Altered cortical beta-band oscillations reflect motor system degeneration in amyotrophic lateral sclerosis

Abstract

Continuous rhythmic neuronal oscillations underpin local and regional cortical communication. The impact of the motor system neurodegenerative syndrome amyotrophic lateral sclerosis (ALS) on the neuronal oscillations subserving movement might therefore serve as a sensitive marker of disease activity. Movement preparation and execution are consistently associated with modulations to neuronal oscillation beta (15-30 Hz) power. Cortical beta-band oscillations were measured using magnetoencephalography (MEG) during preparation for, execution, and completion of a visually cued, lateralized motor task that included movement inhibition trials. Eleven "classical" ALS patients, 9 with the primary lateral sclerosis (PLS) phenotype, and 12 asymptomatic carriers of ALS-associated gene mutations were compared with age-similar healthy control groups. Augmented beta desynchronization was observed in both contra- and ipsilateral motor cortices of ALS patients during motor preparation. Movement execution coincided with excess beta desynchronization in asymptomatic mutation carriers. Movement completion was followed by a slowed rebound of beta power in all symptomatic patients, further reflected in delayed hemispheric lateralization for beta rebound in the PLS group. This may correspond to the particular involvement of interhemispheric fibers of the corpus callosum previously demonstrated in diffusion tensor imaging studies. We conclude that the ALS spectrum is characterized by intensified cortical beta desynchronization followed by delayed rebound, concordant with a broader concept of cortical hyperexcitability, possibly through loss of inhibitory interneuronal influences. MEG may potentially detect cortical dysfunction prior to the development of overt symptoms, and thus be able to contribute to the assessment of future neuroprotective strategies. Hum Brain Mapp 38:237-254, 2017. © 2016 Wiley Periodicals, Inc.

Keywords: biomarker; inhibition; magnetoencephalography; motor neurone disease; neuroimaging; neurophysiology.

Figure 1

Task design schematic demonstrating visual cues instructing lateralized motor preparation for 1 or 2 s (dependent on cue style) followed by response only to green Go targets. [Color figure can be viewed at http://wileyonlinelibrary.com.]

Figure 2

Behavioral data detailing task performance across groups. A: PLS patients responded slower than older controls, P = 0.016. B: A trend towards more NoGo errors was made by pre symptomatic carriers. C: Inverse efficiency, a global measure of task performance (RT/NoGo proportion correct) was impaired in the PLS group (P = 0.022). Median values within participants, mean values within group, error bars = SEM. Uncorrected for multiple comparisons. [Color figure can be viewed at http://wileyonlinelibrary.com.]

Figure 3

MEG data epoched around cue (t ₀ = laterality cue appearance). A: Neural correlate of task performance across frequency range demonstrated in TFR (averaged across healthy controls only) from motor cortex ROIs, contra/ipsilateral relative to effector limb. Beta‐band (15:30 Hz) power decrease (desynchronization) from baseline (more intense desynchronization = deeper blue) occurs during motor preparation, particularly in the contralateral hemisphere, maximally 600 ms post‐cue. B,C: Beta‐band power (relative to baseline) within contralateral motor cortex ROI, group comparisons. D: ALS patients show deeper beta desynchronization. t ₀ + 500 ms: t ₀ + 700 ms. t stats in red/blue, motor cortex mask used for statistics in green. Cluster correction within motor cortex mask, not across groups. Vertical lines denote appearance of visual laterality cue and go/Nogo target. [Color figure can be viewed at http://wileyonlinelibrary.com.]

Figure 4

MEG data epoched around movement initiation ( t ₀ = response execution). A: TFR (averaged across healthy controls) demonstrates movement ERD (in blue) as a bilateral motor cortical event. B,C: Beta‐band power (relative to baseline) within contralateral motor cortex ROI; group comparison. Vertical lines indicate cue onset prior to median RT (distribution overlaid). D: AGCs show deeper beta desynchronization [t ₀: t ₀ + 200 ms] relative to controls, with a directionally similar trend in ALS patients. t stats in red/blue, motor cortex mask used for statistics in green. Vertical lines denote appearance of Go target and timing of response. [Color figure can be viewed at http://wileyonlinelibrary.com.]

Figure 5

MEG data epoched around movement completion (t ₀ = finger replaced). A: TFR (averaged across healthy controls only) from motor cortex ROIs relative to effector limb. Post Movement Beta Rebound (PMBR, in red) is predominantly contralateral in healthy controls. B,C: Beta‐band power (relative to baseline) within contralateral motor cortex ROI; group comparison. Black bar = timespan of significant group difference via cluster permutation testing, P < 0.05. Vertical lines denote appearance of Go target, timing of response initiation and completion. D: ALS patients demonstrate lower beta power in early transition to PMBR [t ₀ + 500 ms: t ₀ + 700 ms]. t‐stats in blue, motor cortex mask used for statistics in green. E: Beta‐band power lateralization evolves over time around a response, PLS patients show a diminished DoL during PMBR (black bar, P < 0.05). F: PMBR DoL appraised in whole‐brain data [t ₀ + 1 s: t ₀ + 2.5 s]. PLS patients again show significantly reduced DoL. t‐stats in ROY‐BIG‐BL. [Color figure can be viewed at http://wileyonlinelibrary.com.]

Figure 6

MEG data epoched around NoGo target presentation (t ₀ = target appearance). Beta‐band power increases over a diffuse cortical network in successfully inhibited NoGo trials, demonstrated in (A) healthy controls, beta power NoGo > Go. [t ₀ + 100 ms: t ₀ + 300 ms]. B: AGCs contrasted against HC, NoGo trials only, reveals increased beta power in posterior cortical regions. No significant differences are noted in ALS or PLS patients. t‐stats in ROY‐BIG‐BL. Whole brain cluster‐permutation correction. [Color figure can be viewed at http://wileyonlinelibrary.com.]