Molecular Mechanisms of Epileptic Encephalopathy Caused by KCNMA1 Loss-of-Function Mutations

Abstract

The gene kcnma1 encodes the α-subunit of high-conductance calcium- and voltage-dependent K⁺ (BK) potassium channel. With the development of generation gene sequencing technology, many KCNMA1 mutants have been identified and are more closely related to generalized epilepsy and paroxysmal dyskinesia. Here, we performed a genetic screen of 26 patients with febrile seizures and identified a novel mutation of KCNMA1 (E155Q). Electrophysiological characterization of different KCNMA1 mutants in HEK 293T cells, the previously-reported R458T and E884K variants (not yet determined), as well as the newly-found E155Q variant, revealed that the current density amplitude of all the above variants was significantly smaller than that of the wild-type (WT) channel. All the above variants caused a positive shift of the I-V curve and played a role through the loss-of-function (LOF) mechanism. Moreover, the β4 subunit slowed down the activation of the E155Q mutant. Then, we used kcnma1 knockout (BK KO) mice as the overall animal model of LOF mutants. It was found that BK KO mice had spontaneous epilepsy, motor impairment, autophagic dysfunction, abnormal electroencephalogram (EEG) signals, as well as possible anxiety and cognitive impairment. In addition, we performed transcriptomic analysis on the hippocampus and cortex of BK KO and WT mice. We identified many differentially expressed genes (DEGs). Eight dysregulated genes [i.e., (Gfap and Grm3 associated with astrocyte activation) (Alpl and Nlrp10 associated with neuroinflammation) (Efna5 and Reln associated with epilepsy) (Cdkn1a and Nr4a1 associated with autophagy)] were validated by RT-PCR, which showed a high concordance with transcriptomic analysis. Calcium imaging results suggested that BK might regulate the autophagy pathway from TRPML1. In conclusion, our study indicated that newly-found point E155Q resulted in a novel loss-of-function variant and the dysregulation of gene expression, especially astrocyte activation, neuroinflammation and autophagy, might be the molecular mechanism of BK-LOF meditated epilepsy.

Keywords: BK channel; KCNMA1; autophagy; epilepsy; loss-of-function variants; neuroinflammation.

FIGURE 1

Location and sequencing of the KCNMA1 variants. (A) Predicted transmembrane topology of KCNMA1 depicting the location of the variants. (B) DNA sequencing identified the mutations in the constructed hBKα plasmid. The mutation sites are marked by a red square.

FIGURE 2

Electrophysiological characterization of variant E155Q. (A) The E155Q variant occur at an evolutionarily conserved amino acid residues. (B) Representative macroscopic currents of WT and mutant BK channels with variant E155Q from whole-cell patch experiments in the presence of 1 and 10 μm Ca²⁺. (C) The I-V curves of WT and E155Q mutant BK channels are shown at 1 and 10 μm Ca²⁺. The I-V curves are fitted by Boltzmann function (solid lines) with V1/2 and slope factor at nominal at 1 μm Ca²⁺ [97.5 ± 4.1 mV, 21.4 ± 2.1 WT, and 113.4 ± 6.4 mV, 33.2 ± 3.3 p.(E155Q)] and at 10 μm Ca²⁺ [30.4 ± 2.4 mV, 15.9 ± 0.8 WT, and 34.2 ± 2.9 mV, 14.6 ± 1.6 p.(E155Q)]. (D) Scatter plots of voltage at half-maximal activation (V1/2) for WT and variants. (E) The current density of WT and E155Q mutant BK channels are shown at 1 and 10 μm Ca²⁺. The data are presented as mean ± SEM. (Compared with WT-1um, *p < 0.05, **p < 0.01, ***p < 0.001. Compared with WT-10um, #p < 0.05, ##p < 0.01, ###p < 0.001. n = 6–14/group).

FIGURE 3

Electrophysiological characterization of variant R458T. (A) The R458T variant occur at an evolutionarily conserved amino acid residues. (B) Representative macroscopic currents of WT and mutant BK channels with variant R458T from whole-cell patch experiments in the presence of 1 and 10 μm Ca²⁺. (C) The I-V curves of WT and R458T mutant BK channels are shown at 1 and 10 μm Ca²⁺. The I-V curves are fitted by Boltzmann function (solid lines) with V1/2 and slope factor at nominal at 1 μm Ca²⁺ [97.5 ± 4.1 mV, 21.4 ± 2.1 WT, and 111.9 ± 6.3 mV, 27.5 ± 2.4 p.(R458T)] and at 10 μm Ca²⁺ [30.4 ± 2.4 mV, 15.9 ± 0.8 WT, and 48.2 ± 2.6 mV, 13.7 ± 1.1 p.(R458T)]. (D) Scatter plots of voltage at half-maximal activation (V1/2) for WT and variants. (E) The current density of WT and R458T mutant BK channels are shown at 1 and 10 μm Ca²⁺. The data are presented as mean ± SEM. (Compared with WT-1um, *p < 0.05, **p < 0.01, ***p < 0.001. Compared with WT-10um, #p < 0.05, ##p < 0.01, ###p < 0.001. n = 6–14/group).

FIGURE 4

Electrophysiological characterization of variant E884K. (A) The E884K variant occur at an evolutionarily conserved amino acid residues. (B) Representative macroscopic currents of WT and mutant BK channels with variant E884K from whole-cell patch experiments in the presence of 1 and 10 μm Ca²⁺. (C) The I-V curves of WT and E884K mutant BK channels are shown at 1 and 10 μm Ca²⁺. The I-V curves are fitted by Boltzmann function (solid lines) with V1/2 and slope factor at nominal at 1 μm Ca²⁺ [97.5 ± 4.1 mV, 21.4 ± 2.1 WT, and 112.0 ± 6.7 mV, 34.8 ± 4.2 p.(E884K)] and at 10 μm Ca²⁺ [30.4 ± 2.4 mV, 15.9 ± 0.8 WT, and 39.3 ± 2.2 mV, 19.3 ± 1.4 p.(E884K)]. (D) Scatter plots of voltage at half-maximal activation (V1/2) for WT and variants. (E) The current density of WT and E884K mutant BK channels are shown at 1 and 10 μm Ca²⁺. The data are presented as mean ± SEM. (Compared with WT-1um, *p < 0.05, **p < 0.01, ***p < 0.001. Compared with WT-10um, #p < 0.05, ##p < 0.01, ###p < 0.001. n = 6–14/group).

FIGURE 5

Construction of BK KO mice with spontaneous epilepsy. (A) Schematic outlining the generation of BK knockout mice using the CRISPR/Cas9 system. The targeting sites of KCNMA1 (gene encoding the α subunit of BK, BKα) are shown. (B,C) BK KO mice were established by breeding BK^+/− males and females. The targeted fragment of KCNMA1 was amplified by PCR using genomic DNA templates, and the BK channel deletion was confirmed by sequencing. Genome sequencing of BK KO mice showed a frameshift mutation (- 16 bp) in exon 4. (D) Spontaneous epileptic behavior of BK KO mice. The proportion of different seizure stages in BK KO/WT mice was observed for 2 h (***p < 0.001, n = 3).

FIGURE 6

In vivo multichannel EEG recording of mice. (A,B) FP signals and spectral heat maps from a representative WT (black) and BK KO (red) mice are shown, respectively. (C) Spectral analysis of PSD values on different frequency δ, θ, α, β, and γ waves in each group. (D) The PSD of mice in each group (Compared with control group, *p < 0.05, **p < 0.01, ***p < 0.001, n = 4).

FIGURE 7

Gait analysis was performed by DigiGait imaging system. (A–C) Schematic diagram of WT and BK KO mouse footprints. (D) Print area (cm²), mean intensity and Swing speed (cm/s) of the right front (RF), the right hind (RH), the left front (LF), the left hind (LH) limb were chosen as the observation index. The Data are presented as means ± SEM. (Compared with control group, *p < 0.05, **p < 0.01, ***p < 0.001, n = 5).

FIGURE 8

Transcriptome profiling in the hippocampus and cortex tissues of mice. (A) Heatmap of the DEGs. (B) Enriched Biological Process pathway, cellular component and molecular function in GO analysis (p < 0.05). (C) KEGG enrichment analysis of DEGs. The intensity of the color depends on the p value. The size of plot depends on the gene count. (n = 3).

FIGURE 9

Differential expression of mRNAs between the cortex and hippocampus of WT (Black) and BK KO (Red) mice validated by RT-PCR. Gfap and Grm3 associated with astrocyte activation, Alpl, and Nlrp10 associated with neuroinflammation, Efna5 and Reln associated with epilepsy, Cdkn1a and Nr4a1 associated with autophagy. Gfap, Grm3, Nlrp10, Alpl in cortex, and Efna5, Reln, Cdkn1a, Nr4a1 in hippocampus. Ns (no significant difference, p > 0.05), ∗p < 0.05, ∗∗p < 0.01.

FIGURE 10

Autophagy in BK KO mice was abnormal. Calcium imaging found that the activation of BK channel could activate lysosomal trpml1 (autophagy key calcium channel). (A) Co labeling of microglia, LC3B, IBA-1, and LAMP1 in control mice. (B) Co labeling of microglia, LC3B, IBA-1, and LAMP1 in BK KO mice. (C) NS1619 was applied to HEK293T transfected with BK channel and TRPML1-GCaMP3 in order to detect its regulation of lysosomal calcium outflow (^∗∗ p < 0.01, n = 6).