1986). To examine which striatal
compartment, the patch or the matrix, might contain neuronal
populations projecting to the cortex multisynaptically, we
applied double immunostaining for rabies virus and calbindin
D28k, a marker for the matrix compartment. As shown in Fig. 5,
a number of rabies-labeled neurons were distributed in both the
patch and the matrix compartments. The present data strongly
support the idea that there is an open loop circuit from the
ventral striatum/the striatal patch compartment to the MI, in
addition to the somatotopically arranged closed loop involving
the MI and the striatal matrix compartment.
The neuronal labeling from the MI was also obtained in the
STN, such that zones representing the hindlimb, forelimb, and
orofacial part appeared to be allocated mediolaterally in the
lateral aspect of the nucleus (see also Kelly and Strick, 2004;
Miyachi et al., 2005). This somatotopic arrangement is in good
accordance with the distribution patterns of direct cortico-
subthalamic inputs from the MI (Nambu et al., 1996) and of
subthalamic zones responsive to somatosensory stimuli
(DeLong et al., 1985). Regarding the proximal vs. distal
forelimb representations, the sites of dense labeling from the
distal forelimb representation were located more laterally than
those from the proximal forelimb representation (Fig. 6B).
Nambu et al. (1996) have reported the occurrence of weaker
corticosubthalamic terminals in the medial STN, in addition to
thedensetermination inthe lateral STN.The present retrograde
transneuronal tracing also resulted in lighter neuronal labeling
in the medial STN after rabies injections into the MI.
It is generally accepted that the basal ganglia constitute
multiple parallel loop circuits with the frontal lobe via the
thalamus (Alexander et al., 1990; Parent and Hazrati, 1995).
Diverse information derived from the cortex is conveyed to the
striatum and processed through the direct (striato-GPi) and
indirect (striato-external pallidum-STN-GPi) pathways before
being sent back to the cortex(Gerfen, 1992; Parent and Hazrati,
1993; Bolam et al., 2000). Moreover, some cortical signals
reach the GPi through the hyperdirect (corticosubthalamic)
pathway (Nambu et al., 1996, 2000, 2002a). The present results
suggest that the cortico-basal ganglia motor loop not only
involving the MI and the putamen, but also involving the MI
and STN is parceled out into somatotopically distinct closed
subloops (Fig. 6C), whereas the ventral (‘limbic’) striatum
provides divergent multisynaptic inputs to the entire MI.
Education, Culture, Sports, Science and Technology of Japan.
Alexander, G.E., DeLong, M.R., 1985. Microstimulation of the primate neos-
triatum. II. Somatotopic organization of striatal microexcitable zones and
their relation to neuronal response properties. J. Neurophysiol. 53, 1417–
Alexander, G.E., Crutcher, M.D., DeLong, M.R., 1990. Basal ganglia-thala-
mocortical circuits: parallel substrates for motor, oculomotor, ‘‘prefrontal’’
and ‘‘limbic’’ functions. Prog. Brain Res. 85, 119–146.
Bolam,J.P., Hanley,J.J., Booth,P.A.,Bevan,M.D., 2000.Synapticorganisation
of the basal ganglia. J. Anat. 196, 527–542.
Darian-Smith, C., Darian-Smith, I., Cheema, S.S., 1990. Thalamic projections
to sensorimotor cortex in the macaque monkey: use of multiple retrograde
fluorescent tracers. J. Comp. Neurol. 299, 17–46.
DeLong, M.R., Crutcher, M.D., Georgopoulos, A.P., 1985. Primate globus
pallidus and subthalamic nucleus: functional organization. J. Neurophysiol.
Eblen, F., Graybiel, A.M., 1995. Highly restricted origin of prefrontal cortical
inputs to striosomes in the macaque monkey. J. Neurosci. 15, 5999–6013.
Ferry, A.T., Ongur, D., An, X., Price, J.L., 2000. Prefrontal cortical projections
to the striatum in macaque monkeys: evidence for an organization related to
prefrontal networks. J. Comp. Neurol. 425, 447–470.
Flaherty, A.W., Graybiel, A.M., 1993. Two input systems for body representa-
tions in the primate striatal matrix: experimental evidence in the squirrel
monkey. J. Neurosci. 13, 1120–1137.
Flaherty, A.W., Graybiel, A.M., 1994. Input-output organization of the sensor-
imotor striatum in the squirrel monkey. J. Neurosci. 14, 599–610.
Gerfen, C., 1984. The neostriatal mosaic: compartmentalization of corticos-
triatal input and striatonigral output systems. Nature 311, 461–464.
Gerfen, C., 1985. The neostriatal mosaic. I. Compartmental organization of
projections from the striatum to the substantia nigra in the rat. J. Comp.
Neurol. 236, 454–476.
Gerfen, C., 1992. The neostriatal mosaic: multiple levels of compartmental
organization. Trends Neurosci. 15, 133–139.
S. Miyachi et al./Neuroscience Research 56 (2006) 300–308 307
Graf, W., Gerrits, N., Yatim-Dhiba, N., Ugolini, G., 2002. Mapping the
oculomotor system: the power of transneuronal labelling with rabies virus.
Eur. J. Neurosci. 15, 1557–1562.
Grove, E.A., Domesick, V.B., Nauta, W.J., 1986.Light microscopic evidenceof
a study employing the anterograde tracer Phaseolus vulgaris leucoagglu-
tinin (PHA-L). Brain Res. 367, 379–384.
Haber, S.N., Lynd, E., Klein, C., Groenewegen, H.J., 1990. Topographic
an anterograde tracing study. J. Comp. Neurol. 293, 282–298.
Hedreen, J.C., DeLong, M.R., 1991. Organization of striatopallidal, striatoni-
gral, and nigrostriatal projections in the macaque. J. Comp. Neurol. 304,
Hoover, J.E., Strick, P.L., 1993. Multiple output channels in the basal ganglia.
Science 259, 819–821.
outputs to primary motor cortex as revealed by retrograde transneuronal
transport of herpes simplex virus type 1. J. Neurosci. 19, 1446–1463.
T., 2003. Cross-reactive antigenicity of nucleoproteins of lyssaviruses
recognized by a monospecific antirabies virus nucleoprotein antiserum
on paraffin sections of formalin-fixed tissues. Pathol. Int. 53, 525–533.
Kelly, R.M.,Strick,P.L.,2000.Rabiesasa transneuronaltracer ofcircuitsin the
central nervous system. J. Neurosci. Methods 103, 63–71.
Kelly, R.M., Strick, P.L., 2004. Macro-architecture of basal ganglia loops with
the cerebralcortex:use ofrabies virusto reveal multisynaptic circuits. Prog.
Brain Res. 143, 449–459.
frontal and parietal cortex in the Rhesus monkey. Science 187, 660–662.
Mesulam, M.M., Mufson, E.J., Levey, A.I., Wainer, B.H., 1983. Cholinergic
innervation of cortex by the basal forebrain: cytochemistry and cortical
connections of the septal area, diagonal band nuclei, nucleus basalis
(substantia innominata), and hypothalamus in the rhesus monkey. J. Comp.
Neurol. 214, 170–197.
Miyachi, S., Lu, X., Inoue, S., Iwasaki, T., Koike, S., Nambu, A., Takada, M.,
2005. Organization of multisynaptic inputs from prefrontal cortex to
primary motor cortex as revealed by retrograde transneuronal transport
of rabies virus. J. Neurosci. 25, 2547–2556.
Moschovakis, A.K., Gregoriou, G.G., Ugolini, G., Doldan, M., Graf, W.,
Guldin, W., Hadjidimitrakis, K., Savaki, H.E., 2004. Oculomotor areas
of the primate frontal lobes: a transneuronal transfer of rabies virus and
[14C]-2-deoxyglucose functional imaging study. J. Neurosci. 24, 5726–
Nambu, A., Takada, M., Inase, M., Tokuno, H., 1996. Dual somatotopical
reversed body-map transformations from the primary motor cortex and the
supplementary motor area. J. Neurosci. 16, 2671–2683.
Y., Hasegawa, N., 2000. Excitatory cortical inputs to pallidal neurons via the
subthalamic nucleus in the monkey. J. Neurophysiol. 84, 289–300.
Nambu, A., Tokuno, H., Takada, M., 2002a. Functional significance of the
cortico-subthalamo-pallidal ‘hyperdirect’ pathway. Neurosci. Res. 43, 111–
Nambu, A., Kaneda, K., Tokuno, H., Takada, M., 2002b. Organization of
corticostriatal motor inputs in monkey putamen.J. Neurophysiol. 88, 1830–
Parent, A., Hazrati, L.N., 1993. Anatomical aspects of information processing
in primate basal ganglia. Trends Neurosci. 16, 111–116.
Parent, A., Hazrati, L.N., 1995. Functional anatomy of the basal ganglia. I.
The cortico-basal ganglia-thalamo-cortical loop. Brain Res. Rev. 20,
Sidibe, M., Bevan, M.D., Bolam, J.P., Smith, Y., 1997. Efferent connections of
the internal globus pallidus in the squirrel monkey. I. Topography and
synaptic organization of the pallidothalamic projection. J. Comp. Neurol.
Smith, Y., Hazrati, L.N., Parent, A., 1990. Efferent projections of the
subthalamic nucleus in the squirrel monkey as studied by the PHA-L
anterograde tracing method. J. Comp. Neurol. 294, 306–323.
from the somatic motor areas of the frontal cortex in the macaque monkey:
supplementary motor area, and the premotor cortex. Exp. Brain Res. 120,
Tokuno, H., Inase,M., Nambu, A., Akazawa,T., Miyachi, S., Takada, M., 1999.
Corticostriatal projections from distal and proximal forelimb representa-
tions of the monkey primary motor cortex. Neurosci. Lett. 269, 33–36.
S. Miyachi et al./Neuroscience Research 56 (2006) 300–308 308