Connectivity-based segmentation of the striatum in Huntington's disease: Vulnerability of motor pathways

https://doi.org/10.1016/j.nbd.2011.02.010Get rights and content

Abstract

The striatum, the primary site of degeneration in Huntington's disease (HD), connects to the cerebral cortex via topographically organized circuits subserving unique motor, associative and limbic functions. Currently, it is not known whether all cortico-striatal circuits are equally affected in HD. We aimed to study the selective vulnerability of individual cortico-striatal circuits within the striatum in HD, and hypothesized that motor cortico-striatal pathways would be most affected, consistent with HD being a primarily motor disorder. Diffusion Tensor Imaging (DTI) tractography was used to identify connections between the striatum and seven major cortical regions in 12 HD patients and 14 matched controls. The striatum of both groups was parcellated into subregions based on connectivity with the cerebral cortex. Volumetric and DTI microstructural measures of Fractional Anisotropy (FA) and Mean Diffusivity (MD) were obtained within each subregion and compared statistically between groups. Tractography demonstrated the topographic organization of cortical connections in the striatum of both controls and HD patients. In HD patients, the greatest difference from controls in volume, FA and MD was observed in M1 and S1 subregions of the caudate and putamen. Motor symptoms correlated with volume and MD in sensorimotor striatal subregions, suggesting that sensorimotor striatal degeneration is closely related to motor dysfunction. DTI tractography provides a novel approach to sensitively examine circuit-specific abnormalities in HD and has identified that the motor cortico-striatal circuit is selectively vulnerable in HD.

Research Highlights

► DTI tractography identifies cortico-striatal pathways in the human brain in vivo. ► Connectivity-based striatal segmentation enables investigation of Huntington's disease. ► We report in vivo circuit-specific striatal neurodegeneration in Huntington's disease.

Introduction

Degeneration of medium spiny neurons in the striatum is the principal neuropathological feature of Huntington's disease (HD) (Hersch and Ferrante, 1997). Neurons of the striatum receive topographically organized inputs from the cerebral cortex, with distinct cortical regions projecting to similarly discrete areas of the striatum (Alexander et al., 1986, Draganski et al., 2008, Lehericy et al., 2004, Lehéricy et al., 2004). These largely segregated circuits are understood to subserve unique motor, associative and limbic functions. Two of the circuits most relevant to HD include the motor and dorsolateral prefrontal cognitive loop. The motor loop, which regulates movement preparation and execution, connects the putamen with premotor, supplementary motor and primary sensorimotor cortical regions. The cognitive loop, on the other hand, connects the caudate with the frontal and posterior parietal cortices and functions in working memory, attention and cognitive flexibility (Alexander et al., 1986, Postuma and Dagher, 2006). Currently, it is not clear if all cortico-striatal circuits are affected equally in HD, or if there is some selective vulnerability.

Until recently, the in vivo examination of individual cortico-striatal circuits has not been possible. Diffusion Tensor Imaging (DTI) tractography enables tracing of white matter pathways in vivo (Jones, 2008) and has recently been used to demonstrate individual cortico-striatal circuits in humans. Draganski et al. (2008) used DTI to reveal, for the first time in humans, topographically organized and partially segregated cortical connections within the striatum. Another study recently used tractography to provide insight into the relationship between structure and function, demonstrating that the number of fibers connecting the striatum with subcortical limbic areas predicted novelty seeking, while fibers between the striatum and prefrontal cortex predicted reward dependence differences (Cohen et al., 2008). DTI tractography-based approaches have previously been used to parcellate subcortical grey matter nuclei into connectivity-based subregions with distinct cortical connectivity profiles (Behrens et al., 2003, Draganski et al., 2008), and remarkable consistency in connectivity profiles has been observed across a range of grey and white matter regions (Behrens et al., 2003, Chao et al., 2009, Doron and Gazzaniga, 2008, Hofer and Frahm, 2006, Park et al., 2008, Ramnani et al., 2006, Zarei et al., 2007, Zarei et al., 2006). The precise knowledge of human cortico-striatal connectivity elicited by these recent studies now presents the exciting possibility of studying individual cortico-striatal circuits in HD.

In addition to enabling cortico-striatal circuits to be examined, DTI also provides measures of microstructure that may provide insight into circuit-specific degeneration. DTI measures water movement, which is constrained by structural features such as axonal membranes and myelin, as well as architectural factors such as axonal density and fiber crossings. Two measures, Fractional Anisotropy (FA) and Mean Diffusivity (MD) have shown particular utility for examining striatal microstructure in HD (Basser and Pierpaoli, 1996). FA measures the degree to which structures in a voxel are coherently oriented. In WM, high FA indicates more coherent fiber organization, with degeneration associated with a decrease in FA. In striatal grey matter in HD however, degeneration has been shown to increase FA, particularly in the putamen (Douaud et al., 2009, Kloppel et al., 2008, Magnotta et al., 2009, Rosas et al., 2006), possibly due to the selective loss of efferent pathways leading to an overall increase in coherence (Douaud et al., 2009). MD, a measure of overall water diffusion, is increased in grey and white matter where cell membrane structure is compromised or cell density reduced, consistent with findings of increased MD in the striatum in HD (Douaud et al., 2009, Magnotta et al., 2009, Mascalchi et al., 2004, Rosas et al., 2006, Seppi et al., 2006, Sritharan et al., 2010, Vandenberghe et al., 2009). Several studies have demonstrated associations between striatal DTI metrics and measures of motor and neuropsychological function, disease duration, and proximity to diagnosis in HD, which suggests a relationship between microstructural damage and functional decline (Magnotta et al., 2009, Mascalchi et al., 2004, Seppi et al., 2006). Although no studies have so far utilized such an approach, examination of DTI measures in individual cortico-striatal circuits could provide significant insights into circuit-specific or selective degeneration in HD.

Given that HD is a primarily motor disorder, we hypothesized that the putamen motor loop (Alexander et al., 1986) would be selectively vulnerable. Interestingly, recent neuroimaging findings demonstrate that sensory and motor cerebral cortices show the most severe atrophy in HD, which suggests that the motor cortico-striatal circuit is particularly affected (Douaud et al., 2009, Rosas et al., 2005, Rosas et al., 2002, Rosas et al., 2008). To directly examine selective vulnerability, we aimed to parcellate the caudate and putamen in HD patients and matched controls based on connectivity with the cerebral cortex using DTI tractography, then to subsequently examine volume, FA and MD within individual striatal subregions. Finally we aimed to determine if striatal subregion measures of volume and microstructure were related to the severity of motor symptoms. We hypothesized that if the motor circuit was most affected in HD, volume and microstructural abnormalities would be most severe in motor regions of the striatum, and that these abnormalities would correlate with motor symptoms.

Section snippets

Participant characterisation and image acquisition

This study was approved by the ethics committees of the Howard Florey Institute and Monash University, and written informed consent was obtained from all participants. Data for 12 genetically tested and clinically diagnosed HD patients and 14 healthy controls was used in the study. All individuals were right-handed, and controls had no history of neurological disorder. Clinical and demographic characteristics are shown in Table 1. Based on judgment of the assessing neurologist, all HD

Cortical connectivity pattern of the striatum in controls and HD patients

The caudate showed connectivity with all cerebral cortical regions (Fig. 2A). The entire caudate showed connectivity with PF, and moreover, the anterior caudate was connected almost exclusively to PF. PM/SM, M1, S1 and PP connections overlapped significantly and were observed primarily in the posterior half of the caudate (Fig. 3A). TEM connections were variably located, primarily in the lateral anterior caudate, and overlapping with PF connections. OCC connections were interspersed throughout

Discussion

We hypothesized that the motor cortico-striatal circuit would be most affected in clinical HD patients. Using tractography, we successfully identified individual cortico-striatal circuits within the striatum, performed connectivity-based parcellation of the striatum into subregions, and compared subregion normalized volume and microstructural measures between HD patients and controls.

Conclusions

DTI connectivity-based parcellation represents a novel approach to study specificity of neurodegenerative mechanisms in the HD brain. Previous efforts to understand striatal degeneration in HD have examined volume and microstructural measures in the complete anatomical structure of interest, or in arbitrarily delineated segments (e.g. anterior/posterior). Accumulating in vivo evidence presented here and elsewhere of the topographic organization of cortico-striatal connections strongly suggests

Acknowledgments

This study was funded by grant number 234247 from the National Health and Medical Research Council (Australia). We sincerely thank all participants for their time and effort. We also thank the Brain Research Institute (Austin Hospital) for the use of the 3 T GE scanner. This study was carried out at Howard Florey Institute and School of Psychology, Psychiatry and Psychological Medicine, Monash University, Clayton, Victoria 3800, Australia.

References (35)

  • Y. Chao

    Probabilistic topography of human corpus callosum using cytoarchitectural parcellation and high angular resolution diffusion imaging tractography

    Hum. Brain Mapp.

    (2009)
  • M. Cohen

    Connectivity-based segregation of the human striatum predicts personality characteristics

    Nat. Neurosci.

    (2008)
  • B. Draganski

    Evidence for segregated and integrative connectivity patterns in the human basal ganglia

    J. Neurosci.

    (2008)
  • S. Hersch et al.
  • S. Kloppel

    White matter connections reflect changes in voluntary-guided saccades in pre-symptomatic Huntington's disease

    Brain

    (2008)
  • S. Lehericy

    3-D diffusion tensor axonal tracking shows distinct SMA and pre-SMA projections to the human striatum

    Cereb. Cortex

    (2004)
  • S. Lehéricy

    Diffusion tensor fiber tracking shows distinct corticostriatal circuits in humans

    Ann. Neurol.

    (2004)
  • Cited by (76)

    View all citing articles on Scopus
    View full text