The underlying pathophysiology of tics in Tourette syndrome is a topic of major scientific interest. To date, there is an absence of consensus among researchers regarding the precise anatomic location responsible for tics. The goal of this article is to review the current understanding of these brain circuits and data supporting specific anatomic regions. In summary, current scientific evidence supports the likelihood of multiple areas of abnormality within cortico-basal ganglia-thalamocortical (CBGTC) circuitry or their connected brain regions. A reasonable anatomic hypothesis is that a disruption anywhere within specific circuitry can ultimately lead to the development of a tic disorder.
Key points
- •
Major areas of the movement-oriented cortico-basal ganglia-thalamocortical (CBGTC) circuit include the cortex, basal ganglia (dorsal and ventral striatum, globus pallidus interna and externa, amygdala, substantia nigra pars reticulate, and subthalamic nucleus), and thalamus. Other brain regions (eg, cerebellum, hippocampus, and hypothalamus) that interconnect with the CBGTC circuit can also impact messages passing through this circuit.
- •
Numerous investigative approaches (clinical, therapeutic trials, imaging, genetic, post-mortem, etc., and animal models) have been utilized to identify the underlying anatomic abnormality in tic disorders.
- •
Evidence exists to support a role for multiple underlying pathophysiological sites in tic disorders.
- •
Investigators have attempted to simplify the conceptual involvement of CBGTC circuitry in movement disorders through the discussion of goal-directed versus. habitual circuits, direct versus. indirect circuits, and inhibitory hyperdirect pathways. All, however, tend to oversimplify the complexity of the CBGTC pathway.
- •
Current scientific evidence supports multiple areas of abnormality within CBGTC circuitry or their connected brain regions. Hence, until proven otherwise, a reasonable anatomic hypothesis for tics is that a disruption anywhere within the CBGTC circuitry can lead to an aberrant message arriving at the primary motor cortex and the enabling of a tic.
Introduction
Tourette syndrome (TS) is a common neuropsychiatric disorder characterized by the presence of chronic motor and phonic tics. The goal of this article is to review our current understanding of brain circuits and anatomic regions that are likely contributors to tic disorders. Our basic approach assumes that tics are a movement disorder and therefore associated with an abnormality within the cortico-basal ganglia-thalamocortical (CBGTC) circuitry. Evidence supporting the possibility that tics could be associated with a sensory abnormality is also reviewed. As most are aware, the precise underlying neuroanatomic location for tic symptoms remains unknown. This article contains 4 major sections including (1) a simplified, useful response to a patient’s or parent’s question pertaining to, “Where are tics coming from?;” (2) an overview of CBGTC circuitry, that is, circuits associated with movement disorders; (3) a discussion of specific brain regions participating in CBGTC circuits, their movement-related functions, and evidence supporting their involvement in tic disorders; and (4) a discussion of tics via the perspective of three often cited cortical-basal ganglia-circuits including Goal Directed and Habitual, Direct and Indirect, and Inhibitory.
Simplified approach to explaining the pathophysiology of tics
Many patients and their families inquire about the underlying etiology for tics (see Dr Fernandez’s article, “ Genetics of Tourette Syndrome ,” in this issue) and ask, “where are tics coming from?” Recognizing that most questions eminate from individuals with limited scientific backgrounds, how should one respond? One beneficial approach is to draw a circuit (call it a racetrack), containing inserted boxes, representing different brain regions, around the track ( Fig. 1 ). The discussion that follows should then explain that the message for a movement starts at the cortex (“starting gate”), proceeds to another part of the brain (basal ganglia), then to the thalamus, and finally terminates back in the cortex (finish line). At each of the new brain locations along the circuit, it passes its message on to a new brain cell (new “race car’) by releasing a chemical substance called a neurotransmitter. One should also note that other parts of the brain (eg, cerebellum) provide input to the diagrammed brain regions and can influence the message. Thus, accepting that an intact message is required for a normal movement to occur, disrupting the message anywhere around the circuit, or disrupting connecting brain regions, could produce an abnormal message and movement.
Although the “anywhere in the racetrack” (CBGTC circuit) concept is presented as a simplified explanation for tics, there is expanding scientific evidence to support this concept. For example,: (1) this article describes investigations supporting the involvement of multiple brain regions; (2) a study analyzing cases of tics causally attributed to preceding brain lesions has suggested that a common network map, composed of the insular cortices, cingulate gyrus, striatum, globus pallidus interna (GPi), thalamus, and cerebellum, all provide insight into tic-inducing lesions ; (3) MRI studies have reported nodal changes in different parts of the CBGTC circuit ; and (4) in a study investigating the network dynamics leading to a tic, relevant predictors included the primary motor cortex, the prefrontal basal ganglia loop, and the amygdala-mediated social processing network. In summary, the described simplistic hypothesis and illustration (see Fig. 1 ) fairly accurately summarize the circuits and their neurotransmitter systems (see Harvey S. Singer and Justin Pellicciotti’s article, “ The Role of CBGTC Synaptic Neurotransmission in the Pathophysiology of Tics ,” in this issue) that are hypothesized to be pathophysiological factors underlying tic disorders.
Overview of cortical-basal ganglia-thalamo-cortical circuits
Motor control is obtained via a complex network of pathways involving a series of parallel and interconnected networks of neurons in the frontal cortex, basal ganglia, and thalamus identified as CBGTC circuitry ( Fig. 2 ). These complex circuits link various functional areas of the cortex (association, motor, limbic, sensory) with the basal ganglia and the thalamus. In addition, these areas receive information from, and send projections to, other regions in the brain, such as the motivational and reward regions of the ventral striatum; the cerebellum, which fine tunes motor coordination; the hippocampus, which stores motor memories; the hypothalamus, which regulates hormonal and emotional responses; the amygdala, which processes emotional and motivational aspects of motor behaviors; and the substantia nigra and the ventral tegmental area, which produce dopamine, a neurotransmitter that modulates reward and learning systems. Thus, the goal of this article is to provide both a basic anatomic understanding of the individual CBGTC components as well as to present scientific evidence supporting their role in the pathophysiology of TS.
One potential limitation to a motor-oriented CBGTC circuit approach when discussing the pathophysiology of tics is the recognition that individuals with tic disorders also have a variety of sensory issues. For example, there is significant support for a sensory cortical involvement in tic syndrome based on the presence of premonitory urges preceding tics, hypersensitivity to external stimuli, and abnormalities in the sensorimotor integration perceptual process. Nevertheless, despite the initial appearance that this weakens our motor-focused approach, there are data in TS patients documenting the finding that the motor system has a suppressive effect on the sensory system. More specifically, the excitability of the sensory cortex is suppressed by excessive stimulation (post repetitive transcranial magnetic stimulation, of the motor cortex).
Major Components of the Cortical-Basal Ganglia-Thalamo-Cortical Circuit and Evidence Supporting Their Involvement in Tics
Cortex
Anatomy
Inputs
Cortical pyramidal neurons, especially those located within the frontal cortex, receive inputs from gamma-aminobutyric acid (GABA)ergic interneurons, glutamatergic projections from the thalamus, dopaminergic (DA) projections from the ventral tegmental area (VTA), serotoninergic inputs from the median raphe, and noradrenergic inputs from the locus coeruleus. Mesocortical dopamine inputs from the VTA influence prefrontal cortical pyramidal neurons directly and indirectly via modulation of GABAergic interneurons.
Outputs
Functionally, investigators have proposed 3 primary cortical-striatal projections including the associative/cognitive, motor, and limbic pathways. (1) The associative/cognitive pathway arises in dorsolateral prefrontal and parietal regions and projects to the caudate; (2) the motor pathway projects from motor/supplementary motor (SMA) cortical areas to the putamen; and (3) the limbic (emotional) pathway projects from orbital and medial prefrontal regions to the ventral striatum (see Fig. 2 ). These pathways provide the striatum with excitatory glutamatergic projections that make synaptic contact with the dendritic spines on GABAergic striatal medium-sized spiny neurons (MSNs) as well as on striatal interneurons. In addition, cortical regions send direct projections to the subthalamic nucleus (STN), substantia nigra (SN), ventral tegmental area (VTA), globus pallidus externa (GPe), cerebellum (CB), and thalamus. Cortical projections to the thalamus include associative cortical inputs to the ventral median nucleus (VMN), premotor inputs to the ventral anterior (VA) and ventral lateral anterior (VLa) nuclei, and motor inputs to ventral lateral posterior (VLp) nuclei (see section on cortical-basal ganglia-circuits).
Evidence supporting cortical involvement in tics
Clinical data supporting cortical involvement in the pathophysiology of TS is based in part on its involvement in sensorimotor disorders, as well its role in coexisting neuropsychological problems such as executive dysfunction, cognitive inhibitory deficits, attention-deficit hyperactivity disorder, obsessive-compulsive disorder (OCD), anxiety, suicidality, mood issues, and disruptive behaviors. , As previously noted, support for a sensory cortical involvement in TS is based on the presence of premonitory urges preceding the tics. This premonitory urge, thought by some to be the driving force for tics, arises in mesial and lateral premotor areas, the insula, SMAs, anterior cingulate cortex, and primary sensorimotor cortex. Post-mortem studies show a greater number of changes in prefrontal regions (Brodmann area 9) rather than in the caudate, putamen, or ventral striatum. , Imaging studies report larger dorsal prefrontal and parietal-occipital areas and cortical thinning in frontal, parietal, and sensorimotor areas. A voxel-based morphology and structural covariance network mapping study showed decreased gray matter volume in the anterior cingulate cortex (ACC) and increased structural covariance between the ACC and inferior frontal cortex, posterior cingulate cortex, and the motor cerebellum. Combined diffusion tensor imaging and paired-pulse TMS studies identified impaired frontal-mediated motor cortex inhibition and abnormalities in motor pathways. White matter changes have included increases in the right frontal cortex and decreases in the deep left frontal, orbital, and medial frontal cortex, cingulate, and temporal areas. Alterations of the corpus callosum include changes in size and reduced white matter connectivity. Diffusion tensor imaging and functional connectivity studies identified increased connections between left postcentral to precuneus connectivity and widespread immature functional connectivity. , Additional studies have suggested that a fronto-parietal network disconnection may contribute to tic severity. , In a study using functional connectivity (fcMRI) and a resting state electroencephalogram determination of topographic organization, TS patients were found to have disruptions in the left temporal lobe. In a study evaluating functional connectivity in patients receiving thalamic deep brain stimulation, tic reduction was associated with positive changes in the sensorimotor cortex, bilateral insula, and inferior frontal cortex. Children with TS were reported to suppress tics via a distributed brain circuit involving the right superior frontal gyrus and the left precuneus. Another study using resting state functional MRI (fMRI) to evaluate dynamic functional characteristics identified 2 recurrent functional brain styles, with TS patients spending more time in state 2. PET studies demonstrated areas of cortical hypermetabolism in premotor cortex and primary motor cortex, anterior cingulate, and Broca’s area. Transcranial magnetic stimulation (TMS) with single and paired pulse stimulation in motor cortex reduced inhibition. , Repetitive TMS of the SMA at a frequency of 1 Hz reduced tic severity in some, but not all individuals.
Animal models
Animal models further support a role for cortical alterations in the production/control of tics. In the Hdc knockout mouse tic model, white matter abnormalities have been identified. In mice, tic-like movements along with increased exploration, sniffing, and paw-licking behaviors were elicited when picrotoxin (a GABA-A antagonist) was infused into the sensorimotor cortex. Optogenetic techniques, methodologies that use light to modulate genetically encoded proteins in living organisms, have also been used to investigate tic-like movements. , Using this approach in mice, repeated hyperactivation of connectivity between the orbitofrontal cortex and ventral medial striatum produced an increase in grooming activity. In a transgenic mouse (D1CT), following activation of dopamine D1 receptor-expressing neurons (located in layers II and III of sensorimotor cortex, layer II of the piriform cortex, and the intercalated nucleus of the amygdala), animals demonstrated compulsive behaviors (repetitive climbing, digging, and leaping), biting behaviors, and jerking movements (flurries of twitches of the head, limbs, and trunk). ,
Basal ganglia
Anatomy
The basal ganglia are composed of the following gray nuclei: caudate, putamen, globus pallidus interna (GPi) and externa (GPe), amygdala, substantia nigra pars reticulate (SNpr), and subthalamic nucleus (STN).
Striatum
The dorsal (caudate, putamen) and ventral striatum primarily contain MSNs that virtually all use GABA as their principal neurotransmitter. In addition to cortical glutamatergic inputs, MSNs also receive glutamatergic input from the intralaminar nuclei of the thalamus and basolateral amygdala; dopaminergic input from the substantia nigra pars compacta (SNpc) and ventral tegmental area (VTA); GABA, substance P, dynorphin, and enkephalin inputs from neighboring MSNs; and acetylcholine, GABA, or peptidergic input from striatal interneurons (giant aspiny cholinergic interneurons and several subtypes of medium-sized GABAergic interneurons). The dorsal striatum integrates information from the cortex, ventral striatum, amygdala, and thalamus with the goal of facilitating voluntary and involuntary action selection and motor outcome. , The ventral striatum (nucleus accumbens (NAc) and olfactory tubercle) serves as the integrating site for incoming information pertaining to emotions, motivation, vigor, reward, attention, and autonomic factors. The NAc receives inputs from the prefrontal cortex, basolateral amygdala, hippocampus, paraventricular and interior thalamus, dorsomedial hypothalamus, SNpc, VTA, and dorsal and median raphe nuclei.
Subthalamic nucleus
The STN is populated mainly by glutamatergic projection neurons, which receive their major afferents from the cerebral cortex, thalamus, globus pallidus externa (GPe), and the brainstem. Projections from STN mainly innervate the globus pallidus (GPi and GPe/SNpr), the substantia nigra, striatum, and the pedunculopontine tegmental nucleus (PPTg).
Amygdala
The amygdala regulates various stress and reward-related behaviors. Retrograde labeling studies have demonstrated connections with the striatum, thalamus, and sensorimotor processing regions of the brain. The amygdala has dense projections to the NAc that taper off as they enter the dorsolateral striatum (DLS). It is believed that glutamatergic projections from the amygdala to the VTA have a critical role in reinforcement of reward-related tasks, and habit formation, whereas the thalamo-amygdala pathway has been demonstrated to regulate reward-based learning.
Supporting evidence for basal ganglia involvement in tics syndrome
Evidence supporting basal ganglia involvement includes its association with other movement disorders and major anatomic roles in several proposed CBGTC circuits (eg, Direct and Indirect, Goal and Habitual, and Hyperdirect–to be discussed). Clinical studies: Both bilateral GPe and GPi deep brain stimulation have been effective in reducing tics. Postmortem studies have shown a 50% to 60% decrease of parvalbumin-positive and choline acetyltransferase-positive cholinergic interneurons. It is hypothesized that the deficit of these striatal interneurons causes impaired cortical and thalamocortical neuron firing. Imaging studies of the caudate and putamen have shown inconsistent results. Volumetric studies have also shown abnormalities in the amygdala. , Electrophysiology studies: Measurement of individual neuronal firing rates within both the anterior GPe and GPi were shown to be tic related.
Animal studies
In human primates, rats, and mice, investigations have shown that disruption of the glutamate/GABA balance within the striatum causes tic-like behaviors. , , In a monkey model, the unilateral injection of a GABA antagonist into the nucleus accumbens generated vocal tics.
Thalamus
The motor thalami (Mthal), primarily the ventral anterior (VA), and ventrolateral (VL) nuclei have a significant role in integrating the control of movements. The Mthal receives input from the cortex (pyramidal neurons in layers V and VI within associative, premotor, and motor cortical areas), basal ganglia (SNpr and GPi), cerebellum (dentate nucleus), and basolateral amygdala. It provides excitatory innervation to dendrites of cortical pyramidal neurons located in layers I and II and to a lesser extent in layer V. In addition, there are short subcortical loops between the thalamus and striatum: motor information passing from the Mthal to the striatum and sensory information from the intralaminar nuclei (centromedian parafasicular complex [CM-Pf]), medial posterior nucleus (POm), lateral posterior (LP), and lateral dorsal (LD) nuclei to the striatum. The CM-Pf complex integrates the motor, associative, and limbic input/output into the CBGTC circuitry. The centromedian (CM) thalamus sends and receives projections from the subthalamic nucleus, striatum, and cortex, while the parafascicular (Pf) thalamus is linked with associative and limbic pathways. Together, the CM-Pf is thought to modulate striatal activity by targeting MSNs and cholinergic interneurons. In three subjects receiving deep brain stimulation (DBS) therapy, electrophysiological recordings showed that spontaneous motor tics are preceded by repetitive coherent thalamo-cortical discharges. DBS therapy in TS patients effectively reduced tics and modulated beta frequency oscillations. Thalamic local field potentials (5–15 Hz) in DBS TS patients were felt to correlate with long-term improvement.
Cerebellum
Although the cerebellum is not considered part of the CBGTC circuit, recent studies support a significant influence of the cerebellum on motor behaviors and decision-making processes. , , The cerebellum receives inputs from the cortex (via the pontine nucleus [PN] and olivocerebellar [OC] pathways), basal ganglia (originates from STN and is transmitted via a disynaptic connection through the peduculopontine tegmental nucleus [PPTg]). It sends excitatory projections directly to the thalamus, amygdala, dorsal raphe nucleus (DRN), and VTA, and indirectly to the basal ganglia (via disynaptic connections from the dentate nucleus to thalamic motor and intralaminar nuclei) and amygdala. The cerebellum has been implicated as a site of abnormality in TS based on studies in both humans and animal models. In a study of drug naïve children with both TS and OCD, higher functional anisotropy was found in the cerebellar peduncles. In animal models, studies have shown that both neurons in the cerebellar cortex and dentate nucleus have increased abnormal discharges and blood flow immediately preceding tics. A cerebellar role is also supported by computational model analyses that reproduce anatomic and functional features of the CBGTC and basal ganglia-cerebellar-thalamocortical networks.
Basic cortico-basal ganglia-thalamocortical pathways and their explaination of tics
Historically, investigators have attempted to simplify the complex motor control CBGTC circuitry by emphasizing three primary pathways: the goal-directed and habitual pathway, which controls deliberate (goal-directed) and automatic (non-goal-directed) actions; the direct and indirect pathway, which facilitates and inhibits movements respectively, and the hyperdirect pathway, which suppresses unwanted impulses. Although each can be individually useful in interpreting a defined anatomic or biochemical alteration, these relatively simplistic models do not fully capture the integrated and broad spectrum of activities required for motor control.
Goal-Directed and Habitual Circuits
This circuitry ( Fig. 3 ) divides movements into 2 categories, goal-directed and habitual, and is restricted to corticostriatal pathways. , Goal-directed movements are defined as consciously performed, controlled, flexible actions that adapt quickly, and are undertaken to achieve a desired purposeful outcome. , The goal-directed circuit originates from pyramidal cells in the orbitofrontal and ventral-medial prefrontal cortex and projects caudate. Habitual behaviors are automatically performed, slow to change, repetitive, stimulus-responsive actions, driven by reinforcement, that have no link to a current or future goal. , The habitual circuit originates in pyramidal cells in the premotor/supplementary motor area (SMA) and projects to the putamen.
