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
. 2015 May:77:141-54.
doi: 10.1016/j.nbd.2015.02.016. Epub 2015 Mar 10.

Sleep impairment and reduced interneuron excitability in a mouse model of Dravet Syndrome

Franck Kalume  1 , John C Oakley  2 , Ruth E Westenbroek  3 , Jennifer Gile  4 , Horacio O de la Iglesia  5 , Todd Scheuer  3 , William A Catterall  6
Affiliations

Sleep impairment and reduced interneuron excitability in a mouse model of Dravet Syndrome

Franck Kalume et al. Neurobiol Dis. 2015 May.

Abstract

Dravet Syndrome (DS) is caused by heterozygous loss-of-function mutations in voltage-gated sodium channel NaV1.1. Our mouse genetic model of DS recapitulates its severe seizures and premature death. Sleep disturbance is common in DS, but its mechanism is unknown. Electroencephalographic studies revealed abnormal sleep in DS mice, including reduced delta wave power, reduced sleep spindles, increased brief wakes, and numerous interictal spikes in Non-Rapid-Eye-Movement sleep. Theta power was reduced in Rapid-Eye-Movement sleep. Mice with NaV1.1 deleted specifically in forebrain interneurons exhibited similar sleep pathology to DS mice, but without changes in circadian rhythm. Sleep architecture depends on oscillatory activity in the thalamocortical network generated by excitatory neurons in the ventrobasal nucleus (VBN) of the thalamus and inhibitory GABAergic neurons in the reticular nucleus of the thalamus (RNT). Whole-cell NaV current was reduced in GABAergic RNT neurons but not in VBN neurons. Rebound firing of action potentials following hyperpolarization, the signature firing pattern of RNT neurons during sleep, was also reduced. These results demonstrate imbalance of excitatory vs. inhibitory neurons in this circuit. As predicted from this functional impairment, we found substantial deficit in homeostatic rebound of slow wave activity following sleep deprivation. Although sleep disorders in epilepsies have been attributed to anti-epileptic drugs, our results show that sleep disorder in DS mice arises from loss of NaV1.1 channels in forebrain GABAergic interneurons without drug treatment. Impairment of NaV currents and excitability of GABAergic RNT neurons are correlated with impaired sleep quality and homeostasis in these mice.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest: None

Figures

Fig. 1
Fig. 1
Unaltered total sleep duration in WT and DS mice. A, Examples of EEG-EMG recordings from a WT mouse and a hypnogram generated for sleep (S)/wake (W) scoring. EEG trace labels correspond to electrode positions illustrated in B. B, Drawing of a mouse skull illustrating the sites of EEG electrode placement. C, Magnified segments of wake and sleep EEG-EMG traces in the box in panel A. Scale bars: 1 s, 500 μV. D, Average total sleep duration for WT and DS mice.
Fig. 2
Fig. 2
Less defined sleep EEG waveforms in DS mice than WT mice. Examples of wake, NREM sleep, and REM sleep EEG-EMG from WT (Left) and DS (Right) mice illustrating the less prominent and less sustained EEG delta waves during NREM sleep in a DS mouse, and the similarly less defined EEG theta waves during REM sleep of the same mouse. The stereotypic mouse behaviors for the EEG traces are characterized by activity in the associated EMG traces and depicted in adjacent photographs extracted from videos of experiments with the WT mouse. EEG trace labels correspond to electrode positions in Fig. 1B.
Fig. 3
Fig. 3
Decreased EEG power densities in DS mice during sleep. A, Unaltered mean power density profiles for combined wake and sleep states for WT and DS mice. B, Mean NREM sleep power density profiles and power ratios for DS and WT mice. (Left) Power density spectra for DS and WT mice. Inset shows mean ± s.e.m. of the integral of power in the delta frequency range. (Right) Ratio of DS/WT power density versus frequency. C, Mean REM sleep power density profiles and power ratios for DS and WT mice. (Left) Mean REM sleep power density profiles for DS and WT mice. (Right) Ratios of HET/WT power density in the theta frequency range.
Fig. 4
Fig. 4
Decreased incidence of sleep spindles during NREM sleep in DS mice. A, Examples of sleep spindles in WT and DS mice. EEG-EMG trace labels correspond to electrode positions in Fig. 1B. Spindles are marked by black bar above EEG traces. B, Average number of spindles per 8-h recording period for WT and DS mice.
Fig. 5
Fig. 5
Increased incidence of brief wakes and interictal spikes during sleep in DS mice. A, Examples of NREM sleep EEG illustrating brief wakes observed in both WT and DS records. EEG-EMG trace labels correspond to electrode positions in Fig. 1B. B, Mean ± s.e.m. of brief wakes in NREM and REM sleep per 8-h recording period. C, Mean ± s.e.m. of spikes in Wake, NREM, and REM sleep per 8-h recording period.
Fig. 6
Fig. 6
Expression of NaV1.1, NaV1.2, NaV1.3, and NaV1.6 channels in the reticular nucleus of the thalamus of P14 WT, HET, and HOMO knockout NaV1.1 mice. A1, B1, C1, Staining with anti-NaV1.1 antibody showed labeling of cells bodies in the reticular nucleus of WT and HET but not HOMO knockout mice. A2, B2, C2, Corresponding GAD 67 staining and A3, B3, C3, merged images of NaV1.1 and GAD 67 staining. D1, E1, F1, Staining with anti-NaV1.2 antibody illustrating staining throughout the RNT region that did not differ between WT and HET or HOMO knockout animals. D2, E2, F2, Corresponding GAD 67 staining and D3, E3, F3, merged images of NaV1.2 and GAD 67 staining. G1, H1, I1, Staining with anti-NaV1.3 antibody demonstrating labeling of cell bodies in the reticular nucleus of WT, HET and HOMO knockout animals. G2, H2, I2, Corresponding GAD 67 staining and G3, H3, I3, merged images for NaV1.3 and GAD 67 staining. J1, K1, L1, Staining with anti-NaV1.6 showing expression of this channel in cell bodies and dendrites. J2, K2, L2, Corresponding GAD 67 staining and J3, K3, L3, merged images of NaV1.6 and GAD 67 staining.
Fig. 7
Fig. 7
Decreased sodium currents in inhibitory but not excitatory thalamic neurons. A, (Left) Micrograph of an RNT neuron. Scale bar: 20 μm. (Right) Representative sodium current traces, from a RNT neuron, evoked with 50-ms depolarizations from a holding potential of −90 mV to + 30 mV in 5-mV increments. For clarity, only the first 5 ms of the traces are illustrated. B, Mean current-voltage relationships recorded from RNT neurons for WT (filled circles), HET (DS, filled squares) and HOMO knockout (filled triangles) mice. C, Voltage-dependence of activation of the sodium conductance in RNT neurons of NaV1.1 WT, HET (DS), and HOMO knockout mice. The half-activation voltage for sodium current in cells of WT animals was −36.1.3 ± 0.6 mV (k= 4.4 ± 0.6 mV, n= 9, p> 0.05) compared to −34.0 ± 0.3 (k= 5.6 ± 0.3, n= 16) and −29.6 ± 1 (k=6.3 ± 0.6, n=11) in cells of HET and HOMO knockout animals, respectively. D, Voltage dependence of sodium current inactivation in GABAergic RNT neurons measured using a 20-ms test pulse to 0 mV following a 100-ms prepulse to a variable potential (5-mV increments). The half-inactivation voltages were −49.4 ± 0.5 mV (k= 5.3 ± 0.2, n=8, p> 0.05), −51.9 ± 0.2 (k= 5.4 ± 0.2, n= 16), and −50.8 ± 0.4 (k= 6.0 ± 0.4, n= 11) for cells of WT, HET (DS), and HOMO knockout mice, respectively. E, (Left) Micrograph of an excitatory thalamocortical neuron of the ventrobasal nucleus of the thalamus. Scale bar: 20 μm. (Right) Representative sodium current traces evoked in a thalamocortical neuron using a family of 50-ms depolarizations from −90 to 0 mV in 5 mV increments. F, Mean sodium current vs. voltage relationships from thalamocortical neurons of WT, HET (DS), and HOMO Knockout mice. G, Voltage-dependence of activation of the sodium conductance in thalamocortical neurons from NaV1.1 WT, HET (DS), and HOMO knock-out mice. H, Voltage-dependence of sodium current inactivation in thalamocortical neurons.
Fig. 8
Fig. 8
Reduced rebound firing of RNT neurons in DS mice. A, Examples of rebound bursts of action potentials from WT, HET (DS), and HOMO knockout neurons for hyperpolarization with −80 pA current. B, Mean number of action potentials (+/− s.e.m) during the rebound plotted against the amount of hyperpolarizing current applied (bottom axis) and voltage reached at end of hyperpolarization (top axis).
Fig. 9
Fig. 9
Reduced EEG power density during rebound from sleep deprivation in DS mice. A, Power density spectra for NREM sleep in WT mice after 5-h normal sleep (Control) and after 5-h sleep deprivation (SD). B, Power density spectra for REM sleep in DS mice after 5 h normal sleep (Control) and after 5-h sleep deprivation (SD). Boxes in A and B indicate the delta and theta frequency bands respectively. C, Rebound REM sleep EEG activity for DS mice after 5h of normal sleep (Control) and after 5h of sleep deprivation (SD). D, Rebound REM sleep EEG activity for DS mice after 5 h of sleep deprivation (SD). Boxes in C and D indicate the delta and theta frequency bands respectively. The power density measured under control conditions in these experiments is less than observed in Fig. 3 because here we recorded EEG for 3 h, beginning 5 h after the onset of the light period for homeostatic sleep regulation, whereas we recorded EEGs during 8 h of sleep from the onset of the light period for sleep architecture studies (See Materials and Methods). E and F, Decreased mean NREM (E) and REM (F) sleep EEG power density in conditional forebrain interneuron specific Scn1a KO mice. Boxes in E and F indicate the delta and theta frequency bands respectively
Fig. 10
Fig. 10
Mild abnormalities of circadian behavioral phenotypes in forebrain interneuron specific Scn1a KO mice. A, Wheel-running activity in a representative control mouse (left) and conditional HET mouse (right). Representative actograms of wheel-running activity display wheel revolutions as black bars on 48-h periods. White and gray areas represent the light and dark period, respectively. Light pulses of 30 min duration were applied to the mouse cage at CT 16 (white circle) and control pulses were similarly applied (dark circle). B, Phase angle and Period of the locomotor activity. C, Mild reduction of activity in the conditional KO mice compared to controls under LD condition (39.2 ± 3.6 in Hetflx/DlxCre+ wheel turns/10 min, n=5 vs. 50.3 ± 1.6 Hetflx/DlxCre− mice, n=5, p< 0.001) and DD condition (51.3 ± 4.5 in Hetflx/DlxCre+ wheel turns/10 min, n=5 vs. 68.6 ± 7.0 Hetflx/DlxCre− mice, n=5, p< 0.001). D, No significant effects of light pulses on conditional HET mice circadian phase shifting.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Anderson MP, et al. Thalamic CaV3.1 T-type Ca2+ channel plays a crucial role in stabilizing sleep. Proc Natl Acad Sci U S A. 2005;102:1743–8. - PMC - PubMed
    1. Arzimanoglou A. Dravet syndrome: from electroclinical characteristics to molecular biology. Epilepsia. 2009;50(Suppl 8):3–9. - PubMed
    1. Asano E, et al. Effect of sleep on interictal spikes and distribution of sleep spindles on electrocorticography in children with focal epilepsy. Clin Neurophysiol. 2007;118:1360–8. - PMC - PubMed
    1. Bazil CW. Epilepsy and sleep disturbance. Epilepsy Behav. 2003;4(Suppl 2):S39–45. - PubMed
    1. Beenhakker MP, Huguenard JR. Neurons that fire together also conspire together: is normal sleep circuitry hijacked to generate epilepsy? Neuron. 2009;62:612–32. - PMC - PubMed

Publication types

MeSH terms

Substances