Elsevier

NeuroImage

Volume 46, Issue 4, 15 July 2009, Pages 958-966
NeuroImage

In vivo evidence for the selective subcortical degeneration in Huntington's disease

https://doi.org/10.1016/j.neuroimage.2009.03.044Get rights and content

Abstract

Although Huntington's disease is largely considered to be a subcortical disease, there is no clear consensus on whether all deep grey matter loss is a direct downstream consequence of the massive degeneration of the medium-size spiny neurons in the striatum. Our aim was to characterise in vivo such preferential degeneration by analysing various distinct diffusion imaging measures including mean diffusivity, anisotropy, fibre orientation (using the information of the principal diffusion direction) and white matter tractography. All results converged to demonstrate the selective degeneration of connections in subcortical grey and white matter, degeneration which was likely to originate with the death of the striatal medium-size spiny neurons. Indeed, we found a significant increase of MD and FA in all the subcortical grey matter structures involved in the cortico-striato-thalamo-cortical loops. The atypical striatal and pallidal increase of FA was concurrent to a decrease of the dispersion of the fibre orientation, unambiguously characterising a preferential loss of connections along specific radiating directions from these structures while some others are comparatively spared. Analysis of striatal and pallidal white matter tracts revealed that striato-pallidal projections were the most affected. The ability of DTI to uncover the impact of such neurodegenerative disease on some specific neuronal/axonal populations is a further step towards the future definition of a surrogate marker of this disease. Beyond Huntington's disease, we prove here that diffusion imaging technique, associated to adequate methodological analyses, can provide insight into any neurodegenerative disorder for which some neuronal populations or connections are selectively targeted over others.

Introduction

Huntington's disease (HD) is a lethal autosomic dominant neurodegenerative disease that predominantly affects subcortical grey matter structures. Post mortem, it is characterised at the cellular level by the progressive degeneration of the striatal medium-size spiny neurons, which represent the great majority of the neuronal populations of the striatum (Graveland et al., 1985, Vonsattel et al., 1985). In vivo, the subcortical impact of HD is macroscopically depicted by the atrophy of the striatum, the pallidum and the thalamus–referred to hereafter as the “subcortical grey nuclei” or SGN–but also of the substantia nigra as measured by conventional structural MRI (Jernigan et al., 1991, Aylward et al., 2004, Fennema-Notestine et al., 2004, Douaud et al., 2006).

It remains unclear whether all deep grey matter loss can be attributed to the original death of striatal spiny neurons. The primary degeneration in the striatum may spread to the other subcortical nuclei composing the basal ganglia loops via the preferential loss of striato-pallidal projections. Being able to characterise such preferential degeneration in basal ganglia connections would be crucial to determine whether protecting or replacing striatal cells would be enough to prevent extrastriatal cell loss (Bachoud-Levi et al., 2000, McBride et al., 2006). Nevertheless, this question has never been addressed in vivo before and, contrary to post mortem investigations, would benefit from access to data at an early stage of the disease, permit a whole-brain exploration of the pathophysiological process and offer a unique way to monitor its progression and the effect of experimental treatments. However, this is not a question that can be answered by conventional imaging techniques. Indeed, these techniques are unable to distinguish regional brain atrophy that is general or non-specific in its cellular targets from atrophy that affects specific cell populations.

On the contrary, we hypothesised that diffusion tensor imaging (DTI) would have the potential to make such a distinction in cases where the connections of different populations of neurons have different orientational organization. First, mean diffusivity (MD) tends to be increased in pathological tissue–grey or white matter–where the density of cellular membranes is reduced. In tissue that is preferentially organized along a particular orientation, such as along axonal fibres, fractional anisotropy (FA) is commonly found to be decreased in case of a pathological process as a result of the reduction in cellular boundaries that hinder diffusion (Pierpaoli and Basser, 1996, Beaulieu, 2002). No systematic effect is observed on the main orientation of the underlying tissue, represented in practice by the orientation of the first eigenvector of the tensor (also called principal diffusion direction or PDD). However in deep grey matter, where there are multiple populations of cells and connections with different orientational preferences, a selective degeneration of some of these connections would make this tissue appear less isotropic, which corresponds to an atypical increase of FA. Subsequently, this preferential degeneration would also be characterised by a better coherence of these multiple orientations within these structures, i.e. by a decrease in the spatial dispersion of the PDD. A systematic shift of orientation of the PDD away from the more vulnerable connections should also be seen (Pierpaoli et al., 2001).

In HD, if the loss of subcortical grey matter originates from the death of medium-size spiny neurons in the striatum, then one organized population of neurons and axons should be comparatively more affected than the others. We thus would expect to see an increase in MD in deep grey matter structures as a marker of degeneration, accompanied by an increase in FA as evidence for the selectivity of this degeneration. Concurrently, fibre orientations should be spatially distributed more homogeneously, i.e. we should be able to show a decrease in their dispersion. In the putamen for instance, where connections to the pallidum (medio-lateral orientation) should degenerate more rapidly than the cortico-subcortical projections (dorso-ventral direction), we should be able to demonstrate a systematic change in fibre orientation towards this superior/inferior axis. Some selectivity in the degeneration of subcortical white matter tracts would also be expected. That is, for each subcortical structure, its connections (reconstructed using tractography) that would be directly affected by the initial death of striatal spiny neurons should be found more abnormal than the others.

HD therefore here serves as an exceptional pathological model to assess whether DTI technique combined to a specific methodology will be able to demonstrate a selective degeneration. This would clearly be of benefit for a better understanding and monitoring of other neurodegenerative diseases such as multiple sclerosis, in which an increase of FA in the basal ganglia was first witnessed (Ciccarelli et al., 2001).

Section snippets

Materials and methods

This study was part of the MIG-HD (Multicentric Intracerebral Grafting in HD) protocol and was approved by the ethics committee of Henri Mondor Hospital in Créteil. All subjects gave their written informed consent.

Whole-brain and ROI investigation of MD and FA: a first step towards the evidence of selective degeneration in the SGN

We found a significant increase in MD in the HD patients primarily located in the subcortical grey matter. More precisely, higher MD values were found bilaterally in the anterior and posterior putamen, the pallidum, the ventral striatum and the thalamus (centro-median nucleus and right ventro-lateral nucleus) (Fig. 2). Additional increase of MD was found in the corona radiata, more precisely in the frontal and parietal white matter, mainly around the central sulcus. All the significant results

Discussion

Although HD is mainly considered to be a subcortical disease, there is no clear consensus on whether all the deep grey matter loss is directly related to the original death of the medium-size spiny neurons of the striatum. It is essential to be capable of characterising in vivo such preferential degeneration as it would give a better insight into the evolving process of this neurodegenerative disease. To question this matter, we have used several distinct diffusion imaging measures, some of

Acknowledgments

We thank Drs. Jean-François Mangin and Edouard Duchesnay at the SHFJ for providing helpful advices. We would also like to thank Drs. Jill O'Reilly and Michael Chappell for their help with editing the English grammar. We are finally grateful to Prof. Stephen Smith for being magnanimous concerning the time spent writing this manuscript. This study is part of the MIG-HD program and is supported by the Délégation Régionale à la Recherche Clinique (PHRC P001106), the Association Française contre les

References (49)

  • ManginJ.F. et al.

    Distortion correction and robust tensor estimation for MR diffusion imaging

    Med. Image Anal.

    (2002)
  • MitchellI.J. et al.

    The selective vulnerability of striatopallidal neurons

    Prog. Neurobiol.

    (1999)
  • ParentA. et al.

    The pallidointralaminar and pallidonigral projections in primate as studied by retrograde double-labeling method

    Brain Res.

    (1983)
  • PierpaoliC. et al.

    Water diffusion changes in Wallerian degeneration and their dependence on white matter architecture

    Neuroimage

    (2001)
  • WakaiM. et al.

    A histometrical study on the globus pallidus in Huntington's disease

    J. Neurol. Sci.

    (1993)
  • AlbinR.L. et al.

    Preferential loss of striato-external pallidal projection neurons in presymptomatic Huntington's disease

    Ann. Neurol.

    (1992)
  • AlexanderG.E. et al.

    Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, “prefrontal” and “limbic” functions

    Prog. Brain Res.

    (1990)
  • AylwardE.H. et al.

    Onset and rate of striatal atrophy in preclinical Huntington disease

    Neurology

    (2004)
  • BeaulieuC.

    The basis of anisotropic water diffusion in the nervous system—a technical review

    NMR Biomed.

    (2002)
  • BehrensT.E. et al.

    Non-invasive mapping of connections between human thalamus and cortex using diffusion imaging

    Nat. Neurosci.

    (2003)
  • CiccarelliO. et al.

    Investigation of MS normal-appearing brain using diffusion tensor MRI with clinical correlations

    Neurology

    (2001)
  • Cointepas Y., Mangin J.F., Garnero L., Poline J.B., Benali H. BrainVISA: Software platform for visualization and...
  • DuchesnayE. et al.

    Population classification based on structural morphometry of cortical sulci

  • Fennema-NotestineC. et al.

    In vivo evidence of cerebellar atrophy and cerebral white matter loss in Huntington disease

    Neurology

    (2004)
  • Cited by (170)

    • Evaluation of early microstructural changes in the R6/1 mouse model of Huntington's disease by ultra-high field diffusion MR imaging

      2021, Neurobiology of Aging
      Citation Excerpt :

      The development of new diffusion MRI techniques has opened a new realm with a great number of parameters associated with microstructural anomalies in HD (Rosas et al., 2006). Diffusion tensor imaging (DTI), a mono-exponential Gaussian imaging technique, has been applied to obtain parameters mostly related to WM changes during the disease in mouse models (Douaud et al., 2009; Gatto, Rodolfo G. et al., 2020; Magnotta et al., 2009), whereas other more complex models like neurite orientation dispersion and density imaging (NODDI) are contributing with additional information in different diseases at the clinical level (Kamiya et al., 2020), with few but growing number of examples in the context of HD preclinical models. Thus, a more comprehensive approach combining the different diffusion MRI techniques could be applied to increase our understanding of the intricate cellular changes in the early stages of HD.

    View all citing articles on Scopus
    View full text