Introduction

Kir4.1 is an ATP-sensitive inwardly-rectifying potassium ion channel encoded by the KCNJ10 gene and expressed exclusively in glial cells. It plays a key role in regulating spatial K⁺ buffering of astrocytes. Kir4.1 loss-of-function results in increased extracellular K⁺ level and impaired glutamate uptake that cause neuronal hyperexcitability and seizures. A functional decrease in Kir4.1 expression has been reported in human acquired and genetic epilepsies, and in rodent models of epilepsy (1-3). Thus, it is hypothesized that upregulation of Kir4.1 in astrocytes will counteract pathogenic rises in extracellular K⁺ and glutamate to prevent transition to seizures.

Aims and objectives

We aim to develop a novel astrocyte gene therapy suitable for pharmacoresistent epilepsies including focal, multifocal and genetic epilepsies. If effective it can be adapted to enter clinical trials to bring transformative benefits to epileptic patients.

Method

All animal procedures were approved by the local ethical committee and performed in accordance with the United Kingdom Animals Scientific Procedures Act (1986) and associated guidelines. The mouse strain C57BL/6 was used; and anaesthetised via inhalation of 4-5% isoflurane (induction) followed by 1-2% (maintenance) during all surgeries performed.

To investigate whether Kir4.1 over-expression could reduce seizure susceptibility, we injected adeno-associated viral (AAV) vectors unilaterally into the mouse somatosensory and visual cortex, either AAV9-gfaABC1D-Kir4.1-tdTomato or AAV9-gfaABC1D-tdTomato that used the astrocyte-specific gfaABC1D promoter to drive transgene expression. 3 weeks later, electrographic recordings were performed in awake head-fixed mice and acute seizures were induced by focal injection of picrotoxin to the transduced area of cortex.

In parallel, animals were made chronically epileptic by intrahippocampal injection of kainic acid into the right hippocampus. 2 weeks after acute status epilepticus, telemetry devices were implanted to record baseline electrocorticography (ECoG). Then animals were treated with AAV-PHP.eB-gfaABC1D-EGFP-Kir4.1 or AAV-PHP.eB-gfaABC1D-EGFP vector injected into the right hippocampus. From 2 weeks post-treatment, ECoG was continued to record for another 4 weeks and analysed using a semi-automated program to quantify seizures.

Results

In the mouse model of acute seizures, Kir4.1 pre-treatment in cortical astrocytes significantly reduced the number of acute seizures (0.7500 ± 0.4787) during the 80-minute post-picrotoxin period compared to tdTomato control (6.000 ± 0.7071). Data are presented as mean ± SEM; P = 0.0014, unpaired t test with Welch’s correction, n = 4 per group, males only.

In the mouse model of chronic epilepsy, Kir4.1 over-expression in hippocampal astrocytes significantly reduced seizure frequency normalised to baseline levels compared to GFP control. P = 0.033, mixed-effects analysis; Kir4.1: n = 8, 3 females and 5 males; GFP: n = 10, 3 females and 7 males.

Conclusions

We conclude that astrocytes can be a promising target for disease-modifying treatment of epilepsy; and AAV-Kir4.1 astrocyte gene therapy is effective in reducing seizure frequency in rodent models of acute and chronic epilepsy.