Antiepileptic Drugs Elevate Astrocytic Kir4.1 Expression in the Rat Limbic Region

Inwardly rectifying potassium (Kir) channel subunits Kir4.1 are specifically expressed in astrocytes and regulate neuronal excitability by mediating spatial potassium buffering. In addition, it is now known that astrocytic Kir4.1 channels are closely involved in the pathogenesis of epilepsy. Here, to explore the role of Kir4.1 channels in the treatment of epilepsy, we evaluated the effects of the antiepileptic drugs, valproate, phenytoin, phenobarbital and ethosuximide, on Kir4.1 expression in astrocytes using immunohistochemical techniques. Repeated treatment of rats with valproate (30–300 mg/kg, i.p., for 1–10 days) significantly elevated the Kir4.1 expression levels in the cerebral cortex, amygdala and hippocampus. Up-regulation of Kir4.1 expression by valproate occurred in a dose- and treatment period-related manner, and did not accompany an increase in the number of astrocytes probed by glial fibrillary acidic protein (GFAP). In addition, repeated treatment with phenytoin (30 mg/kg, i.p., for 10 days) or phenobarbital (30 mg/kg, i.p., for 10 days) also elevated Kir4.1 expression region-specifically in the amygdala. However, ethosuximide (100 mg/kg, i.p., for 10 days), which can alleviate absence but not convulsive seizures, showed no effects on the astrocytic Kir4.1 expression. The present results demonstrated for the first time that the antiepileptic drugs effective for convulsive seizures (valproate, phenytoin, and phenobarbital) commonly elevate the astrocytic Kir4.1 channel expression in the limbic regions, which may be related to their antiepileptic actions.


INTRODUCTION
Epilepsy is a chronic neurologic disease characterized by recurrent convulsive and/or nonconvulsive seizures, affecting approximately 70 million people worldwide (nearly 1% of the population) (Banerjee et al., 2009;Ngugi et al., 2010;Zack and Kobau, 2017). Various antiepileptic drugs, which predominantly act on the neuronal ion channels (e.g., blockers of voltage-gated Na + and Ca 2+ channels) and the inhibitory GABAergic system (e.g., stimulants of GABA A receptor/Cl − channel complex and inhibitors of GABA transaminase), are currently used in the treatment of epilepsy (Meldrum and Rogawski, 2007). Therapy with these standard antiepileptic drugs provides adequate control in about 70% of epilepsy patients; however, the remaining 30% of patients still suffer from refractory (treatment-resistant) symptoms and are sometimes subjected to surgical treatments (e.g., ablation of seizure foci, deep brain stimulation and vagus nerve stimulation) (Mattson, 1998).
It is now known that Kir4.1 plays an important role in inducing and developing epilepsy (epileptogenesis). Kir4.1 knockout mice showed severe motor impairment (e.g., ataxia and tremor), epileptic symptoms (e.g., jerky movements and convulsive seizures), and early mortality (Kofuji et al., 2000;Neusch et al., 2001;Djukic et al., 2007). In addition, astrocytic Kir4.1 expression was reported to be reduced (down-regulated) in the brain regions related to seizure foci in patients with epilepsy and animal models of epilepsy (Ferraro et al., 2004;Inyushin et al., 2010;Das et al., 2012;Heuser et al., 2012;Steinhäuser et al., 2012;Harada et al., 2013). Furthermore, it has been shown that loss-of-function mutations (i.e., missense and nonsense mutations) in the human KCNJ10 gene encoding Kir4.1 caused the epileptic disorders known as "EAST/SeSAME" syndrome (Bockenhauer et al., 2009;Scholl et al., 2009;Reichold et al., 2010). Patients with EAST/SeSAME syndrome manifested generalized tonic-clonic seizures (GTCSs) within a few months after birth, in addition to sensorineural deafness, ataxia and electrolyte imbalance. Therefore, it is likely that Kir4.1 channels are closely involved in the pathogenesis of epilepsy. However, the roles of Kir4.1 channels in the treatment of epilepsy or the influences of antiepileptic drugs on Kir4.1 expression are still unknown.
Besides the acute neural inhibition, repeated treatments with antiepileptics are known to exert, to some extent, prophylactic effects in chronic epilepsy, although the underlying mechanisms remain unclear (Iudice and Murri, 2000;Michelucci, 2006;Torbic et al., 2013). This makes a hypothesis that antiepileptics may enhance Kir4.1 expression to prevent epileptogenesis. In the present study, therefore, we evaluated the effects of the antiepileptic drugs, valproate, phenytoin, phenobarbital and ethosuximide, on astrocytic Kir4.1 expression to explore the potential role of Kir4.1 expression in the treatment of epilepsy.

Animals
Male 6-week-old SD rats (Japan SLC, Shizuoka, Japan) were used. The animals were kept in air-conditioned rooms (24 ± 2 • C and 50 ± 10% relative humidity) under a 12-h light/dark cycle (light FIGURE 1 | Schematic illustration of a brain section selected for quantitative analysis of immunoreactivity (IR) of Kir4.1 or GFAP. Filled squares in each brain region indicate the areas analyzed for counting of Kir4.1-IR-or GFAP-IR-positive cells, which were set according to the rat brain atlas (Paxinos and Watson, 2007). Motor, motor cortex; S1BF, primary somatosensory cortex barrel field; PRh-Ect, perirhinal-ectorhinal cortex; MePV, medial amygdaloid nucleus posteroventral part; MePD, medial amygdaloid nucleus posterodorsal part; BLA, basolateral amygdaloid nucleus anterior part; BMP, basomedial amygdaloid nucleus posterior part; CA1 medial or lateral, CA3 and DG, hippocampal CA1 medial, CA1 lateral, CA3 and dentate gyrus; L, Lateral coordinates (mm); H, Horizontal coordinates from interaural line (mm). on: 8:00 a.m.) and allowed ad libitum access to food and water. The animal care methods complied with the Guide for the Care and Use of Laboratory Animals of the Ministry of Education, Science, Sports and Culture of Japan. The experimental protocols of this study were approved by the Animal Research Committee of Osaka University of Pharmaceutical Sciences.

Drug Treatments and Brain Sampling
Animals (6 rats/group) were intraperitoneally injected with a daily dose of an antiepileptic drug as followed; valproate (30, 100, and 300 mg/kg), phenytoin (30 mg/kg), phenobarbital (30 mg/kg), or ethosuximide (100 mg/kg) for 10 days. To evaluate the time-course, animals were treated with valproate (300 mg/kg) for 1 or 5 day(s). The test doses of each drug were set to anticonvulsive doses in rodents, according to previous papers (Walton and Treiman, 1989;Lothman et al., 1991;Löscher, 1999;Gören and Onat, 2007). Twenty-four hours after the last drug treatment, the animals were deeply anesthetized with pentobarbital (80 mg/kg, i.p.), transcardially perfused with ice-cold phosphate-buffered saline (PBS) and then with 4% paraformaldehyde solution. The brains were then removed from the skull and placed in fresh fixative for at least 24 h.

Drugs
Sodium valproate, phenytoin, phenobarbital, and ethosuximide were purchased from Sigma-Aldrich. Other common laboratory reagents were also obtained from commercial sources.

Statistical Analysis
All data are expressed as the mean ± S.E.M. Comparisons between two groups were performed by Student's t-test. Statistical significance of differences among multiple groups was determined by one-way ANOVA followed by Tukey's post hoc test. A P-value of less than 0.05 was considered statistically significant.

Effects of Valproate on Astrocytic Kir4.1 Expression
We first confirmed the expression pattern of Kir4.1 in rat brains using the immunofluorescence double staining method. As reported previously , confocal laser microscopic analysis revealed that Kir4.1-IR was specifically expressed in astrocytes (somata and processes of stellate-shaped cells) probed by GFAP (Figure 2A).

DISCUSSION
Evidence is accumulating that the dysfunction (reduced function or expression) of astrocytic Kir4.1 channels causes epileptic disorders, including not only EAST/SeSAME syndrome with KCNJ10 mutations (Bockenhauer et al., 2009;Scholl et al., 2009;Reichold et al., 2010), but also idiopathic epilepsy (Das et al., 2012;Heuser et al., 2012;Steinhäuser et al., 2012). These findings suggest that enhancement of Kir4.1 channel activities can prevent the development of epilepsy (epileptogenesis) by facilitating astrocytic spatial potassium buffering. The present study demonstrated for the first time that several antiepileptic drugs, which are commonly effective for GTCSs in patients, enhance the astrocytic Kir4.1 expression in the limbic regions. Valproate significantly elevated the astrocytic Kir4.1 expression in the amygdala, hippocampus and cerebral cortex, in a doseand time-dependent manner. Phenytoin and phenobarbital also increased the Kir4.1 expression in the amygdala region. In addition, up-regulation of Kir4.1 expression by these agents did not accompany the increase in the number of astrocytes (astrogliosis). Limbic structures such as the amygdala have been generally recognized as sites closely related to epileptogenesis in animal models of epilepsy (McNamara, 1984;Morimoto et al., 2004). Moreover, human limbic regions are also involved in seizure generation not only in temporal lobe epilepsy, the most common type of adult localization-related epilepsy, but also in epilepsy induced by autoimmune encephalitis (Tatum, 2012;Melzer et al., 2015). Thus, our results suggest that the elevation of astrocytic Kir4.1 expression in limbic regions by the antiepileptic drugs contributes to their antiepileptic actions. Indeed, in our preliminary studies using audiogenic seizure susceptible Lgi1 L385R mutant rats (Baulac et al., 2012;Fumoto et al., 2014), repeated treatment with valproate alleviated epileptogenesis (development of seizure susceptibility) of the Lgi1 L385R mutant rats which exhibited down-regulation of astrocytic Kir4.1 expression (Kinboshi et al., 2017b).
Valproate inhibits GABA transaminase and increases GABA levels, thereby enhancing inhibitory GABAergic activities (Vajda and Eadie, 2014). Phenobarbital also activates the GABAergic system by prolonging the opening time of chloride ion channels within GABA A receptors. In addition, both valproate and phenytoin possess an inhibitory action against voltage-gated Na + channels. All these actions of antiepileptic drugs reduce neural excitability and contribute to an acute inhibitory action on seizure induction. Besides the acute actions, repeated treatments with these antiepileptics are known to exert, to some extent, prophylactic effects in chronic epilepsy, although such usage are sometimes limited by their side effects and/or drug interactions (e.g., enzyme-inducing properties) (Iudice and Murri, 2000;Michelucci, 2006;Torbic et al., 2013). Indeed, valproate reportedly had the potential to prevent epileptogenesis although the underlying mechanisms remain unclear (Silver et al., 1991;Bolanos et al., 1998;Hashimoto et al., 2003). The present fact that the up-regulation of Kir4.1 channels by antiepileptics was mostly manifested after repeated treatments suggests that the elevated expression of Kir4.1 channels may contribute to the seizure-preventive (prophylactic) actions of these agents.
Ethosuximide specifically alleviates absence seizures and does not affect (or sometimes worsen) GTCSs. It inhibits the low threshold T-type Ca 2+ currents in thalamic neurons, although other mechanisms (e.g., inhibition of the non-inactivating Na + currents and the Ca 2+ -activated K + currents) are also proposed (Crunelli and Leresche, 2002). Interestingly, ethosuximide failed to affect Kir4.1 expression in any brain regions examined. Therefore, Kir4.1 channels may not be involved in preventive effects of ethosuximide on absence seizures. This is consistent with our previous findings that down-regulation of Kir4.1 expression was observed only in the GTCSs model (e.g., Noda epileptic rats), but not in the absence seizure model (Groggy rats), implying that pathophysiological alterations of Kir4.1 are not linked to non-convulsive absence seizures (Harada et al., , 2014Ohno et al., 2015).
In conclusion, we evaluated the effects of the antiepileptic drugs, valproate, phenytoin, phenobarbital and ethosuximide, on expressional levels of astrocytic Kir4.1 channels in rats. Valproate, phenytoin and phenobarbital, which commonly alleviate GTCSs, significantly increased Kir4.1 expression in the limbic regions (e.g., amygdala) without affecting the number of astrocytes. Upregulation of Kir4.1 channels by valproate occurred in a doseand treatment period-dependent manner. In contrast, treatment of rats with ethosuximide, which selectively ameliorates absence seizures, did not affect Kir4.1 expression. The present results demonstrated for the first time that antiepileptics (e.g., valproate, phenytoin and phenobarbital) up-regulate astrocytic Kir4.1 channels in the amygdala, which may contribute to their clinical efficacy in chronic epilepsy. However, it remains uncertain how these antiepileptics elevated the expression of Kir4.1 channels region-specifically in the limbic regions. Further studies are required to clarify the mechanisms underlying the control of astrocytic Kir4.1 expression by antiepileptic drugs.