Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Mar;30(3):e14206.
doi: 10.1111/cns.14206. Epub 2023 Apr 18.

Sleep waves in a large-scale corticothalamic model constrained by activities intrinsic to neocortical networks and single thalamic neurons

Martynas Dervinis  1 , Vincenzo Crunelli  1
Affiliations

Sleep waves in a large-scale corticothalamic model constrained by activities intrinsic to neocortical networks and single thalamic neurons

Martynas Dervinis et al. CNS Neurosci Ther. 2024 Mar.

Abstract

Aim: Many biophysical and non-biophysical models have been able to reproduce the corticothalamic activities underlying different EEG sleep rhythms but none of them included the known ability of neocortical networks and single thalamic neurons to generate some of these waves intrinsically.

Methods: We built a large-scale corticothalamic model with a high fidelity in anatomical connectivity consisting of a single cortical column and first- and higher-order thalamic nuclei. The model is constrained by different neocortical excitatory and inhibitory neuronal populations eliciting slow (<1 Hz) oscillations and by thalamic neurons generating sleep waves when isolated from the neocortex.

Results: Our model faithfully reproduces all EEG sleep waves and the transition from a desynchronized EEG to spindles, slow (<1 Hz) oscillations, and delta waves by progressively increasing neuronal membrane hyperpolarization as it occurs in the intact brain. Moreover, our model shows that slow (<1 Hz) waves most often start in a small assembly of thalamocortical neurons though they can also originate in cortical layer 5. Moreover, the input of thalamocortical neurons increases the frequency of EEG slow (<1 Hz) waves compared to those generated by isolated cortical networks.

Conclusion: Our simulations challenge current mechanistic understanding of the temporal dynamics of sleep wave generation and suggest testable predictions.

Keywords: biophysical model; delta waves; sleep spindles; slow (<1 Hz) waves; thalamic reticular nucleus.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

FIGURE 1
FIGURE 1
Slow (<1 Hz) and delta oscillations in the isolated neocortical model. EEG (top trace in A, C, F, I, L) and corresponding color‐coded membrane potential graphs for the indicated neuronal populations (bottom traces in A, C, F, I, L), EEG autocorrelograms (B, D, G, J, M) and EEG power graphs (E, H, K) of the indicated activity. (A, B), with a large gKL the cortical model does not elicit any activity. (C–E), Decreasing gKL of ND and EF cells results in spontaneous slow (<1 Hz) oscillations in the entire cortical network. (F–H), Further reducing gKL of ND cells leads to delta waves in the isolated cortical model. (I–K), When gKL of other cortical neurons (except ND and EF neurons) is slightly decreased a non‐regular activity pattern is observed, that is, there is a break‐down of the oscillations. (L, M), The network shows a desynchronized EEG and tonic firing in all neuronal populations when gKL of all cell types (except ND cells) is further decreased. L4 PY, pyramidal neurons in cortical layer 4; L4 IN, interneurons in cortical layer 4; L5 PY, pyramidal neurons in cortical layer 5; L5 IN, interneurons in cortical layer 5; L6 PY, pyramidal neurons in cortical layer 6; L6 IN, interneurons in cortical layer 6; L2/3 PY, pyramidal neurons in cortical layers 2 and 3; L2/3 IN, interneurons in cortical layers 2 and 3.
FIGURE 2
FIGURE 2
Experimental and simulated membrane potential dynamics in cortical neurons during slow (<1 Hz) oscillations in the isolated cortical network. (A, B), Simultaneous local field potential and membrane potential dynamics of an RS neuron recorded in vitro (experiment) during slow (<1 Hz) oscillations and its simulated activity in the isolated cortical network (simulation). The left‐hand traces show the initial stage of the oscillation while the right‐hand traces show the oscillation at a later stage (increased neuromodulatory drive). (C, D), Normalized membrane potential distribution plots for the experimental and simulated slow (<1 Hz) oscillations of an RS neuron with the typical peaks of the up‐ and down‐states. (E, F), Same as (C, D) but for an EF neuron. (G, H), Same as (A, B) for an EF neuron. (I, J), Same as (A, B) but for an FS neuron. (K, L), Same as (C, D) but for an FS neuron. Note the lack of a clear bimodal membrane potential distribution in the plots of both the experimental and simulated data of the FS neurons. (M, N), Same as (C, D) but for an ND neuron. (O, P), Same as (A, B) but for an ND neuron. Experimental data are reproduced with permission from Lorincz et al. (2015).
FIGURE 3
FIGURE 3
Experimental and simulated membrane potential dynamics of TCFO neurons during intrinsically generated slow (<1 Hz) oscillations. (A), Tonic firing (top trace), quiescence (middle trace), and delta oscillations (bottom trace) are observed at different level of membrane polarization (injected current at the bottom of each trace) in a cat ventrobasal TC neuron recorded in a thalamic slice maintained in a standard recording solution. (B), After application of 100 μM trans‐ACPD (a metabotropic glutamate receptor, mGluR, agonist) the same neuron shows slow (<1 Hz) oscillations between the quiescent state and the delta oscillations (bottom trace). (C), Simulation in the TCFO neuron model of the experimental activity shown in (A). (D), Simulation of experimental activity shown in (B) was obtained after reducing gKL of the TCFO neuron model to mimic the effect of mGluR activation in vitro. Dashed lines on the left of each trace indicate −60 mV. (A) and (B) are reproduced with permission from Zhu et al. (2006).
FIGURE 4
FIGURE 4
Slow (<1 Hz) oscillations in the full corticothalamic model. (A), EEG showing the rhythmic pattern of slow (<1 Hz) oscillations. (B), EEG (top trace) and color‐coded membrane potential plots of the indicated cortical and thalamic neuronal populations during the five cycles of the slow (<1 Hz) oscillation highlighted in (A) (note the two separate color‐scales for the cortical and thalamic neurons). Below are the corresponding membrane potential waveforms of the two neurons indicated by the red arrow on the left of the corresponding color‐coded plots. (C), EEG (top trace), AP rastergrams of the firing in each neuronal population for the slow (<1 Hz) oscillation cycle highlighted in (B). Red dashed vertical line represents the first AP of the up‐state in each population. The latency (indicated in red below each rastergram) is measured relatively to the first AP of the cycle in the TCFO neuron that fires first (time zero). Below the rastergrams are the corresponding membrane potential waveforms of that cycle for the indicated neuron. (D), Cross‐correlations of EEG and membrane potential (black trace) and EEG and APs (color trace) for the indicated neuronal populations, calculated over a 485 s‐long simulation. Shaded regions are 95% confidence intervals. (E), Enlargement of the highlighted sections of the membrane potential color plot of the TCFO neurons. L4 FS, FS neuron in cortical layer 4; L4 IB, IB neuron in cortical layer 4; L4 IN, interneurons in cortical layer 4; L4 PY, pyramidal neurons in cortical layer 4; L5 FS, FS neuron in cortical layer 5; L5 IN, interneurons in cortical layer 5; L5 PY, pyramidal neurons in cortical layer 5; L5 RS, RS neuron in cortical layer 5; NRTFO, first‐order NRT neurons; NRTHO, higher‐order NRT neurons; TCFO, first‐order TC neurons; TCHO, higher‐order TC neurons.
FIGURE 5
FIGURE 5
Delta waves in the full corticothalamic model. (A), EEG showing a period of rhythmic delta waves. (B), EEG (top trace) and color‐coded membrane potential plots of the indicated cortical and thalamic neuronal populations during the section of delta waves highlighted in (A) (note the two separate color‐scales for the cortical and thalamic neurons). Below are the corresponding membrane potential waveforms of the two neurons indicated by the red arrow on the left of the corresponding color‐coded plots. (C), EEG (top trace), AP rastergrams of the firing of each neuronal population for the delta wave cycle highlighted in (B). Red dashed vertical line represents the first AP of the up‐state in each population. The latency (indicated in red below each rastergram) is measured relatively to the first AP of the cycle in the TCFO neuron that the fires first (time zero). Below the rastergrams are the corresponding membrane potential waveforms of that cycle for the indicated neuron. (D), Cross‐correlations of EEG and membrane potential (black trace) and EEG and APs (color trace) for the indicated neuronal populations, calculated over a 485 s‐long simulation. Shaded regions are 95% confidence intervals. Dashed vertical line indicates zero lag. (E), Distribution of the first AP in a delta wave cycle with respect to the EEG for all APs of the indicated neuronal populations. Shaded regions are 95% confidence intervals. Dashed vertical line indicates zero lag. L4 FS, FS neuron in cortical layer 4; L4 IN, interneurons in cortical layer 4; L4 PY, pyramidal neurons in cortical layer 4; L4 RS, RS neuron in cortical layer 4; L5 FS, FS neuron in cortical layer 5; L5 IB, IB neuron in cortical layer 5; L5 IN, interneurons in cortical layer 5; L5 PY, pyramidal neurons in cortical layer 5; NRTFO, first‐order NRT neurons; NRTHO, higher‐order NRT neurons; TCFO, first‐order TC neurons; TCHO, higher‐order TC neurons.
FIGURE 6
FIGURE 6
Sleep spindles in the full corticothalamic model. (A), EEG showing the rhythmic pattern of sleep spindles. (B), EEG (top trace) and color‐coded membrane potential plots of the indicated cortical and thalamic neuron populations during the sleep spindle highlighted in (A) (note the two separate color‐coded scales for the cortex and the thalamus). Below are the corresponding membrane potential waveforms of the two neurons indicated by the red arrow on the left of the corresponding color‐coded plots. (C), EEG (top trace), AP rastergrams of the firing of the first AP in each neuronal population for the onset of sleep spindles highlighted in (B). Red dashed vertical line represents the first AP of the up‐state in each population. The latency (indicated below each rastergram) is measured relatively to the first AP of the cycle in an NRTFO neuron (time zero). Below the rastergrams are the corresponding membrane potential waveforms of that cycle for the indicated neuron. (D), Cross‐correlations of EEG and membrane potential (black trace) and EEG and APs (color trace) for the indicated neuronal populations, calculated over a 485 s‐long simulation. Shaded regions are 95% confidence intervals. Dashed vertical line indicates the zero lag. (E), AP distribution with respect to the EEG for all APs of the indicated neuronal populations. Shaded regions are 95% confidence intervals. Dashed vertical line indicates the zero lag. L4 FS, FS neuron in cortical layer 4; L4 IB, IB neuron in cortical layer 4; L4 IN, interneurons in cortical layer 4; L4 PY, pyramidal neurons in cortical layer 4; L5 FS, FS neuron in cortical layer 5; L5 IN, interneurons in cortical layer 5; L5 PY, pyramidal neurons in cortical layer 5; L5 RS, RS neuron in cortical layer 5; NRTFO, first‐order NRT neurons; NRTHO, higher‐order NRT neurons; TCFO, first‐order TC neurons; TCHO, higher‐order TC neurons.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Contreras D, Steriade M. Spindle oscillation in cats: the role of corticothalamic feedback in a thalamically generated rhythm. J Physiol. 1996;490:159‐179. - PMC - PubMed
    1. Destexhe A, Contreras D, Steriade M. Spatiotemporal analysis of local field potentials and unit discharges in cat cerebral cortex during natural wake and sleep states. J Neurosci. 1999;19:4595‐4608. - PMC - PubMed
    1. Steriade M, Timofeev I, Grenier F. Natural waking and sleep states: a view from inside neocortical neurons. J Neurophysiol. 2001;85:1969‐1985. - PubMed
    1. Dang‐Vu TT, Schabus M, Desseilles M, et al. Spontaneous neural activity during human slow wave sleep. Proc Natl Acad Sci USA. 2008;105:15160‐15165. - PMC - PubMed
    1. Vyazovskiy V, Olcese U, Lazimy YM, et al. Cortical firing and sleep homeostasis. Neuron. 2009;63:865‐878. - PMC - PubMed

Publication types