Rationale:
GRIN2B mutations have been identified in individuals with severe neurodevelopmental disorders resulting in epilepsy, intellectual disability and severe autism. Seizure types in patients include focal, tonic clonic and absence. GRIN2B encodes the GluN2B subunit of the N-methyl-D-aspartate (NMDA) receptor and dominates subunit expression during late embryogenesis and early postnatal development, playing a fundamental role in circuit formation. Understanding how GRIN2B mutations in animal models result in altered physiological mechanisms, could lead to novel therapeutic strategies.
Methods:
We implanted GRIN2B heterozygous deletion (
Grin2B-/+) Long Evans rats and wild-type littermate controls (
Grin2B+/+) with 14-channel skull-surface EEG probes and recorded continuously for 72 hrs. We identified and quantified spike and wave discharges (SWDs), with oscillatory frequencies between 5 and 10 Hz, which are the electrophysiological correlate of absence seizures. We assessed the sleep-wake cycle, and we calculated spectral properties and correlation between electrodes using imaginary coherence. We also attempted SWD blockade with the t-type calcium channel blocker, ethosuximide (100 mg/kg), and the NMDA receptor antagonist, memantine (5 mg/kg).
Results:
Over the 24 hr period, Grin2b
+/- rats had more SWDs (Mann-Whitney U test, p = 0.0002), Grin2b
+/- rats had a longer average SWD duration (Mann-Whitney U test, p = 0.0009), and the total SWD time was also longer when compared to Grin2b
+/+ rats (Mann-Whitney U test, p = 0.0003). Treatment with ethosuximide significantly reduced the number of SWDs (Mixed effects ANOVA, p = 0.031), while both ethosuximide and memantine significantly reduced the average duration of SWDs (Mixed effects ANOVA, ethosuximide p < 0.001, Memantine p = 0.001) and total duration of SWDs (Mixed effects ANOVA, ethosuximide p = 0.002, memantine p = 0.006).
We find Grin2b+/- spent equivalent amounts of time awake and in NREM sleep compared to Grin2b+/+ littermates (Two-sample unpaired t-test, p = 0.26; p = 0.18). However, Grin2b+/- rats had altered sleep-wake distributions, whereby they had significantly more wake bouts (Mann-Whitney U test, p = 0.008) that were shorter in duration (Welch two-sample t-test, p = 0.027), and more NREM bouts (Mann-Whitney U test, p = 0.0086) that were of similar length to Grin2b+/+ rats (Two-sample unpaired t-test, p = 0.14). Notably, Grin2b+/- animals spent less time in REM sleep (Two-sample unpaired t-test, p = 0.002), which was due to fewer REM sleep bouts (Welch two-sample t-test, p = 0.0005) that were similar in duration for both groups (Two-sample unpaired t-test, p = 0.48).
Grin2b+/- rats also displayed significantly increased spectral power across frequency bands (Two-way ANOVA, p = 0.02) and imaginary coherence (Two-way ANOVA, p = 0.04) when compared to
Grin2b+/+ animals.
Conclusions:
These data show Grin2B-/+ rats have an increased propensity for absence seizures and altered sleep, which may be clinically relevant phenotypes for studying the underlying mechanisms and developing new therapeutics for GRIN2B encephalopathy.
Funding: Simons Initiative for the Developing Brain