The thalamocortical network as a single slow wave-generating unit

https://doi.org/10.1016/j.conb.2014.09.001Get rights and content

During non-REM sleep the EEG is dominated by slow waves which result from synchronized UP and DOWN states in the component neurons of the thalamocortical network. This review focuses on four areas of recent progress in our understanding of these events. Thus, it has now been conclusively demonstrated that the full expression of slow waves, both of natural sleep and anesthesia, requires an essential contribution by the thalamus. Furthermore, the modulatory role of brainstem transmitters, the function of cortical inhibition and the relative contribution of single neocortical neurons to EEG slow waves have started to be carefully investigated. Together, these new data confirm the view that a full understanding of slow waves can only be achieved by considering the thalamocortical network as a single functional and dynamic unit for the generation of this key EEG rhythm.

Introduction

Slow waves are one of the most ubiquitous features of the EEG in mammals and have been a focus of neuroscience research for many decades. Whilst slow waves are inextricably linked with sleep physiology, dominating the EEG during the deeper stages of non-NREM sleep (i.e. slow wave sleep), they are also present during certain types of anesthesia and in response to sensory stimuli. As such, the relationship between slow waves and states of sleep and consciousness is a complex one that necessitates a detailed understanding at the cellular and network level of the mechanisms that lead to slow waves and their neuronal counterparts, that is UP and DOWN states. To this end, over the last 30 years an array of in vivo and in vitro electrophysiological studies has shed significant light on the intrinsic and synaptic events that lead to slow waves [1, 2, 3]. Such studies have naturally focused on thalamocortical interactions and have highlighted central roles for both the neocortex and thalamus in shaping the overall manifestation of slow waves. Indeed, the requirement for a particular architecture and connectivity capable of generating slow waves becomes evident if one considers non-neocortical areas such as the piriform cortex which has a laminar structure that is distinct from the neocortex, lacks thalamic input and neither generates typical slow waves [4, 5] or exhibits coherence with neocortical territories in low frequency EEG bands in vivo [5]. However, in the intact brain, several other regions have their activity phase-locked to EEG slow waves, including the hippocampus [6], striatum [7] and cerebellum [8] as well as brainstem nuclei that are the main sources of modulatory transmitters such as the locus coeruleus for noradrenaline [9], the dorsal raphe for serotonin [10] and the pedonculopontine nucleus for acetylcholine [11].

Given that slow waves are also difficult to fully characterize due to their multiple forms of intrinsic expression (i.e. waveform, frequency and amplitude) and contextual manifestation (i.e. in vivo versus in vitro models and anesthesia versus natural sleep), the identification of the brain regions and mechanisms responsible for initiating, propagating, terminating and modulating slow waves remains challenging at the experimental, computational and theoretical levels. Notwithstanding this, important progress has been made in recent years, and this review will bring together recent advances in our understanding of the generation of slow wave and UP and DOWN states with the aim of developing a concise and coherent framework for explaining how this brain rhythm develops during natural sleep and anesthesia. For the sake of clarity, in the remainder of this article the use of the word ‘generation’ refers to the fundamental creation of the overall, coordinated oscillation or rhythm, as opposed to the shaping by individual current sources of the specific LFP or EEG signal in different regions.

Section snippets

The thalamus is as essential as the cortex for the full expression of slow waves

One area that has often been proposed to play a key role in slow wave generation is the thalamus [12]. Four main lines of evidence support this assertion. First, the firing of thalamocortical (TC) neurons is tightly associated with EEG slow waves. Together with the strong afferent and efferent connections with neocortical layers [13, 14, 15] that are involved in slow waves [16, 17••, 18], this suggests that thalamic nuclei can control UP and DOWN state dynamics in neocortical circuits. Second,

Intrinsic and network mechanisms of UP and DOWN states in the neocortex

Within this framework of the thalamocortical circuit being a unified slow wave-generating unit, it still remains of great significance to fully understand the intrinsic and/or synaptic origin of UP and DOWN states in the different thalamic and cortical neuronal populations. While the ability of NRT and TC neurons in both sensory and intralaminar thalamic nuclei to generate intrinsic rhythmic UP and DOWN states was established a few years ago [22, 23] (and extensively reviewed in [12]), some

Neocortical inhibitory mechanisms involved in slow waves

Although slow waves in the neocortex are widely accepted to be mainly due to a finely tuned dynamic balance between excitation and inhibition, it is somewhat surprising that, relatively fewer studies have investigated in detail the inhibitory mechanisms involved in this brain rhythm. Nevertheless, a recent study in ferret slices has shown that GABA-A receptor-mediated fast inhibition is crucial for maintaining the appropriate balance of persistent excitatory and inhibitory synaptic activity

Neuromodulation of slow waves

As mentioned earlier, rhythmic firing that is phase-locked to EEG slow waves and UP and DOWN state dynamics occur not only in cortex and thalamus but also in many other brain regions, which interestingly include those brainstem nuclei that are the main source of modulatory transmitters such as noradrenaline [9], serotonin [10] and acetylcholine [11]. In view of the known role that these transmitters play in regulation of brain states [56, 57, 58] and their effects on slow waves as described in

Conclusions

Over the last few years a number of important and conclusive studies has significantly furthered our understanding of slow wave generation. Thus, the widely accepted view that the slow waves observed in natural sleep and during anesthesia are generated entirely within neocortical territories is no longer tenable: neither the isolated cortex (nor the isolated thalamus) can express identical slow waves to those observed in vivo but their full expression requires an intact thalamocortical network

Conflict of interest statement

Nothing declared.

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

Acknowledgements

Our work in this topic is supported by the Wellcome Trust (grant 91882 to VC), Hungarian Scientific Research Fund (OTKA NF 105083 to MLL), National Brain Research Programme (grant KTIA_NAP_13-2-2014-0014 to MLL) and Human Frontier Science Program fellowship (LT001009/2010L to MLL).

References (65)

  • M. Steriade et al.

    The slow (<1 Hz) oscillation in reticular thalamic and thalamocortical neurons: scenario of sleep rhythm generation in interacting thalamic and neocortical networks

    J Neurosci

    (1993)
  • M. Steriade et al.

    Intracellular analysis of relations between the slow (<1 Hz) neocortical oscillation and other sleep rhythms of the electroencephalogram

    J Neurosci

    (1993)
  • M.V. Sanchez-Vives et al.

    Rhythmic spontaneous activity in the piriform cortex

    Cerebral Cortex

    (2008)
  • H. Manabe et al.

    Olfactory cortex generates synchronized top-down inputs to the olfactory bulb during slow-wave sleep

    J Neurosci

    (2011)
  • E.A. Stern et al.

    Spontaneous subthreshold membrane potential fluctuations and action potential variability of rat corticostriatal and striatal neurons in vivo

    J Neurophysiol

    (1997)
  • H. Ros et al.

    Neocortical networks entrain neuronal circuits in cerebellar cortex

    J Neurosci

    (2009)
  • O. Eschenko et al.

    Noradrenergic neurons of the locus coeruleus are phase locked to cortical up-down states during sleep

    Cerebral Cortex

    (2012)
  • J. Mena-Segovia et al.

    Cholinergic brainstem neurons modulate cortical gamma activity during slow oscillations

    J Physiol

    (2008)
  • V. Crunelli et al.

    The slow (1 Hz) rhythm of non-REM sleep: a dialogue between three cardinal oscillators

    Nat Neurosci

    (2010)
  • R.M. Bruno et al.

    Cortex is driven by weak but synchronously active thalamocortical synapses

    Science

    (2006)
  • E.G. Jones

    Synchrony in the interconnected circuitry of the thalamus and cerebral cortex

    Ann N Y Acad Sci

    (2009)
  • C.M. Constantinople et al.

    Deep cortical layers are activated directly by thalamus

    Science

    (2013)
  • M.V. Sanchez-Vives et al.

    Cellular and network mechanisms of rhythmic recurrent activity in neocortex

    Nat Neurosci

    (2000)
  • R. Beltramo et al.

    Layer-specific excitatory circuits differentially control recurrent network dynamics in the neocortex

    Nat Neurosci

    (2013)
  • R. Cossart et al.

    Attractor dynamics of network UP states in the neocortex

    Nature

    (2003)
  • A. Slézia et al.

    Phase advancement and nucleus-specific timing of thalamocortical activity during slow cortical oscillation

    J Neurosci

    (2011)
  • M. Ushimaru et al.

    Differentiated participation of thalamocortical subnetworks in slow/spindle waves and desynchronization

    J Neurosci

    (2012)
  • M. Sheroziya et al.

    Global intracellular slow-wave dynamics of the thalamocortical system

    J Neurosci

    (2014)
  • S.W. Hughes et al.

    Cellular mechanisms of the slow (<1 Hz) oscillation in thalamocortical neurons in vitro

    Neuron

    (2002)
  • K.L. Blethyn et al.

    Neuronal basis of the slow (<1 Hz) oscillation in neurons of the nucleus reticularis thalami in vitro

    J Neurosci

    (2006)
  • J.F.A. Poulet et al.

    Thalamic control of cortical states

    Nat Neurosci

    (2012)
  • E.F. Civillico et al.

    Spatiotemporal properties of sensory responses in vivo are strongly dependent on network context

    Front Syst Neurosci

    (2012)
  • Cited by (130)

    View all citing articles on Scopus
    4

    Equal contribution.

    View full text