Contributed equally
Potassium channels can be affected by epileptic seizures and serve a crucial role in the pathophysiology of epilepsy. Dimethylation of histone 3 lysine 9 (H3K9me2) and its enzyme euchromatic histone-lysine N-methyltransferase 2 (G9a) are the major epigenetic modulators and are associated with gene silencing. Insight into whether H3K9me2 and G9a can respond to epileptic seizures and regulate expression of genes encoding potassium channels is the main purpose of the present study. A total of 16 subtypes of potassium channel genes in pilocarpine-modelled epileptic rats were screened by reverse transcription-quantitative polymerase chain reaction, and it was determined that the expression ATP-sensitive inward rectifier potassium channel 10 (Kcnj10) increased in hippocampus and insular cortex, while the expression of most of the other subtypes decreased. The total level of H3K9me2 decreased in the model group compared with the control. The Kcnj10 gene encoding the Kir4.1 channel was selected to analyse changes in H3K9me2 in the promoter region by the chromatin immuno-precipitation method. Anti-H3K9me2 and anti-G9a antibodies were used to identify the modified DNAs. Five primers were designed across the promoter region of the Kcnj10 gene. In epileptic hippocampi, the relative abundance of H3K9me2 and G9a in the promoter region of Kcnj10 decreased markedly. Removal of the H3K9me2 repressive mark resulted in decreased transcriptional inhibition of the Kcnj10 gene and therefore increased its expression. In the cultured C6 cells, specific inhibition of the enzymatic activity of G9a by 2-(Hexahydro-4-methyl-1H-1,4-diazepin-1-yl)-6,7-di-methoxy-N-(1-(phenyl-methyl)-4-piperidinyl)-4-quinazolinamine tri-hydrochloride hydrate (bix01294) resulted in upregulation of the expression of Kir4.1 proteins. The present study demonstrated that H3K9me2 and G9a are sensitive to epileptic seizure activity during the acute phase of epilepsy and can affect the transcriptional regulation of the Kcnj10 channel.
Epilepsies are disorders of neuronal excitability, characterized by spontaneous and recurrent seizures. Potassium ion channels serve an important role in the pathogenesis of epilepsy. They are the core targets for excitability regulation and are involved in potassium homeostasis. Potassium ion channels adjust the cell membrane resting potential and determine the amplitude and frequency of action potentials, and buffer the extracellular potassium ions. Over 80 subtypes of potassium ion channels have been identified to date on the surface of neuronal and glial cells (
An increasing amount of evidence suggests that epigenetic mechanisms are involved in the regulation of potassium channels in various neural pathogeneses (
One of the major epigenetic modulations is histone H3 at lysine 9 (H3K9) methylation. It includes mono-, di- and tri-methylations, which are catalysed by several histone methyltransferases, such as histone-lysine N-methyltransferase EHMT2 (G9a) and histone-lysine N-methyltransferase SUV39H1. H3K9me2 is the main repressive mechanism of the transcriptional regulation of gene expression, and is commonly associated with gene silencing. G9a mediates H3K9 mono- and di-methylation (H3K9me1/2) and targets euchromatic loci (
All experimental procedures in the present study were approved by the Animal Ethics and Use Committees of Ningxia Medical University (Yinchuan, China).
Healthy male Sprague Dawley rats (weight, 280–300 g; aged 8–12 weeks) obtained from the Animal Centre of Ningxia Medical University were reared at 6 per cage at 24.0±0.2°C in a humidified atmosphere (54±5%). Rats were fed with a standard rodent animal feed, were specific-pathogen-free and kept under a natural circadian rhythm, with natural lighting and regular ventilation. Rats were randomly divided into healthy control and pilocarpine epilepsy groups. Totally 40 rats were used, with 25 in the epilepsy group and 15 in the control group. More specifically, 15 epileptic rats and 5 control rats for western blotting; 5 epileptic rats and 5 control rats for RT-qPCR and 5 epileptic rats and 5 control rats for ChIP assay (
Rat brains were dissected, and hippocampal and insular cortex tissues were isolated and processed immediately for protein or RNA extraction. Each assay included 5 rats. For western blotting, a Total Protein Extraction kit (Nanjing KeyGen Biotech. Co., Ltd., Nanjing, China) was used, and the Bicinchoninic Acid assay was used for protein quantification (Nanjing KeyGen Biotech. Co., Ltd.) according to the manufacturer's protocol. The following antibodies were used: Anti-EHMT2/G9A (dilution, 1:1,000; cat. no. ab40542), anti-Kcnj10 (dilution, 1:1,000; cat. no. ab192404), anti-histone H3 (di methyl K9; dilution, 1:1,000, cat. no. ab1220; all from Abcam, Cambridge, MA, USA) and anti-GAPDH (dilution, 1:5,000; cat. no. bsm-0798m; BIOSS, Beijing, China). Peroxidase-conjugated goat anti-rabbit IgG (cat. no. bs-0295G) and goat anti-mouse IgG (cat. no. bs-0296G) were used as the secondary antibodies (both dilutions, 1:10,000; BIOSS). All antibodies were diluted with 5% skimmed milk (BD Biosciences, Franklin Lakes, NJ, CA, USA).
Total RNA was purified from rat brain tissues with TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to manufacturer's protocol. The RNA was reverse-transcribed to cDNA using a Reverse Transcription system (Pomega Corporation, Madison, WI, USA) according to the manufacturer's protocol: 70°C for 5 min, 25°C for 5 min, 42°C for 60 min, and 70°C for 15 min. A total of 16 primer pairs specific to different potassium channel subtypes were designed and tested by the melting curve analysis (
A total of 5 epileptic and 5 control rats were used for ChIP assay. Anti-histone H3 (di methyl K9) antibody (dilution, 1:1,000; cat. no. ab1220; Abcam) and anti-EHMT2/G9A (dilution, 1:1,000; cat. no. ab40542; Abcam) antibodies, and an EZ-ChIP kit (Haoran Biological Technology Co., Ltd., Shanghai, China) were used. DNA fragments were sonicated to 100–500 bp. RT-qPCR was performed as described in the previous section. A total of 5 primer pairs specific to the Kcnj10 promoter region were designed, and the sequences are presented in
The C6 glioma cell line (Shanghai Jing Kang Biological Engineering. Co., Ltd., Shanghai, China;
Statistical analyses were performed using SPSS (version 17.0; SPSS, Inc., Chicago, IL, USA). All data are presented as the mean ± standard error. Differences between groups were analysed using Student's t-test or one-way analysis of variance with Dunnett's post hoc test. P<0.05 was considered to indicate a statistically significant difference.
A total of 16 potassium channel subtypes including voltage-gated, inwardly rectifying, ATP sensitive, M type and Ca2+−stimulated, were screened by RT-qPCR. In the insular cortex, and in hippocampus, 5 h after seizure, Kcnj10 was upregulated (
The expression of Kir4.1 was tested by western blotting with five epileptic and five control rats. Tissues were analysed 1, 5 and 24 h following an epileptic seizure. The protein expression levels of the Kir4.1 were upregulated at 24 h following an epileptic seizure in the insular cortex (
In order to analyse the mechanism of epigenetic regulation of the potassium channel genes, the expression of H3K9me2 and its enzyme G9a was tested by a western blotting at three different time points: 1, 5 and 24 h. As presented in
The aforementioned results suggested an association between the H3K9me2 modification and the transcriptional regulation of Kcnj10 gene. ChIP experiments were designed to analyse changes in levels of H3K9me2 in the promoter region of the Kcnj10 gene. A total of 5 primer pairs specific to the Kcnj10 promoter region were designed to anneal across exon1 (
Relative levels of H3K9me2 at primer annealing sites 1, 2, 3, and 4 decreased significantly 5 h following an epileptic seizure in the hippocampus (
In order to confirm whether the observed changes in methylation were catalysed by G9a, ChIP experiments using the anti-G9a antibody were performed to identify on-going methylation events. The abundance of G9a at primer annealing sites 2, 3 and 4 also decreased significantly 5 h following an epileptic seizure in the hippocampus (
Bix01294, a specific G9a inhibitor, was used to verify Kcnj10 regulation by G9a. Following a 1 and 4 µmol/l treatments with bix01294 for 5 h, the expression of G9a and H3K9me2 decreased in C6 cells, while the expression of Kir4.1 (Kcnj10) demonstrated an increase in a dose-dependent manner between 1 and 4 µmol/l doses (
Seizures can cause long-lasting changes in gene expression patterns. Epigenetic mechanisms may alter gene expression in response to seizure. Without causing mutations in gene sequences, epigenetic modifications, including histone methylation, may respond to changes inside the cell caused by seizure and alter gene transcription through chromatin adjustments. The present study investigated the association between epigenetic modification of the potassium channel expression and epileptic activities.
The present study demonstrated that H3K9 dimethylation and its enzyme G9a were sensitive to the seizure activity. Following a 1 h-long seizure interrupted with LiCl/pilocarpine, H3K9me2 and G9a were downregulated. The downregulation lasted for <24 h, was lowest at 5 h following the seizure and then returned to its original level after 1 day. A likely explanation for these observations is that H3K9me2 and G9a were sensitive to the stress caused by seizure, and responded to modulate the acute phase of epilepsy. Similar mechanisms have been previously reported in a study investigating seizures in cocaine addicts (
An alternative mechanism of G9a and H3K9me2 response to epileptic activity is through RE1 silencing transcription factor (REST). REST can respond to epileptic activity within 5 h, and recruit co-repressors including HDACs, KDMs, Brg1 and G9a to become more abundant in dentate granule neurons (
Potassium channels may be regulated by G9a and H3K9me2. In a neuropathic pain model, persistent pain stimulus can lead to a continuous upregulation of G9a and H3K9me2, and downregulation of some subtypes of potassium channels, such as Kcnq4, Kcnd2, Kcnq2 and Kcnq1 in dorsal root ganglion neurons (
Kir4.1 encoded by Kcnj10, is a potassium channel expressed on the membrane of the astrocytes. Kir4.1 is an inwardly rectifying ATP-activated potassium channel that is mainly expressed in astrocytes, and serves a crucial role in the regulation of potassium ion homeostasis during neuronal activity (
The association between G9a and H3K9me2 downregulation, and the upregulation of the Kir4.1 potassium channel suggest that G9a and H3K9me2 mediate epileptic seizure-induced changes in the Kcnj10 expression. ChIP results obtained in the present study confirmed that G9a and H3K9me2 accumulation in the Kcnj10 promoter region decreased 5 h following epileptic activity. Inhibition of G9a and H3K9me2 with bix01294 in astrocyte tumour cell line promoted the expression of Kcnj10 in a dose dependent manner, which further confirmed the inhibitory regulation of G9a and H3K9me2 on Kcnj10 transcription.
The other upregulated gene was Kcnk10, which encodes for arachidonic acid sensitive TREK-2 K(+) potassium channel. The TREK channels are expressed highly in the human central nervous system, and can be activated by temperature, membrane stretch and internal acidosis (
In conclusion, H3K9me2 and G9a are sensitive to epileptic seizure activity during the acute phase of epilepsy and can affect the transcriptional regulation of the Kcnj10 (Kir4.1) channel. Our findings may provide a potential epigenetic link between the regulation of potassium channel subtypes and epileptic development which may be helpful for understanding epilepsy.
The present study was funded by the National Natural Science Foundation of China (grant nos. 31460300 and 31260246); the Ningxia Medical University Foundation of China (grant nos. XY201406 and XY2016055); the National Basic Research Program of China (973 Program; grant no. 2012CB722408) and the 13.5 Major Scientific and Technological Projects of Ningxia Hui Autonomous Region (grant no. 2016BZ07).
Screening of potassium channel subtypes by reverse transcription-quantitative polymerase chain reaction. Sixteen potassium subtypes were screened in (A and B) two brain regions of rats 5 h following epilepsy. Five rats were used in each group (n=5). Data are expressed as the mean ± standard error. *P<0.05 vs. the control group. Kcnt2, potassium channel subfamily T member 2; Kcna5, potassium voltage-gated channel subfamily A member 5; Kcnq2, potassium voltage-gated channel; Kcnq2, subfamily KQT member 2; Kcns2, subfamily S member 2; Kcnh2, subfamily H member 2; Kcna3, subfamily A member 3; Kcnj10, ATP-sensitive inward rectifier potassium channel 10; Kcnj10, ATP-sensitive inward rectifier potassium channel 10; Kcnj16, inward rectifier potassium channel 16; Kcnj4, inward rectifier potassium channel; Kpnb1, importin subunit beta-1; Kpna3, inward rectifier potassium channel alpha-4; Kcnk10 importin subunit potassium channel subfamily K member 10; Kcnab2, voltage-gated potassium channel subunit beta-2; Kctd2, BTB/POZ domain-containing protein KCTD2; Kcnip1, Kv channel-interacting protein 1; Con, control.
Western blotting of Kir4.1. The expression levels of Kir4.1 protein was significantly increased (A) 24 h following epileptic seizure in the insular cortex and (B) after 5 h in the hippocampus. Five rats were used in each group (n=5). Data are expressed as the mean ± standard error. *P<0.05 vs. the control group. Con, control.
Western blotting of H3K9me2 and G9a. The expression levels of H3K9 and G9a were markedly decreased at 5 h post seizure, both in hippocampus and insular cortex, and then increased at the 24-h time point. Five rats were used in each group (n=5). Data are expressed as the mean ± standard error. *P<0.05 vs. the control group. H3K9me2 dimethylation of histone 3 lysine 9; G9a, euchromatic histone-lysine N-methyltransferase 2; Con, control.
ChIP results demonstrating the relative abundance of H3K9me2 and G9a in the promoter region of the Kcnj10 gene. (A) Visualized localization of reverse transcription-quantitative polymerase chain reaction primers in the promoter region of Kcnj10. (B) The relative occupancy of H3K9me2 and (C) G9a in the promoter region of Kcnj10. Genome DNA was used as an input. Five rats were used in each group (n=5). Data are expressed as the mean ± standard error. *P<0.05 vs. the control group. ChIP, chromatin immune-precipitation; H3K9me2 dimethylation of histone 3 lysine 9; G9a, euchromatic histone-lysine N-methyltransferase 2; Kcnj10, ATP-sensitive inward rectifier potassium channel 10; Con, control.
Western blotting of G9a, H3K9me2 and Kcnj10 following inhibition with Bix01294. C6 cells were stimulated with 1,2,4 or 8 µmol/lBix01294 for 5 h. Data are expressed as the mean ± standard error. *P<0.05 vs. the control group. G9a, euchromatic histone-lysine N-methyltransferase 2; H3K9me2 dimethylation of histone 3 lysine 9; Kcnj10, ATP-sensitive inward rectifier potassium channel 10; Bix01294, 2-(Hexahydro-4-methyl-1H-1,4-diazepin-1-yl)-6,7-di-methoxy-N-(1-(phenyl- methyl)-4-piperidinyl)-4-quinazolinamine tri-hydrochloride hydrate.
Records of the behaviour of epileptic rats.
Experiment | Western blotting | RT-qPCR | ChIP | ||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Rat no. | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 |
Time point of death following epilepsy | 1 h | 5 h | 24 h | 5 h | 5 h | ||||||||||||||||||||
Number of pilocarpine injections | 2 | 2 | 1 | 4 | 4 | 1 | 2 | 2 | 3 | 3 | 1 | 1 | 2 | 3 | 4 | 3 | 4 | 2 | 3 | 2 | 3 | 2 | 2 | 3 | 3 |
Seizure latency (min) | 50 | 59 | 28 | 104 | 118 | 29 | 44 | 60 | 88 | 67 | 20 | 24 | 48 | 65 | 106 | 76 | 102 | 38 | 96 | 37 | 62 | 51 | 54 | 63 | 62 |
RT-qPCR, reverse transcription-quantitative polymerase chain reaction; ChIP, chromatin immune-precipitation.
Primer sequences for RT-qPCR and ChIP.
RT-qPCR primer sequence (5′→3′) | ||
---|---|---|
Gene | Forward | Reverse |
KCNT2 | ACATCCCCAAAGCCCACA | TGGGAGCAGATTTTACGAGTTC |
KCNA5 | GAGAATGACCAGGACCGACAC | TCATCAAGGAAGAGGAGAAGCC |
KCNQ2 | GGATTCCGCCGTTTCTCA | ACCCTCATTGGTGTCTCATTCTT |
KCNJ10 | GGGGACGCCACTTTCACAA | GAGATCCTCTGGGGCTACGA |
KCNS2 | TGCCCAAGTTAGCCAAGGT | AATCGTGGAGCACTTTGGC |
KCNJ16 | GGGTCGGAAGTCACCTATGC | CATTGGGGCAGCCTTGG |
KCNH2 | GCCCCTCGGAATGGTTTG | TTCCATCAGATTCCCAACCC |
KCNJ4 | CGGAGATGACAGCGTGGTG | GCGGTCATCGCAGTGGTT |
KCNA3 | TTACTGGGGCAAGCAAAGAAT | TCTTCTGCTTGGAGACACTACCC |
KCNC3 | CAACGCCCAACTGCTACG | GACTCCCGTTCCCTCTTCG |
KPNB1 | TGGATCATTGGCCTAGCTTCTA | AGCCCAAGTGGACAAGTCAGA |
KPNA3 | TGGGACAGGACATCGCAGTT | AGCCACCAGGAAGTCAAAGTT |
KCNK10 | CCCTGCCACAAAATCACCA | ATCCCGCTCTTCGGTTTCT |
KCNAB2 | GCTCCAGCGTGATGTGCC | GAAGAAGGGGTGGAGACGG |
KCTD2 | CGTTGTCCCGTATCCTTTCC | ACTCGGACAAGGATGAGACTGG |
Kcnip1 | ATGATGTTGTCGTCCTCCTGAC | CATCAACAAGGACGGATACATAAAC |
GAPDH | GGTCCACCACTGACACGTTG | ACAACAGCCTCAAGATCATCAG |
1-KCNJ10 | AAGAAGGAGGGAAGAACA | ACATAGTGGATGGGAAGAT |
2-KCNJ10 | TGAAGGAACCAGGACGAG | ACAGGGAACAACGAAAAC |
3-KCNJ10 | GGTAGTGGGACAGATGAAGA | GACAAACCCTTATCTGATTC |
4-KCNJ10 | TGCCAACCGCCAACTACA | TCCCTCGTGCTTTCATCG |
5-KCNJ10 | CTCTGCCCTCTACACTCAT | GATAATCTCCCATTCCTAAC |
RT-qPCR, real time-quantitative polymerase chain reaction; ChIP, chromatin immune-precipitation.