Spatial Remapping of Cortico-striatal Connectivity in Parkinson's Disease

Rick C. Helmich 1 2 3 Loes C. Derikx 2 3 Maaike Bakker 1 2 3 Rene Scheeringa 2 3 Bastiaan R. Bloem 1 3 Ivan Toni 0 2 3 0 Donders Institute for Brain, Cognition and Behaviour, Centre for Cognition, Radboud University Nijmegen , 6500 HB Nijmegen, the Netherlands 1 Department of Neurology and Parkinson Centre Nijmegen (ParC), Radboud University Nijmegen Medical Centre , 6500 HB Nijmegen, the Netherlands 2 Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen , 6500 HB Nijmegen, the Netherlands 3 The Author 2009. Published by Oxford University Press. All rights reserved. For permissions , please Parkinson's disease (PD) is characterized by striatal dopamine depletion, especially in the posterior putamen. The dense connectivity profile of the striatum suggests that these local impairments may propagate throughout the whole cortico-striatal network. Here we test the effect of striatal dopamine depletion on cortico-striatal network properties by comparing the functional connectivity profile of the posterior putamen, the anterior putamen, and the caudate nucleus between 41 PD patients and 36 matched controls. We used multiple regression analyses of resting-state functional magnetic resonance imaging data to quantify functional connectivity across different networks. Each region had a distinct connectivity profile that was similarly expressed in patients and controls: the posterior putamen was uniquely coupled to cortical motor areas, the anterior putamen to the pre--supplementary motor area and anterior cingulate cortex, and the caudate nucleus to the dorsal prefrontal cortex. Differences between groups were specific to the putamen: although PD patients showed decreased coupling between the posterior putamen and the inferior parietal cortex, this region showed increased functional connectivity with the anterior putamen. We conclude that dopamine depletion in PD leads to a remapping of cerebral connectivity that reduces the spatial segregation between different cortico-striatal loops. These alterations of network properties may underlie abnormal sensorimotor integration in PD. Introduction Parkinsons disease (PD) is characterized by a degeneration of dopaminergic cells in the midbrain (Braak et al. 2003), which leads to dopamine depletion in the striatum (Brooks and Piccini 2006). This neurochemical alteration impairs neuronal processing in the basal ganglia (Rivlin-Etzion et al. 2006), which propagates, through the dense cortico-striatal connections (Houk and Wise 1995), to altered activity in other brain regions (van Eimeren and Siebner 2006). This indicates that taking a network perspective on PD is fundamental for understanding the pathophysiology of this disease (He et al. 2007). Previous neuroimaging studies in PD have described patterns of spatial covariance between different brain regions during performance of a task (Monchi et al. 2004), as well as steadystate differences in brain activity during rest (Eckert et al. 2007). These patterns of coactivations might suggest the presence of a functional circuit (Postuma and Dagher 2006), but networks are better defined on the basis of the structure of temporal interactions between regions (functional connectivity; He et al. 2007). Accordingly, electrophysiological studies have used this approach to describe altered connectivity patterns in PD (Williams et al. 2002; Stoffers et al. 2008), but these methods have very limited spatial coverage and are mostly blind to subcortical structures. Previous functional magnetic resonance imaging (fMRI) studies have focused on altered connectivity related to performance of a specific task (Rowe et al. 2002; Helmich et al. 2009), but this approach confines the findings to a particular cognitive process. In contrast, here we study the temporal coupling between intrinsic blood oxygen level--dependent (BOLD) fluctuations over the whole brain, testing whether striatal dysfunction in PD alters functional connectivity both within and between different cortico-striatal circuits. Using intrinsic BOLD fluctuations to study functional connectivity of the human brain is a relatively novel experimental approach, supported by empirical evidence detailing the specific spatial and temporal structure of these fluctuations (Biswal et al. 1995; Damoiseaux et al. 2006; Fox and Raichle 2007). These intrinsic fluctuations engage specific cerebral assemblies on a time scale of several seconds (Biswal et al. 1995), and they are thought to reflect the hemodynamic consequences of slow variations in transient neuronal dynamics that propagate through anatomically connected networks (Ghosh et al. 2008; He et al. 2008; Honey et al. 2007, 2009). The huge metabolic load of these intrinsic fluctuations suggests that they are functionally relevant (Fox and Raichle 2007), possibly by normalizing or consolidating synaptic weights within a cerebral network (Pinsk and Kastner 2007; Balduzzi et al. 2008). In addition, it has been shown that alterations in these intrinsic fluctuations can be used as a marker of network dysfunction (Li et al. 2002; Greicius et al. 2004; Sheline et al. 2009). Here we compare intrinsic fluctuations measured in PD patients and healthy controls, focusing on 3 distinct corticostriatal loops involving the posterior putamen, the anterior putamen, and the caudate nucleus. This parcellation rests on 2 facts. First, these cortico-striatal loops have been clearly described in macaques (Alexander et al. 1986), and they have recently been confirmed in healthy humans using both diffusion tensor imaging (Lehericy, Ducros, Van de Moortele, et al. 2004; Draganski et al. 2008) and resting-state fMRI (Di Martino et al. 2008; Zhang et al. 2008; Kelly et al. 2009). In macaques, these loops remain largely segregated in terms of functional processing and anatomical connectivity (Alexander et al. 1986; Hoover and Strick 1993). For example, whereas the head of the caudate receives massive projections from the prefrontal cortex, the posterior putamen connects to the primary motor cortex and the supplementary motor area (SMA) (Alexander et al. 1986). Second, these loops respect the regionally specific pattern of dopamine depletion observed in PD. That is, although the posterior putamen is heavily depleted of dopamine, the anterior putamen and the caudate nucleus are relatively spared (Kish et al. 1988; Guttman et al. 1997; Nurmi et al. 2001; Bruck et al. 2006). Accordingly, we test the hypothesis that PD patients show altered cortico-striatal connectivity, and that this alteration follows the specific spatial pattern of dopamine depletion occurring in this disease. This implies that functional connectivity within the cortico-striatal loop passing through the posterior putamen should decrease, whereas connectivity with the anterior putamen and the caudate nucleus should remain relatively intact. Furthermore, given that dopamine depletion might cause patho (...truncated)