Neurophysiological control of voluntary movement

Get Complete Project Material File(s) Now! »

Chapter 3 Does Response Inhibition Have Pre- and Postdiagnostic Utility in Parkinson’s Disease?

This chapter has been reported as a review article in Journal of Motor Behavior, MacDonald and Byblow, Does response inhibition have pre- and postdiagnostic utility in Parkinson’s disease? (2015); 47(1); 29-45. Reprinted with permission of Taylor & Francis Ltd.
Terminology: The current chapter uses DA to denote dopamine agonist, as opposed to dopamine.

Abstract

Parkinson’s disease (PD) is the second most prevalent degenerative neurological condition worldwide. Improving and sustaining quality of life is an important goal for Parkinson’s patients. Key areas of focus to achieve this goal include earlier diagnosis and individu-alized treatment. In this review the authors discuss impulse control in PD and examine how measures of impulse control from a response inhibition task may provide clinically useful information (i) within an objective test battery to aid earlier diagnosis of PD and (ii) in postdiagnostic PD, to better identify individuals at risk of developing impulse control disorders with dopaminergic medication.

 Introduction

Parkinson’s disease (PD) is a relatively common progressive neurodegenerative disease af-fecting about 6.3 million people worldwide (Apaydin et al., 2002; Braak et al., 2004; EPDA, 2014). The pathological process that underlies PD relentlessly progresses to the full-blown clinical syndrome over several years (Braak et al., 2004). PD is associated with the degenera-tion of dopaminergic nigrostriatal neurons in the substantia nigra pars compacta (SNpc) (Apaydin et al., 2002; Cameron et al., 2010; Gradinaru et al., 2009)), as well as several other nuclei of the brainstem (Grinberg et al., 2010). Although motor symptoms are common landmarks for diagnosis and staging, PD also involves disturbances in cognitive, limbic and autonomic systems.
Currently there is no definitive test for PD, making diagnosis relatively subjective. Clinical presentation of the disease can fluctuate over time, complicating disease monitoring and treatment evaluation. In addition, the pathology of this disease is only diagnostic in the brainstem, which is difficult to study during life, adding to the challenge of unbiased and objective monitoring. The time between the onset of neurodegeneration and the ability to clinically diagnose PD is termed the preclinical phase (Gaenslen et al., 2011; Truong and Wolters, 2009). During the preclinical phase unspecific symptoms appear in isolation. For this reason, such symptoms are only confirmed retrospectively after clinical diagnosis, but it is also possible for healthy individuals to exhibit many of these unspecific symptoms as part of normal ageing (e.g., impaired sense of smell, disturbed sleep, balance disturbance, muscle stiffness). Although the simple occurrence of select symptoms is not sufficient for a PD diagnosis (Gaenslen et al., 2011), the presentation of certain symptoms in a predicted chronological order (Braak et al., 2003; Przuntek et al., 2004) may identify a higher risk for PD. It has been proposed that at least two positive tests on potential preclinical symptoms may be considered sufficient evidence to diagnose a patient as clinically possible to have PD and begin protective therapy (Truong and Wolters, 2009). Affirmation of this probable diagnosis may be acquired through a response to PD treatment. During the preclinical phase, irretrievable dopaminergic neuron loss in SNpc has not yet reached the threshold for diagnosis. This phase presents an opportunity for treatment optimization through earlier diagnosis using tools that are sensitive enough to identify early changes in biological systems. The aim of earlier diagnosis would be to enable early neuroprotective therapies to prevent or delay further neuron degeneration (Berg and Poewe, 2012).
Once diagnosed, there are further challenges presented to patients and clinicians from treatment side effects. The two most common forms of treatment are medication and surgery (e.g., deep brain stimulation, pallidotomy). One prevalent and detrimental side effect of dopaminergic medication treatment is the development of impulse control disorders, which manifest as impulsive and compulsive behaviours in a variety of contexts. There is currently no available screening method to identify patients at high risk of developing impulse control disorders. Treatment decisions would be aided by an objective method to predict and classify an individual’s risk of impulsivity with dopaminergic treatment.
In this review we focus on impulse control in PD and how objective measures of im-pulse control could be used clinically pre- and postdiagnosis. The review has three main aims. First, to review and summarize functional changes in frontostriatal and basal ganglia-thalamocortical networks in PD and their effect on impulse control, specifically response inhibition. Second, to introduce the potential of standardized response inhibition paradigms from motor behaviour research to provide useful measures within an objective test battery to identify insidious motor and nonmotor changes during the preclinical phase, and aid earlier diagnosis. Third and finally, we introduce the idea of combining information obtained from response inhibition tasks with genetic analysis in the postdiagnostic phase of PD to better identify mechanisms which predispose individuals to impulse control disorders with dopaminergic treatment.

Impulse control in Parkinson’s disease

Execution of premature or inappropriate responses reflects poor impulse control (Duque and Ivry, 2009). There are two aspects of impulse control: cognitive/psychological and motor/behavioural (Nemoda et al., 2011). Impulsive decision making (i.e., dysfunctional cognitive impulse control) manifests as an inability to evaluate the potential consequences of a decision and modify the decision accordingly, and has been associated with inferior structural integrity of white matter projections between the prefrontal cortex (PFC) and striatum (Peper et al., 2013). The inability to suppress an unnecessary action (i.e., impaired response inhibition) is an example of the motor/behavioural aspect (Nemoda et al., 2011).
There is a general decrease in impulse control with ageing, seen in motor (Braver and Barch, 2002; Fisk and Sharp, 2004), and cognitive impulse control (Fisk and Sharp, 2004). The age-related deterioration of motor impulse control is exacerbated with basal ganglia (BG) dysfunction as is evident in focal dystonia (Stinear and Byblow, 2004) and PD (Bokura et al., 2005; Cooper et al., 1994; Gauggel et al., 2004; Obeso et al., 2011). A dose-dependent inverted-U relationship (Goldman-Rakic et al., 2000) has been hypothesized to explain observed behavioural changes in impulse control associated with changes in dopamine neurotransmission in PFC (Robbins and Arnsten, 2009). For those with lower levels of dopamine neurotransmission, increasing PFC dopamine concentration promotes better motor (Congdon et al., 2009) and cognitive impulse control (Diamond et al., 2004) and decreased impulsivity (Farrell et al., 2012).

 Evidence for motor tests as biomarkers for Parkinson’s disease

Deterioration of the motor system occurs well in advance of the clinical diagnosis of PD. Evidence of subtle motor deficits exists during the preclinical stage. Therefore salient tests of motor function could become possible biomarkers as part of a larger test battery or di-agnostic algorithm. Years before clinical diagnosis patients subjectively report increased stiffness, slowness of movement, changes in gait pattern, reduced arm swing, tremor and postural imbalance (de Lau et al., 2006; Gaenslen et al., 2011). Movement tasks are capable of providing objective measures of these abnormalities which may be sensitive but not neces-sarily specific to preclinical PD in isolation. However the presentation of these subtle motor abnormalities following specific nonmotor symptoms in a predicted chronological order may identify a higher risk for PD. Tasks of visuomotor control (Hocherman and Giladi, 1998) and handwriting (Horstink and Morrish, 1999) have identified deficits in the asymptomatic hand of patients in the early stage of PD. Movement tasks which are able to detect subtle deterioration of motor control show greatest potential as biomarkers as part of a larger test battery.
Motor deficits are present in de novo PD patients compared to healthy age-matched controls. For example, de novo PD patients show impaired upper limb performance during bimanual (Ponsen et al., 2006) and unimanual (Pfann et al., 2001; Ponsen et al., 2008) movements. During a complex bimanual circle-drawing task, measures of success rate and accuracy are lower in PD patients, especially in the nondominant hand (Ponsen et al., 2006). However, even simple unimanual movement tasks reveal impairments in modulation of muscle activity and coordination in de novo patients (Pfann et al., 2001; Ponsen et al., 2008). Pfann et al. (2001) demonstrated that electromyography (EMG) can reveal an inability to modify movements according to task demands (impaired motor set) in de novo PD patients during a rapid elbow flexion task. EMG traces of patients reveal an impaired modulation of muscle bursting during upper limb movements of both single-joint (Hallett and Khoshbin, 1980; Vaillancourt et al., 2004) and multijoint movements (Farley et al., 2004). EMG activity patterns are therefore a potentially sensitive motor measure for identification of PD, with one study finding 90 % sensitivity at differentiating individuals with PD from healthy controls during single-joint point-to-point flexion movements (Robichaud et al., 2009). Of note, one healthy control that demonstrated abnormal EMG was subsequently diagnosed with PD 30 months later. Ultimately, evidence exists that noninvasive and objective motor measures from EMG have the potential to reveal subclinical neurophysiological abnormalities during the preclinical stage.

 Response inhibition in Parkinson’s disease

Functional changes in basal ganglia-thalamocortical networks with Parkinson’s disease Nuclei of the BG are an integral part of the sensorimotor system, forming a cortico-subcortico-cortical loop involved in planning, executing and cancelling responses. Response inhibition (RI) depends critically upon interactions between the right frontal cortex (in particular, the right inferior frontal gyrus) and the BG (Aron et al., 2007b; Aron and Poldrack, 2006; Coxon et al., 2012; Jahfari et al., 2011; Robbins, 2007; Swann et al., 2011; Zandbelt and Vink, 2010). Striatal gray matter atrophy has been linked to impaired RI in PD (O’Callaghan et al., 2013). The striatum is the main input to the BG, receiving projections from the cerebral cortex, midbrain and thalamus (Figure 3.1). The striatum is divided into dorsal (caudate nucleus and putamen, separated by the internal capsule) and ventral regions (Kandel et al., 2013). BG sensorimotor networks comprise the motor portions of the putamen (i.e., dorsal striatum (Alexander and Crutcher, 1990a; Kandel et al., 2013).
Projecting from the dorsal striatum are two parallel pathways through the BG: the direct and indirect pathways (Figure 3.1). Both pathways modulate thalamic output to the cortex by either increasing (direct) or decreasing (indirect) thalamocortical drive (Alexander and Crutcher, 1990a; Danion and Latash, 2011). The subthalamic nucleus (STN) has been recognized as another significant input structure of the BG (Coizet et al., 2009; Lanciego et al., 2004). A third hyperdirect pathway projecting from the cortex can rapidly decrease thalamocortical drive (i.e., an inhibitory pathway) without synapsing onto the striatum, but rather directly onto STN (Nambu et al., 2002). Figure 3.1 illustrates examples of hyperdirect pathways from the inferior frontal cortex and the presupplementary motor area (preSMA) connecting directly to STN. Evidence suggests that hyperdirect pathways are critical for successful response inhibition (Aron et al., 2007a; Aron and Poldrack, 2006; Coxon et al., 2012; King et al., 2012).
The main BG output nuclei are internal globus pallidus (GPi) and substantia nigra pars reticulata (SNpr). These two structures provide sustained inhibitory input onto thalamocor-tical neurons. The general consensus is that movement initiation (through facilitation of the Figure 3.1 Basal ganglia sensorimotor network. Model of the cortico-basal ganglia-thalamocortical sensorimotor networks involved in movement execution and inhibitory control. The direct, indirect and hyperdirect pathways are represented at a basic level. Grey boxes indicate nuclei of functional basal ganglia; arrows indicate facilitatory input; filled circles indicate inhibitory input; open diamond indicates dopamine-dependent input. IFC: inferior frontal cortex; SNpc: substantia nigra pars compacta; GPe: globus pallidus exter-nus; STN: subthalamic nucleus; GPi: globus pallidus internus; SNpr: substantia nigra pars reticulata; preSMA: presupplementary motor area; M1: primary motor cortex. direct pathway) is an active process, requiring a pause in tonic inhibition and disinhibition of the thalamus. The default state of the motor system is analogous to driving with the brakes on, or the ‘hold your horses’ model (Ballanger et al., 2009). The spatial and tempo-ral recruitment of the direct, indirect and hyperdirect pathways gives rise to the complex functionality of the BG.
The balance of neuronal output from the three BG pathways is usually maintained by dopamine through dopaminergic projections from SNpc to all BG nuclei, but most signif-icantly through the dense connections to the dorsal and ventral striatum (Bjorklund and Dunnett, 2007). The origins of the direct and indirect pathways are from separate pop-ulations of striatal medium spiny neurons. The direct pathway originates from neurons expressing predominantly D1 dopamine receptors (DRD1), and the indirect pathway from neurons expressing predominantly D2 receptors (DRD2). Dopaminergic projections from SNpc differentially activate the two striatal neuronal populations due to the different dom-inant postsynaptic receptors. Dopamine binding to DRD1 facilitates the direct pathway, whereas dopamine binding at DRD2 suppresses the indirect pathway (Danion and Latash, 2011; Gerfen et al., 1990; Kandel et al., 2013; Obeso et al., 2013). The presence of dopamine therefore modulates movement control by reinforcing any cortically initiated activation of BG-thalamocortical networks leading to disinhibition of thalamocortical neurons (Alexander and Crutcher, 1990a).
In PD, RI is disrupted by dopamine disturbance within the sensorimotor BG-thalamocortical network. The degeneration of nigrostriatal dopaminergic neurons reduces the dopaminergic modulation through reduced striatal dopamine levels (Kandel et al., 2013). Reduced striatal dopamine results in less activation of striatal DRD1 and DRD2 affecting the direct and indirect pathways, respectively. Through decreased activation of DRD2s, striatal inhibitory projections to GPe become more active through reduced suppression. There is increased striatal inhibition of GPe, subsequent disinhibition of STN and therefore STN hyperactivity within the indirect pathway. Simultaneously, striatal projections to GPi and SNpr become less active through decreased DRD1 activation (Alexander and Crutcher, 1990a). As a result of increased STN excitation and reduced striatal inhibition, there is increased activity of GPi and SNpr, and a subsequent increase in tonic inhibition of the thalamocortical projection neurons. Therefore, with PD, maladaptive modulation of tonic inhibition of the thalamus through the separate striatal neuron subpopulations and subsequent BG pathways (Albin et al., 1989) has a crucial impact on motor impulse control (i.e., response inhibition).

Functional changes in frontostriatal networks with Parkinson’s disease

In PD, abnormal modulation of striatal and PFC dopamine results in dysfunction within frontostriatal networks (Shepherd, 2013) and impaired impulse control. Networks involving the PFC and striatum are implicated not only in motor control, but also in several cognitive functions, including memory, action selection, behaviour reinforcement, and contextual conditioning (Alexander and Crutcher, 1990a; Goto and Grace, 2005; Pennartz et al., 2009). Dysfunction of frontostriatal networks can therefore contribute to a multitude of symptoms. As with many biological systems the key to optimizing function is balance. The optimal range for dopaminergic neurotransmission for frontostriatal networks is best illustrated by the inverted-U relationship. For example, the optimal range of PFC dopaminergic neurotrans-mission is surrounded by that above or below optimal (Figure 3.2), resulting in decreased PFC function. However the relationship between dopamine and functional performance is also complex, and includes several contributing factors.
Reducing dopamine neurotransmission can impair certain functions while enhancing others. The functional state of PFC relates to RI performance. RI tasks are sensitive to cognitive deficits as well as motor impairments, and are essentially cognitive-motor tasks. In addition to proficient motor impulse control, RI also requires efficient attention control (199, 1997; Tachibana et al., 1997) and cognitive flexibility (Cooper et al., 1994), both of which recruit PFC. Decreasing prefrontal dopamine promotes cognitive flexibility in young (Blasi et al., 2005; Colzato et al., 2010b) and older (Fallon et al., 2013) healthy individuals and PD patients (Cools et al., 2010), at the expense of cognitive stability. These findings are in line with Dual State Theory (Durstewitz and Seamans, 2008) whereby a DRD1 dom-inated system contains high-energy barriers between neural state representations. This system therefore favors stability and perseveration. Conversely, a DRD2 dominated system favors switching between representational states through low energy barriers. Direct neu-rophysiological evidence for Dual State Theory is supported by rodent models (Goto and Grace, 2005). The inverted-U relationship explains why equivalent changes to dopaminergic neurotransmission can have opposite effects on cognitive functions.
In early PD, dopamine concentrations are decreased within regions of the SN and stria-tum, yet paradoxically dopamine concentration is increased in PFC (Kaasinen et al., 2001). This counterintuitive increase in PFC dopamine could reflect compensation within the mesocorticolimbic (MCL) network (Rakshi et al., 1999; Zigmond et al., 1990). It could be due to reduced striatal dopamine levels, given the wellstudied inverse relationship between MCL and nigrostriatal dopaminergic systems (Akil et al., 2003; Carr and Sesack, 2000; Carter and Pycock, 1980; Jahanshahi et al., 2010; Kolachana et al., 1995; Pycock et al., 1980; Roberts et al.,

1 Introduction 
1.1 Overview of the thesis
2 Review of Literature 
2.1 Neurophysiological control of voluntary movement
2.2 Inhibition of voluntary movement
2.3 Parkinson’s disease
3 Does Response Inhibition Have Pre- and Postdiagnostic Utility in Parkinson’s Disease?
3.1 Introduction
3.2 Impulse control in Parkinson’s disease
3.3 Response inhibition tasks in preclinical Parkinson’s disease
3.4 Preclinical Parkinson’s disease: Response inhibition tasks as a component of an objective test battery
3.5 Postdiagnostic Parkinson’s disease
3.6 Conclusion
4 Overview of Experimental Techniques andMethods 
4.1 Transcranial magnetic stimulation
4.2 Pharmacological intervention
4.3 Neuropsychological assessment
4.4 Computational modelling
4.5 Experimental setup for the anticipatory response inhibition task
4.6 Data analysis for the anticipatory response inhibition task
5 Uncoupling Response Inhibition 
5.1 Introduction
5.2 Methods
5.3 Results
5.4 Discussion
6 The Fall and Rise of Corticomotor Excitability with Cancellation and Reinitiation of Prepared Action 
6.1 Introduction
6.2 General methods
6.3 Experiment 1
6.4 Experiment 2
6.5 General discussion
6.6 Conclusion
7 An Activation ThresholdModel for Response Inhibition 
7.1 Introduction
7.2 Methods
7.3 Results
7.4 Discussion
8 Dopamine Gene Profiling Can Predict Impulse Control and Effects of Dopamine Agonist Ropinirole 
8.1 Introduction
8.2 Methods
8.3 Results
8.4 Discussion
8.5 Supplementarymaterial
9 General Discussion
9.1 Summary of main findings
9.2 Potential limitations
9.3 Future directions
References
GET THE COMPLETE PROJECT