SUMMARY

Sleep is a behavioral state ideal for studying functional connectivity because it minimizes many sources of between-subject variability that confound waking analyses. This is particularly important for potential connectivity studies in mental illness where cognitive ability, internal milieu and active psychotic symptoms can vary widely across subjects. We, therefore, sought to adapt techniques applied to magnetoencephalography for use in high-density electroencephalography (EEG), the gold-standard in brain-recording methods during sleep. Autoregressive integrative moving average modeling was used to reduce spurious correlations between recording sites (electrodes) in order to identify functional networks. We hypothesized that identified network characteristics would be similar to those found with magnetoencephalography, and would demonstrate sleep stage-related differences in a control population. We analysed 60-s segments of low-artifact data from seven healthy human subjects during wakefulness and sleep. EEG analysis of eyes-closed wakefulness revealed widespread nearest-neighbor positive synchronous interactions, similar to magnetoencephalography, though less consistent across subjects. Rapid eye movement sleep demonstrated positive synchronous interactions akin to wakefulness but weaker. Slow-wave sleep (SWS), instead, showed strong positive interactions in a large left fronto-temporal-parietal cluster markedly more consistent across subjects. Comparison of connectivity from early SWS to SWS from a later sleep cycle indicated sleep-related reduction in connectivity in this region. The consistency of functional connectivity during SWS within and across subjects suggests this may be a promising technique for comparing functional connectivity between mental illness and health.

Background

Sleep disturbance plays an important role in major depressive disorder (MDD). Prior investigations have demonstrated that slow wave activity (SWA) during sleep is altered in MDD; however, results have not been consistent across studies, which may be due in part to sex-related differences in SWA and/or limited spatial resolution of spectral analyses. This study sought to characterize SWA in MDD utilizing high-density electroencephalography (hdEEG) to examine the topography of SWA across the cortex in MDD, as well as sex-related variation in SWA topography in the disorder.

Methods

All-night recordings with 256 channel hdEEG were collected in 30 unipolar MDD subjects (19 women) and 30 age and sex-matched control subjects. Spectral analyses of SWA were performed to determine group differences. SWA was compared between MDD and controls, including analyses stratified by sex, using statistical non-parametric mapping to correct for multiple comparisons of topographic data.

Results

As a group, MDD subjects demonstrated significant increases in all-night SWA primarily in bilateral prefrontal channels. When stratified by sex, MDD women demonstrated global increases in SWA relative to age-matched controls that were most consistent in bilateral prefrontal regions; however, MDD men showed no significant differences relative to age-matched controls. Further analyses demonstrated increased SWA in MDD women was most prominent in the first portion of the night.

Conclusions

Women, but not men with MDD demonstrate significant increases in SWA in multiple cortical areas relative to control subjects. Further research is warranted to investigate the role of SWA in MDD, and to clarify how increased SWA in women with MDD is related to the pathophysiology of the disorder.

In a recent series of experiments, we demonstrated that a visuomotor adaptation task, 12 hours of left arm immobilization, and rapid transcranial magnetic stimulation (rTMS) during waking can each induce local changes in the topography of electroencephalographic (EEG) slow wave activity (SWA) during subsequent non-rapid eye movement (NREM) sleep. However, the poor spatial resolution of EEG and the difficulty of relating scalp potentials to the activity of the underlying cortex limited the interpretation of these results. In order to better understand local cortical regulation of sleep, we used source modeling to show that plastic changes in specific cortical areas during waking produce correlated changes in SWA during sleep in those same areas. We found that implicit learning of a visuomotor adaptation task induced an increase in SWA in right premotor and sensorimotor cortices when compared to a motor control. These same areas have previously been shown to be selectively involved in the performance of this task. We also found that arm immobilization resulted in a decrease in SWA in sensorimotor cortex. Inducing cortical potentiation with repetitive transcranial magnetic stimulation (rTMS) caused an increase in SWA in the targeted area and a decrease in SWA in the contralateral cortex. Finally, we report the first evidence that these modulations in SWA may be related to the dynamics of individual slow waves. We conclude that there is a local, plasticity dependent component to sleep regulation and confirm previous inferences made from the scalp data.

The existence of normal sleep in patients in a vegetative state is still a matter of debate. Previous electrophysiological sleep studies in patients with disorders of consciousness did not differentiate patients in a vegetative state from patients in a minimally conscious state. Using high-density electroencephalographic sleep recordings, 11 patients with disorders of consciousness (six in a minimally conscious state, five in a vegetative state) were studied to correlate the electrophysiological changes associated with sleep to behavioural changes in vigilance (sustained eye closure and muscle inactivity). All minimally conscious patients showed clear electroencephalographic changes associated with decreases in behavioural vigilance. In the five minimally conscious patients showing sustained behavioural sleep periods, we identified several electrophysiological characteristics typical of normal sleep. In particular, all minimally conscious patients showed an alternating non-rapid eye movement/rapid eye movement sleep pattern and a homoeostatic decline of electroencephalographic slow wave activity through the night. In contrast, for most patients in a vegetative state, while preserved behavioural sleep was observed, the electroencephalographic patterns remained virtually unchanged during periods with the eyes closed compared to periods of behavioural wakefulness (eyes open and muscle activity). No slow wave sleep or rapid eye movement sleep stages could be identified and no homoeostatic regulation of sleep-related slow wave activity was observed over the night-time period. In conclusion, we observed behavioural, but no electrophysiological, sleep wake patterns in patients in a vegetative state, while there were near-to-normal patterns of sleep in patients in a minimally conscious state. These results shed light on the relationship between sleep electrophysiology and the level of consciousness in severely brain-damaged patients. We suggest that the study of sleep and homoeostatic regulation of slow wave activity may provide a complementary tool for the assessment of brain function in minimally conscious state and vegetative state patients.

Slow waves are the most prominent electroencephalographic (EEG) feature of non-rapid eye movement (NREM) sleep. During NREM sleep, cortical neurons oscillate approximately once every second between a depolarized upstate, when cortical neurons are actively firing, and a hyperpolarized downstate, when cortical neurons are virtually silent (Steriade et al., 1993a; Destexhe et al., 1999; Steriade et al., 2001). Intracellular recordings indicate that the origins of the slow oscillation are cortical and that cortico-cortical connections are necessary for their synchronization (Steriade et al. 1993b; Amzica and Steriade, 1995; Timofeev and Steriade, 1996; Timofeev et al., 2000). The currents produced by the near-synchronous slow oscillation of large populations of neurons appear on the scalp as EEG slow waves (Amzica and Steriade, 1997).

Despite this cellular understanding, questions remain about the role of specific cortical structures in individual slow waves. Early EEG studies of slow waves in humans were limited by the small number of derivations employed and by the difficulty of relating scalp potentials to underlying brain activity (Brazier 1949; Roth et al 1956). Functional neuroimaging methods offer exceptional spatial resolution but lack the temporal resolution to track individual slow waves (Maquet, 2000; Dang-Vu et al., 2008). Intracranial recordings in patient populations are limited by the availability of medically necessary electrode placements and can be confounded by pathology and medications (Nir et al., 2010; Cash et al., 2009; Wenneberg 2010).

Source modeling of high-density EEG recordings offers a unique opportunity for neuroimaging sleep slow waves. So far, the results have challenged several of the influential topographic observations about slow waves that had persisted since the original EEG recordings of sleep. These recent analyses revealed that individual slow waves are idiosyncratic cortical events and that the negative peak of the EEG slow wave often involves cortical structures not necessarily apparent from the scalp, like the inferior frontal gyrus, anterior cingulate, posterior cingulate and precuneus (Murphy et al., 2009). In addition, not only do slow waves travel (Massimini et al., 2004), but they often do so preferentially through the areas comprising the major connectional backbone of the human cortex (Hagmann et al., 2008). In this chapter we will review the cellular, intracranial recording and neuroimaging results concerning EEG slow waves. We will also confront a long held belief about peripherally evoked slow waves, also known as K-complexes, namely that they are modality-independent and do not involve cortical sensory pathways. The analysis included here is the first to directly compare K-complexes evoked with three different stimulation modalities within the same subject on the same night using high-density EEG.

Slow waves and sleep spindles are the two main oscillations occurring during NREM sleep. While slow oscillations are primarily generated and modulated by the cortex, sleep spindles are initiated by the thalamic reticular nucleus (TRN), and regulated by thalamo-reticular and thalamo-cortical circuits. In a recent high-density electroencephalographic (hd-EEG) study we found that 18 medicated schizophrenics had reduced sleep spindles compared to healthy and depressed subjects during the first NREM episode. Here we investigated whether spindle deficits were: a) present in a larger sample of schizophrenic patients; b) consistent across the night; c) related to antipsychotic medications; d) suggestive of impairments in specific neuronal circuits. Whole night hd-EEG recordings were performed in 49 schizophrenics, 20 non-schizophrenic patients on antipsychotics and 44 healthy subjects. In addition to sleep spindles, several parameters of slow waves were assessed. Schizophrenics had whole-night deficits in spindle power (12–16 Hz) and in slow (12–14 Hz) and fast (14–16 Hz) spindle amplitude, duration, number and integrated spindle activity (ISA) in prefrontal, centroparietal and temporal regions. ISA and spindle number had the largest effect sizes (ES≥2.21). By contrast, no slow wave deficits were found in schizophrenics. These results indicate that spindle deficits i) can be reliably established in schizophrenics, ii) are stable across the night, iii) are unlikely to be due to antipsychotic medications, and iv) point to deficits in TRN and thalamo-reticular circuits.

A state-space formulation is introduced for estimating multivariate autoregressive (MVAR) models of cortical connectivity from noisy, scalp recorded EEG. A state equation represents the MVAR model of cortical dynamics while an observation equation describes the physics relating the cortical signals to the measured EEG and the presence of spatially correlated noise. We assume the cortical signals originate from known regions of cortex, but that the spatial distribution of activity within each region is unknown. An expectation maximization algorithm is developed to directly estimate the MVAR model parameters, the spatial activity distribution components, and the spatial covariance matrix of the noise from the measured EEG. Simulation and analysis demonstrate that this integrated approach is less sensitive to noise than two-stage approaches in which the cortical signals are first estimated from EEG measurements, and next an MVAR model is fit to the estimated cortical signals. The method is further demonstrated by estimating conditional Granger causality using EEG data collected while subjects passively watch a movie.

Synaptic modification supporting memory formation is thought to depend on gene expression and protein synthesis. Disrupting either process around the time of learning prevents the formation of long-term memory. Recent evidence suggests that memory also becomes susceptible to disruption upon retrieval. Whether or not the molecular events involved in the formation of new memory are the same as what is needed for memory to persist after retrieval has yet to be determined. In the present set of experiments, rats were given inhibitors of protein or messenger ribonucleic acid (mRNA) synthesis into the amygdala just after training or retrieval of fear memory. Results showed that blocking mRNA or protein synthesis immediately after learning prevented the formation of long-term memory, while stability of memory after retrieval required protein, but not mRNA, synthesis. These data suggest that the protein needed for memory reconsolidation after retrieval may be transcribed from pre-existing stores of mRNA.