Oscillatory cortical drive to isometrically contracting muscle in Unverricht-Lundborg type progressive myoclonus epilepsy (ULD)

Cortico-muscular coherence and brain networks in familial adult myoclonic epilepsy and progressive myoclonic epilepsy

OBJECTIVE

Familial Adult Myoclonic Epilepsy (FAME) presents with action-activated myoclonus, often associated with epilepsy, sharing various features with Progressive Myoclonic Epilepsy (PMEs), but with slower course and limited motor disability. We aimed our study to identify measures suitable to explain the different severity of FAME2 compared to EPM1, the most common PME, and to detect the signature of the distinctive brain networks.

METHODS

We analyzed the EEG-EMG coherence (CMC) during segmental motor activity and indexes of connectivity in the two patient groups, and in healthy subjects (HS). We also investigated the regional and global properties of the network.

RESULTS

In FAME2, differently from EPM1, we found a well-localized distribution of beta-CMC and increased betweenness-centrality (BC) on the sensorimotor region contralateral to the activated hand. In both patient groups, compared to HS, there was a decline in the network connectivity indexes in the beta and gamma band, which was more obvious in FAME2.

CONCLUSIONS

In FAME2, better localized CMC and increased BC in comparison with EPM1 patients could counteract the severity and the spreading of the myoclonus. Decreased indexes of cortical integration were more severe in FAME2.

SIGNIFICANCE

Our measures correlated with different motor disabilities and identified distinctive brain network impairments.

Cortico-muscular and cortico-cortical coherence changes resulting from Perampanel treatment in patients with cortical myoclonus

OBJECTIVE

To investigate the mechanisms by which Perampanel (PER) reduces the severity of action myoclonus, we studied on MEG signals the changes occurring in cortico-muscular coherence (CMC) and cortico-cortical connectivity in patients with progressive myoclonus epilepsies.

METHODS

The subjects performed an isometric extension of the hand; CMC and cortico-cortical connectivity were assessed using autoregressive models and generalized partial-directed coherence. The contralateral (Co) sensors showing average CMC values >0.7 of the maximum (set to 1) were grouped as central (C) regions of interest (ROI), while adjacent sensors showing CMC values >0.3 were grouped as Surrounding (Sr) ROIs.

RESULTS

Under PER treatment, CMC decreased on Co C and Sr ROIs, but also on homologous ipsilateral (Ip) ROIs; out-degrees and betweenness centrality increased in Co ROIs and decreased in Ip ROIs. The flow from Ip to Co ROIs and from activated muscles to Ip C ROI decreased.

CONCLUSION

The improvement of myoclonus corresponded to decreased CMC and recovered leadership of the cortical regions directly involved in the motor task, with a reduced interference of ipsilateral areas.

SIGNIFICANCE

Our study highlights on mechanisms suitable to treating myoclonus and suggests the role of a reduced local synchronization together a better control of distant synaptic effects.

MEG Insight into the Spectral Dynamics Underlying Steady Isometric Muscle Contraction

To gain fundamental knowledge on how the brain controls motor actions, we studied in detail the interplay between MEG signals from the primary sensorimotor (SM1) cortex and the contraction force of 17 healthy adult humans (7 females, 10 males). SM1 activity was coherent at ∼20 Hz with surface electromyogram (as already extensively reported) but also with contraction force. In both cases, the effective coupling was dominant in the efferent direction. Across subjects, the level of ∼20 Hz coherence between cortex and periphery positively correlated with the “burstiness” of ∼20 Hz SM1 (Pearson r ≈ 0.65) and peripheral fluctuations (r ≈ 0.9). Thus, ∼20 Hz coherence between cortex and periphery is tightly linked to the presence of ∼20 Hz bursts in SM1 and peripheral activity. However, the very high correlation with peripheral fluctuations suggests that the periphery is the limiting factor. At frequencies <3 Hz, both SM1 signals and ∼20 Hz SM1 envelope were coherent with both force and its absolute change rate. The effective coupling dominated in the efferent direction between (1) force and the ∼20 Hz SM1 envelope and (2) the absolute change rate of the force and SM1 signals. Together, our data favor the view that ∼20 Hz coherence between cortex and periphery during isometric contraction builds on the presence of ∼20 Hz SM1 oscillations and needs not rely on feedback from the periphery. They also suggest that effective cortical proprioceptive processing operates at <3 Hz frequencies, even during steady isometric contractions.

SIGNIFICANCE STATEMENT Accurate motor actions are made possible by continuous communication between the cortex and spinal motoneurons, but the neurophysiological basis of this communication is poorly understood. Using MEG recordings in humans maintaining steady isometric muscle contractions, we found evidence that the cortex sends population-level motor commands that tend to structure according to the ∼20 Hz sensorimotor rhythm, and that it dynamically adapts these commands based on the <3 Hz fluctuations of proprioceptive feedback. To our knowledge, this is the first report to give a comprehensive account of how the human brain dynamically handles the flow of proprioceptive information and converts it into appropriate motor command to keep the contraction force steady.

A clinical and neurophysiological motor signature of Unverricht-Lundborg disease

OBJECTIVES

Unverricht-Lundborg disease (ULD) is the most common form of progressive myoclonus epilepsy. Cerebellar dysfunction may appear over time, contributing along with myoclonus to motor disability. The purpose of the present work was to clarify the motor and neurophysiological characteristics of ULD patients.

METHODS

Nine patients with genetically proven ULD were evaluated clinically (medical history collected from patient charts, the Scale for the Assessment and Rating of Ataxia and Unified Myoclonus Rating Scale). Neurophysiological investigations included EEG, surface polymyography, long-loop C-reflexes, somatosensory evoked potentials, EEG jerk-locked back-averaging (JLBA) and oculomotor recordings. All patients underwent brain MRI. Non-parametric Mann-Whitney tests were used to compare ULD patients' oculomotor parameters with those of a matched group of healthy volunteers (HV).

RESULTS

Myoclonus was activated by action but was virtually absent at rest and poorly induced by stimuli. Positive myoclonus was multifocal, often rhythmic and of brief duration, with top-down pyramidal temporospatial propagation. Cortical neurophysiology revealed a transient wave preceding myoclonus on EEG JLBA (n=8), enlarged somatosensory evoked potentials (n=7) and positive long-loop C-reflexes at rest (n=5). Compared with HV, ULD patients demonstrated decreased saccadic gain, increased gain dispersion and a higher frequency of hypermetric saccades associated with decreased peak velocity.

CONCLUSION

A homogeneous motor pattern was delineated that may represent a ULD clinical and neurophysiological signature. Clinical and neurophysiological findings confirmed the pure cortical origin of the permanent myoclonus. Also, oculomotor findings shed new light on ULD pathophysiology by evidencing combined midbrain and cerebellar dysfunction.

The network sustaining action myoclonus: a MEG-EMG study in patients with EPM1

Background

To explore the cortical network sustaining action myoclonus and to found markers of the resulting functional impairment, we evaluated the distribution of the cortico-muscular coherence (CMC) and the frequency of coherent cortical oscillations with magnetoencephalography (MEG). All patients had EPM1 (Unverricht-Lundborg) disease known to present with prominent and disabling movement-activated myoclonus.

Methods

Using autoregressive models, we evaluated CMC on MEG sensors grouped in regions of interests (ROIs) above the main cortical areas. The movement was a repeated sustained isometric extension of the right hand and right foot. We compared the data obtained in 10 EPM1 patients with those obtained in 10 age-matched controls.

Results

As expected, CMC in beta band was significantly higher in EPM1 patients compared to controls in the ROIs exploring the sensorimotor cortex, but, it was also significantly higher in adjacent ROIs ipsilateral and contralateral to the activated limb. Moreover, the beta-CMC peak occurred at frequencies significantly slower and more stable frequencies in EPM1 patients with respect to controls. The frequency of the beta-CMC peak inversely correlated with the severity of myoclonus.

Conclusions

the high and spatially extended beta-CMC peaking in a restricted range of low-beta frequencies in EPM1 patients, suggest that action myoclonus may result not only from an enhanced local synchronization but also from a specific oscillatory activity involving an expanded neuronal pool. The significant relationship between beta-CMC peak frequency and the severity of the motor impairment can represent a useful neurophysiological marker for the patients’ evaluation and follow-up.

Cerebellar Involvement in Patients with Mild to Moderate Myoclonus Due to EPM1: Structural and Functional MRI Findings in Comparison with Healthy Controls and Ataxic Patients

EPM1 (epilepsy, progressive myoclonic 1; Unverricht-Lundborg disease, OMIM #254800) is the most frequent form of progressive myoclonus epilepsy. Previous findings have suggested that its pathophysiology mainly involves the cerebellum, but the evaluation of cerebellar dysfunction is still unsatisfactory. The aim of this study was to assess the structural and functional involvement of the cerebellum in EPM1. We used voxel-based morphometry and spatially unbiased infra-tentorial template analyses of structural magnetic resonance imaging (MRI) scans, and functional MRI (fMRI) scans during block and event-related go/no-go motor tasks to study 13 EPM1 patients with mild to moderate myoclonus. We compared the results with those obtained in 12 age-matched healthy controls (HCs) and in 12 patients with hereditary spinocerebellar ataxia (SCA). Structural analyses revealed different patterns of atrophic changes in the EPM1 and SCA patients: in the former, they involved both cerebrum and cerebellum but, in the latter, only the cerebellum. During fMRI, block and event-related go/no-go tasks similarly activated the cerebellum and cerebrum in the EPM1 patients and HCs, whereas both tasks revealed much less cerebellar activation in the SCA patients than in the other two groups. Volumetric evaluation of the EPM1 patients showed that the cerebellum seemed to be marginally involved in a widespread atrophic process, and fMRI showed that it was not functionally impaired during motor tasks.

[Magnetoencephalography: a method for the study of brain function in neurosurgery]

Magnetoencephalography (MEG) is a non-invasive method for the study of electro-magnetic brain activity. Using multi-channel recordings the topography of the magnetic field can be recorded above the scalp with a temporal resolution of less than one millisecond. The method is suitable for the description and localization of cortical brain functions. The magnetic field strength that can be measured at up to 300 sensors is in the range of a few femto Tesla (10(-15) T) to somepico Tesla (10(-12) T). In order to measure these low magnetic fields highly sensitive SQUID-detectors are used on the one hand. On the other hand appropriate shielding equipment is employed to reduce effects of noise. Besides brain responses evoked by internal and external events (event-related magnetic fields), state-dependant oscillatory brain activity MEG can be recorded (spontaneous activity). Slow cortical oscillations in the range of 1 to 4 Hz are generated by damage of brain tissue and in the surrounding of brain tumors. In neurosurgery these activities can be used to monitor therapeutic success. Furthermore, oscillatory activities provide information about cortical regions involved in motor control. The measurement of motor related activities allows for the identification of recovery processes and reorganization after brain injury. Event-related magnetic brain responses are used in pre-surgical diagnosis and planning of treatment in epilepsy. In addition, they can be utilized to assess alterations in the functional organization of the cortex following injuries, tumor growth and neurosurgical interventions.

Protein aggregation as a possible cause for pathology in a subset of familial Unverricht–Lundborg disease

Loss of function mutations in the gene (CSTB) encoding human cystatin B, a widely expressed cysteine protease inhibitor, are responsible for a severe neurological disorder known as an Unverricht-Lundborg disease (EPM1). EPM1 had been linked to chromosome 21q22.3 in Finnish families and it is an autosomal recessive inherited disorder with a homozygous minisatellite expansion in the cystatin B gene (stefin B gene). This disease is difficult to treat because it is refractory to most antiepileptic drugs and using multiple medications had been unsuccessful so far. To come a step closer to understanding of the nature of this disease, especially about the events on the molecular level, in vitro properties of missense EPM1 mutant G4R were determined. It was observed that the mutant has a prolonged lag phase of fibrillation at the same protein stability, which could indicate it were more toxic to the cells. Similar experiments with the N-terminal fragment of 67 aminoacid residues are ongoing, showing higher propensity to aggregate. Therefore, a hypothesis is launched that at least in a subset of Unverricht-Lundborg disease patients, cystatin B protein may aggregate in the cell. Protein aggregation can be secondary to external insults or aging, however, inherited forms of neurodegenerative diseases, such as familial Parkinson's, Huntington's or familial Alzheimer's disease, are directly linked to the mutant proteins aggregation. Protein aggregates in the form of amyloid plaques, neurofibrilary tangles, intra-cytoplasmic or intra-nuclear inclusions lead to increased production of the reactive oxygen species and dysfunction of the ubiquitine/proteasome system. Finally, mitochondrial dysfunction and cell death are observed. Certainly, it remains to be checked by experiments whether overexpression in cell culture of the missense mutants G4R and N-terminal fragment to residue 68 lead to cellular inclusions and the accompanying changes characteristic for the conformational disorders.

Myoclonus: current concepts and recent advances

Myoclonus presents as a sudden brief jerk caused by involuntary muscle activity. An organisational framework is crucial for determining the medical significance of the myoclonus as well as for its treatment. Clinical presentations of myoclonus are divided into physiological, essential, epileptic, and symptomatic. Most causes of myoclonus are symptomatic and include posthypoxia, toxic-metabolic disorders, reactions to drugs, storage disease, and neurodegenerative disorders. The assessment of myoclonus includes an initial screening for those causes that are common or easily corrected. If needed, further testing may include clinical neurophysiological techniques, enzyme activities, tissue biopsy, and genetic testing. The motor cortex is the most commonly shown myoclonus source, but origins from subcortical areas, brainstem, spinal, and peripheral nervous system also occur. If treatment of the underlying disorder is not possible, treatment of symptoms is worthwhile, although limited by side-effects and a lack of controlled evidence.

Synchronous cortical oscillatory activity during motor action

Oscillations of the motor cortex interact with similar activity of the spinal motoneuron pool in the 15-30 Hertz frequency range. Recent observations have demonstrated how this interaction affects the firing of single corticospinal neurons. The interaction, reflected as corticomuscular coherence, occurs for both distal and proximal muscles and it constitutes one connection in a larger web of oscillatory interactions, including several other motor areas in the cortex, thalamus, and cerebellum. New results cast light on the possible functional significance of this interaction. The rhythmic interaction may reveal interesting information in several motor disorders, including essential tremor, Parkinson's disease, myoclonus epilepsy, and mirror movements.