Elsevier

Epilepsy Research

Volume 70, Issues 2–3, August 2006, Pages 229-238
Epilepsy Research

Examining the role of common genetic variation in the γ2 subunit of the GABAA receptor in epilepsy using tagging SNPs

https://doi.org/10.1016/j.eplepsyres.2006.05.009Get rights and content

Abstract

Introduction

Mutations in the γ2 subunit gene of the GABAA receptor, GABRG2, have been shown to cause generalised epilepsy syndromes in rare familial cases. Here we set out to examine whether common variation in GABRG2 predisposes to the development of common, complex forms of epilepsy in two large independent cohorts.

Methods

We have applied a tagging single nucleotide polymorphism (tSNP) technique allowing us to satisfactorily represent common variation in the gene. However, to ensure maximal representation of functional variation and in particular in cases of low minor allele frequency (MAF), we have identified and forced known functional variation as tagging SNPs. We examined the association between tagging SNPs and subtypes of epilepsy in two independent cohorts; the first consisted of 677 cases and 384 healthy controls, the second of 684 cases and 277 healthy controls.

Results

We failed to detect any variation that conferred an increased risk of disease development in both cohorts.

Discussion

Our results suggest that common variants of strong effect in GABRG2 do not appear to play a role in the development of common, complex forms of epilepsy. This report illustrates a number of important features of study design in genetic association studies, including the simultaneous use of map and sequence-based techniques and the necessity of replication before robust conclusions can be drawn from results.

Introduction

γ-Aminobutyric acid (GABA) is the principal inhibitory neurotransmitter in the brain and, as such, alterations in GABAergic function may underlie the increased neuronal excitability associated with epilepsy. GABA mediates its effects in the central nervous system via its action on specific GABA receptors. GABAA receptors mediate fast synaptic inhibition and are heteropentameric structures composed of subunits that belong to eight different classes (α, β, γ, δ, ɛ, θ, π, ρ) (Sieghart and Sperk, 2002). Each subunit may have several subtypes and the α1, β2 and γ2 subunits are the most abundantly expressed in the brain (Sieghart and Sperk, 2002). The genes for these subunits are primarily located in four clusters scattered over the genome, thought to represent an evolutionary replication of a single original cluster (Russek, 1999).

The role of the GABA neurotransmitter in epilepsy has been confirmed by the fact that mutations in GABA receptor genes cause rare familial epilepsy syndromes. Mutations in the gene for the γ2 subunit, GABRG2, have been shown to cause generalised epilepsy with febrile seizures plus (Baulac et al., 1999; Harkin et al., 2002) and childhood absence epilepsy with febrile seizures in large pedigrees with epilepsy (Wallace et al., 2001, Kananura et al., 2002). Although these kindreds account for only a very small proportion of individuals with epilepsy, it is reasonable to speculate that variation in this gene may predispose to more common, complex forms of epilepsy. Indeed, several studies have attempted to examine this hypothesized role. A synonymous variant (rs211037) in GABRG2 was reported to associate with febrile seizures in the Taiwanese population (Chou et al., 2003). However, a follow-up study failed to replicate this association in a population of European ancestry (Cavalleri et al., 2005). Other reports have failed to show association of particular variants in GABRG2 with either febrile seizures (Nakayama et al., 2003), childhood absence epilepsy (Kananura et al., 2002, Lu et al., 2002) or severe myoclonic epilepsy of infancy (Madia et al., 2003). There have been no studies to date on the role of this gene in localisation-related epilepsies.

Several features of previous studies examining GABRG2 illustrate a number of the problems in the methodology with which association-based studies are being applied to epilepsy. Firstly, each study, by focusing on a subset of putatively functional variants, employed a ‘sequence-based’ approach. Such an approach, if negative, does not rule out a contribution from other variation across the gene or region in question. We, and others, have recently called for the merging of sequence-based with map-based techniques in an attempt to examine all common variation across a candidate gene or region (Cavalleri et al., 2005, Tan et al., 2004). Secondly, analyses have tended to be performed in single cohorts of relatively small size. Small cohorts lack the power to detect functional variants with anything but very strong effects. Variation of marginal to medium effect may therefore be missed. Finally, none of the previous studies attempted to replicate detected associations prior to publication. This is of particular importance when a priori evidence is scarce as is the situation when the functionality of the associating variant is unknown. By replicating positive findings prior to publication more robust conclusions can be made and the incidence of false positives in the literature would decrease.

Here we set out to examine, using a comprehensive study design involving tagging SNPs, whether common variation in GABRG2 predisposes to the development of common, complex forms of epilepsy.

Section snippets

Methods

Ethical permission was obtained from the relevant institutional bodies, and all patients provided written informed consent.

Results

We found six tSNPs to be sufficient to satisfy our tagging criteria. The six tSNPs selected were tSNP1: rs209348, tSNP2: rs209353, tSNP3: rs211037, tSNP4: rs211029, tSNP5: rs766349 and tSNP6: rs989694. The average haplotype r2 using these tSNPs was 0.94 with a single minimum value of 0.63. Average SNP drop values were 0.88 with a single minimum of 0.63. These six tSNPs therefore represent the vast majority of common variation across the GABRG2 gene with minimal loss of power.

We tested the

Discussion

This study represents the first combined use of HapMap data with haplotype-based tagging technology in a study seeking to map genetic variation predisposing to the development of epilepsy. Using this method, association can be detected not only at the level of the individual SNP, but also at the haplotype level. Assessment of haplotypes is important as it allows detection of variants interacting in a cis-like manner, i.e. combinations of alleles on the same chromosome acting together to

Acknowledgements

We would like to thank all the individuals who donated DNA and participated in this study. Special thanks to Prof. E. O’Brien and Dr. A Stanton for use of control DNA. Peter Kinirons and C.P. Doherty are supported by the Higher Education Authority of Ireland (PRTLI funded Program for Human Genomics). Gianpiero L. Cavalleri is supported by the Annals of Human Genetics Studentship in Human Population Genetics. David B. Goldstein is a Royal Society/Wolfson research Merit Award holder. This study

References (18)

There are more references available in the full text version of this article.

Cited by (17)

  • GABRG2 rs211037 polymorphism and epilepsy: A systematic review and meta-Analysis

    2013, Seizure
    Citation Excerpt :

    Studies have suggested that mutations such as R43Q, Q40X, K289M, and IVS6+2T in this gene are involved in childhood absence epilepsy (CAE), febrile seizures (FS), generalized epilepsy with febrile seizures plus (GEFS+), and Dravet syndrome.6,7 Previous studies examined whether the 588C>T Asn196Asn exon 5 polymorphism (rs211037) is related to susceptibility to FS or idiopathic generalized epilepsy (IGE) in different populations, however the results were inconsistent (Table 1 and Fig. 1).10–18 To shed light on the association between rs211037 and susceptibility to epilepsy, we carried out a systematic review and meta-analysis.

View all citing articles on Scopus
1

These authors contributed equally.

View full text