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

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

Possible Contribution of the Basal Ganglia

1991; Inglis & Winn, 1995; Jones, 1991).

**4.2.2 Narcolepsy-like symptoms** 

PD (Takakusaki 2008).

Brainstem System to the Pathogenesis of Parkinson's Disease 447

Several mechanisms are postulated in relation to the basal ganglia regulation of sleep. On one hand, recent brain imaging studies revealed that damage of brainstem, particularly the reticular formation, is critically involved in the pathogenesis of RBD (Unger et al., 2010). The brainstem damage could also explain some non-motor symptoms in this disease, which often precede diagnosis, such as autonomic dysfunction and sleep disorders. On the other hand, roles of dopaminergic influence on the basal ganglia in the control of sleep-wake behavior are suggested (Mena-Segovia J et al., 2008). It is also possible that basal ganglia efferents to the non-specific thalamic nuclei may affect awake-sleep states by modulating the activity of ascending reticular activating system (ARAS). Since the classical study of Moruzzi and Magoun (1949), the pontomesencephalic reticular formation has been known to comprise the ARAS. The PPN has been considered as a part of the ARAS (Garcia-Rill,

The SNr has a direct projection to the thalamic nuclei (Parent et al., 1983; Pare´ et al., 1990) in addition to the PPN (Fig.7D). Because PPN has dense projections to the midbrain dopaminergic neurons, activity of the PPN neurons may affect awake-sleep states by modulating dopaminergic systems projecting to the basal ganglia and extra-basal ganglia areas. Consequently, our idea is that basal ganglia output from the SNr may affect awake– sleep cycles by modulating the activity of the ARAS through dual mechanisms (Fig.7D). One is through direct nigro-thalamic projection, and the other, which is considered in this study,

The presence in PD patients of narcolepsy-like features, such as daytime REM sleep intrusions associated with visual hallucinations, has led some authors to suggest that a mechanism similar to that of narcolepsy might underlie excessive daytime sleepiness (EDS) in PD (Arnulf et al., 2000). Thannickal et al. (2007) demonstrated that a massive loss of orexin neurons was found in PD patients and suggested that it was a cause of the narcolepsy-like symptoms. However, Compta et al. (2009) showed that orexin-A level was normal in the cerebrospinal fluid and it was unrelated to severity of sleepiness or the cognitive status of PD patients. Therefore alternative mechanisms other than dysfunction of orexin neurons might be responsible for EDS and the disturbance of sleep architecture in PD. In animal experiments, midbrain strucutes, including the SNr, the PPN and the MLR, receive orexinergic efferents from the perifornical lateral hypothalams (Nambu et al., 1999; Peyron et al., 1998). Therefore it is interesting to elucidate how orexinergic projections to the midbrain are involved in alteration of sleep-awake states. Then we examined effects of injections of orexin-A into the MLR, PPN and the SNr upon motor behaviors in decerebrate cats (Takakusaki et al., 2005). Microinjections of orexin into the MLR facilitated locomotion, while those into either the PPN or the SNr suppressed PPNinduced muscular atonia. The latter effects were reversed by subsequent injection of bicuculline into the PPN. Thus the excitability seems to be higher in the locomotor system than in the atonia system in the presence of orexin. On the contrary the excitability of the muscle tone inhibitory system may be higher than that of the locomotor system in the absence of orexin. Accordingly GABAergic projection from the SNr to the PPN/MLR area (BG-BS system) may underlie orexin-mediated vigilance state regulation and its dysfunction may be one of pathophysiological mechanisms of narcolepsy-like features of

is though indirect connections via the PPN (Takakusaki et al. 2004c, 2006).

stimuli applied to the lateral part of the SNr (blue circles in Fig.7C) completely abolished the PPN-induced REM with atonia (Fig.7Bb). On the other hand, stimulation of the SNr, denoted by black squares in Fig.7C, only inhibited REM (Fig.7Bd). It was also observed that stimuli applied to the sites which were indicated by red circles in Fig.7C did not block REM but attenuated the muscular atonia (Fig.7Bc), i.e., REM without atonia (Sanford, 1994), which is considered to be relevant to RBD in human (Culebras & Moore, 1989), was evoked. These findings indicate that neurons in the PPN that are responsible for generation of REM sleep are affected by GABAergic projections from the SNr.

Fig. 7. Possible mechanisms of basal ganglia modulation of REM sleep

A. Experimental design in decerebrate cats. B. (a) Stimulation of the PPN induced REM and atonia. (b) ~ (d) Effects of stimuli applied to various sites of the SNr on the PPN-induced REM and atonia. (b) Stimulation of the sites indicated by blue circles in C inhibited both REM and atonia. (c) Stimulation of the sites indicated by black squares in C. only inhibited REM. (d) Stimulation of the sites indicated by red circles in C inhibited atonia but not REM (REM without atonia). C. Effective sites for modulating the eye movements and muscular atonia were located in the lateral part of the SNr. D. Possible mechanisms of basal ganglia modulation of REM sleep. Abbreviations, ARAS; ascending reticular activating system, REM; rapid eye movement.

#### **4.2 Possible mechanisms of sleep disturbances in Parkinson's disease 4.2.1 Disturbance of REM sleep (REM sleep behavior disorder; RBD)**

Sleep disturbance is one of early signs of PD (Askenasy, 2001**;** Ferini-Strambi & Zucconi, 2000; Larsen & Tandberg, 2001). There are several reports suggesting that nearly half number of PD patients who were diagnosed as idiopathic RBD but free of neurodegenerative diseases had developed PD (Boeve, 2007; Iranzo, 2006; Schenck, 1996). Postuma et al. (2010) conclude that severity of REM atonia loss in idiopathic RBD predicts PD. Neuronal loss in the PPN was reported in PD (Hirsch et al., 1987; Jellinger, 1988; Zweig et al., 1989), and loss of cholinergic neurons in the PPN is possibly related to disability of PD patients (Rinne et al., 2008).

stimuli applied to the lateral part of the SNr (blue circles in Fig.7C) completely abolished the PPN-induced REM with atonia (Fig.7Bb). On the other hand, stimulation of the SNr, denoted by black squares in Fig.7C, only inhibited REM (Fig.7Bd). It was also observed that stimuli applied to the sites which were indicated by red circles in Fig.7C did not block REM but attenuated the muscular atonia (Fig.7Bc), i.e., REM without atonia (Sanford, 1994), which is considered to be relevant to RBD in human (Culebras & Moore, 1989), was evoked. These findings indicate that neurons in the PPN that are responsible for generation of REM

sleep are affected by GABAergic projections from the SNr.

Fig. 7. Possible mechanisms of basal ganglia modulation of REM sleep

**4.2 Possible mechanisms of sleep disturbances in Parkinson's disease 4.2.1 Disturbance of REM sleep (REM sleep behavior disorder; RBD)** 

REM; rapid eye movement.

patients (Rinne et al., 2008).

A. Experimental design in decerebrate cats. B. (a) Stimulation of the PPN induced REM and atonia. (b) ~ (d) Effects of stimuli applied to various sites of the SNr on the PPN-induced REM and atonia. (b) Stimulation of the sites indicated by blue circles in C inhibited both REM and atonia. (c) Stimulation of the sites indicated by black squares in C. only inhibited REM. (d) Stimulation of the sites indicated by red circles in C inhibited atonia but not REM (REM without atonia). C. Effective sites for modulating the eye movements and muscular atonia were located in the lateral part of the SNr. D. Possible mechanisms of basal ganglia modulation of REM sleep. Abbreviations, ARAS; ascending reticular activating system,

Sleep disturbance is one of early signs of PD (Askenasy, 2001**;** Ferini-Strambi & Zucconi, 2000; Larsen & Tandberg, 2001). There are several reports suggesting that nearly half number of PD patients who were diagnosed as idiopathic RBD but free of neurodegenerative diseases had developed PD (Boeve, 2007; Iranzo, 2006; Schenck, 1996). Postuma et al. (2010) conclude that severity of REM atonia loss in idiopathic RBD predicts PD. Neuronal loss in the PPN was reported in PD (Hirsch et al., 1987; Jellinger, 1988; Zweig et al., 1989), and loss of cholinergic neurons in the PPN is possibly related to disability of PD Several mechanisms are postulated in relation to the basal ganglia regulation of sleep. On one hand, recent brain imaging studies revealed that damage of brainstem, particularly the reticular formation, is critically involved in the pathogenesis of RBD (Unger et al., 2010). The brainstem damage could also explain some non-motor symptoms in this disease, which often precede diagnosis, such as autonomic dysfunction and sleep disorders. On the other hand, roles of dopaminergic influence on the basal ganglia in the control of sleep-wake behavior are suggested (Mena-Segovia J et al., 2008). It is also possible that basal ganglia efferents to the non-specific thalamic nuclei may affect awake-sleep states by modulating the activity of ascending reticular activating system (ARAS). Since the classical study of Moruzzi and Magoun (1949), the pontomesencephalic reticular formation has been known to comprise the ARAS. The PPN has been considered as a part of the ARAS (Garcia-Rill, 1991; Inglis & Winn, 1995; Jones, 1991).

The SNr has a direct projection to the thalamic nuclei (Parent et al., 1983; Pare´ et al., 1990) in addition to the PPN (Fig.7D). Because PPN has dense projections to the midbrain dopaminergic neurons, activity of the PPN neurons may affect awake-sleep states by modulating dopaminergic systems projecting to the basal ganglia and extra-basal ganglia areas. Consequently, our idea is that basal ganglia output from the SNr may affect awake– sleep cycles by modulating the activity of the ARAS through dual mechanisms (Fig.7D). One is through direct nigro-thalamic projection, and the other, which is considered in this study, is though indirect connections via the PPN (Takakusaki et al. 2004c, 2006).

### **4.2.2 Narcolepsy-like symptoms**

The presence in PD patients of narcolepsy-like features, such as daytime REM sleep intrusions associated with visual hallucinations, has led some authors to suggest that a mechanism similar to that of narcolepsy might underlie excessive daytime sleepiness (EDS) in PD (Arnulf et al., 2000). Thannickal et al. (2007) demonstrated that a massive loss of orexin neurons was found in PD patients and suggested that it was a cause of the narcolepsy-like symptoms. However, Compta et al. (2009) showed that orexin-A level was normal in the cerebrospinal fluid and it was unrelated to severity of sleepiness or the cognitive status of PD patients. Therefore alternative mechanisms other than dysfunction of orexin neurons might be responsible for EDS and the disturbance of sleep architecture in PD. In animal experiments, midbrain strucutes, including the SNr, the PPN and the MLR, receive orexinergic efferents from the perifornical lateral hypothalams (Nambu et al., 1999; Peyron et al., 1998). Therefore it is interesting to elucidate how orexinergic projections to the midbrain are involved in alteration of sleep-awake states. Then we examined effects of injections of orexin-A into the MLR, PPN and the SNr upon motor behaviors in decerebrate cats (Takakusaki et al., 2005). Microinjections of orexin into the MLR facilitated locomotion, while those into either the PPN or the SNr suppressed PPNinduced muscular atonia. The latter effects were reversed by subsequent injection of bicuculline into the PPN. Thus the excitability seems to be higher in the locomotor system than in the atonia system in the presence of orexin. On the contrary the excitability of the muscle tone inhibitory system may be higher than that of the locomotor system in the absence of orexin. Accordingly GABAergic projection from the SNr to the PPN/MLR area (BG-BS system) may underlie orexin-mediated vigilance state regulation and its dysfunction may be one of pathophysiological mechanisms of narcolepsy-like features of PD (Takakusaki 2008).

Possible Contribution of the Basal Ganglia

**5. Concluding thoughts** 

**6. Acknowledgment** 

**7. References** 

0022-510X.

both dopaminergic and cholinergic systems.

Brainstem System to the Pathogenesis of Parkinson's Disease 449

with information relating to reward and reinforcement by modulating the excitability of

The basal ganglia controls various function by acting on thalamocortical loop (C-BG loop) and the brainstem (BG-BS system). There are two key mechanisms for the operation by the basal ganglia circuit. One is sequential information processing, which would enhance the temporal contrast of the excitability of the target systems so that only the selected motor program could be appropriately executed, whereas other competing programs can be cancelled. The other is sustained output from the basal ganglia, which may control the degree of freedom of the excitability of the target systems during movements. We suggest that following roles can be played by the BG-BS system. First this system is involved in the automatic or unconscious control of movements that accompany voluntary movements. Second, the BG-BS systems may be involved in the maintenance of arousal and attentive states and in the regulation of REM sleep. Because output from the basal ganglia is thought to be overactive in PD, dysfunction of the BG-BS system in addition to that of C-BG loop can

This work is supported by Grants-in-Aid for Challenging Exploratory Research (Project # 23650202) and Priority Areas "Emergence of Adaptive Motor Function through Interaction between Body, Brain and Environment (Area #454)" from the Japanese Ministry of Education, Culture, Sports, Science and Technology to K.T. We express sincere appreciation

Aarsland, D., Bronnick, K., Williams-Gray, C., Weintraub, D., Marder, K., Kulisevsky, J.,

Arnulf, I., Bonnet, AM., Damier, P., Bejjani, BP., Seilhean, D., Derenne, JP. & Agid, Y. (2000)

Alessandro, S., Ceravolo, R., Brusa, L., Pierantozzi, M., Costa, A., Galati, S., Placidi, F.,

Andujar, J-E. & Drew, T. (2007) Organization of the projections from the posterior parietal

*Neurology*, Vol. 55, No.2, (July 25), pp. 281-288, ISSN 0028-3878 .

*Neurol.* Vol. 504, No.1, (Sep. 1), pp. 17–41, ISSN 0092-7317.

Burn, D., Barone, P., Pagonabarraga, J., Allcock, L., Santangelo, G., Foltynie, T., Janvin, C., Larsen, JP., Barker, RA. & Emre, M. (2010) Mild cognitive impairment in Parkinson disease: a multicenter pooled analysis. *Neurology*, Vol. 75,

Hallucinations, REM sleep, and Parkinson's disease: a medical hypothesis.

Romigi, A., Iani, C., Marzetti, F. & Peppe, A. (2010) Non-motor functions in parkinsonian patients implanted in the pedunculopontine nucleus: focus on sleep and cognitive domains. *J. Neurol. Sci.,* Vol. 289, No. 1-2, (Feb. 15) pp. 44-48, ISSN

cortex to the rostral and caudal regions of the motor cortex of the cat. *J. Comp.* 

be seriously involved in motor and non-motor functions in this disease.

No.12,(September 21), pp. 1062-1069, ISSN 0028-3878.

to Ms. Mihoko Ebisawa for preparation of this manuscript.

### **4.3 Disturbances of arousal state, attention and cognition**

Behavioral arousal requires an activation of dopaminergic projections arising from the SNc to the striatum and the ventral tegmental area (VTA) to the prefrontal cortex and the limbic system. The nigrostriatal projection is responsible for basal ganglia related motor functions. The mesocortical projection contributes to volitional expression and attention, and the mesolimbic projection is involved in emotional expression. On the other hand, ARAS plays a major role in the electroencephalographic arousal. An activation of the two arousal systems is required to maintain arousal state that enables alert, attention and cognition (Jones 1991). Because PPN has dense cholinergic and non-cholinergic excitatory connections with dopamine (DA) neurons in the SNc and other basal ganglia nuclei (Futami et al., 1994; Kitai, 1998; Takakusaki et al., 1996), these projections appear to play a role in more specific subcortical integration of motor and non-motor functions such as behavioral arousal, attention and reward (Kitai, 1998). For example, an injection of muscimol into the PPN reduced the speed and amount of arm movements and delayed the onset of movements but the accuracy was rather maintained (Matsumura and Kojima, 2001). Moreover, Kojima et al. (1997) demonstrated that kainic acid-induced lesion in the unilateral PPN induced hemiparkisonism which was observed in the contralateral side of the injection. From these findings they suggest that the PPN may thus facilitate the voluntary limb movements through its excitatory connections with the DA neurons.

Midbrain DA neurons are also involved in the predictive reward which is specifically linked with reinforcement behaviors. DA neurons are activated by rewarding events that are better than predicted, remain uninfluenced by events that are worse than predicted (Hikosaka et al., 2000; Schultz, 1998). Kobayashi et al. (2002) demonstrated that PPN neurons showed multi-modal activities during saccade tasks in alert monkey; their activities were related to the arousal levels, execution and preparation of movements, the level of task performance, and reward. Therefore the PPN may serve as an integrative interface between the various signals required for performing purposive behaviors (Kobayashi et al., 2004). We postulate that the PPN facilitates, possibly via dopaminergic systems, the central processes for motor command generation and extrinsic sensory processing by modulating arousal and attentive states.

In non-human primate, limited lesions of the striatum induce deficits in rule acquisition (Divac 1972), cognition (Taylor et al., 1990), working memory performance (Goldman-Rakic, 1987) and selected attention (Battig et al., 1962). Laplane et al. (1984) reported a patient with restricted bilateral pallidal lesions who was appeared apathetic and unconcerned or attention deficits, and his affect was flattened and emotional responses were blunted in the absence of any motor disorder or pure psychic akinesia. These symptoms were also described in progressive supranuclear palsy (PSP) in which major lesions were observed in the subcortical areas including the PPN. Because loss of cholinergic PPN neurons were observed not only in PSP (75-80%) but also PD (43-57%) (Hirsch et al., 1987; Jellinger, 1988; Zweig et al., 1987, 1989), the loss of cholinergic PPN neurons in both diseases could attribute to attentive and cognitive impairments and sleep deficiencies in these diseases (Scarnati & Florio, 1997).

Both neuroanatomical (von Krosigk et al., 1992; Smith & Bolam, 1990) and electrophysiological (Häusser & Yung, 1994; Saitoh et al., 2004; Paladini et al., 1999) studies demonstrated that dopaminergic neurons, as well as cholinergic neurons, receive GABAergic inhibitory effects from the basal ganglia, particularly from the SNr. Consequently a BG-BS system appears to involve the interdigitation of motor information with information relating to reward and reinforcement by modulating the excitability of both dopaminergic and cholinergic systems.
