Saccade reprogramming in Friedreich ataxia reveals impairments in the cognitive control of saccadic eye movement

Abstract

Although cerebellar dysfunction has known effects on motor function in Friedreich ataxia (FRDA), it remains unclear the extent to which the reprogramming of eye movements (saccades) and inhibition of well-learned automatic responses are similarly compromised in affected individuals. Here we examined saccade reprogramming to assess the ability of people with FRDA to respond toward unexpected changes in either the amplitude or direction of an “oddball” target. Thirteen individuals with genetically confirmed FRDA and 12 age-matched controls participated in the study. The saccade reprogramming paradigm was used to examine the effect of an unpredictable “oddball” target on saccade latencies and accuracy when compared to a well-learned sequence of reciprocating movements. Horizontal eye movements were recorded using a scleral search coil eye tracking technique. The results showed a proportionally greater increase in latencies for reprogrammed saccades toward an oddball-direction target in the FRDA group when compared to controls. The FRDA group were also less accurate in primary saccade gain (i.e. ratio of saccade amplitude to target amplitude) when reprogramming saccades toward an unexpected change in direction. No significant group differences were found on any of the oddball-amplitude targets. Significant correlations were revealed between latency and disease severity as measured by the Friedreich Ataxia Rating Scale. These findings provide further support to the view that cognitive changes in FRDA may arise from disruption of cerebellar connections to cortical structures.

Introduction

Early views of cerebellar function proposed that the input to the cerebellum from sensory systems was devoted exclusively to motor learning and the control of voluntary movement (Glickstein, 1998, Ito, 2001, Stein and Glickstein, 1992, Wolpert et al., 1998). However, recent studies in both humans and non-human primates supports the involvement of the cerebellum in higher cognitive functions conventionally ascribed to the prefrontal and posterior parietal cortices (for reviews, see Ramnani, 2012, Strick et al., 2009). The reverse cerebellar diaschisis hypothesis posits that functional changes in cortical regions are secondary to disruption to connectivity with the cerebellum, resulting in the cerebellar cognitive affective syndrome (Schmahmann & Sherman, 1998). Evidence for a cerebellar role in non-motor functions has been demonstrated by clinical and neuroimaging studies reporting cerebellar involvement in several cognitive functions including attention, language, executive functions and visuospatial cognition (Ito, 2008; Schmahmann, 2004; Stoodley & Schmahmann, 2009). However, to date little is known about the complex functional domains that require cerebellar processing, and so further examination of the contribution of the cerebellum to higher order processes is clearly warranted.

Important insights into the role of the cerebellum in cognitive functions are being gained by studying individuals with primary cerebellar pathology such as Friedreich ataxia (FRDA). FRDA is the most common form of inherited ataxia with a prevalence of approximately 1 in 29,000 (Delatycki, Williamson, & Forrest, 2000). FRDA is a multi-system autosomal-recessive disease due to mutations in the FXN gene on chromosome 9q13. Symptoms typically emerge around puberty with an average age of onset of 10 years (Delatycki et al., 1999). In 98% of mutant alleles, there is an expansion of a GAA trinucleotide repeat in intron 1 of the gene whilst in the other 2% there is a point mutation/deletion (Voncken, Ionnou, & Delatycki, 2004). FRDA is characterized by a diverse range of clinical features including progressive gait ataxia, spasticity, impaired vibration sense and proprioception, and eye movement disturbances including fixation instability. The pathology associated with FRDA is a consequence of a reduction in frataxin, a mitochondrial protein involved in iron sulfur cluster synthesis (Campuzano et al., 1996, Vaubel and Isaya, 2013). The major sites of pathology in FRDA include the dorsal root ganglia and posterior columns of the spinal cord, spinocerebellar tracts, and dentate nucleus of the cerebellum (Junck et al., 1994, Pandolfo, 2003). Although the primary regions of pathology are associated with areas supporting motor function, recent studies have demonstrated specific cognitive impairments that may be associated with cerebellar involvement in FRDA, the source of which requires further investigation (Corben et al., 2011, Corben et al., 2011, Corben et al., 2010, Fielding et al., 2010, Hocking et al., 2010).

Cognitive functions in FRDA have been a relatively under-researched aspect of the disease, possibly because the primary regions of pathology fall outside of regions conventionally thought to subserve cognition (Wollmann, Barroso, Monton, & Nieto, 2002). However, slowed information processing speed has been consistently reported in FDRA (see Corben et al., 2006, for a review), alongside impairments to working memory, sustained attention and other aspects of executive functioning (Klopper et al., 2011, White et al., 2000, Wollmann et al., 2002). More recently we have demonstrated reprogramming impairments using an experimental paradigm which incorporated movements with occasional “oddball” targets that change in amplitude or direction, or both, in individuals with FRDA (Corben et al., 2010). We found that individuals with FRDA were slower overall than controls in movement execution but were significantly less able to adjust movements to accommodate changes to both direction and extent. These findings suggest a specific impairment in the inhibition of well-learned automatic responses associated with stimulus–response incompatibility in FRDA. However, a limitation of assessing cognitive function in tasks that require a motor response in FRDA is the confounding effects of limb inertia and motor impairments.

Detection of an infrequent target stimulus involves widespread neural activation in regions involved in the executive control system (Loose et al., 1090, Schweizer et al., 2007). The “oddball” paradigm involves the inhibition of a planned response based on prior experience, and the initiation of an unexpected new response. This requires a switch of attention which evokes activation in prefrontal and parietal cortices (Clark et al., 2001, Clark et al., 2000, Strange et al., 2000). However, the ability to adapt to changes in task demands is also thought to involve the cerebellum, most probably via cerebellar feedforward loops facilitating connectivity with fronto-parietal regions critical to this function (Berger et al., 2005, Ramnani, 2012, Strick et al., 2009). In addition, the cerebellum, in conjunction with frontal and posterior parietal cortices, is crucial to controlling, monitoring and optimizing eye movements (Curtis & D’Esposito, 2003). The ocular motor system is increasingly viewed as a model motor system in so far as the neural circuitry involved in the generation of saccadic eye movements has been shown to critically depend on the integrity of the cerebellum, especially for adaptation for the amplitude and direction of saccades (Collins et al., 2008, Desmurget et al., 1998). Our recent studies have revealed a range of deficits in the higher order, or cognitive control of saccadic eye movements in individuals with FRDA (Fielding et al., 2010, Hocking et al., 2010). Significantly, the cognitive component of saccadic latency was strongly associated with measures of disease severity and duration, suggesting that ocular motor measures may be sensitive surrogate markers in evaluating future therapeutic trials in FRDA (Fielding et al., 2010, Hocking et al., 2010). However, the functional contribution of the cerebellum to the reprogramming of saccades that require greater cognitive control in the ability to select between and inhibit competing responses has not yet been determined.

The present study aimed to examine cerebellar contributions to the reprogramming of saccades and inhibition of well-learned automatic responses using a saccadic oddball paradigm in individuals with Friedreich ataxia. The oddball paradigm was employed to examine the capacity to inhibit a prepotent response and respond toward unexpected changes in either the direction or amplitude of an oddball target when compared to a well-learned sequence of reciprocating movements (Winograd-Gurvich et al., 2006a, Winograd-Gurvich et al., 2006b, Gurvich et al., 2007). It was hypothesized that cerebellar pathology in participants with FRDA would manifest in a proportionally greater increase in latencies for reprogramming of saccades toward oddball targets reflected in a more pronounced impairment in the ability to inhibit and switch to a new response. We explored the relationship between saccade reprogramming measures and clinical and genetic measures to identify potential biomarkers of disease severity and progression in individuals with FRDA.