Aberrant reward processing in Parkinson's disease is associated with dopamine cell loss

Abstract

Dopamine has been implicated in reward-related impulsivity, but the exact relationship between dopamine, reward and impulsivity in humans remains unknown. We address this question in Parkinson's disease (PD), which is characterized by severe dopamine depletion. PD is associated primarily with motor and cognitive inflexibility, but can also be accompanied by reward-related impulsivity. This paradoxical symptom of PD has often been attributed to dopaminergic overstimulation by antiparkinson medication, which is necessary to relieve the motor and cognitive inflexibility. However, factors other than medication may also contribute to aberrant impact of reward. Here we assess whether cognitive inflexibility and aberrant reward impact in PD are two sides of the same coin, namely dopamine cell loss. To measure dopamine cell loss, we employed ¹²³I-FP-CIT Single Photon Emission Computed Tomography (SPECT) in 32 PD patients (10 never-medicated patients and 22 patients after withdrawal of all medication for > 12 h) and related the values to behavior on a rewarded task-switching paradigm. Dopamine cell loss was associated not only with cognitive inflexibility (under low reward), but also with aberrant impact of reward. These effects could not be attributed to medication use. Relative to controls (n = 26), aberrant reward processing in PD was particularly expressed as reduced capacity to maintain (i.e., repeat) the current task-set under high reward. Our findings demonstrate that factors intrinsically related to PD may underlie the paradoxical symptoms of inflexibility and reward-related impulsivity in PD. The present results concur with observations that low baseline dopamine states predispose to drug and other addictions.

Introduction

Dopamine has long been implicated in reward motivation and impulsivity. However, the precise nature of the relationship between dopamine, reward and impulsivity remains unclear. Here we address this issue by assessing reward processing in Parkinson's disease (PD), a neurodegenerative disorder characterized by severe dopamine loss. PD can be accompanied by compulsive drug taking (i.e., dopamine dysregulation syndrome) and/or impulse control problems, such as pathological gambling, compulsive shopping, and hyper-sexuality (Evans et al., 2009, Voon et al., 2011, Weintraub et al., 2010). According to recent understanding, such impulsive–compulsive behavior may result from the common dopaminergic medication that is prescribed to treat the motor and cognitive symptoms of PD, such as the D2/3 receptor agonists and levodopa (Gallagher et al., 2007, Pontone et al., 2006, Weintraub et al., 2010). In particular, we and others have put forward the dopamine overdose hypothesis, stating that medication doses that are necessary to remedy severe dopamine depletion in the dorsal striatum might detrimentally overdose relatively intact dopamine levels in the ventral striatum (Cools et al., 2001a, Cools et al., 2003, Cools et al., 2007a, Swainson et al., 2000). Remediation of dorsal striatal dopamine would lead to restoration of motor and cognitive inflexibility, whereas overdosing of ventral striatal dopamine would contribute to aberrant reward sensitivity and impulsivity. This theory concurs with the implication of dopamine in reward and incentive motivation (Dagher and Robbins, 2009), and with the well-known dopamine-releasing properties of drugs and other rewards of abuse (Boileau et al., 2006, Di Chiara and Imperato, 1988, Leyton et al., 2002).

However, factors unrelated to medication have also been implicated in aberrant reward sensitivity and impulsivity in PD, for instance premorbid personality and genetic vulnerability (Dagher and Robbins, 2009, Evans et al., 2009). Furthermore, in apparent contradiction to the overdose hypothesis, trait impulsivity has been associated with low dopamine function (Buckholtz et al., 2010, Cools et al., 2007b, Dalley et al., 2007). Specifically, pathological gambling and other addictions have been hypothesized to reflect self-medication of a reward-deficient state (Blum et al., 2000, Reuter et al., 2005). If true, then the aberrant impact of reward in PD might be related, at least in part, to degeneration of dopamine cells. This implies that the dopamine-depleted state of PD should be accompanied not only by motor and cognitive inflexibility, but also by aberrant impact of reward.

Here we investigated associations between, on the one hand, cognitive inflexibility and reward impact, and on the other hand, the degree of dopamine cell loss of mild PD patients. Cognitive inflexibility was anticipated to be positively associated with dopamine cell loss. Furthermore, aberrant reward impact was hypothesized to be greatest in patients with the greatest dopamine cell loss. To test this hypothesis, dopamine cell loss was quantified using pre-synaptic dopamine transporter (DaT) imaging with ¹²³I-FP-CIT Single Photon Emission Computed Tomography (SPECT). Patients performed a behavioral task that allowed us to quantify cognitive inflexibility as well as the impact of reward. Specifically, we used a task-switching paradigm where each trial was preceded by either a high or low reward cue (Aarts et al., 2010). Cognitive inflexibility was indexed by the ability to switch between tasks under low reward. Previous task-set switching studies in PD have found that switch deficits were restricted to switching between well-established sets, and did not extend to new, to-be-learned sets (Cools et al., 2001b, Lewis et al., 2005, Slabosz et al., 2006). Because the best established task (here, the arrow task) is known to evoke larger switch costs than the least established task (here, the word task) (Wylie and Allport, 2000), we expected cognitive inflexibility in PD to be most pronounced in the arrow task.