Introduction

Epilepsy is a group of heterogeneous disorders characterized by paroxysms resulting from bioelectric hyperexcitation of neuronal networks of the brain. Recently, it has been well recognized that dysfunctions of ion channels expressed in the brain contribute to such hyperexcitation and hence closely associate with the pathogenesis of epilepsy. Accordingly, to date, a number of mutations of the genes encoding ion channels have been identified in various types of epilepsy.

Gamma-aminobutyric acid receptor type A (GABAA receptor) is one such ion channel where mutations have been identified in several epilepsy phenotypes. GABAA receptor, which is a ligand-gated chloride ion channel, serves as a major component of the neuronal inhibitory system in adult brain. It is considered that majority of the receptors in the brain function as a pentamer consisting of α1, β2 and γ2 subunits.

Mutations of the genes encoding the subunits of GABAA receptors were so far found in the genes encoding α1 (GABRA1), δ (GABRD) and γ2 subunits (GABRG2).1, 2, 3, 4 A mutation of GABRA1 was found in autosomal dominant juvenile myoclonic epilepsy, a rare inherited idiopathic epilepsy phenotype.2 Several variants of GABRD were not reported as causes of epilepsy but suggested to associate with susceptibility to genetic epilepsy with febrile seizures plus (GEFS+).3 In contrast, mutations in GABRG2 have been reported as causes of a wide spectrum of epilepsies, from Dravet syndrome to milder conditions such as childhood absence epilepsy, GEFS+ and FS+.1, 4, 5, 6, 7, 8

Although major progress has been made in tying GABRG2 to some categories of epilepsy (that is, GEFS+, childhood absence epilepsy and Dravet syndrome), its role in most kinds of childhood epilepsy such as benign epilepsy of childhood with centrotemporal spikes, generalized tonic-clonic seizures (GTCS), partial epilepsy and so on remains poorly understood. Accordingly, this study further investigates the role of GABRG2 in the pathogenesis of childhood epilepsy.

Materials and methods

Patients

Our study included 140 pediatric patients who had been diagnosed with epilepsy at various departments of neurology in regional tertiary pediatric hospitals (Table 1). The epilepsy phenotypes included Dravet syndrome, GEFS+, GTCS, partial epilepsy, juvenile myoclonic epilepsy, benign familial neonatal seizures, childhood absence epilepsy, benign epilepsy of childhood with centrotemporal spikes, epilepsy with continuous spikes and waves during slow sleep and progressive myoclonus epilepsy (PME). We also recruited 48 healthy Japanese volunteers as the ethnic matched control group. Each patient or parent/guardian signed an informed consent form approved by the Ethics Review Committee of Fukuoka University or similar committees of the participating institutions.

Table 1 Clinical and genetic characteristics of the patients
Full size table

Genetic analysis

Using QIAamp DNA Blood kit (Qiagen, Hilden, Germany), genomic DNAs were prepared from ethylenediaminetetraacetic acid-treated whole blood samples. Genetic abnormalities were sought within all 10 exons of GABRG2 and their flanking intronic splice sites by a direct sequencing method with an automatic sequencer as described earlier.9, 10 Details of the PCR conditions and the primers used are available on request. Reference sequences of mRNA were based on information available from RefSeq, accession numbers: Human GABRG2, NM_198904.

Statistical analysis

To determine whether the polymorphisms were involved in the pathogenesis of epilepsy, we determined the genotype and allele frequency of the polymorphisms in our patients and in the control population. Data analysis was performed by Fisher’s exact test using an SPSS software package (version 13.0). All P-values were two-tailed. The significance level was considered to be 5%.

Results

Mutational analysis for the 140 patients revealed one novel missense mutation (c.236A>G: p.N40S, Figure 1) in one of six patients with GTCS. The mutation was heterozygous and not found in other patients of the cohort or in the 48 ethnic matched control individuals. The DNA of the parents of the patient was not available. The patient was a girl whose first seizure occurred when she was 15-years old, and after that re-occurred once a month. She exhibited the typical seizure and electroencephalographic phenotype and susceptibility to anti-epileptic drugs of GTCS supporting that the diagnosis of GTCS. She had no mental retardation or other neurological disorder. Family history of febrile seizures or epilepsy was not found in parents of the patient. The Asn residue at the position of 40, which was replaced with Ser by the mutation, is a highly conserved amino acid throughout many species (Figure 1).

Figure 1
figure 1

GABRG2 mutation and alignment of affected amino acid. Arrow indicates where mutation occurs. Rectangular box represents the corresponding amino acid to the amino acid where the mutation occurs, that is highly conserved throughout different species.

Full size image

Three exonic variants were found on exon 3 (rs11135176, n=74), exon 5 (rs211037, n=105) and exon 10 (c.1254 C/T: p.D379D, n=3), and one intronic variant (IVS1+18delT, n=1) was found (Figure 2; Table 1). Variants of rs11135176 and rs211037 were also found in the control population. None of the variants led to any amino-acid changes in the protein sequence. Fisher’s exact test showed no significant differences (P>0.05) in the genotype and allele frequency between the two groups (Table 2).

Figure 2
figure 2

Four GABRG2 variants are identified. Arrows indicate where variants occur. (a) Variant in intron 1. Among these variants there are two known SNPs (b, c) (rs211037 and rs11135176) and one new SNP (d) (c.1254 C>T).

Full size image
Table 2 Distribution of polymorphisms in the GABRG2 gene in patients and controls
Full size table

Discussion

This study reports one novel missense mutation of GABRG2 (c.236 A>G: p.N40S) in a female with the typical GTCS phenotype. The mutation was heterozygous and not found in 48 ethnic matched control samples. No other relevant genetic variations were found in the 140 patients. To our knowledge, this is the first report of GABRG2 mutation in GTCS providing compelling evidence of the involvement of GABRG2 in epilepsy.

GABAA receptors are the major inhibitory neurotransmitter receptors in the central nervous system, and several anti-epileptic drugs including benzodiazepines, barbiturates and neurosteroids act by enhancing GABAA receptor currents. These inhibitory receptors are pentamers formed by assembly of multiple subunit subtypes and the α1β2γ2 receptor is the most abundant receptor isoform. The γ2 subunit contributes to receptors involved in both phasic and tonic inhibition and underlies the benzodiazepine sensitivity of both modes of inhibition.11 It is also critical for receptor trafficking, clustering and synaptic maintenance.12, 13 Up to date, at least seven mutations of GABRG2 including missense mutations, nonsense mutations and splice-site mutation have been associated with a broad spectrum of epilepsies.1, 4, 5, 6, 7, 8, 14

The N40S mutation identified in this study affects a highly conserved Asn at residue 40 of the mature γ2 subunit, thus, the mutation is adjacent to the first one of the two high-affinity benzodiazepine-binding domains of the γ2 subunit (Lys-41-Trp-82 in the mature γ2 subunit). Wallace et al.15 had suggested that R43Q mutation in the benzodiazepine-binding site can attenuate benzodiazepine sensitivity of GABAA receptor. Another study on this mutation was shown to accelerate deactivation of the receptor.16 These lines of evidence support that the N40S mutation should attenuate the GABAA receptor functions whereby increasing the intracortical excitability of the brain. This notion accords with the recent findings with the GABRG2 mutations in the N-terminal domain.17

In addition, Kang and Macdonald18 suggest that with heterozygous expression, the R43Q mutation may result in impaired receptor trafficking and increased retention of the receptor in intracellular compartments, including the ER. This reduced cell surface expression would result in decreased inhibitory GABAA receptor current in vivo, and consequently, an increase in neuronal excitability and epilepsy. Several subsequent studies have identified the retention of mutant receptors in the ER14 and shown that GABRG2 mutations have reduced trafficking either to the membrane surface with relatively normal function18, 19 or to the surface with impaired function.15, 20 It is believed that the main electrophysiological deficit of GABAA receptor resulting from the mutation is due to intracellular trafficking abnormality of channel molecules. Thus, the N40S mutation may result in an aberrant trafficking of the GABAA receptor. We anticipate that this mutation N40S may contribute to the patient’s pathogenesis of GTCS.

Recently, Wang et al.21 reported that the GABRG2 polymorphism rs211014A allele was higher in the febrile seizures group (P<0.005), suggesting that GABRG2 polymorphisms may predict susceptibility of febrile seizures. Another study22 found that children with the GABRG2 (SNP:rs211037)-C allele had a higher incidence of idiopathic generalized epilepsies, indicating that the GABRG2 (SNP:rs211037)-C allele is a candidate genetic marker for idiopathic generalized epilepsies. In this study, three exonic variants and one intronic variant are identified. Among these variants, c.1254C/T is a novel polymorphism. None of these variants leads to any amino-acid changes in the protein sequence. However, in this study, there are no significant differences for these polymorphisms in genotype and allele frequency between patients and control population, suggesting that they are not involved in the etiology of Japanese childhood epilepsy.

In summary, one novel missense mutation of GABRG2 (N40S) was identified in a patient with GTCS. To our knowledge, there have been no previous reports of mutation of GABRG2 in GTCS. This finding indicates that mutation GABRG2 can underlie the pathogenesis of GTCS and thus reinforces the involvement of GABAA receptors in epilepsy. However, the contribution of mutations of GABRG2 per se to the pathogenesis of epilepsy remains unclear because we do not have sufficient patients. Further studies will involve gathering more cases especially GTCS, progressive myoclonus epilepsy patients and performing the GABRG2 analysis.