Altered cerebellar feedback projections in Asperger syndrome

Abstract

It has been proposed that the biological basis of autism spectrum disorder includes cerebellar ‘disconnection’. However, direct in vivo evidence in support of this is lacking. Here, the microstructural integrity of cerebellar white matter in adults with Asperger syndrome was studied using diffusion tensor magnetic resonance tractography. Fifteen adults with Asperger syndrome and 16 age–IQ–gender-matched healthy controls underwent diffusion tensor magnetic resonance imaging. For each subject, tract-specific measurements of mean diffusivity and fractional anisotropy were made within the inferior, middle, superior cerebellar peduncles and short intracerebellar fibres. No group differences were observed in mean diffusivity. However, people with Asperger syndrome had significantly lower fractional anisotropy in the short intracerebellar fibres (p < 0.001) and right superior cerebellar (output) peduncle (p < 0.001) compared to controls; but no difference in the input tracts. Severity of social impairment, as measured by the Autistic Diagnostic Interview, was negatively correlated with diffusion anisotropy in the fibres of the left superior cerebellar peduncle. These findings suggest a vulnerability of specific cerebellar neural pathways in people with Asperger syndrome. The localised abnormalities in the main cerebellar outflow pathway may prevent the cerebral cortex from receiving those cerebellar feedback inputs necessary for a successful adaptive social behaviour.

Introduction

Autism spectrum disorder (ASD, including autism and Asperger syndrome) is a relatively common neurodevelopmental disorder characterised by a triad of repetitive and stereotypic behaviour, impaired communication and striking deficits in social reciprocity (WHO, 1993). Evidence implicating the cerebellum in the pathophysiology of autism was originally put forward by Ornitz et al. in the 1960s and 1970s (Ornitz, 1985). Their studies showed that autistic children exhibited clinical and neurophysiologic deficits indicative of cerebellar dysfunction (Ciaranello and Ciaranello, 1995). Some years later, the role of the cerebellum was further suggested by neuropathological studies reporting a reduction of the Purkinje cell density in the cerebellar cortex in both children and adults from across the autistic spectrum (Bauman and Kemper, 1985, Lee et al., 2002, Ritvo et al., 1986, Vargas et al., 2005, Williams et al., 1980) (for a review see Palmen et al., 2004b). Other neuropathological abnormalities of the cerebellum have been described in autism; these include cytoplasmatic inclusions in Purkinje cells (Bailey et al., 1998), a reduction of the number of cells (Kemper and Bauman, 1998) and ectopic grey matter (Bailey et al., 1998) in the deep cerebellar nuclei, and an intense neuroinflammatory process extending to the cerebellar white matter (Vargas et al., 2005).

Reports of structural differences in the cerebellum as determined by magnetic resonance imaging have been less consistent. In some studies the midsagittal area of vermis (lobules VI–VII) has been reported to be smaller in children with autism (Carper and Courchesne, 2000, Courchesne et al., 1994, Kates et al., 1998, Kaufmann et al., 2003). Other studies reported the midsagittal area of vermis to be larger (Courchesne et al., 1994) or similar to controls in both children (Kleiman et al., 1992, Piven et al., 1997) and adults (Garber and Ritvo, 1992, Piven et al., 1997) with autism. An increase of the total cerebellar volume has also been found in both adults and children (but not in very young children aged < 35 months (Hazlett et al., 2005) with autism (Herbert et al., 2003, Palmen et al., 2005, Palmen et al., 2004a, Piven et al., 1997, Sparks et al., 2002), but this enlargement seems to be proportional to the enlargement of the total cerebral volume, and not specific to the cerebellum (Palmen et al., 2004a, Piven et al., 1997). Volumetric studies of the cerebellum in adults with Asperger syndrome found no differences as compared to age matched controls (McAlonan et al., 2002). Segmentation studies of the cerebellum white matter showed volume reduction in children with autism (Boddaert et al., 2004, McAlonan et al., 2005) and adults with Asperger syndrome (McAlonan et al., 2002). Segmentation studies of the cerebellum grey matter showed volume reduction in adults with Asperger syndrome (Abell et al., 1999, McAlonan et al., 2002) but no differences in children with autism (McAlonan et al., 2005). Preliminary analysis of structural “connectivity” found a negative correlation between the volume of frontal lobe and cerebellar vermis in children with autism (Carper and Courchesne, 2000). There is also evidence of a correlation between volumetric differences in cerebellum and brain activation during cognitive and motor tasks (Allen and Courchesne, 2003, Allen et al., 2004). Overall these studies raise the possibility that people with ASD have abnormalities in the development of cerebello-frontal networks but direct in vivo anatomical evidence to support this hypothesis is still lacking.

One technique for exploring white matter pathways in vivo is diffusion tensor magnetic resonance imaging (DT-MRI) tractography. This is a non-invasive neuroimaging technique which has been used to reconstruct three-dimensional trajectories of white matter tracts (Basser et al., 2000, Catani et al., 2002, Conturo et al., 1999, Croxson et al., 2005, Lehericy et al., 2004, Mori et al., 2000, Poupon et al., 2000, Rushworth et al., 2006) and probe the microstructural integrity of white matter in a wide range of conditions including epilepsy (Concha et al., 2005), and schizophrenia (Jones et al., 2006). Tractography has also been applied to study cerebellar white matter anatomy in the healthy population (Ramnani et al., 2006), and in subjects with cerebellar disorders including phenytoin users with cerebellar atrophy (Lee et al., 2003), spinocerebellar degeneration disease (Taoka et al., 2007) and people with cerebellar congenital malformations (Lee et al., 2005). To date, however, the microstructural integrity of cerebellar neural systems in people with ASD has not been directly investigated in vivo with tractography. We used DT-MRI tractography to study if people with Asperger syndrome had differences in the microstructural integrity of cerebellar white matter and to determine whether white matter abnormalities are diffuse to all cerebellar connections or specific to single tracts that are relevant to cognition and behaviour.