Genetic reduction of the α1 subunit of Na/K-ATPase corrects multiple hippocampal phenotypes in Angelman syndrome

Abstract

Angelman syndrome (AS) is associated with symptoms that include autism, intellectual disability, motor abnormalities, and epilepsy. We recently showed that AS model mice have increased expression of the alpha1 subunit of Na/K-ATPase (α1-NaKA) in the hippocampus, which was correlated with increased expression of axon initial segment (AIS) proteins. Our developmental analysis revealed that the increase in α1-NaKA expression preceded that of the AIS proteins. Therefore, we hypothesized that α1-NaKA overexpression drives AIS abnormalities and that by reducing its expression these and other phenotypes could be corrected in AS model mice. Herein, we report that the genetic normalization of α1-NaKA levels in AS model mice corrects multiple hippocampal phenotypes, including alterations in the AIS, aberrant intrinsic membrane properties, impaired synaptic plasticity, and memory deficits. These findings strongly suggest that increased expression of α1-NaKA plays an important role in a broad range of abnormalities in the hippocampus of AS model mice.

Figure 1. Genetic reduction of α1-NaKA corrects increased expression of ankyrin-G and NaV1.6 in AS model mice

Western blot analysis of hippocampal homogenates from wild-type (WT), Angelman syndrome (AS), α1-NaKA heterozygous knockout (NaKA) and AS/NaKA heterozygous knockout (dKO) mice. A) Representative blots of ankyrin-G, NaV1.6, α1-NaKA, and actin. B) Cumulative data for the blots probed for α1-NaKA and actin. C) Cumulative data for the blots probed with antibodies to ankyrin-G and actin. D) Cumulative data for the blots probed for NaV1.6 and actin. For all panels, actin was used as a loading control and quantification was done as a ratio to wild-type on the same blot. (WT: n=6; AS: n=7; α1-NaKA: n=7; dKO: n=7). * denotes statistical significance (p<0.05) and ** denotes statistical significance (p<0.01) both from wild-type with an ANOVA (Tukey’s post-hoc test).

Figure 2. AIS length in AS model mice is restored to wild-type length in hippocampal CA1 and CA3 pyramidal neurons by genetically reducing α1-NaKA levels

Immunostaining analysis of ankyrin-G expression and AIS length in hippocampal and cortical tissue from wild-type (WT), Angelman syndrome (AS), α1-NaKA heterozygous knockout (α1-NaKA) and AS/α1-NaKA heterozygous knockout (dKO) mice. A–C) Coronal brain slices containing the hippocampal formation and somatosensory cortex were immunostained with antibodies targeted against ankyrin-G. AIS length was quantified in CA1 (A), CA3 (B), and layer II/III somatosensory cortex (C) in tissue from each of the four genotypes. WT: n=240 cells/region, 2 mice; AS: n=239 cells/region, 2 mice; α1-NaKA: n=240 cells/region, 2 mice; dKO: n=239 cells/region, 2 mice. Average, region-specific AIS length data is provided in box and whisker plots. ***denotes statistical significance (p<0.001) with ANOVA. Fluorescence intensity (F.I.) plots provide a comparison of representative AIS ankyrin-G immunosignal strength and AIS length of cells contained in the expanded views, shown to the right of the contextual image.

Figure 3. Genetic reduction of α1-NaKA corrected alterations in the intrinsic membrane properties of hippocampal CA1 pyramidal neurons in AS model mice

Intrinsic membrane properties of hippocampal CA1 pyramidal neurons in WT, AS, α1-NaKA heterozygous knockout (α1-NaKA) and dKO mice measured by current clamp without current injection (I=0). A) Resting membrane potential. B) Membrane time constant. C) Membrane input resistance. D) Threshold potential. E) Action potential amplitude. F) Action potential full width at half-max. G) Maximal rate of rise (dV/dt) of the action potential. H) Rheobase (amount of current injection needed to induce an action potential at the 5 msec time point). WT: n=28 cells, 8 mice; AS: n=28 cells, 8 mice; α1-NaKA: n=24 cells, 7 mice; dKO: n=27 cells, 8 mice. *** denotes statistical significance (p<0.001) with ANOVA.

Figure 4. Genetic reduction of α1-NaKA corrects deficits in hippocampal LTP, contextual fear memory, and spatial memory displayed by AS mice

A) Hippocampal slices from WT, AS, α1-NaKA heterozygous knockout mice (NaKA) and dKO mice were stimulated with two trains of high-frequency stimulation (HFS, indicated by the arrows). WT: n=10 slices, 5 mice; AS: n=10 slices, 5 mice; α1-NaKA: n=10 slices, 5 mice; dKO: n=16 slices, 8 mice. p<0.05 for group p<0.0001 for time p<0.0001 for interaction RM-ANOVA. B) Sample traces of typical field excitatory postsynaptic potentials (fEPSPs) recorded before (black) and 60 min after (red) HFS. C) Mice were trained using a standard contextual fear conditioning paradigm and tested for long-term memory measured as % time freezing seven days after training. WT: n=12 mice; AS: n=11 mice; α1-NaKA: n=11 mice; dKO: n=11 mice. Results are displayed as the average of % freezing during the entire 5 minute test. *** denotes statistical significance of p<0.001 in ANOVA (post-test tukey). D) Results are displayed as % freezing along the entire test, minute by minute. *** denotes statistical significance of p<0.001 for interaction of group and treatment in RM-ANOVA. E) Mice were trained using a standard Morris water maze paradigm and tested for spatial memory of platform location in the probe test (platform removed). n=8 for WT, AS, and dKO; n=5 for α1-NaKA. Results are displayed as % of time spent in each quadrant along the entire probe test (Quadrant time occupancy). * denotes statistical significance of p<0.05 for interaction of genotype and quadrant in 2way-ANOVA. ** denotes statistical significance of p<0.01 for the trained quadrant in Bonferroni post-test. F) Results are displayed as a number of platform location crosses during the probe test. ** denotes statistical significance of p<0.01 in ANOVA (post-test tukey).