Dynamic Activity Model of Movement Disorders: The Fundamental Role of the Hyperdirect Pathway

Schematic illustration of cortically induced dynamic activity changes of the output nuclei of the basal ganglia (the internal segment of the globus pallidus, GPi and the substantia nigra pars reticulata, SNr) in the healthy and diseased states. The height of the dam along the time course controls the expression of voluntary movements. Its alterations could cause a variety of movement disorders, such as Parkinson's disease and hyperkinetic disorders. © 2023 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society. Malfunction of the basal ganglia (BG) results in the origin of movement disorders such as Parkinson's disease (PD), dystonia, Huntington's disease (HD), levodopa-induced dyskinesia (LID), and hemiballism. So far, two models have been proposed to explain their pathophysiology, especially for PD, based on recordings of neuronal activity in animal models and patients.1 The "firing rate model" proposes that the mean firing rates in the output nuclei of the BG, the internal segment of the globus pallidus (GPi) and substantia nigra pars reticulata (SNr), are increased, and suppress thalamocortical activity, resulting in bradykinesia.2 In the "firing pattern model," abnormal firing patterns in the BG, such as oscillatory and synchronized activity, disturb movement-related neuronal signals.3 Alternatively, we propose the "dynamic activity model," based on alterations of cortically induced dynamic activity in the BG, aiming to explain the pathophysiology of movement disorders in a unified manner. To explore how the BG control voluntary movements, we have long been studying the response in the GPi/SNr induced by cortical stimulation, which is presumed to mimic cortical excitation to initiate voluntary movements (Fig. 1). In the normal state, electrical stimulation in the motor and prefrontal cortices induces a triphasic response consisting of early excitation (Fig. 1B, magenta), inhibition (blue), and late excitation (green) in the GPi/SNr of monkeys, rodents, and probably humans. Each component is mediated by the cortico-subthalamo (STN)-GPi/SNr hyperdirect, cortico-striato (Str)-GPi/SNr direct, and cortico-Str-external pallido (GPe)-STN-GPi/SNr indirect pathways, respectively (Fig. 1A).4-8 These three pathways would work during the execution of voluntary movements in a similar manner (Fig. 1B,C).5 When voluntary movements are about to be initiated, the signals through the hyperdirect pathway first reach the GPi/SNr, inhibit thalamocortical activity, reset cortical activity related to ongoing movements, and prepare for the next action. Then, the signals through the direct pathway reach the GPi/SNr, disinhibit thalamocortical activity, and release an appropriate movement at an appropriate time. Finally, the signals through the indirect pathway reach the GPi/SNr, inhibit thalamocortical activity, and stop the movement released by the direct pathway. The conduction time through these pathways is short enough to affect voluntary movements (<100 ms). In the spatial domain, inhibitory inputs through the direct pathway terminate in a relatively small, limited area in the GPi/SNr (Fig. 1C, blue), whereas excitatory inputs through the hyperdirect (magenta) and indirect pathways (green) terminate over a wide area,9, 10 thereby forming inhibitory-center and excitatory-surround spatial organization in the GPi/SNr. Inhibition in the center area would release a selected movement, whereas excitation in the surrounding area would continuously inhibit other unintended movements. Indeed, activation of the hyperdirect and indirect pathways suppressed movements, whereas activation of the direct pathway facilitated movements.4, 7, 8, 11, 12 We have also been examining how cortically induced response patterns in the GPi/SNr are altered in various movement disorder models (Fig. 2, plotted in the hyperkinetic-hypokinetic and hypertonic-hypotonic plane) and would like to discuss their pathophysiology based on the dynamic activity model. Monkeys treated with a dopaminergic neurotoxin, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP),13 and mice treated with another dopaminergic neurotoxin, 6-hydroxydopamine (6-OHDA),14 exhibit features of bradykinesia that can be equated with those observed in PD patients. In these models, cortically induced inhibition in the GPi/SNr mediated by the direct pathway was greatly diminished (Fig. 2, bradykinesia in PD), whereas early and late excitation, mediated by the hyperdirect and indirect pathways, respectively, was slightly enhanced or little changed. The extent of the decrease in inhibition paralleled the severity of bradykinesia.13 Systemic administration of l-DOPA ameliorated bradykinesia and restored cortically induced inhibition, manifesting a normal triphasic response in the GPi.13 Thus, in PD, significant reduction of cortically induced inhibition in the GPi/SNr leads to diminished capacity of releasing the intended movements, resulting in bradykinesia.13, 14 The reduced inhibition in the GPi/SNr stems from hypoactivity of dopamine D1 receptor-expressing Str direct pathway neurons.12, 17 Enhanced STN activity could additionally make the inhibition less probable.18 The DYT1 (DYT-TOR1A) dystonia mouse model overexpressing human abnormal torsinA exhibits hyperlocomotion, dystonic-like abnormal movements, and sustained co-contractions of agonist and antagonist muscles.15 In this model, cortical stimulation induced early excitation, followed by long-lasting inhibition and reduced late excitation in the entopeduncular nucleus (EPN, the rodent equivalent of the GPi) and GPe (Fig. 2, dystonia).15 This suggests that neurotransmission through the direct and indirect pathways is enhanced. In addition, their spontaneous firing rates were reduced with bursts and pauses. Somatotopy was overlapped, and a single EPN/GPe neuron tended to receive multiple cortical inputs.15 One patient with cervical dystonia exhibited similar cortically induced long-lasting inhibition in the GPi/GPe when examined during stereotactic surgery.16 Thus, in dystonia, enhancement of cortically induced inhibition in the temporal and spatial domains in the GPi strongly disinhibits thalamic and cortical activity and causes sustained muscle contractions even if the original cortical excitation is weak.9, 15 Overlapped somatotopy causes coactivation of agonist and antagonist muscles and motor overflow, that is, unintentional muscle contractions at distant sites. Furthermore, "sensory trick," a reduction in muscle contractions by touching the involved/adjacent body parts, could be explained by distinct cortically induced early excitation in the GPi. Motor cortical excitation evoked by sensory inputs from these body parts would reset ongoing cortical activity and muscle contractions through the hyperdirect pathway. The LID mouse model was developed by daily administration of l-DOPA to 6-OHDA-treated mice. In the parkinsonian state, cortically induced inhibition was largely suppressed in the SNr (Fig. 2, bradykinesia in PD),14 whereas after l-DOPA injection, when the mice showed abnormal involuntary movements, cortically induced inhibition was recovered and enhanced, and late excitation was largely suppressed (Fig. 2, LID in PD).14 Thus, marked enhancement of cortically induced inhibition in the GPi/SNr releases unintended movements at random times, and the suppression of late excitation cannot stop once-released movements, manifesting LID.14 The most significant pathology of HD is observed in the striatum, where Str-GPe indirect pathway neurons degenerate in the early stage, followed by the degeneration of Str-GPi/SNr direct pathway neurons.19 Selective elimination of Str-GPe indirect pathway neurons by an immunotoxin-mediated cell-targeting method in transgenic mice caused motor hyperactivity and could be a model of the early stage of HD.20 In these mice, motor cortical stimulation induced early excitation and inhibition without late excitation in the SNr (Fig. 2, chorea in HD).7 Similarly, bicuculline or gabazine (GABAA receptor antagonist) injection into the GPe suppressed Str-GPe indirect pathway neurotransmission and induced chorea in monkeys.21, 22 Thus, the loss of cortically induced late excitation in the GPi/SNr cannot stop once-released movements, manifesting involuntary movements.7 Blocking STN activity by lesioning or muscimol (GABAA receptor agonist) injection induces hemiballism in monkeys.4, 23 In this state, cortical stimulation induced long-lasting inhibition without early and late excitation in the GPi/GPe (Fig. 2, hemiballism).4 Besides, spontaneous firing rates were decreased, and bursts and pauses were observed in the GPi/GPe, as the STN is a main excitatory source.4 Thus, enhancement of cortically induced inhibition in the GPi easily releases movements, and suppression of early and late excitation cannot reset nor stop movements, manifesting involuntary movements.4 The "dynamic activity model" can explain some aspects of firing rate/pattern changes associated with movement disorders.1 Cortically induced early excitation is enhanced, and inhibition is greatly diminished in the GPi in PD (Fig. 2, bradykinesia in PD). If the excitation is repetitively induced, spontaneous firing rates in the GPi could be increased. Likewise, in dystonia, cortically induced inhibition in the GPi could be strong enough to decrease spontaneous firing rates. On the other hand, various degrees of oscillatory activity, such as low-beta oscillation in PD and theta oscillation in dystonia, have been observed in the GPi/GPe and STN.3, 16, 24 Alterations of the interaction in the cortico-GPi/GPe/STN and STN-GPe derived from altered dynamic activity could be the origin of such oscillatory activity,24, 25 determining their oscillation frequency. So far, several motor manifestations are not simply explained by the available models. Here, we discuss two principal but unexplained motor signs of PD, that is, tremor and rigidity. Tremor at rest predominantly involves distal limb muscles and is characterized by a 4–6 Hz alternating activity in agonist/antagonist pairs. In MPTP-treated monkeys, oscillatory activity in the BG did not usually develop low-frequency tremors.24 In PD patients, oscillatory activity is typically synchronized with tremors in the STN and thalamus,3, 26-28 the latter of which receives direct inputs from the cerebellum, but not from the BG. A lesion in the STN or thalamus abolished tremor, suggesting their causal role.27, 28 Strong, synchronized oscillatory activity in the BG could spread to the cerebello-thalamic systems through the disynaptic projection from the STN to the cerebellar cortex via the pontine nuclei29 and eventually induce PD tremors. Alternatively, tremor could be initiated by oscillatory activity in the cortico-BG loop and amplified by the cortico-cerebellar interactions.30 Finally, tremor-related activity is transferred to the spinal cord through the cortex. Velocity-dependent increased resistance to passive stretching associated with the cogwheeling phenomenon is typical of rigidity in PD. The enhanced basal neuronal activity in the STN and its response via the hyperdirect pathway may reset cortical activity, enhance long-latency stretch reflex, and provide one relevant mechanism for rigidity.31 In addition, descending pathways from the BG to the spinal cord, that is, from the GPi/SNr and STN to the pedunculopontine tegmental nucleus (PPN) and midbrain locomotor region, also control movements, such as locomotion, gait, and posture.32 The GPi/SNr send inhibitory projections to the PPN, and PPN activity decreases muscle tone.32 In the PD state, cortically induced excitation is dominant in the GPi/SNr (Fig. 2, bradykinesia in PD), which could suppress PPN activity, leading to increased muscle tone, rigidity. On the other hand, in hemiballism, enhancement of cortically induced inhibition without excitation in the GPi (Fig. 2, hemiballism) could induce opposite effects in the cortex and PPN to those in PD, leading to decreased muscle tone. The motor symptoms are quite different between dystonia, LID, and hemiballism, which could be explained by different muscle tone, different nature of imbalance between the three pathways, different spontaneous firing rates, and presence/absence of overlapped somatotopy in the GPi/SNr. Stereotactic surgery targeting the STN and GPi, such as deep brain stimulation (DBS) and thermoablation, improves cardinal motor features of PD. In MPTP-treated monkeys, STN blockade by muscimol injection ameliorated bradykinesia, and at the same time, suppressed cortically induced early and late excitation, and restored cortically induced inhibition in the GPi (Fig. 2, PD + STN block).13 Restored cortically induced inhibition in the GPi through the direct pathway enables the release of an intended movement by disinhibiting thalamocortical activity.13 The beneficial effects of STN-DBS are not meant to be so simple and may include excitation or inhibition of various neuronal elements, such as afferent and efferent axons and soma in the STN, but at least one effect is likely to disrupt signal transmission through the STN and suppress cortically induced early and late excitation in the GPi.33-35 The "dynamic activity model" gives us better prospects of movement disorders and a more comprehensive view of their underlying pathophysiology. However, it may oversimplify complex disease conditions as it has been characterized only in animal models. Other neurotransmitters, including serotonin and noradrenaline, as well as dopamine, are disturbed in MPTP-treated monkeys and PD patients,36, 37 and neurodegeneration is observed not only in the BG but also in the brainstem and cerebral cortex in PD patients,38 contributing to a plethora of symptoms. Besides, the "dynamic activity model" is based on responses induced by artificial electrical stimulation in the cerebral cortex, and we need to examine whether cortical excitation to initiate voluntary movements could cause similar movement-related activity in the GPi/SNr, and whether its alterations could disturb actual movements. In sum, the "dynamic activity model" implies that the balance between the hyperdirect, direct, and indirect pathways plays an essential role in motor dysfunction in movement disorders. Possible future therapeutic approaches for PD will include adaptive DBS that delivers stimulation to the STN based on movement-related activity in the motor cortex,39 chemogenetic40/optogenetic suppression of STN activity, and optogenetic activation of the direct pathway.11 (1) Research project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript: A. Writing of the first draft, B. Review and Critique. A.N.: 1A, 3A S.C.: 1A, 3B H.S.: 1A, 3B N.H.: 1A, 3B J.A.O.: 1A, 3B We thank S. Sato, H. Isogai, N. Suzuki, K. Awamura, K. Miyamoto, T. Sugiyama, and M. Hayashi, for technical support; and Y. Yamagata for critical reading of the manuscript. Financial disclosure related to research covered in this article. Atushi Nambu—funding sources: Japan Agency for Medical Research and Development (AMED); Japan Society for the Promotion of Science (JSPS); Ministry of Education, Culture, Sports, Science and Technology (MEXT). Satomi Chiken—funding sources: Japan Society for the Promotion of Science (JSPS), Core Research for Evolutional Science and Technology (CREST). Hiromi Sano—funding sources: Japan Society for the Promotion of Science (JSPS); Ministry of Education, Culture, Sports, Science and Technology (MEXT). Nobuhiko Hatanaka—funding source: Japan Society for the Promotion of Science (JSPS). José A. Obeso—funding sources: Fundacíón HM Hospitales; CIBERNED, Instituto Carlos III, Ministerio de Educación y Ciencias. Full financial disclosure for the previous 12 months. Atushi Nambu—funding sources: Japan Agency for Medical Research and Development (AMED), Japan Society for the Promotion of Science (JSPS), Ministry of Education, Culture, Sports, Science and Technology (MEXT), Eisai, Kyocera; publishing royalties: Igakushoin, MEDSi; honoraria: Otsuka Pharmaceutical, Kyowa Kirin, AMED, JSPS, Japan Science and Technology Agency (JST), FP, Tokyo Metropolitan Institute for Medical Sciences, International Parkinson and Movement Disorder Society, Japanese Basal Ganglia Society, Narishige, Japan Society for Stereotactic and Functional Neurosurgery, Japanese Society of Clinical Neurophysiology; employment: National Institute for Physiological Sciences, The University of Tokyo, Tamagawa University, Nagoya City University, Mie University, Kanazawa Medical University, Kyoto University, Kindai University. Satomi Chiken—funding sources: Japan Society for the Promotion of Science (JSPS), Core Research for Evolution Science and Technology (CREST); employment: National Institute for Physiological Sciences, Tokyo Metropolitan University. Hiromi Sano—funding sources: Japan Society for the Promotion of Science (JSPS), Ministry of Education, Culture, Sports, Science and Technology (MEXT); employment: Fujita Health University. Nobuhiko Hatanaka—funding sources: Japan Society for the Promotion of Science (JSPS); employment: National Institute for Physiological Sciences, Aichi Gakuin University. José A Obeso—funding sources: Fundacíon HM Hospitales; CIBERNED, Instituto Carlos III, Ministerio de Educación y Ciencia; honoraria: Insightec, Biogen, Laboratorios Estebe, Editorial Viguera; employment: HM Hospitales. No new data were generated or analyzed in this study.