Down-Regulation of Astrocytic Kir4.1 Channels during the Audiogenic Epileptogenesis in Leucine-Rich Glioma-Inactivated 1 (Lgi1) Mutant Rats

The dysfunction of astrocytic inwardly rectifying potassium (Kir) 4.1 channels, which mediate the spatial potassium-buffering function of astrocytes, is known to be involved in the development of epilepsy. Here, we analyzed the Kir4.1 expressional changes in Leucine-Rich Glioma-Inactivated 1 (Lgi1) mutant rats, which is a model of autosomal dominant lateral temporal lobe epilepsy in humans, to clarify the role of astrocytic Kir4.1 channels in Lgi1-related epileptogenesis. Priming acoustic stimulation (at postnatal day 16) conferred seizure susceptibility on Lgi1 mutant rats, which evoked audiogenic seizures with test stimulation at eight weeks. In the seizure-susceptible Lgi1 mutant rats (before test stimulation), astrocytic Kir4.1 expression was down-regulated region-specifically in the cerebral cortex, hippocampus, and amygdala. In addition, prophylactic treatments of Lgi1 mutant rats with valproic acid (VPA, 30 mg/kg and 200 mg/kg) for two weeks prevented both the development of seizure susceptibility and the down-regulation of Kir4.1 expression in astrocytes. The present study demonstrated for the first time that the astrocytic Kir4.1 expression was reduced in the Lgi1-related seizure model, suggesting that the down-regulation of Kir4.1 channels in astrocytes is involved in audiogenic epileptogenesis caused by Lgi1 mutation. In addition, VPA seemed to have a prophylactic effect on Lgi1-related seizures.


Introduction
Astrocytes are a major component of glial cells in the brain that form tripartite synapses in conjunction with neuronal components (presynaptic nerve terminal and postsynaptic neuronal membrane), and actively regulate the excitability and plasticity of neurons. They maintain extracellular ion homeostasis, metabolize neurotransmitters (e.g., glutamate and GABA), and secrete various neuroactive substances (e.g., gliotransmitters, neurotrophic factors, and cytokines) [1][2][3]. Specifically, the spatial potassium (K + ) buffering by astrocytes' functions is a clearance mechanism of excessive 2. Results

Audiogenic Seizure Induction
Wild-type (WT) rats and Lgi1 mutant rats were divided into four groups. Groups A and B did not receive the acoustic priming stimulation without (Group A) and with (Group B) the test stimulation (130 dB, 10 kHz, one minute) at the age of eight weeks, respectively ( Figure 1). Groups C and D received the acoustic priming stimulation (130 dB, 10 kHz, five minutes) at postnatal day (P) 16 without (Group C) and with (Group D) the test stimulation at eight weeks, respectively.
In all of the non-primed Group B animals (either WT or Lgi1 mutant rats), the acoustic test stimulation at eight weeks did not cause any behavioral changes (Table 1). In contrast, in all of the primed Group D animals (either WT or Lgi1 mutant rats), the acoustic test stimulation at eight weeks evoked wild running behavior. Moreover, Lgi1 mutant rats (n = 8), except for one, showed generalized tonic-clonic seizures (GTCSs) following wild running, although none of the WT rats in Group D exhibited GTCSs (Table 1). 20, x; doi: www.mdpi.com/journal/ijms evoked wild running behavior. Moreover, Lgi1 mutant rats (n = 8), except for one, showed generalized tonic-clonic seizures (GTCSs) following wild running, although none of the WT rats in Group D exhibited GTCSs (Table 1).

Expressional Changes in Astrocytic Kir4.1 during Audiogenic Epileptogenesis
We first confirmed the expression pattern of Kir4.1 in WT F344 rats using the immunofluorescent double-staining techniques. As in other strains (Wistar and SD rats) [14,20,29], Kir4.1immunoreactivity (IR) was mostly co-stained with glial fibrillary acidic protein (GFAP), which is a specific marker for astrocytes, in stellate-shaped cells (Figure 2A top), illustrating that Kir4.1 channels are specifically expressed in astrocytes in F344 rats. In addition, Kir4.1 expression in astrocytes was considerably reduced in the Lgi1 mutant rats that received priming stimulation at P16 (

Expressional Changes in Astrocytic Kir4.1 during Audiogenic Epileptogenesis
We first confirmed the expression pattern of Kir4.1 in WT F344 rats using the immunofluorescent double-staining techniques. As in other strains (Wistar and SD rats) [14,20,29], Kir4.1-immunoreactivity (IR) was mostly co-stained with glial fibrillary acidic protein (GFAP), which is a specific marker for astrocytes, in stellate-shaped cells (Figure 2A top), illustrating that Kir4.1 channels are specifically expressed in astrocytes in F344 rats. In addition, Kir4.1 expression in astrocytes was considerably reduced in the Lgi1 mutant rats that received priming stimulation at P16 (Figure 2A bottom)  Topographical mapping analysis for the changes in Kir4.1 and GFAP expression was performed by counting the number of Kir4.1-IR-positive and GFAP-IR-positive cells stained by the avidin-biotin complex (ABC) method. Nineteen brain regions were analyzed, including the cerebral cortex, hippocampus, amygdala, thalamus, and hypothalamus, which were reported to be activated with audiogenic seizures in Lgi1 mutant rats [28] ( Figure 2B). Kir4.1-IR was mostly expressed in stellateshaped cells, and there was no morphological difference in Kir4.1-IR or GFAP-IR positive cells between WT and Lgi1 mutant rats ( Figure 2C).
Neither priming stimulation at P16 nor test stimulation at eight weeks affected the number of astrocytes (GFAP-IR-positive cells) per se either in WT or Lgi1 mutant rats of groups A-D ( Figure 3). Topographical mapping analysis for the changes in Kir4.1 and GFAP expression was performed by counting the number of Kir4.1-IR-positive and GFAP-IR-positive cells stained by the avidin-biotin complex (ABC) method. Nineteen brain regions were analyzed, including the cerebral cortex, hippocampus, amygdala, thalamus, and hypothalamus, which were reported to be activated with audiogenic seizures in Lgi1 mutant rats [28] ( Figure 2B). Kir4.1-IR was mostly expressed in stellate-shaped cells, and there was no morphological difference in Kir4.1-IR or GFAP-IR positive cells between WT and Lgi1 mutant rats ( Figure 2C).
Neither priming stimulation at P16 nor test stimulation at eight weeks affected the number of astrocytes (GFAP-IR-positive cells) per se either in WT or Lgi1 mutant rats of groups A-D ( Figure 3).   Figure S1). Although adiogenic seizures induced by the test stimulation tended to further reduce Kir4.1 expression in some brain regions (e.g., cerebral cortex, hippocampus, and amygdala) of the Lgi1 mutant rats, these changes between Group C and Group D were not statistically significant (figures 4 and 5). We also checked the changes in Kir4.1 expression by three-way (i. mutation; ii. priming stimulation; iii. test stimulation) ANOVA, which revealed a statistical significance with the two factors, mutation and priming stimulation, but not with the test stimulation, in most brain regions.  Figure S1). Although adiogenic seizures induced by the test stimulation tended to further reduce Kir4.1 expression in some brain regions (e.g., cerebral cortex, hippocampus, and amygdala) of the Lgi1 mutant rats, these changes between Group C and Group D were not statistically significant (Figures 4 and 5). We also checked the changes in Kir4.1 expression by three-way (i. mutation; ii. priming stimulation; iii. test stimulation) ANOVA, which  In contrast to Lgi1 mutant rats, WT rats showed no changes in Kir4.1 expression with or without the priming stimulation at P16 and the test stimulation at eight weeks in all of the examined brain regions (Figures 4 and 5). In contrast to Lgi1 mutant rats, WT rats showed no changes in Kir4.1 expression with or without the priming stimulation at P16 and the test stimulation at eight weeks in all of the examined brain regions (figures 4 and 5).

Prophylactic Actions of VPA on Seizure Susceptibility in Lgi1 Mutant Rats
We next examined the prophylactic effects of VPA on the audiogenic seizure susceptibility in Lgi1 mutant rats. Lgi1 mutant rats were treated with VPA (30 and 200 mg/kg, i.p.) for two weeks from one day before priming stimulation. Under these conditions, both 30 mg/kg and 200 mg/kg of VPA effectively prevented GTCS induction ( Figure 6 and Table 2). Only one out of five Lgi1 mutant rats

Prophylactic Actions of VPA on Seizure Susceptibility in Lgi1 Mutant Rats
We next examined the prophylactic effects of VPA on the audiogenic seizure susceptibility in Lgi1 mutant rats. Lgi1 mutant rats were treated with VPA (30 and 200 mg/kg, i.p.) for two weeks from one day before priming stimulation. Under these conditions, both 30 mg/kg and 200 mg/kg of VPA effectively prevented GTCS induction ( Figure 6 and Table 2). Only one out of five Lgi1 mutant rats (20%) exhibited GTCSs with test stimulation, whereas three out of four Lgi1 mutant rats (75%) treated with saline showed GTCSs.

Expressional Changes in Astrocytic Kir4.1 with Treatment of VPA
Similar to the results of prior experiments, the Kir4.1 expression rate significantly decreased in the cerebral cortex (Au1 and Pir), hippocampus (CA1, CA2, and DG), and amygdala (BMP) in the primed Lgi1 mutant rats treated with saline ( Figure 7). However, the prophylactic treatment with VPA (30 mg/kg and 200 mg/kg, i.p.) for two weeks significantly elevated the Kir4.1 expression rate in a dose-related manner in the cerebral cortex (Pir), hippocampus (CA1 and CA2), and amygdala (BMP) (Figure 7).

Discussion
It has been shown that the hypofunction of Kir4.1 channels in astrocytes was involved in various epileptic disorders including not only epilepsy models in animals [13,14], but also EAST/SeSAME syndrome with KCNJ10 mutations [10][11][12] and TLE in humans [15][16][17]. Here, we demonstrated for the first time that astrocytic Kir4.1 expression was markedly reduced in the cerebral cortex, hippocampus, and amygdala during the audiogenic epileptogenesis in Lgi1 mutant rats, which is a rat model of human ADLTE. The brain regions, which showed a down-regulation of Kir4.1 expression, were consistent with the regional distribution of Lgi1 expression [30]. The hippocampus and amygdala were also suggested to be involved in generation of audiogenic seizures in Fragile X mental retardation 1 (Fmr1) knockout mice [31]. It should be noted that the down-regulation of astrocytic Kir4.1 in Lgi1 mutant rats was obtained without any seizure experience (before applying acoustic test stimulation) after the development of audiogenic epileptogenesis. The present results suggested that the reduced expression of astrocytic Kir4.1 expression is involved in Lgi1-related epileptic disorders. The dysfunction of astrocytic Kir4.1 channels elevates the extracellular levels and K + and glutamate by suppressing the special K + buffering function of astrocytes, which increases the excitability of neurons [4][5][6][7]. In addition, we previously showed that down-regulation of the Kir4.1 channel expression enhanced the expression of BDNF, which is a key modulator of epileptogenesis, in astrocytes [18,19]. Knockdown of the BDNF gene [32][33][34][35] or inhibition of the BDNF receptor TrkB [36,37] has been shown to suppress seizure development. Thus, it seems possible that the down-

Discussion
It has been shown that the hypofunction of Kir4.1 channels in astrocytes was involved in various epileptic disorders including not only epilepsy models in animals [13,14], but also EAST/SeSAME syndrome with KCNJ10 mutations [10][11][12] and TLE in humans [15][16][17]. Here, we demonstrated for the first time that astrocytic Kir4.1 expression was markedly reduced in the cerebral cortex, hippocampus, and amygdala during the audiogenic epileptogenesis in Lgi1 mutant rats, which is a rat model of human ADLTE. The brain regions, which showed a down-regulation of Kir4.1 expression, were consistent with the regional distribution of Lgi1 expression [30]. The hippocampus and amygdala were also suggested to be involved in generation of audiogenic seizures in Fragile X mental retardation 1 (Fmr1) knockout mice [31]. It should be noted that the down-regulation of astrocytic Kir4.1 in Lgi1 mutant rats was obtained without any seizure experience (before applying acoustic test stimulation) after the development of audiogenic epileptogenesis. The present results suggested that the reduced expression of astrocytic Kir4.1 expression is involved in Lgi1-related epileptic disorders. The dysfunction of astrocytic Kir4.1 channels elevates the extracellular levels and K + and glutamate by suppressing the special K + buffering function of astrocytes, which increases the excitability of neurons [4][5][6][7]. In addition, we previously showed that down-regulation of the Kir4.1 channel expression enhanced the expression of BDNF, which is a key modulator of epileptogenesis, in astrocytes [18,19]. Knockdown of the BDNF gene [32][33][34][35] or inhibition of the BDNF receptor TrkB [36,37] has been shown to suppress seizure development. Thus, it seems possible that the down-regulation of astrocytic Kir4.1 channels participates in the development of epilepsy in Lgi1 mutant rats.
Epileptogenesis generally shows a time lag until the development of seizures following trigger events (e.g., stroke, traumatic brain injury, and status epilepticus) [38]. In audiogenic epileptogenesis caused by Lgi1 mutation, astrocytic Kir4.1 expression also required time to be down-regulated. Although the mechanisms underlying the down-regulation of Kir4.1 expression by Lgi1 mutation are still uncertain, it has been recently demonstrated that Kir4.1 channel expression in astrocytes was down-regulated by increased glutamate signaling [39]. Since previous studies have suggested that the synaptic release of glutamate is increased by Lgi1 mutation [40,41], this might be involved in the down-regulation of astrocytic Kir4.1 expression in Lgi1 mutant rats. In addition, Lgi1 is known to regulate the expression and/or clustering of potassium channels (e.g., Kv1.1 channels) through the interaction with the membrane-anchored proteins (ADAM22 and ADAM23) [23,26]. Thus, disruptions of the Lgi1-ADAM22/23 interaction in Lgi1 mutant rats might also influence the Kir4.1 channel expression in astrocytes. Further studies are required to clarify the mechanisms underlying the interaction of Lgi1 with Kir4.1 channels in the Lgi1-related epileptogenesis.
VPA is widely used as an antiepileptic drug that is effective for both generalized and partial seizures. It enhances GABAergic neurotransmission by inhibiting the GABA metabolic enzyme, GABA transaminase [19,42]. In addition, VPA also blocks voltage-gated sodium channels and T-type calcium channels. These actions are considered to cause neuronal inhibition and contribute to its anticonvulsive activity. Nonetheless, several studies have suggested that VPA shows prophylactic effects on seizure generation and might have the potential to prevent epileptogenesis [43][44][45]. In consistent with this possibility, we have previously shown that repeated treatment of animals with VPA enhanced astrocytic Kir4.1 expression in the cerebral cortex and limbic regions (hippocampus and amygdala), which seems to inhibit epileptogenesis via facilitating the spatial potassium buffering function of astrocytes. In the present study, we confirmed that VPA could alleviate the epileptogenesis in Lgi1 mutant rats with elevating astrocytic Kir4.1 expression. Our results suggest that VPA can prevent or alleviate Lgi1-related epileptogenesis by normalizing the down-regulation of the astrocytic Kir4.1 channel.
In conclusion, we analyzed the astrocytic Kir4.1 expression during audiogenic epileptogenesis in Lgi1 mutant rats. Kir4.1 expression in astrocytes was significantly decreased by the postnatal priming stimulation in the cerebral cortex, hippocampus, and amygdala. In addition, prophylactic treatments with VPA effectively alleviated audiogenic seizure susceptibility and prevented the reduced expression of astrocytic Kir4.1 channels in Lgi1 mutant rats. The present study suggests that the down-regulation of Kir4.1 channels in astrocytes is involved in audiogenic seizure generation caused by Lgi1 mutation. Moreover, VPA seemed to have a prophylactic effect on Lgi1-related epileptogenesis. However, this study evaluated the Kir4.1 levels solely by counting IR-positive cells, but not by calculating the IR area [46], so that compensatory changes represented by increased Kir4.1-IR in the remaining positive cells could not be excluded. Further analyses of Kir4.1 expression at cellular and subcellular levels are necessary to delineate the mechanisms underlying the Kir4.1 down-regulation by Lgi1 mutation and its relationship to epileptogenesis.

Animals
F344-Lgi1 m1Kyo rats (NBRP Rat No. 0656), carrying a heterozygous missense mutation (L385R/+) generated by N-ethyl-N-nitrosourea (ENU) mutagenesis [27], were obtained from the National BioResource Project-Rat (NBRP-Rat, http://www.anim.med.kyoto-u.ac.jp). The housing conditions of the rats and 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 procedures used were approved by the Animal Research Committee of Kyoto University (Med Kyo 14560, 6 November used for phenotypic analyses.

Audiogenic Seizure Induction
Animals divided into four groups (Group A-D) were individually exposed to auditory stimulation using a previously reported method [27,28] (Figure 1). Priming stimulation was performed in P16 rats with a sound stimulus (130 dB, 10 kHz, five minutes). Test stimulation was performed with a sound stimulus (130 dB, 10 kHz, one minute) at eight weeks. Within two hours after seizure induction or a sham procedure, the animals were deeply anesthetized with pentobarbital (80 mg/kg, i.p.) that was transcardially perfused with ice-cold phosphate-buffered saline (PBS), and then with 4% paraformaldehyde solution. The brains were removed from the skull and placed in fresh fixative for 24 h.

Drug Treatments
To determine whether an antiepileptic agent modified the pathophysiological alterations in Kir4.1 channels in Lgi1-related epileptogenesis, we studied the prophylactic actions of VPA in Lgi1 mutant rats, with a focus on its regulation of Kir4.1 channel expression. From one day before priming stimulation, animals were intraperitoneally injected with VPA (30 mg/kg and 200 mg/kg) (Sigma-Aldrich) or saline for 14 continuous days ( Figure 6). Then, the animals were subjected to the audiogenic seizure test four weeks after the last VPA treatment, and brain samples were obtained as described previously.

Statistical Analysis
All of the data were expressed as the mean ± S.E.M. Comparisons between two groups (expressional changes of Kir4.1 determined by immunohistochemical analysis in each region) were performed by a parametric Student's t-test. Statistical significance of differences among multiple groups (expressional changes of Kir4.1 or GFAP determined by immunohistochemical analysis in each region) 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.