**3. Basal ganglia control of movements and motor disturbances by the basal ganglia dysfunction**

It is established that the C-BG loop is required for volitional movement (Delong & Wichmann, 2007; Middleton & Strick, 2000). Neural circuits between the prefrontal cortex and the caudate nucleus (cognitive loop) are involved in the regulation of complex, visuallyguided limb movements and the planning and programming those movements. Neural

Possible Contribution of the Basal Ganglia

locomotion in decerebrate cats

PAG; periaqueductal grey, RN; red nucleus.

Brainstem System to the Pathogenesis of Parkinson's Disease 441

Fig. 4. Nigral stimulus effects on PPN/MLR-induced muscle tone suppression and

injection of muscimol into the lateral part of the SNr. C. (a) MLR-activated stepping movements. (b) Inhibition of MLR-activated stepping by SNr stimulation. (c) Locomotion induced by an injection of muscimol into the medial part of the SNr. D. Changes in step cycles and gait onset of the MLR-activated locomotion following changes in stimulus intensity (a) and frequency (b) applied to the SNr. Abbreviations, III; oculomotor nerve,

**3.2 Role of the C-BG loop and BG-BS system in relation to the gait control** 

As illustrated in Fig.5A, output from the SNr and the internal segment of globus pallidus (GPi) is regulated by hyper-direct, direct and indirect pathways in the basal ganglia circuits (Delong & Wichmann, 2007; Hikosaka et al., 2000; Numbu, 2004). Neurons in the GPi/SNr inhibit target neurons in the thalamus and brainstem with their tonic, high background activity, thus preventing unnecessary movements. To initiate movements, motor commands from the motor cortices first increase the basal ganglia output by an activation of the hyper-direct pathway to the subthalamic nucleus (STN) so that the excitability of target systems would be further reduced (Fig.5Ba). Signals via the direct pathway from the striatum to the GPi/SNr remove this sustained inhibition, resulting in a disinhibition of the target systems (Fig.5Bb). The phasic activity of GABAergic output neurons in the striatum, which are mostly silent, interrupts the tonic GPi/SNr inhibition, and movements are allowed to occur. Finally, signals through the indirect pathway, involving the external segment of the globus pallidus (GPe) and the STN, can further re-enhance the inhibition of the target systems (Fig.5Bc). This sequential information processing, an enhancement of tonic inhibition and disinhibition, would enhance the temporal contrast of the excitability of the target systems so that only the selected motor

A. (a) Experimental design in decerebrate cat preparation. (b) Effective stimulus sites in the SNr for inhibition of the PPN (filled squares) and the MLR (open squares) effects. (c) Effective stimulus sites for evoking muscular atonia (PPN; red circles) and locomotion (MLR; blue circles) in the mesopontine tegmentum. B. (a) PPN-induced muscular atonia. (b) Inhibition of the PPN-induced atonia by SNr stimulation. (c) Muscular atonia induced by an

circuits between motor cortical areas, including the primary motor cortex, PM and SMA, and the putamen (motor loop) contribute to the regulation of voluntary, discrete, ipsilateral limb movements. In addition the BG-BS system may control automatic and steady-state locomotor movements. This section first refers how the BG-BS system controls posture and locomotion and then considers how BG-BS contributes to pathophysiological mechanisms of motor disturbances in PD.

#### **3.1 BG-BS system controls postural muscle tone and locomotion**

How do the basal ganglia control locomotion and muscle tone via the BG-BS system? To answer this question, we employed decerebrate cat preparation where only the substantia nigra was preserved, whereas most basal ganglia structures were removed (Fig.4A). Then we examined how GABAergic output from the basal ganglia to the brainstem modulated the PPN/MLR-activated movements by manipulating the activity of neurons in the substantia nigra pars reticulata (SNr). Repetitive stimuli applied to the ventrolateral part of the PPN (red circles in Fig.4Ac) abolished muscle tone (Fig.4Ba). While stimulation of the lateral part of the SNr (filled squares in Fig.4Ab) alone did not alter muscle tone, it completely blocked the PPN-effects (Fig.4Bb). Stimulation of the MLR (blue circles in Fig.4Ac) increased muscle tone and evoked stepping movements on stationary surface (Fig.4Ca). But stimulation of the medial part of the SNr, indicated by filled squares in Fig.4Ab, arrested the MLR-activated locomotion (Fig.4Cb). It was re-established after termination of the SNr stimulation. In immobilized decerebrate preparation, MLR stimulation first depolarized the membrane potential and then generated rhythmic membrane oscillations associating with bursting firing, which corresponded to step cycles, in hindlimb motoneurons (fictive locomotion). During SNr stimulation, the rhythmic oscillation was arrested, and membrane potential was maintained at the depolarizing state in both extensor and flexor motoneurons, indicating that SNr stimulation co-contracts agonistic and antagonistic muscles, leading to disturbing rhythmic limb movements. Because these effects by SNr stimulation were blocked by injections of bicuculline (a GABAA-receptor antagonist) into the PPN/MLR, the SNr stimulus effects can be mediated by GABAergic projections to these areas. These findings suggest that the BG-BS system controls steady state (e.g., rhythmic limb movements) and dynamic state (e.g., initiation and termination) of locomotion. It should be noted that the effects of SNr stimulation was depend on stimulus parameters. Step cycles and onset of MLR-activated locomotion was prolonged by increasing SNr stimulus intensity (Fig.4Da). However frequency more than 140 Hz was less effective while frequencies between 50 and 100 Hz were prominently inhibited locomotor activity (Takakusaki et al., 2003a). Similar findings were observed in the nigral control of muscle tone (Takakusaki et al., 2004c). Frequency between 50-100 Hz is almost the same as the spontaneous firing rates SNr neurons in alert monkey (Hikosaka & Wurtz, 1985). A frequency within this range can be a critical determinant in the control of muscle tone, locomotion and saccadic eye movements (Hikosaka et al., 2000).

Next we injected muscimol, a GABAA receptor agonist, into the SNr in order to inhibit neuronal activity. It was observed that muscimol injections into the lateral part of the SNr suppressed postural muscle tone (Fig.4Bc) and those into the medial part of the SNr evoked locomotion (Fig.4Cc). Judging from the relationship between effective stimulus sites in the SNr (Fig.4Ab) and those in the mesopontine tegmentum (Fig.4Ac), there exist functional topography in the nigrotegmental projections; lateral SNr controls muscle tone and the medial SNr controls locomotion. It follows that the BG-BS system can control locomotion and postural muscle tone independently.

circuits between motor cortical areas, including the primary motor cortex, PM and SMA, and the putamen (motor loop) contribute to the regulation of voluntary, discrete, ipsilateral limb movements. In addition the BG-BS system may control automatic and steady-state locomotor movements. This section first refers how the BG-BS system controls posture and locomotion and then considers how BG-BS contributes to pathophysiological mechanisms of

How do the basal ganglia control locomotion and muscle tone via the BG-BS system? To answer this question, we employed decerebrate cat preparation where only the substantia nigra was preserved, whereas most basal ganglia structures were removed (Fig.4A). Then we examined how GABAergic output from the basal ganglia to the brainstem modulated the PPN/MLR-activated movements by manipulating the activity of neurons in the substantia nigra pars reticulata (SNr). Repetitive stimuli applied to the ventrolateral part of the PPN (red circles in Fig.4Ac) abolished muscle tone (Fig.4Ba). While stimulation of the lateral part of the SNr (filled squares in Fig.4Ab) alone did not alter muscle tone, it completely blocked the PPN-effects (Fig.4Bb). Stimulation of the MLR (blue circles in Fig.4Ac) increased muscle tone and evoked stepping movements on stationary surface (Fig.4Ca). But stimulation of the medial part of the SNr, indicated by filled squares in Fig.4Ab, arrested the MLR-activated locomotion (Fig.4Cb). It was re-established after termination of the SNr stimulation. In immobilized decerebrate preparation, MLR stimulation first depolarized the membrane potential and then generated rhythmic membrane oscillations associating with bursting firing, which corresponded to step cycles, in hindlimb motoneurons (fictive locomotion). During SNr stimulation, the rhythmic oscillation was arrested, and membrane potential was maintained at the depolarizing state in both extensor and flexor motoneurons, indicating that SNr stimulation co-contracts agonistic and antagonistic muscles, leading to disturbing rhythmic limb movements. Because these effects by SNr stimulation were blocked by injections of bicuculline (a GABAA-receptor antagonist) into the PPN/MLR, the SNr stimulus effects can be mediated by GABAergic projections to these areas. These findings suggest that the BG-BS system controls steady state (e.g., rhythmic limb movements) and dynamic state (e.g., initiation and termination) of locomotion. It should be noted that the effects of SNr stimulation was depend on stimulus parameters. Step cycles and onset of MLR-activated locomotion was prolonged by increasing SNr stimulus intensity (Fig.4Da). However frequency more than 140 Hz was less effective while frequencies between 50 and 100 Hz were prominently inhibited locomotor activity (Takakusaki et al., 2003a). Similar findings were observed in the nigral control of muscle tone (Takakusaki et al., 2004c). Frequency between 50-100 Hz is almost the same as the spontaneous firing rates SNr neurons in alert monkey (Hikosaka & Wurtz, 1985). A frequency within this range can be a critical determinant in the control of

**3.1 BG-BS system controls postural muscle tone and locomotion** 

muscle tone, locomotion and saccadic eye movements (Hikosaka et al., 2000).

and postural muscle tone independently.

Next we injected muscimol, a GABAA receptor agonist, into the SNr in order to inhibit neuronal activity. It was observed that muscimol injections into the lateral part of the SNr suppressed postural muscle tone (Fig.4Bc) and those into the medial part of the SNr evoked locomotion (Fig.4Cc). Judging from the relationship between effective stimulus sites in the SNr (Fig.4Ab) and those in the mesopontine tegmentum (Fig.4Ac), there exist functional topography in the nigrotegmental projections; lateral SNr controls muscle tone and the medial SNr controls locomotion. It follows that the BG-BS system can control locomotion

motor disturbances in PD.

Fig. 4. Nigral stimulus effects on PPN/MLR-induced muscle tone suppression and locomotion in decerebrate cats

A. (a) Experimental design in decerebrate cat preparation. (b) Effective stimulus sites in the SNr for inhibition of the PPN (filled squares) and the MLR (open squares) effects. (c) Effective stimulus sites for evoking muscular atonia (PPN; red circles) and locomotion (MLR; blue circles) in the mesopontine tegmentum. B. (a) PPN-induced muscular atonia. (b) Inhibition of the PPN-induced atonia by SNr stimulation. (c) Muscular atonia induced by an injection of muscimol into the lateral part of the SNr. C. (a) MLR-activated stepping movements. (b) Inhibition of MLR-activated stepping by SNr stimulation. (c) Locomotion induced by an injection of muscimol into the medial part of the SNr. D. Changes in step cycles and gait onset of the MLR-activated locomotion following changes in stimulus intensity (a) and frequency (b) applied to the SNr. Abbreviations, III; oculomotor nerve, PAG; periaqueductal grey, RN; red nucleus.

#### **3.2 Role of the C-BG loop and BG-BS system in relation to the gait control**

As illustrated in Fig.5A, output from the SNr and the internal segment of globus pallidus (GPi) is regulated by hyper-direct, direct and indirect pathways in the basal ganglia circuits (Delong & Wichmann, 2007; Hikosaka et al., 2000; Numbu, 2004). Neurons in the GPi/SNr inhibit target neurons in the thalamus and brainstem with their tonic, high background activity, thus preventing unnecessary movements. To initiate movements, motor commands from the motor cortices first increase the basal ganglia output by an activation of the hyper-direct pathway to the subthalamic nucleus (STN) so that the excitability of target systems would be further reduced (Fig.5Ba). Signals via the direct pathway from the striatum to the GPi/SNr remove this sustained inhibition, resulting in a disinhibition of the target systems (Fig.5Bb). The phasic activity of GABAergic output neurons in the striatum, which are mostly silent, interrupts the tonic GPi/SNr inhibition, and movements are allowed to occur. Finally, signals through the indirect pathway, involving the external segment of the globus pallidus (GPe) and the STN, can further re-enhance the inhibition of the target systems (Fig.5Bc). This sequential information processing, an enhancement of tonic inhibition and disinhibition, would enhance the temporal contrast of the excitability of the target systems so that only the selected motor

Possible Contribution of the Basal Ganglia

Brainstem System to the Pathogenesis of Parkinson's Disease 443

Fig. 6. Hypothetical models for movement control by the basal ganglia

ganglia control of voluntary movements, locomotion and muscle tone. C.

of freedom of movements may be restricted.

A. GABAergic basal ganglia projections to the thalamocortical neurons are involved in the volitional control of movements, while those to the MLR and the PPN may be responsible for the automatic control of locomotion and muscle tone. B. Normal operation of the basal

Pathophysiological changes in the activities of the cortico-BG loop and BG-BS system in PD. Loss of dopamine results in an increase in the basal ganglia output to the cerebral cortex, the limbic system and the brainstem. Consequently, voluntary and cognitive activities of the cerebral cortex and emotional expression can be reduced. Reduced cortical output results in bradykinesia and hypokinesia. Reduced activity in the MLR-locomotor system may induce gait failure. Inhibition of the PPN-muscle tone inhibitory system may induce hypertonus. D. Motor disturbances in basal ganglia disorders. (a) Parkinson disease. (b) Huntington's disease. Regardless of whether an increase or a decrease in the basal ganglia output, degree

Given the above consideration, we propose a hypothetical model in Fig.6A for the basal ganglia control of movements. The motor cortical neurons that receive basal ganglia output may control the velocity and the amount of voluntary movement (Turner & Anderson, 1997), which is indicated in the ordinate on the left of the graph in Fig.6Ba. The GABAergic basal ganglia output to the MLR reduced the drive from the MLR, resulting in disruption of the activity of CPGs in the spinal cord. Basal ganglia efferents to the MLR may therefore control the locomotor pattern (ordinate on the right in Fig.6Ba). In addition, basal ganglia efferents to the PPN may determine the level of muscle tone via the muscle tone control

program could be initiated, executed and terminated at the appropriate timing, whereas other competing programs can be cancelled (Hikosaka et al., 2000; Numbu, 2004). This is the "first key mechanism" of movement control by the basal ganglia.

The above mechanisms may act on brainstem networks, including the locomotor system and muscle tone control systems (Fig.5A, lower right). Therefore the brainstem networks could be combined with basal ganglia motor circuits. In this "hybrid model", output of the basal ganglia controls the MLR for locomotion and the PPN for muscle tone via GABAergic projection. When locomotor movement is being prepared, tonic activity of SNr neurons would continuously inhibit both systems. When a trigger signal occurred, the hyper-direct pathway would enhance the inhibition. Then the direct pathway would release the activity of these systems, resulting in an initiation of locomotion that would be followed by a smooth reduction of the level of muscle tone. To terminate the locomotion, the direct pathway would inhibit each system, resulting in a cessation of rhythmic locomotor movements and an accompanying increase in the level of muscle tone (muscle cocontraction). A parallel organization from the SNr to the MLR/PPN would be therefore assist regulation of the level of muscle tone which was appropriate for the initiation and termination of locomotion.

Fig. 5. Hybrid model of C-BG loop and BG-BS system

A. Left; basal ganglia motor circuits. Lower right; BG-BS system for controlling locomotion and muscle tone. B and C. Changes in the basal ganglia output and in the excitability of target systems following sequential information processing of hyper-direct (a), direct (b) and indirect (c) pathways. When excitability of target systems goes beyond the threshold, movements occur. Excitability of direct and indirect pathways is modified by dopaminergic projections from the SNc to the striatum. B. Normal condition. C. Parkinson's disease. Abbreviations, D1 and D2; D1 and D2-dopamine receptors, DA; dopamine, enk; enkephaline, Glu; glutamate, GPe; external segment of globus pallidus subP; substance P, STN; subthalamic nucleus

program could be initiated, executed and terminated at the appropriate timing, whereas other competing programs can be cancelled (Hikosaka et al., 2000; Numbu, 2004). This is the "first

The above mechanisms may act on brainstem networks, including the locomotor system and muscle tone control systems (Fig.5A, lower right). Therefore the brainstem networks could be combined with basal ganglia motor circuits. In this "hybrid model", output of the basal ganglia controls the MLR for locomotion and the PPN for muscle tone via GABAergic projection. When locomotor movement is being prepared, tonic activity of SNr neurons would continuously inhibit both systems. When a trigger signal occurred, the hyper-direct pathway would enhance the inhibition. Then the direct pathway would release the activity of these systems, resulting in an initiation of locomotion that would be followed by a smooth reduction of the level of muscle tone. To terminate the locomotion, the direct pathway would inhibit each system, resulting in a cessation of rhythmic locomotor movements and an accompanying increase in the level of muscle tone (muscle cocontraction). A parallel organization from the SNr to the MLR/PPN would be therefore assist regulation of the level of muscle tone which was appropriate for the initiation and

key mechanism" of movement control by the basal ganglia.

Fig. 5. Hybrid model of C-BG loop and BG-BS system

A. Left; basal ganglia motor circuits. Lower right; BG-BS system for controlling locomotion and muscle tone. B and C. Changes in the basal ganglia output and in the excitability of target systems following sequential information processing of hyper-direct (a), direct (b) and

indirect (c) pathways. When excitability of target systems goes beyond the threshold, movements occur. Excitability of direct and indirect pathways is modified by dopaminergic projections from the SNc to the striatum. B. Normal condition. C. Parkinson's disease. Abbreviations, D1 and D2; D1 and D2-dopamine receptors, DA; dopamine, enk;

enkephaline, Glu; glutamate, GPe; external segment of globus pallidus subP; substance P,

termination of locomotion.

STN; subthalamic nucleus

Pathophysiological changes in the activities of the cortico-BG loop and BG-BS system in PD. Loss of dopamine results in an increase in the basal ganglia output to the cerebral cortex, the limbic system and the brainstem. Consequently, voluntary and cognitive activities of the cerebral cortex and emotional expression can be reduced. Reduced cortical output results in bradykinesia and hypokinesia. Reduced activity in the MLR-locomotor system may induce gait failure. Inhibition of the PPN-muscle tone inhibitory system may induce hypertonus. D. Motor disturbances in basal ganglia disorders. (a) Parkinson disease. (b) Huntington's disease. Regardless of whether an increase or a decrease in the basal ganglia output, degree of freedom of movements may be restricted.

Given the above consideration, we propose a hypothetical model in Fig.6A for the basal ganglia control of movements. The motor cortical neurons that receive basal ganglia output may control the velocity and the amount of voluntary movement (Turner & Anderson, 1997), which is indicated in the ordinate on the left of the graph in Fig.6Ba. The GABAergic basal ganglia output to the MLR reduced the drive from the MLR, resulting in disruption of the activity of CPGs in the spinal cord. Basal ganglia efferents to the MLR may therefore control the locomotor pattern (ordinate on the right in Fig.6Ba). In addition, basal ganglia efferents to the PPN may determine the level of muscle tone via the muscle tone control

Possible Contribution of the Basal Ganglia

loop are controlled separately in dystonia.

basal ganglia diseases.

symptoms in PD.

Brainstem System to the Pathogenesis of Parkinson's Disease 445

background of PD and HD. Consequently, dysfunction of the BG-BS system together with that of the C-BG loop may underlie the pathogenesis of the motor disturbances in these

Dystonia is a syndrome characterized by abnormal postures, muscle spasms and tremor, due to involuntary muscle co-contractions. Some dystonia are task specific, and patients only develop muscular co-contraction when performing skilled movements such as writing (Van der Kamp et al. 1989). By using positron emission tomography an inappropriate overactivity of the basal ganglia projections to the premotor and dorsal prefrontal cortex has been observed (Brooks 1995). However the activity of the primary sensorimotor and caudal premotor cortices is rather attenuated (Hutchins et al. 1988). Although alterations of noradrenaline and DA levels in brainstem structures have been reported in two cases (Hornykiewicz et al., 1986), most studies, by contrast, have not found abnormalities in the brainstem. This evidence suggests that the activity of the BG-BS system and that of the C-BG

Recently PPN/MLR area became one of targets of deep brain stimulation (DBS) for neurosurgical therapy for PD (PPN-DBS) (Stefani et al., 2007; Pierantozzi et al., 2008; Alessandro et al., 2010). Low frequency stimulation (~25Hz) applied to the above area ameliorated postural disturbance and gait failure. On the other hand, DBS applied to the SNr (SNr-DBS) with high frequency (135-190 Hz), which possibly intervened to the output from the SNr, also ameliorated axial symptoms such as gait akinesia and postural disturbances (Chasetan et al. 2009). Although evidence of the PPN-DBS and the SNr-DBS is still limited, these clinical findings agree well with our results suggesting that the BG-BS

Disturbances in cognitive and psychotic processes have been observed in patients with degenerative disorders that involve primarily the basal ganglia such as PD (Mellers et al., 1995; Taylor et al., 1986) and HD (McHugh & Folsten, 1975). Awake-sleep states were also impaired in PD (Bliwise et al., 2000; Eisensehr et al., 2001). It is also reported that PD is preceded and accompanied by daytime sleep attacks, nocturnal insomnia, REM sleep behavior disorder, hallucinations and depression, symptoms which are frequently as troublesome as the motor symptoms of this disease. All these symptoms are present in narcolepsy (Thannical et al., 2007). These clinical evidences corroborate that the basal ganglia and their connections with the brainstem are also involved in the expression of nonmotor function. In this section, we focus on the roles played by the BG-BS system in the regulation of vigilance states, arousal state, attention and cognition in relation to non-motor

Cholinergic neurons in the PPN and laterodorsal tegmental nucleus are thought to be involved in not only the maintenance of arousal state but also generation of REM sleep (Datta and Siwek, 2002; Koyama & Sakai, 2000; Maloney et al., 1999). Therefore, we elucidated how GABAergic SNr-PPN projection altered the activities of the REM generator and the muscle tone inhibitory system (Takakusaki et al., 2004c). Summary of the results are shown in Fig.7. Stimulation of inhibitory region of the PPN induced REM which was associated with muscular atonia in decerebrate cats (REM and atonia; Fig.7Ba). Conditioning

system contributes to the postural and locomotor synergies in human.

**4.1 Does output of the basal ganglia modulate sleep?** 

**4. Disturbances of non-motor functions in Parkinson's disease** 

systems (abscissa). Because the basal ganglia output is variable in a normal condition (Fig.6Ba), the degree of freedom for the amount and the velocity of movement, the locomotor velocity, and the muscle tone, can be large. Each parameter can take any of the coordinates within the frame in Fig.6Ba. For example, when a subject needs to adapt to a heavy load during walking, the subject may unconsciously select an appropriate gait pattern which is associated with a higher level of muscle tone and slower walking speed. Such a gait pattern could be realized by an increase in sustained SNr output to the PPN and the MLR. This would result in a decrease in the excitabilities of the inhibitory system and locomotor system (Fig.6Bb). Consequently, a sustained basal ganglia output may control the degree of freedom of the excitability of the target systems during movements. This can be the "second key mechanism" of motor control by the basal ganglia.

#### **3.3 Pathophysiological mechanisms of motor disturbances in basal ganglia disorders**

Gait disturbances and postural instability are observed in PD patients (Morris et al., 1994; Murray et al., 1978). These are delays in gait onset including freezing of gait (FOG), an increase in the stance phase in locomotor cycles, tiny steps and a decrease in gait velocity. Neurodegeneration of dopamine neurons in PD patients leads to higher activity in the hyper-direct and indirect pathways (Fig.5Ca and Cc) and lower activity of the direct pathway (Fig.5Cb). Therefore the GABAergic basal ganglia output is thought to be overactive in PD (Fig.6C). Excessive inhibitory actions on the target systems can consequently produce motor disturbances. For example, the excessive inhibition upon thalamocortical neurons may suppress cortical information processing (dysfunction of C-BG loop). A decrease in the output from the primary motor cortex reduces the amount (hypokinesia) and the velocity of movement (bradykinesia), leading to tiny step with a decrease in gait velocity. Moreover, reduced activities in the prefrontal cortex and PM/SMA may disturb planning and motor programming, respectively (Hanakawa et al. 1999). This may further disturb intentional gait control (gait initiation, precise hoot placement and obstacle avoidance), resulting in FOG. In the brainstem, the excessive inhibition of the MLR together with a decrease in cortical excitation of the reticular formation may decrease the activity of locomotor system and then disturb automatic aspect of steady-state gait control (rhythmic limb movements). Similarly, an increase in basal ganglia inhibition together with a decrease in cortical excitation of the PPN may reduce the activity of inhibitory system, which, in turn, facilitates excitatory systems. As a result, muscle tone would be increased (hypertonus). Therefore muscular rigidity, one of the most prominent symptoms of PD, can be the result of inhibition of the muscle tone inhibitory system that reduces the inhibition to α- and γ-motoneurons. As shown in Fig.4C, MLR-activated locomotion was arrested but the muscle tone was maintained at higher level during the period of SNr stimulation, indicating that muscle rigidity is a cause of gait disturbances. We postulate that dysfunction of the BG-BS system can be the primary basis for gait impairments of PD.

In contrast, an output from the basal ganglia is decreased in Huntington's disease (HD) because of increased activity of the direct pathway. This may extremely facilitate cortical information processing, thus unnecessary motor programs cannot be cancelled, resulting in hyperkinesia and involuntary movements (Chorea). The decrease in the basal ganglia output to the PPN may reduce muscle tone (hypotonus). It should be noted, regardless of PD or HD, the degree of freedom of movements would be reduced and restricted. The frame moves to the upper right for PD (Fig.6Da) and to lower left for HD (Fig.6Db). From these considerations, the reduction of the degree of freedom of movements could exist in the

systems (abscissa). Because the basal ganglia output is variable in a normal condition (Fig.6Ba), the degree of freedom for the amount and the velocity of movement, the locomotor velocity, and the muscle tone, can be large. Each parameter can take any of the coordinates within the frame in Fig.6Ba. For example, when a subject needs to adapt to a heavy load during walking, the subject may unconsciously select an appropriate gait pattern which is associated with a higher level of muscle tone and slower walking speed. Such a gait pattern could be realized by an increase in sustained SNr output to the PPN and the MLR. This would result in a decrease in the excitabilities of the inhibitory system and locomotor system (Fig.6Bb). Consequently, a sustained basal ganglia output may control the degree of freedom of the excitability of the target systems during movements. This can be the "second

**3.3 Pathophysiological mechanisms of motor disturbances in basal ganglia disorders**  Gait disturbances and postural instability are observed in PD patients (Morris et al., 1994; Murray et al., 1978). These are delays in gait onset including freezing of gait (FOG), an increase in the stance phase in locomotor cycles, tiny steps and a decrease in gait velocity. Neurodegeneration of dopamine neurons in PD patients leads to higher activity in the hyper-direct and indirect pathways (Fig.5Ca and Cc) and lower activity of the direct pathway (Fig.5Cb). Therefore the GABAergic basal ganglia output is thought to be overactive in PD (Fig.6C). Excessive inhibitory actions on the target systems can consequently produce motor disturbances. For example, the excessive inhibition upon thalamocortical neurons may suppress cortical information processing (dysfunction of C-BG loop). A decrease in the output from the primary motor cortex reduces the amount (hypokinesia) and the velocity of movement (bradykinesia), leading to tiny step with a decrease in gait velocity. Moreover, reduced activities in the prefrontal cortex and PM/SMA may disturb planning and motor programming, respectively (Hanakawa et al. 1999). This may further disturb intentional gait control (gait initiation, precise hoot placement and obstacle avoidance), resulting in FOG. In the brainstem, the excessive inhibition of the MLR together with a decrease in cortical excitation of the reticular formation may decrease the activity of locomotor system and then disturb automatic aspect of steady-state gait control (rhythmic limb movements). Similarly, an increase in basal ganglia inhibition together with a decrease in cortical excitation of the PPN may reduce the activity of inhibitory system, which, in turn, facilitates excitatory systems. As a result, muscle tone would be increased (hypertonus). Therefore muscular rigidity, one of the most prominent symptoms of PD, can be the result of inhibition of the muscle tone inhibitory system that reduces the inhibition to α- and γ-motoneurons. As shown in Fig.4C, MLR-activated locomotion was arrested but the muscle tone was maintained at higher level during the period of SNr stimulation, indicating that muscle rigidity is a cause of gait disturbances. We postulate that dysfunction of the BG-

key mechanism" of motor control by the basal ganglia.

BS system can be the primary basis for gait impairments of PD.

In contrast, an output from the basal ganglia is decreased in Huntington's disease (HD) because of increased activity of the direct pathway. This may extremely facilitate cortical information processing, thus unnecessary motor programs cannot be cancelled, resulting in hyperkinesia and involuntary movements (Chorea). The decrease in the basal ganglia output to the PPN may reduce muscle tone (hypotonus). It should be noted, regardless of PD or HD, the degree of freedom of movements would be reduced and restricted. The frame moves to the upper right for PD (Fig.6Da) and to lower left for HD (Fig.6Db). From these considerations, the reduction of the degree of freedom of movements could exist in the background of PD and HD. Consequently, dysfunction of the BG-BS system together with that of the C-BG loop may underlie the pathogenesis of the motor disturbances in these basal ganglia diseases.

Dystonia is a syndrome characterized by abnormal postures, muscle spasms and tremor, due to involuntary muscle co-contractions. Some dystonia are task specific, and patients only develop muscular co-contraction when performing skilled movements such as writing (Van der Kamp et al. 1989). By using positron emission tomography an inappropriate overactivity of the basal ganglia projections to the premotor and dorsal prefrontal cortex has been observed (Brooks 1995). However the activity of the primary sensorimotor and caudal premotor cortices is rather attenuated (Hutchins et al. 1988). Although alterations of noradrenaline and DA levels in brainstem structures have been reported in two cases (Hornykiewicz et al., 1986), most studies, by contrast, have not found abnormalities in the brainstem. This evidence suggests that the activity of the BG-BS system and that of the C-BG loop are controlled separately in dystonia.

Recently PPN/MLR area became one of targets of deep brain stimulation (DBS) for neurosurgical therapy for PD (PPN-DBS) (Stefani et al., 2007; Pierantozzi et al., 2008; Alessandro et al., 2010). Low frequency stimulation (~25Hz) applied to the above area ameliorated postural disturbance and gait failure. On the other hand, DBS applied to the SNr (SNr-DBS) with high frequency (135-190 Hz), which possibly intervened to the output from the SNr, also ameliorated axial symptoms such as gait akinesia and postural disturbances (Chasetan et al. 2009). Although evidence of the PPN-DBS and the SNr-DBS is still limited, these clinical findings agree well with our results suggesting that the BG-BS system contributes to the postural and locomotor synergies in human.
