The Basal Ganglia in Motor Control
Non-motor Aspects of Parkinson's Disease
Eye Movement Control Systems in Movement Disorders
The Basal Ganglia in Motor Control
The basal ganglia are a group of interconnected subcortical structures that play a key role in the selection of appropriate actions. In the movement domain, activity within the basal ganglia is thought to facilitate desired movements and inhibit competing (undesired) motor programs. This process is disrupted in many neurological disorders affecting the basal ganglia. For example, impaired voluntary movement (eg slowed movement) is seen in Parkinson’s disease. In other disorders, such as dyskinesias or dystonia, involuntary movements develop. We are interested in understanding the underlying neural mechanisms of these movement disorders, with an eye to developing new therapies, but to understand healthy motor control and action selection.
Parkinson’s Disease and Levodopa-Induced Dyskinesia
Parkinson’s Disease (PD) is a complex neurological disorder with both motor and non-motor manifestations. Many of the core motor features of the disease are closely linked to progressive loss of midbrain dopamine neurons. For more than 50 years, the gold-standard therapy has been the administration of the dopamine precursor levodopa, which is thought to restore dopamine release and thereby restore many aspects of movement. However, long-term use of levodopa often leads to the development of involuntary movements known as levodopa-induced dyskinesia (LID). While the movement abnormalities in both PD and LID are dopamine-dependent, how chronic changes in dopamine alter cellular and circuit function are still poorly understood.
To better understand the mechanisms underlying PD and LID, we have been primarily focused on the striatum, a structure critical for motor function that receives dense dopaminergic inputs and gives rise to the direct and indirect pathways of the basal ganglia. Prevailing theories have speculated that an imbalance in activity between the direct and indirect pathway medium spiny neurons (dMSNs and iMSNs, respectively), are thought to contribute to the loss of normal voluntary movement in these disorders. Using optogenetically-labeled recordings (“photo-tagging”) of dMSNs and iMSNs in parkinsonian mice, we have shown that dopamine loss and levodopa administration have differential effects on the firing of striatal neurons. To investigate how striatal activity changes with repeated, long-term use of levodopa, we are currently performing longitudinal recordings of calcium signals in dMSNs and iMSNs. During LID in particular, we have shown that levodopa produces heterogeneous effects on dMSN activity, with a distinct subpopulation of dMSNs preferentially activated during LID. We have also shown that this subpopulation is necessary and sufficient to mediate LID.
Using cell type-specific approaches, we have investigated the underlying synaptic adaptations responsible for the aberrant patterns of activity that occur during LID. Key excitatory and inhibitory inputs to dMSNs and iMSNs are reorganized in the mouse model of PD and LID, and contribute to the severity of the disease phenotype. This project utilizes a combination of in vivo and ex vivo electrophysiology, fiber photometry, optogenetics, circuit tracing, and behavior.
Coronal brain slice showing rabies tracing of TRAPed striatal neurons.
Paroxysomal nonkinesigenic dyskinesia
In some cases, involuntary movements develop in parallel with a neurodegenerative process, such as Parkinson’s or Huntington’s Disease. However, in dystonia, a disorder characterized by involuntary co-contraction of antagonist muscles, neuroimaging and postmortem pathology often does not show any overt “lesion” or cell loss. This observation implies that abnormal movements may be driven by abnormal patterns of brain activity or connectivity, rather than a frank loss of cells. Indeed, the fact that neuromodulatory treatments like Deep Brain Stimulation (DBS) can be effective in treatment of dystonia supports this idea. But what cells and what patterns of activity drive dyskinetic movements? Many theorize aberrant activity may arise in movement control circuitry like the basal ganglia and cerebellum. To investigate the connections and patterns of activity that generate involuntary movements, we use a variety of mouse models.
One major project in the lab focuses on the PNKD (paroxysmal nonkinesigenic dyskinesia) mouse model, developed by our UCSF colleague Louis Ptacek. This model is based on a human autosomal dominant disorder of the same name, in which patients experience attacks of involuntary movements lasting hours, after drinking alcohol or coffee, or experiencing extreme stress. Mice with the same gene mutations found in patients have normal-looking brains, but also have these attacks, triggered by administration of alcohol or caffeine. PNKD mice allow us to investigate the neural correlates of dyskinetic attacks, using in vivo single-unit electrophysiological recordings or fluorescent activity reporters in freely moving animals. We recently published work using optogenetically labeled single-unit recordings, as well as DREADDs (designer receptors exclusively activated by designer drugs) to test longstanding theories in the field that speculate that dyskinesia may be facilitated by an imbalance in direct and indirect pathway striatal activity. Currently, we are using this same model to explore physiological mechanisms underlying dystonia, which tends to occur when PNKD mice are treated with alcohol.
Freeze frame of PNKD mouse experiencing orofacial dyskinesia following caffeine injection.
Tardive Dyskinesia
Another project focuses on tardive dyskinesia (TD), a movement disorder that occurs in a subset of people who take a particular class of dopamine-blocking medications over many years. One such drug is the antipsychotic drug haloperidol. TD is characterized by involuntary movements of the jaw, lips and tongue. The fact that these movements develop gradually, after months or more often years of treatment, suggests that neuroplasticity may be a key mechanism. In addition, the condition often persists, even after stopping the offending drug, consistent with the idea of a long-lasting form of plasticity. We are interested in understanding what about these drugs leads to TD, and what forms of plasticity may be responsible.
In mice, chronic haloperidol treatment also induces spontaneous, involuntary abnormal chewing activity (called vacuous chewing movements or VCMs) which resemble the signs of TD in humans. We are currently using a combination of in vivo physiology, genetics, and pharmacology to investigate the neural correlates of orofacial movements, and how they change from healthy to pathological states.
Non-Motor Aspects of Parkinson's Disease
Though Parkinson’s Disease (PD) is classified as a movement disorder, it has both motor and non-motor manifestations. Motor phenotypes have been studied extensively in people with PD, as well as in animal models of PD. Much less is known about the underlying mechanisms of non-motor phenotypes, which include sleep disorders, cognitive impairment, autonomic dysfunction, and psychiatric symptoms like depression and anxiety. While many of the motor aspects of PD respond to dopamine replacement therapies, such as levodopa or dopamine agonist medications, these non-motor aspects typically do not improve, and may even worsen. Why? We are curious whether non-motor phenotypes are mediated by similar or distinct mechanisms, especially the role of neuromodulators and neural activity.
Sleep Disturbances in Parkinson's Disease
One new project in the lab is investigating the circuit and synaptic mechanisms of sleep disorders in mouse models of PD. For this study, we are measuring sleep in mouse models of PD with EEG and EMG. With EEG and EMG, the arousal state of the mouse can be characterized across time as awake, REM (rapid eye movement) or non-REM sleep. We are focusing on a key finding in people with PD, “sleep fragmentation”, in which the brain more frequently switches between arousal states. This phenotype can be reproduced in mouse models of PD. Using a combination of in vivo and ex vivo physiology methods, we are investigating how activity in the basal ganglia and traditional brainstem sleep nuclei may be altered to produce PD-associated sleep fragmentation.
Cognitive Impairment in Parkinson's Disease
Another project is focused on the development of cognitive impairment in PD. People with PD often note increased difficulty with multi-tasking, mental flexibility, organization and planning – all of which are “executive functions”, some of the most complex cognitive functions seen in humans. Executive function has been studied extensively in healthy animals and humans, and is linked to the function of the frontal cortex, the basal ganglia, and dopamine. However, the precise links between PD circuit pathology and executive dysfunction is unclear. We are investigating how pathology in neuromodulatory neurons and their targets may contribute to executive dysfunction in a mouse model of PD, using an operant decision-making task as a readout of behavioral flexibility. To get at the circuit mechanisms, we are using in vivo fiber photometry, electrophysiology, and optogenetics in conjunction with the decision-making task.
Impulse Control Disorder in Parkinson's Disease
Finally, we are interested in the cognitive-behavioral complications of dopamine replacement therapy in PD. In a subset of people with PD, therapy leads to Impulse Control Disorder (ICD). ICD consists of impulsive and/or compulsive behaviors like pathological gambling, compulsive porn use, binge eating or shopping. ICD can be an incredibly disruptive development in PD, and is usually managed by stopping the offending medication or markedly reducing the dose. To better understand why ICD develops in some people with PD, and how it might be prevented or treated, we recently developed a mouse model of PD/ICD. We combined a toxin-based mouse model of PD and chronic treatment with pramipexole, a dopamine agonist medication that is closely associated with ICD in people with PD. We measure impulsivity using a “delay discounting” task, in which a mouse chooses between an immediate, small reward or a delayed, large reward. Just like people, mice tend to prefer large rewards, but they discount the value of the reward according to how long they have to wait for it. As the wait gets longer, they are more likely to choose an immediate, small reward instead. This phenomenon is more pronounced in people with PD/ICD, and in our mouse model of PD/ICD. To date we have found that alterations in striatal activity appear to contribute to this phenomenon, and are investigating how chronic changes in corticostriatal synaptic connectivity may create a vulnerability to dopamine agonist-induced impulsivity in the mouse model.
A mouse conducting operant behavior training.
The Neural Control of Gait
Walking may seem effortless, but it relies on precise coordination across multiple brain regions, such as the basal ganglia and the cerebellum. In neurological disorders that affect these brain areas, such as Parkinson’s disease and cerebellar ataxia, this coordination breaks down, leading to debilitating gait disturbances. Both the basal ganglia and the cerebellum send strong inputs to the motor thalamus, a key hub that relays motor signals to the cortex. However, how these brain pathways work together to support normal walking, and how they fail in disease, remains largely unknown. We seek to understand the role of the motor thalamus in gait and how disruptions in these circuits contribute to parkinsonian walking deficits. To do this, we use high-speed video tracking and advanced machine learning-based gait analysis in freely moving mice, combined with in vivo neural recordings and targeted circuit manipulations. By identifying the neural circuits that control gait and how they are altered in disease, this research will help guide the development of more precise therapies for movement disorders.
Eye Movement Control Systems in Movement Disorders
Picture of head-fixed setup (top) and schematic of experimental set up (bottom).
Progressive supranuclear palsy (PSP) is a complex neurodegenerative disorder affecting at least 20,000 people in the U.S. Clinically, PSP presents with similar hypokinetic movements seen in patients with Parkinson’s disease (PD) and is typically misdiagnosed as PD. Unique to PSP, is the early presentation gaze palsy, i.e, slowed saccadic eye movements, and horizontal square wave jerks. Neuropathologically, postmortem analysis of PSP brain tissue revealed the presence of insoluble aggregates of the microtubule associated protein Tau in several motor control regions, with the earliest affected areas being basal ganglia nuclei such as the subthalamic nucleus (STN), Globus pallidus (GP), striatum, and substantia nigra pars reticulata (SNr). While PSP bear similar clinical presentation to PD, dopamine replacement therapies used to treat PD are largely ineffective in typical PSP. The circuit mechanisms driving the motoric symptoms of PSP are currently unknown, largely due to lack of animal models of disease. Using a Tau transgenic mouse to model PSP we assess:
1. The validity of Tau transgenic mouse to recapitulate the gross locomotor and neuropathological aspects of PSP
2. Whether Tau transgenic mice have impaired saccadic eye movements
3. The involvement of basal ganglia output on saccade parameters in healthy and Tau transgenic mice.