Ataxia and cerebellar hypoexcitability in a mouse model of SCN1B-linked Dravet syndrome

Abstract

Patients with Dravet syndrome (DS) present with severe, spontaneous seizures and ataxia. While most patients with DS have variants in the sodium channel Nav1.1 α subunit gene, SCN1A, variants in the sodium channel β1 subunit gene, SCN1B, are also linked to DS. Scn1b null mice model DS, with spontaneous generalized seizures that start in the second week of life. In Scn1b null cerebellum, neuronal pathfinding is severely altered, and Purkinje cells (PCs) and granule neurons have altered excitability. Here, we show that Scn1b null mice are ataxic. Expression of β1 protein in WT cerebellum, assessed using a CRISPR transgenic mouse model containing an in-frame V5 epitope tag at the β1 C-terminus, is widespread. Scn1b null PCs and interneurons in cerebellar slices have increased thresholds for action potential initiation and decreased repetitive firing frequency compared with WT. Scn1b null PCs have reduced transient and resurgent sodium current densities. We propose that reduced PC excitability underlies the ataxic phenotype of Scn1b mice. In addition, because cerebellar output to other areas of the brain can result in termination of seizures, we propose that PC hypoexcitability exacerbates the severe phenotype of this mouse model.

Keywords: Epilepsy; Genetics; Mouse models; Neuroscience; Sodium channels.

Figure 1. Scn1b null mice are ataxic.

Scn1b null mice display multiple stride differences compared with WT. (A and B) Images of a single stride captured from video recordings of 1 WT (A) and 1 Scn1b null (B) mouse. Top: Top-down view; bottom: side view. (C) Annotated footprints from 3 consecutive strides of 1 WT (left) and 1 Scn1b null (right) mouse. Insets in D–F display the measurement locations for each analysis type. (D) Stride length for forepaw and hind paw steps. (E) Stride width for forepaw and hind paw steps. (F) Step angle for forepaw and hind paw steps. (G) Ratio of step width and step angle. WT: filled circles; Scn1b null: unfilled circles. *P < 0.05; ***P < 0.001 (unpaired t test, 2-tailed).

Figure 2. Scn1b null PCs are hypoexcitable.

(A and B) Representative traces showing evoked repetitive firing of PCs in cerebellar slices from WT (A) or null (B) mice. Repetitive AP firing was evoked by injections of 1,500 ms pulse currents of –60 pA to +180 pA (only selected –40 pA– to 180 pA–evoked responses are shown). (C) Input-output curves of AP firing for WT (black) and null (blue) PCs in response to current injections from –60 pA to 180 pA. Null PCs show reduced AP firing frequencies at all stimulation intensities. Note, cells firing 0 APs or 3 or fewer APs are not included in input-output analyses. The dotted line indicates the range of -20 pA to 180 pA current points. (D) Null PCs show decreased maximal firing frequencies. Values are mean ± SEM of 36 cells from 23 WT mice or 41 cells from 29 null mice, respectively. *P < 0.05, ****P < 0.0001, * ^– **** represent P values from <0.05 to <0.0001 (2-way ANOVA for C, unpaired t test for D, 2-tailed P value).

Figure 3. Scn1b null PCs show aberrant bursting activity.

(A–D) Representative examples of spontaneous (A and B) or evoked (C and D) burst firing from WT (A and C) or null (B and D) PCs under current-clamp. Traces are representative of 7 cells of 43 WT PCs (23 mice) or 19 cells of 41 Scn1b null PCs (29 mice). (E and F) Rates of spontaneous firing for WT (black) or null (blue) PCs. Values in parentheses represent numbers of cells that did or did not fire spontaneously, as indicated. Differences between genotypes are not significant. (G and H) Rates of spontaneous or evoked burst firing for WT (black) or null (blue) PCs. Values in the parentheses represent numbers of cells that did or did not show bursting firing, as indicated. Differences between WT and null cells are significant (P = 0.0043, Fisher’s Exact Test, 2-tailed). (I and J) Percentages of WT (black) or null (blue) PCs that fired 3 or fewer APs versus more than 3 APs. Values in parentheses represent numbers of cells that fired fewer or more than 3 APs, respectively. Differences between genotypes are significant (P = 0.0101, Fisher’s Exact Test, 2-tailed).

Figure 4. Scn1b^–/– PCs have reduced transient and resurgent I_Na densities.

(A) Representative transient I_Na traces from WT (black) or null (blue) PCs. Currents were evoked by a depolarizing pulse from –120 to –30 mV. Transient I_Na was measured at the peak, and persistent I_Na was assessed as the average current between 48 and 50 ms. (B) Maximal transient I_Na density. (C) Persistent I_Na density. (D) Representative resurgent I_Na density traces recorded in response to a repolarizing pulse to –40 mV, following a prepulse to +30 mV. A fast tail current is observed initially, followed by slow activating and inactivating resurgent I_Na. (E) Resurgent I_Na density. (F) Activation and inactivation curves. No significant differences were observed between genotypes. Values are provided in Supplemental Table 3. Gmax, maximum conductance; Vm, membrane potential. Data are presented as means ± SEM for N = 12 WT or 13 null mice. **P < 0.005, *P < 0.05 (unpaired t test, 2-tailed P value).

Figure 5. Scn1b null MLIs have increased AP initiation threshold and reduced firing frequency.

Representative traces showing evoked repetitive firing of MLIs in lobule IV/V in sagittal cerebellar slices from WT (A) or null (B) mice. Repetitive AP firing was evoked by injections of 1,500 ms pulse currents of –60 pA to +180 pA (only selected –60 pA– to +50 pA–evoked responses are shown). Stronger depolarizing current injections were required to evoke repetitive firing in MLIs from null mice. (C) Null MLIs (blue) show reduced AP firing frequencies at all stimulation intensities compared with WT (black).* ^– **** represent P values from <0.05 to <0.0001. (D) Null MLIs show decreased maximal firing frequencies. n = 11 cells from 8 WT mice (black), and n = 11 cells from 9 null mice (blue). ***P = 0.0004, (2-way ANOVA for C, unpaired t test for D, 2-tailed P value).

Figure 6. Scn1b null PCs have reduced sIPSC frequency but not amplitude.

sIPSCs in PCs were recorded from a holding potential of –70 mV with a CsCl-based internal solution in the presence of 10 μM CNQX and 100 μM APV. (A and B) Representative traces showing sIPSCs recorded from P14–20 WT (black) or null (blue) PCs. (C and D) Comparisons of cumulative fraction of interevent intervals (C) or amplitudes (D) of sIPSCs from the same cells shown in A and B. (E and F) Comparisons of differences in mean frequencies (E) or amplitudes (P = 0.0027, Mann-Whitney test, 2-tailed) (F) of sIPSCs between WT and null mice. Recordings from 12 cells from 11 WT mice or 18 cells from 13 null mice.

Figure 7. Scn1b null PCs have increased sEPSC frequency but not amplitude.

sEPSCs in PCs were recorded from a holding potential of –70 mV in the presence of 10 μM bicuculline. (A and B) Representative traces showing sEPSCs recorded from P14–20 WT (black) or null (blue) PCs. (C and D) Comparison of cumulative fraction of interevent intervals (C) or amplitudes (D) of sIPSCs from the same cells shown in A and B. (E and F) Comparisons of differences in mean frequencies (E) or amplitudes (F) of sEPSCs between WT and null mice. Recordings from 14 cells of 8 WT mice or 23 cells of 16 null mice. *P = 0.026 (unpaired t test, 2-tailed).

Figure 8. Scn1b deletion alters short-term synaptic plasticity at CF-PC synapses.

Paired-pulse evoked EPSCs were recorded in PCs by stimulation of parallel fibers (PFs) in the pier of the external molecular layer or climbing fibers (CFs) in white matter at different interstimulus intervals (ISI 40, 60, 100, or 300 ms, as indicated). Amplitudes of EPSCs evoked by field stimulation varied from cell to cell because of individual cell variation and position of the stimulating electrode. WT cells (black) showed paired-pulse facilitation (PPF) at PF-PC synapses (A, left) and paired-pulse depression (PPD) at CF-PC synapses (B, left). Scn1b deletion (blue) promoted PPF at CF-PC synapses (B, right) with no significant effect on PF-PC synapses. Each trace is a representative example of 8 cells from 5 WT mice or 19 cells from 9 Scn1b null mice. (C) Statistical analyses of differences in short-term synaptic plasticity of CF-PC synapses at each indicated ISI between genotypes. EPSC2/EPSC1 ratios > 1 (above the dotted lines) were defined as PPF. EPSC2/EPSC1 ratios < 1 (below the dotted line) were defined as PPD. EPSC2/EPSC1 ratios = 1 (overlapping the dotted line) were defined as no interaction. CF-EPSCs evoked at ISI 60 showed significant changes in paired-pulse stimulation-evoked responses (P < 0.05, Fisher’s Exact Test, 2-tailed).