Sleep and Brain Activity

Marcos G. Frank

• Publisher: Academic Press
• Year: 2012
• Print ISBN: 9780123849953
• eBook ISBN: 9780123849960
• Topic: Biological Sciences
• Subtopic: Neuroscience
• Index: Biological Sciences

In the last few decades, scientists have discovered that far from being a time of neural silence, sleep is characterized by complex patterns of electrical, neurochemical, and metabolic activity in the brain. Sleep and the Brain presents some of the more dramatic developments in our understanding of brain activity in sleep. The book discusses what parts of the brain are active in sleep and how, and presents research on the function of sleep in memory, learning, and further brain development. Coverage encompasses the network and membrane mechanisms responsible for waking and sleeping brain activity, the roles of glial cells in the sleeping brain, the molecular basis of sleep EEG rhythms, and research on songbirds, rodents, and humans indicating the function of sleep. - Collates material dispersed across a wide gamut of primary literature into one place - Focuses on the most interesting and prolific research results on brain activity as it relates to sleep - Practical real data discussion includes functional brain imaging and EEG research

Chapter 1

Neuronal Oscillations in the Thalamocortical System during Sleeping and Waking States

Igor Timofeev

Département de Psychiatrie et de Neurosciences, The Centre de Recherche Université Laval Robert-Giffard (CRULRG), Université Laval, Québec, (QC), G1J 2G3, Canada

All normal brain processes occur over three main states of vigilance: wake, slow-wave sleep, and REM sleep. These states can be subdivided further to passive and active wakefulness, three to four stages of slow-wave sleep, and active and passive REM sleep. There are also abnormal states of the brain like paroxysmal seizure activities or brain activities generated under anesthesia conditions. The states of vigilance by themselves originate via interaction of circadian and homeostatic processes (Achermann & Borbely, 2003; Borbely, Baumann, Brandeis, Strauch, & Lehmann, 1981) with a leading role of interactions between suprachiasmstic nucleus and hypothalamic regions (Fuller, Sherman, Pedersen, Saper, & Lu, 2011; Saper, Scammell, & Lu, 2005). Different brain states are expressed as different forms of global electrographic activities recorded from a brain surface (elecrocorticogram), which are reflected on a head surface and recorded as an electroencephalogram (EEG). These global electrical activities are mediated by synchronous synaptic activities of neurons. Synchronous de- or hyperpolarization of neighboring neurons will generate large amplitude global waves, the asynchronous activities of neurons will not generate the global field potential signals at all, and intermediates neuronal synchrony will produce field potential waves of intermediate amplitude. While the states of vigilance by themselves originate in suprachiasmstic and hypothalamic regions and are transmitted to the other brain structures via ascending activating systems (Steriade & McCarley, 2005), the top level of the brain, primarily the thalamocortical system, generates the electrical activities that are characteristic to different states of vigilance.

Neuronal Synchronization

Neuronal synchronization requires some form of interactions between neurons. I will only briefly overview this aspect. To understand the processes of neuronal synchronization, one has to separate local and long-range synchronization. Local synchronization is required to produce simultaneous de- or hyperpolarization of a local group of neurons that mediates generation of local field potentials; and long-range synchronization synchronizes the local field potentials generated at some distance. The amplitude of field potential recordings with one electrode will provide information on levels of local neuronal synchrony. Multisite recordings are used to investigate long-range synchronization.

The most investigated aspect of neuronal interactions are chemical synaptic interactions (Eccles, 1964). Action potentials generated in excitatory neurons will propagate to synaptic boutons and release excitatory neurotransmitter (mainly glutamate within the thalamocortical system). The amplitude of single-axon induced excitatory postsynaptic potentials (EPSPs) is small from 0.1 to several millivolts with overall mean less than 2 mV (Thomson, West, Wang, & Bannister, 2002). This neurotransmitter will exert depolarizing action on targets. Because each axon forms multiple contacts with target neurons it can produce sufficient local field effects if the target neurons are located in proximity. If several neurons excite the same group of target cells nearly simultaneously, the overall postsynaptic effect will definitely be larger due to spatial summation. That will produce summated effects, which are sensed at the field potential level. Multiple excitatory connections are formed by long-axon neurons; therefore they are well positioned to mediate long-range synchrony. Activities of inhibitory neurons within the thalamocortical system release mainly GABA, an inhibitory neurotransmitter. All known cortical interneurons have a short-axon (Markram et al., 2004; Somogyi, Tamás, Lujan, & Buhl, 1998), which contacts multiple target neurons in a local network. Therefore, cortical interneurons can contribute to local, but not long-range synchronization. This is not the case for other inhibitory cells. During development, GABAergic neurons of thalamic reticular (RE) nucleus form variable patterns of connectivity from a compact, focal projection to a widespread, diffuse projection encompassing large areas of Ventro-Basal complex (VB) (Cox, Huguenard, & Prince, 1996). Indirect action of reticular thalamic neurons that exerts diffuse projections onto thalamocotical neurons could likely be detected as synchronous field potential events over some large cortical areas when thalamocortical neurons will fire action potentials. Neuronal constellations outside the thalamocortical system may also to cortical synchronization. A recent study (Eschenko, Magri, Panzeri, & Sara, 2012) has demonstrated that in rats, the locus coreleus neurons fire in phase with cortical slow waves and they even preceded onsets of cortical neuronal firing, suggesting a contribution of locus coreleus not only in setting up general cortical excitability, but also in influencing the cortical synchronization processes.

The next mechanism contributing to neuronal synchronization is electrical coupling between cells that is mediated by gap junctions. The astrocytic network is tightly connected via gap junctions (Mugnaini, 1986). Gap junctions were also found between multiple groups of cortical interneurons (Galarreta & Hestrin, 1999; Gibson, Beierlein, & Connors, 1999). Electrical coupling was demonstrated between neurons of reticular thalamic nucleus (Fuentealba et al., 2004; Landisman et al., 2002). Dye coupling, presence of spikelets, and modeling experiments suggest the existence of electrical coupling between axons of hippocampal pyramidal cells (Schmitz et al., 2001; Vigmond, Perez Velazquez, Valiante, Bardakjian, & Carlen, 1997). A single study has found spikelets, an accepted signature of electrical coupling, in thalamocortical neurons (Hughes, Blethyn, Cope, & Crunelli, 2002). Indirect data on electrical coupling between thalamocortical neurons and axo-axonal coupling are so far not supported by demonstration of the presence of gap junctions. The gap junctions, mediating electrical coupling, form high resistance contacts between connected cells; therefore, they act as low-pass filters (Galarreta & Hestrin, 2001). Confirmed gap junctions are formed within dendritic arbor of connected cells, therefore, they can contribute to the local synchronization only.

Ephaptic interactions constitute another mechanism of neuronal synchronization. Changes in neuronal membrane potential produce extracellular fields that affect the excitability of neighboring cells (Jefferys, 1995). The extracellular fields produced by a single neuron are weak, however, when a local population of neurons generate nearly simultaneous excitation or inhibition, their summated effects can significantly influence the excitability of neighboring neurons, contributing to local synchronization. External electric field applied with intensities comparable to endogenous fields applied to cortical slices modulated cortical slow oscillation (Frohlich & McCormick, 2010). During seizure activity, the extracellular space reduces and the effects of ephaptic interactions increase (Jefferys, 1995).

Neuronal activities are associated with movement of ions across membrane due to activation of ligand- or voltage-controlled channels. Because extracellular space in the brain is about 20% of total brain volume (Syková, 2004; Sykova & Nicholson, 2008) and an activation of ionic pumps, ionic diffusion, etc., is a time-dependent process, the neuronal activities alter extracellular concentration of implicated ions (Somjen, 2002). Changes are temporal and local, but on a short timescale they affect neuronal excitability. For example, during active states of cortical slow oscillation, due to activation of synaptic currents (primarily NMDA receptor-mediated) and Ca²⁺ gated intrinsic channels, the extracellular concentration of Ca²⁺ decreases from 1.2 mM to 1.0 mM (Crochet, Chauvette, Boucetta, & Timofeev, 2005; Massimini & Amzica, 2001). This leads to a dramatic increase in synaptic failure rates (up to 80%) (Crochet, et al., 2005). Dramatic changes of ionic concentrations occur during seizure activity. In these conditions, the extracellular Ca²⁺ drops to below 0.5 mM (Amzica, Massimini, & Manfridi, 2002; Heinemann, Lux, & Gutnick, 1977; Pumain, Kurcewicz, & Louvel, 1983) and extracellular K⁺ increases to 7–18 mM (Amzica et al., 2002; Moody, Futamachi, & Prince, 1974; Prince, Lux, & Neher, 1973). This results in an impairment of chemical synaptic transmission and axonal conduction of action potentials is impaired too (Seigneur & Timofeev, 2010). The reversals potential for ionic currents with implicated ions are shifted affecting overall neuronal excitability. Just some little increase in extracellular K⁺ concentration and a decrease in extracellular Ca²⁺ and Mg²⁺ concentration may induce slow oscillation in neocortical slices (Reig, Gallego, Nowak, & Sanchez-Vives, 2006). Stronger changes in ion concentration occurring within physiological/pathological range have profound influence on brain network activities.

Therefore, synaptic potentials (primarily excitatory) are in a position to mediate long-range neuronal synchronization. Synaptic activities and all the other mechanisms of neuronal synchronization are local and they contribute to short-range neuronal synchronization. Local synchronous activities of neuronal groups may create propagating or travelling waves of activities that can involve large neuronal territories and can be detected as synchronous activities with shifts in the phase of oscillations (Boucetta, Chauvette, Bazhenov, & Timofeev, 2008; Massimini, Huber, Ferrarelli, Hill, & Tononi, 2004).

Basic Thalamocortical States

There are two basic states of thalamocortical network: silent and active.

During silent states all active conductances (intrinsic and synaptic) are inactive. Basically, during silent states the membrane potential of neurons is mediated by leak current only. This is a theoretical situation and it is unlikely that it occurs in real life. In practice, some inactivating intrinsic currents, for example inward rectifying current, can also be active. Activities of these currents preset so-called resting membrane potential of neurons. The resting membrane potential of cortical neurons is situated somewhere between −70 mV and −80 mV. Given that equilibrium potential for leak current is around −95 mV (McCormick, 1999), an activation of inward rectifying currents appears to play a role in the mediation of the resting membrane potential. The resting membrane potential can be recorded from neurons in nonoscillating slices maintained in vitro; it can also be recorded from cortical neurons during silent phases of sleep slow oscillation (see below). Because very few current are active during network silent states, the input resistance of neurons during silent states is higher as compared to active states (Contreras, Tim...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Preface
• List of Contributors
• Chapter 1. Neuronal Oscillations in the Thalamocortical System during Sleeping and Waking States
• Chapter 2. Corticothalamic Rhythms during States of Reduced Vigilance
• Chapter 3. Glial Modulation of Sleep and Electroencephalographic Rhythms
• Chapter 4. Genetic Mechanisms Underlying Rhythmic EEG Activity during Sleep
• Chapter 5. Evoked Electrophysiological and Vascular Responses across Sleep
• Chapter 6. Sleep and Learning in Birds
• Chapter 7. Phasic Pontine-Wave (P-Wave) Generation
• Chapter 8. Neural Correlates of Human Sleep and Sleep-Dependent Memory Processing
• Chapter 9. Sleep EEG Rhythms and System Consolidation of Memory
• Chapter 10. Sleep Slow Oscillations and Cortical Maturation
• Index